publications of the INSTITUTE of MARINE SCIENCE Volume 5 DECEMBER 1958 Published by Institute of Marine Science The University of Texas Port Aransas, Texas INSTITUTE OF MARINE SCIENCE THE UNIVERSITY OF TEXAS PORT ARANSAS, TEXAS The Institute of Marine Science with laboratories at Port Aransas, Texas is a research division of The University of Texas related to the Departments of Bacteriology, Botany, Chemistry, Geology, Physics, and Zoology and the Meteorology Division of the Department of Aeronautical Engineering at the Main University at Austin. An integrated teaching program in graduate marine science is made up of the course offerings from the several departments and from the Institute. The permanent staff in residence at Port Aransas is primarily concerned with basic research with a general emphasis on functional processes in marine environments. During the summer courses are given at the Institute for graduate students with serious interest in the sea. Staff for 1958 Director: Howard T. Odum. Assistant Director: Louis S. Kornicker. Budget Council: W. F. Blair, Chairman; S. P. Ellison, Jr.; H. C. Bold; and H. T. Odum. ECOLOGY PROGRAM Howard T. Odum, Ph.D., Research Scientist V, Lecturer in Zoology. William McConnell, Ph.D., Research Scientist III. Thomas B. Hellier, Jr., M.S., Research Scientist II. tWalter Abbott, M.S., National Science Foundation Fellow. tRobert Beyers, A.B., Port Aransas Rod and Reel Scholarship. tDick Hunter, A.B., Rachel and Ben Vaughn Scholarship. Emilio Guerra, Laboratory Research Assistant I. *Neal Armstrong, Laboratory Research Assistant I. *Jacque Pirson, Laboratory Research Assistant I. MARINE MICROBIOLOGY PROGRAM Carl H. Oppenheimer, Ph.D., Research Scientist V, Lecturer in Bacteriology. tCarol Volkmann, A.B., Research Scientist I. *Mills Tandy, Laboratory Research Assistant I. Roscoe Lamplugh, Technical Staff Assistant. MARINE GEOLOGY PROGRAM Louis S. Kornicker, Ph.D., Research Scientist IV, Lecturer in Geology. tCharles M. Hoskin, M.A., Research Scientist II. Charles D. Wise, M.S., Research Scientist I. Mrs. Winnie Rice, Curator of Mollusk Collection. tDarrel K. Jones, B.S., Research Scientist I. *Clyde H. Moore, B.S., Research Scientist I. MARINE BOTANY PROGRAM John T. Conover, Ph.D., Research Scientist III, Lecturer in Botany. *Lazern 0. Sorensen, Ph.D., Research Scientist III, Pan American College. *Susan Hamilton, Laboratory Research Assistant I. ICHTHYOLOGY AND PHYSIOLOGY PROGRAM William N. McFarland, Ph.D., Research Scientist III, Lecturer in Zoology. VISITING PROGRAMS 1. Clay Minerals: *Edward Jonas, Ph.D., Department of Geology, The University of Texas. *F. Pearson, A.B., Research Scientist I. *C. Walker, A.B., Research Scientist I. 2. Fish Hybridization: *C. Hubbs, Ph.D., Department of Zoology, The University of Texas. *G. Drewry, A.B., Research Scientist I. 3. Invertebrates: *Bassett Maguire, Ph.D., Department of Zoology, The University of Texas. 4. Shrimp Larvae: tCharles Goodwin, M.S., Texas Game and Fish Commission and A. & M. College. 5. Seiche Effects: W. Armstrong Price, Ph.D., Corpus Christi, Texas. SUPPORTING OPERATIONS Herman Moore, Motorboatman, Launch "Ciencia". Sam Gampert, Building Attendant. *Mrs. Katherine Hunter, A.B., Food Service Supervisor III. Dining Room. OFFICE AND LIBRARY Mrs. Lyla Olive, Senior Secretary. Mrs. Harolene Hadden, Secretary. *Mrs. Anna K. McFarland, Senior Secretary. *Mrs. Gretchen M. Jones, B.S., Clerk Typist. *Summer staff only. t Candidate for graduate degree, part time in residence. publications of the INSTITUTE of MARINE SCIENCE Volume 5 DECEMBER 1958 Published by Institute of Marine Science The University of Texas Port Aransas, Texas Table of Contents Effect of the Microbial Production of Hydrogen Sulfide and Carbon Dioxide on the pH of Recent Sediments, by Carl Oppenheimer and Louis S. Kornicker.. ..Page 5 Comparative Studies on the Metabolism of Marine Waters, by Howard T. Odum and Charles M. Hoskin ----------------····--------···----·-----------------·-·····----------------·-··-··-···-·· 16 Diurnal pH Variation in Texas Bays, and its Application to Primary Production Estimation, by Kilho Park, Donald W. Hood, and Howard T. Odum ___ __________ ___ 47 The Chlorophyll "A" of Communities, by Howard T. Odum, William McConnell, and Walter Abbott --------····---····------------·····-------------·······--···----·-·-----------·--·--·······-··· 65 Seasonal Growth of Benthic Marine Plants as Related to Environmental Factors in an Estuary, by John T. Conover --·-······---------------·-··-···-··-----------------------··-------------97 Artificially Formed Mud Balls, by Louis S. Kornicker, Carl H. Oppenheimer, and John T. Conover ----------------------··-----------------------------------------------------------------·····-·----148 Notes on Blanquilla Reef, the Most Northerly Coral Formation in the Western Gulf of Mexico, by Donald R. Moore-------------------------····-·--··--·-···-----··-···----------·------------151 Evidence for Fossil Bacteria in Phosphate Rocks, by Carl H. Oppenheimer -----------··-156 A Bacterium Causing Tail Rot in the Norwegian Codfish, by Carl H. Oppenheimer_ 160 An Incidence of Pink Oysters in Galveston Bay, Texas, by Richard Y. Morita ______ __ _ 163 The Drop-Net Quadrat, a New Population Sampling Device, by Thomas R. Hellier, Jr. ------------------·-····------------------·····-·········-·--·-----: ____ _ ____ ____ __________ _____________ 165 Fluctuations in the Relative Abundance of Sport Fishes as Indicated by the Catch at Port Aransas, Texas 1952-1956, by Victor G. Springer and Jacques Pirson______ 169 Population Studies of the Shallow Water Fishes of an Outer Beach in South Texas, by Gordon Gunter-···········--------------------·························-····--····-----·------····--------------186 Ecology and Taxonomy of Recent Marine Ostracodes in the Bimini Area, Great Bahama Bank, by Louis S. Kornicker ·------------------------------···············--------······-··-194 Trematode ·Parasites of Donax variabilis at Mustang Island, Texas, by Sewell H. Hopkins ----------------------····--·····-----------------------------------------------------------------------------··-· 301 A Partially Annotated Checklist of the Marine Fishes of Texas, by Hinton D. Hoese 312 Food Habits of Fishes and Larger Invertebrates of Lake Pontchartrain, Louisiana, an Estuarine Community, by Rezneat M. Darnell ------------------------------------------------353 Systematics and Zoogeography of the Clinid Fishes of the Subtribe Labrisomini Hubbs, by Victor G. Springer ------------------------------------------------------------------------------417 Effect of the Microbial Production of Hydrogen Sulfide and Carbon Dioxide on the pH of Recent Sediments CARL H. OPPENHEIMER AND Loms S. KoRNicKER Institute of Marine Science The University of Texas Port Aransas, Texas Introduction Bacteria in recent marine sediments are capable of changing the pH of the environ­ ment during their metabolism by the production and decomposition of organic and inorganic acids and bases. The purpose of this paper is to present data which indicate that the pH of recent sediments may be directly related to the presence of bacterially produced hydrogen sulfide and carbon dioxide gases, which dissociate to their respective weak acids. Whereas the acid nature of carbon dioxide and hydrogen sulfide is well known, the behavior of the gases in the complex mixture of natural sediments can not be readily predicted without experimental evidence. Hydrogen sulfide and carbon dioxide are commonly produced by bacterial activity in recent marine sediments. These two gases dissolve and dissociate in the interstitial water. The production of hydrogen sulfide by the bacterial reduction of sulfates is well known (ZoBell, 1946a; Senez, 1951; and Postgate, 1951). The production of carbon dioxide and carbonates in marine environments has been described by Harvey (1955) and Revelle and Fairbridge (1957). Laboratory and field experiments were conducted on sediments from the shallow marine bays near the Institute of Marine Science. Experiments are described which show that the saturation of sediments with hydrogen sulfide and carbon dioxide gas results in a lowering of the pH of the sediment. Exposure of the gas saturated sediment to air or inert nitrogen gas is shown to result in an increase in the pH of the sediment. The pH of naturally occurring anaerobic sediments was found to increase after exposure to air, in a manner similar to the results of laboratory experiments. This suggests that low pH values, from 5.4 to 7.6, measured in situ in anaerobic sediments can be caused by bacterially produced hydrogen sulfide and carbon dioxide. Bacterial activity might be a major factor in the early diagenesis of sediment. If bacteria can lower pH and Eh, they can affect both physical and chemical properties of minerals. When bacterial activity ceases, the sediments slowly oxidize. The accom­ panying increase in pH and Eh will once more affect mineral properties. The ecological significance of bacterial activity has been pointed out by Mortimer (1941) who has reported changes in the state of certain inorganic compounds during · the fluctuation of sediments from the aerobic to the anaerobic state. La Lou (1957) has shown that the precipitation of carbonates may depend on changes in pH caused by microorganisms, where carbonate is precipitated at the anaerobic-aerobic interface or the pH transition area. The changes in pH in sediments may interfere with nutrient absorption by living organisms by changing the permeability of cellular membranes (Booij and Bungenberg de Jong, 1956, p. 140). Debyser (1952 and 1955) has shown that the in situ pH of reduced sediments may be as low as 6.5, and that the decrease in pH is accompanied by an increase in sulfide. Shepard and Moore ( 1955) have studied the pH, Eh, and organic contents of Texas Coast sediments and show that the increase in org~nic matter is related to an increase in pH and a decrease in Eh. They also show that the sediments near a river environment have a lower pH than those of bay environments. Emery and Rittenberg ( 1952) have related pH to the regeneration of nutrients in sediments of marine basins near the coast of Southern California. They report that many of the cores increased in pH with depth of sediment and that the over-all range in the water and sediments was 7.3 to 8.5. Emery and Rittenberg made pH measurements on sediment samples which were diluted with co~ free, distilled water and found that this treatment did not affect the pH. ZoBell ( 1946b, p. 501) also diluted sediment samples with boiled, distilled water but found that the pH was usually higher for the water plus sediment than for the sediment alone. If hydrogen sulfide or carbon dioxide were diluted or lost during ZoBell's manipulations, one might expect that the pH would rise. It would be interesting to determine the difference between the sediment studied by ZoBell and those by Emery and Rittenberg. Perhaps the difference is due to the differences in hydrogen sulfide and carbon dioxide content. A recent papc>r by Carroll ( 1958) gives data on the changes of pH and Eh which occur in simulated black anaerobic marine mud made with clays. Carroll reports that the pH of the mud rises during incubation, accompanied by a decrease in Eh. The author states that the pH and Eh readings were made each day by removing a small sample. This technique may be reflected by the rather high pH and Eh readings reported, although no mention was made of the method of measuring either Eh or pH. Such an omission of techniques makes it difficult to interpret Carroll's data to compare with the data in our paper. METHODS All measurements of pH and Eh were made with a Beckman Model G, pH meter fitted with standard heavy duty external electrode3. In situ pH measurements in sedi· ments were made by pushing the electrodes directly into the sediment. The electrodes were not pushed beyond eight centimeters because of the possibility of damage. The temperatures of the sediment samples did not vary more than 2° C during the pH meas­urements. As there is no simple relationship between Eh and pH of sediments the Eh and pH are both specified and no corrections made. Experiments fISH DECOMPOSITION EXPERIMENT Sediment was obtained from experiments which were being conducted to study the processes occurring during the decomposition of fishes buried in sediments. This sedi­ment, consisting of fine sand with shell fragments, was collected from the Gulf of Mexico beach. The pH of the sediment was 7.6 before the fish were added. The fish were buried approximately 12 cm. below the surface. The sediment surrounding the buried fish grad­ ually turned b!ack probably because of ferrous sulfide. The pH decreased to 6.3 in five days, from the original value of 7.6. The changes were presumably caused by the activi­ties of the bacteria as they attacked the organic matter of the fish. During this time the Eh of the sediment dropped from the aerobic state, to -300 millivolts (mv.). Hydrogen sulfide gas slowly bubbled up from the area surrounding the fish. A core through the aerobic and anaerobic sediment was collected in a glass tube with an internal diameter of about 4 cm and the sediment was extruded onto a glass plate. The pH of the aerobic and anaerobic part of the core was measured immediately and then at intervals during a three hour period. As shown in Table 1 the pH of the anaerobic section of the TABLE 1 The 'change of pH with Time During Exposure of Sediments to the Air. Aerobic and Anaerobic Sections of a 20 cm. Core Taken from Sediment Adjacent to a Decomposing Fish pH of Core Time in min. Aerobic Area from the Surface Anaerobic Area from 12 cm. Depth 0 .......................................................... 5 ............................................................ 15 ........................................................... . 45 .............. ............................................. . 75 ........................................................... 180 ........................................................... . 7.7 7.7 7.7 7.7 7.7 6.7 7.2 7.4 8.5 9.2 9.2 core increased from 6.7 to 9.2 during the first 72 minutes. One hundred and five minutes later the pH was still 9.2. The pH of the aerobic sediment remained at 7.7 throughout the three hour period. The pH evidently increased 0.4 units, from 6.3 to 6.7, during the time required for coring, extrusion of the core, and measuring of the initial pH of the core. SATURATION OF SEDIMENT WITH HYDROGEN SULFIDE AND CARBON DIOXIDE The -anaerobic section of the core was dried at room temperature for 48 hours. Distilled water -was added to the dry sediment to provide the approximate original water content. The pH of the rehydrated sediment was 8.5. The sediment was separated into two aliquots. Hydrogen sulfide was bubbled through one aliquot and carbon dioxide was bubbled through the other. The gas saturation was carried out in a stoppered bottle with two glass tubes inserted into the stopper. The tube through which the gas was introduced penetrated the sediment in the bottom of the bottle. The other tube did not come into contact with the sediment and served as an outlet for excess gas. The stopper was then removed, a jet of nitrogen gas was blown into the bottle, and pH electrodes were inserted into the sediment. The nitrogen gas flushed air, which could oxidize the reduced sediment, from the sediment and also helped flush hydrogen sulfide and carbon dioxide out of the sediment. The sediment was stirred every 5 minutes and the pH measured at short intervals over a three hour period. The data are presented in Figure L The pH of the rehydrated sediment decreased from 8.5 to 6.8 when saturated with H2S, and from 8.5 to 6.1 when saturated with C02• The pH values then increased at a con­stant rate during nitrogen flushing. A straight line was obtained when the pH values were plotted against the square root of time (Fig. I) . FIELD EXPERIMENTS ·During laboratory experiments described above, the pH of anaerobic sediment rapidly increased when the sediments were exposed to air. The next step was to determine if an 8 pH /TIME in MINUTES FIG. 1. Change of pH with square root of time for hydrogen sulfide and carbon dioxide gas satu­rated sediment after exposure to an atmosphere of nitrogen. TABLE 2 In situ Values of pH and Eh in Millivolts (mv.l in Marine Sediments near Corpus Christi, Texas and Sapelo Island, Georgia Depth of Measurements in cm. Station Surface 4 pH 9.0 7.9 6.6 6.6 6.6 6.6 Aransas Bay near Corpus Christi Eh +144 + 44 +44 -276 pH 9.2 8.6 7.6 6.8 6.7 6.7 Aransas Bay near Corpus Christi Eh +4 -156 -300 pH 8.6 6.6 6.5 Aransas Bay near Corpus Christi Eh -136 -256 pH 8.4 8.0 6.9 6.5 6.5 Aransas Bay near Corpus Christi Eh +74 -36 -36 -146 -236 pH 8.0 7.1 6.8 6.7 6.7 Corpus Christi Bay Eh -186 -286 -306 pH 8.4 7.1 7.0 6.7 6.8 6.7 Mustang Island Eh -226 -256 -156 -56 -91 Sapelo Island Stat:on 1 pH* 8.2 6.9 6.3 5.9 Station 2 pH* 8.4 6.8 5.4 5.4 * (Eh was n ot measured in Sapelo Island sedimenls. Abundant H:!S wa~ present and ii is assumed that lhe core below the sur£ace was anaerobic.) increase in pH occurred when samples of anaerobic sediment collected in natural en­vironments are exposed to air. Several locations containing anaerobic sediments were selected from the shallow marine bays near the lnstitut.e of Marine Science. Measure­ments of pH and Eh were made directly in the sediment in order to obtain in situ pH and Eh measurements. A core was collected in a 5 cm plastic tube in the area adjacent to where the electrodes were inserted. Changes in pH were measured when the core was exposed to the air. Sediments from Corpus Christi Bay, Laguna Madre, Aransas Bay, Mustang Island, Texas and Sapelo Island, Georgia, were examined. The in situ pH and Eh of the sedi­ments are given in Table 2. The pH values appear to vary directly with the Eh values. Whereas the Eh might be expected to decrease with rise in pH (ZoBell, 1946b), a reverse relationship was found. In all anaerobic sediment tested the pH increased when the sedi­ments were exposed to the air. A typical series of measurements made in the field are presented in Figure 2 and Table 3 which show the change in pH with time. The data in l .!5 I .--x 2cm. pH 7.!5 ..!TiME in MINUTES FIG. 2. Change of pH with the square root of time for a core removed from anaerobic sediment on Mustang Island. Measurements were made at the indicated depth measured from the top of the core. Figure 2 are plotted against the square root of time. The amount of pH increase varied from sediment to sediment and suggests that the net increase might be proportional t<> the amount of hydrogen sulfide and carbon dioxide present in the sediment at the time of collection. Anaerobic marsh sediments from Sapelo Island, Georgia, with more acid pH, showed a similar increase in pH when sections of the sediment were exposed to air. It was immediately apparent that the rate at which the pH increased was influenced by the manner in which the core was treated. If the core was broken open and dis­rupted during the pH measurements, the pH changed more rapidly. If the holes in the TABLE 3 The Changes in pH Which Occurred in Disturbed and Undisturbed Cores of Sediments from Laguna Madre After Exposure to the Air pH in an Measuremenls pH in a Disturbed Core Undisturbed Core Core Deplh in situ Eh (millfrolts) pH Minute5 after collection: 15 25 35 60 Minutes after collection: 20 40 Description of Sedimenls Surface +244 8.2 8.2 8.2 Coarse sand 0.5 + 84 7.2 7.2 7.2 Sandy shell 2 -264 7.0 7.B B.0 8.5 8.2 7.6 8.1 Sandy shell 6 -246 7.0 7.7 7.5 7.9 8.1 7.9 7.2 7.3 Clay silt with decaying algae 8 - 266 i .2 7.8 8.1 8.7 8.7 8.4 7.1 7.5 Fine sand sediment where the electrodes were inserted were covered up after the electrodes were removed, the rate of change of pH was reduced. This is illustrated in Table 3 which gives field data on the pH of cores which were disturbed by the intermittent insertion of the electrodes and adjacent cores which were extruded but not otherwise disturbed before the pH was measured. The pH in the undisturbed cores changed more slowly than the pH of the disturbed core. This is considered to be the result of a more rapid release of gases from the disturbed core. During the in situ field experiments, there was some indication that a small but rapid change in pH took place immediately when cores were removed from the sediment. Measurements of pH were made in situ and on cores approximately one minute after they were removed from the adjacent sediment. The pH increased from 0.1 to 0.4 units above the in situ values (Table 4). Measurements were also made on sealed stored cores TABLE 4 Changes in pH in Sediments during Collection pH Measurement One Minute 2-4 hrs. Sample Depth in situ after Collection after Collection Unstoppered Samples: 1 3 6.6 6.8 7 6.6 6.8 2 6 6.5 6.6 8 6.5 6.5 3 6 6.3 6.7 Cores Collected in Plastic Tubes, Stoppered, Extruded in Laboratory: 1 6 6~ 7.3 8 6~ 7.1 2 4 6~ 7.1 8 6.7 7.2 which were collected in plastic tubes, tightly stoppered, stored for from 2 to 4 hours and then extruded. The pH measured one minute after the cores were extruded was 0.4 to 0.6 units above the original in situ values (Table 4b). Discussion The data contained in this paper illustrate ways that pH can be increased or decreased by the addition or removal of hydrogen sulfide and carbon dioxide from sediments. The increase in pH shown in Figure 2, which occurred after natural anaerobic sedi· ments were removed from their environment, was similar to the increase shown in Figure 1 which took place when hydrogen sulfide and carbon dioxide in artificially gas sat­urated sediments escaped from the sediment samples in the laboratory. ' Data from experiments with decomposing fish indicate that the area adjacent to the fish where hydrogen sulfide was observed became more acid. When the sediment was removed from near the fish, the pH rapidly increased (Table 1) to a .level higher than the original pH of the sediment before the decomposition started. DIFFUSION PROCESSES The increase in pH for both laboratory and field experiments was linear with the square root of time. As shown by Hober et al. (1945, p. 10) gas diffusion processes have a linear function with the square root of time. This indicates a similarity between the gas diffusion process and the experimental increase in pH described in this paper. How­ever, since pH is a logarithmic function, the linear change in pH may actually be the result of dissociation of the gases, or an equilibrium between the dissolved and ionic state of the gases. Diffusion rate of gases through sediment is governed by Fick's Law and thus is de­pendent on the amount of interstitial water, density of the gas, temperature, agitation of the -sediment, salt content, amount of colloidal and dissolved organic matter and the abundance, size, and arrangement of interstitial spaces of the sediment. These factors probably account for variations in the amount of pH increase found in various places and depths of sediment when the sediments are exposed to air. The data from Table 3 show how the pH change is more rapid for disturbed cores as compared to undisturbed cores. The permeability and amount of water present may influence the rate of diffusion. Sediments composed of small particles normally have a larger water content than sedi­ments composed of large particles (Pettijohn, 1949: 67), and one might expect more gas to diffuse from finer sediments, such as clays, at a faster rate. However, small particles of one micron and smaller will have a greater proportion of, or perhaps all, the Water in the bound state, and thus the diffusion rate may be lower (See Szent­Gyorgyi, 1957: 35) • . The diffusion coefficient of a gas affects the slope of a curve of pH plotted against the square root of time but does not change the linearity. The variation in slope in Figure 1 suggests that the effects of the diffusion of hydrogen sulfide and carbon dioxide from sediment are different. The C02 saturated sediment has a pH of approximately 6.1 whereas the H2S saturated sediment has a pH of 6.8 as shown in Figure 1. DISSOCIATION CONSTANT FOR HYDROGEN SULFIDE AND CARBON DIOXIDE The dissociation constant for C02 (as H2C03) in sea water, K1 =0.26 to l.12Xl0-6, varies with the chlorinity and temperature as summarized by Harvey (1955, p. 166) . The pH of C02 saturated sea water may be calculated by equation 2 of Harvey (p. 172). With 1 atmosphere partial pressure of C02, at 30° C and chlorinity of 19 parts per thousand the calculated pH is 5. This figure is somewhat lower than the experimental value given in Figure 1. In sediments the partial pressure of the gas niay be lower than !atmosphere and a relatively higher pH should result. The K1 dissociation constant of H2S in fresh water at 25° C is 5.1 X 108• According to Hutchinson (1957, p. 759, water, subjected to a pressure of 1 atm of H2S, will have a pH of approximately 4. The addition of base substances will cause a rise in the pH. Thus the pH will be a function of the continued supply of the weak acid of H2S to the buffering capacity of the system. CHANGES IN PH OBSERVED IN SEDIMENTS The pH increase observed after anaerobic sediments are exposed to atmospheric con­dition is quite rapid. The data in Table 4 show that a small change in pH takes place within the time required to measure the pH (e.g. 1 min.) and that a small change in pH also takes place in sediments stored in plastic tubes which are sealed with rubber stop­pers at end. This change probably occurs as the sediments immediately react to agitation and exposure to air. Therefore one can assume that the low pH is not princi­pally caused by non-gaseous inorganic or organic acids which would diffuse very slowly. However the over-all pH of any sediment will depend not only on the presence of carbon dioxide and hydrogen sulfide but on the balance of all the acid or basic con· stituents of the sediments, for many of the by-products of bacterial activity may be organic acids. The data in Table 2 representing measurements from several locations in the bays show that a low Eh is accompanied by a relatively low pH, as compared with the sur­ face pH. Sediments used in this study were selected from field areas which were rich in sulfides in order to get large variations in Eh within a short core. The senior author has made several hundred in situ measurements of Eh and pH (unpublished) in various sedimentary environments and has found no general trend in Eh as plotted with depth. Some reduced sediments without H2S do not have a low pH and some cores with re­ duced areas near the surface became more oxidized with depth. Debyser (1952, 1955) and Wheatland (1954) have reported that whereas the pH of some anaerobic sediments which contained hydrogen sulfide decreased with depth, others increased. Debyser did not state how his samples were handled during the pH measurements. The data in Table 2 also show that the pH of the surface of the sediments varied from 8 to 9.2. This relatively high pH on the surface of those moist marine sediments may have been caused by photosynthetic activities of the mat of unicellular flagellates, dia· toms and other algae which were present. Carbon dioxide, which is used as a carbon source by the plants, is removed from dissolved bicarbonate ions leaving an excess of hydroxyl ions (Rabinowitch, 1951, p. 899). Some investigators (Rittenberg, Emery, and Orr, 1955, p. 33; Revelle and Fairbridge, 1957, p. 282) have published information which indicates that pH values up to 8.5 are found in anaerobic sediments. The anaerobic sediments studied increased in pH when they were removed from their natural environment. The change was usually quite rapid. It is very possible that pH values of 7 and above reported in the literature for anaerobic sediments may be the result of the pH having increased because of the manner in which the cores were handled before the pH was measured. The microbiologist has been able to determine the number and types of bacteria in sediment but has not been able to relate the population to activity. The amount of pH increase when anaerobic sediments are exposed to the air may be an index of the amount of activity of microorganisms in the undisturbed sediment, a large increase in pH representing a large microbial activity. BACTERIAL ACTIVITY AND PH The data from our experiments indicate that the pH of natural sediments may be lower than that anticipated according to reactions obtained from pure culture work on, sulfate reducing bacteria such as described by Senez (1951 and 1953). According to Senez, the hydrogen for the reduction of sulfate to sulfide by sulfate reducing bacteria may be provided from organic matter or hydrogen gas according to the following re­actions: 2CH CHOHCOOH +2H 0 +2H+ +SO--= 2CH COOH J ~ 4 3 + 2C02 + H2S + 4H20 (1) 2H+ +so~-+ 4H2 = H2S + 4H20 (2) Both reactions from controlled experiments should result in a loss of hydrogen ions and a production of weak acids with a resulting net increase in pH. Revelle and Fairbridge (1957, p. 282) also state that the removal of the strong acid sulfate ion and the re­ placement with sulfide, a weak acid, should result in a rise in pH. However, it must be emphasized that changes in sulfate content alone will produce only small changes in pH as compared to changes in hydrogen ion concentration. The results in this paper of a lowering of pH during fish decomposition and hydrogen sulfide and carbon dioxide production show that sulfate reduction apparently is not accompanied by a de­ crease in hydrogen ion concentration. Perhaps the bacterial reduction of sulfate to sulfide may take place in nature as shown by the following reaction: 2CH CHOHCOOH +2H 0 +Ca++ +SO--= 2CH COOH 3 2 4 3 +CaC03 +H2S +C02 +3H20 (3) The pH according to equation 3 would be lowered by bacterial sulfate reduction, pro".id­ ing the hydrogen sulfide and carbon dioxide remain dissociated in the sediment. In order for this reaction to take place, calcium carbonate would have to precipifate. Shepard and Moore (1955) and Ginsburg (1957) state that carbonates are often foiind in acid sediments so this limitation may not be serious in nature. Emery and Rittenberg (1952) state that pH may be dependent on the bacterial conver­sion of organic matter and the nature of organic by-products illustrated by ,H2A and A in the equations: 4H A +SO--= S--+4A +4H 0 (4) 2 4 2 where A represents some o~idizable organic compound. For example, ifH2A is an alcohol the resulting production of 2 moles of acetic acid for each mole of sulfate reduced should produce a decrease in pH. However, our results indicate that dissolved gases (H2S and C02) are the major factors affecting pH rather than organic acids. The senior author has made pH measurements on an enriched culture containing sulfate reducing bacteria in sediments with lactate as an energy source. The pH de­creased from 8.3 to . 6.5 in three days and then started to rise reaching a pH of 7 .5 in three days. This might be explained by metabolic by-products being accumulated during growth and the initial pHregulation by H2S or C02overshadowed by alkaline metabolic products. Summary l. The pH of sediments decreases when hydrogen sulfide and carbon dioxide are bubbled through the sediment or when the gases are produced in the sediment during the bacterial decomposition of fishes. 2. The sediments saturated with hydrogen sulfide or carbon dioxide increased in pH linea'rly as a function of the square root of time when the sediments were exposed to an atmosphere of nitrogen gas. The increase in pH is considered to be caused by the escap~ of the gases from the sediment. The diffusion rate increased when the sediments were disturbed. 3. The similarity in the manner and rate at which pH increased in sediments im­pregnated with the gases and anaerobic sediments obtained in natural and simulated natural environments, suggests that the change in pH values in the latter is caused by a loss of hydrogen sulfide and carbon dioxide. Low pH values obtained in anaerobic sedi· ments in nature are due at least in part to the presence of hydrogen sulfide and carbon dioxide. Both gases are produced during microbial decomposition of organic matter. 4. The fact that samples of anaerobic sediment seem to increase in pH immediately after removal from the ground casts doubt upon the reliability of numerous pH values reported in the literature from determinations of pH made on sediments carried from the sampling area to the laboratory. In situ determination of pH seems to be the only reliable way of obtaining accurate pH values of anaerobic sediments. LITERATL"RE CITED Carroll, D. 1958. Role of clay minerals in the transportation of iron. Geochim. et cosmoch. Acta, 14: 1-28. Booij, H. L. and H. G. Bungenberg de Jong. 1956. Protoplasmatologia, Vienna, Springer-Verlag, pp. 1-162. Debyser, J. 1952. \'ariation du pH clans l'epaisseur d'une rnse Am·iomarine. C. R. Acad. Sci., Paris, 234: 141-742. ----. 1955. Etude Sedimentologique du Systeme Lagunaire D"Abidjan. (Cote-D'Irnire) Rev. Inst. franc. Petrole, 10: 319-334. Emery, K. 0. and S. C. Rittenberg. 1952. Early diagenesis of California Basin sediments in relation to origin of oil. Bull. Amer. Assoc. Petrol. GeoL 36: 735-807. Ginsburg, R. N. 1957. Early diagenesis and Hthification of shallow-water carbonate sediments in south Florida. Reg. A~pects of Carbonate Deposit: on. Soc. Econ. Paleont. Miner., Puhl., 5: 80-100. Han·ey, H. \\''. 1955. The chemistry and fertility of sea waters. Cambridge, Cambridge Univ. Press, pp. 224. Hober, R. 1945. Physical cht>mistry of cells and tissues. Philadelphia, The Blakiston Co., pp. 676. Hutchinson. G. E. 1957. A treatise on L!mnology. Vol. I. Geography, physics and chemistry. New York, John Wiley&: Sons. pp. 1015. LaLou, C. 1957. Studies on the bacterial precipitation of carbonates in sea water. J. sediment. Petrol., 27: 190-195. Mortimer. C. H. 1941. The exchange of dissoh-ed substances between water and mud in lakes. J. Ecol., 29: 280-329. Pettijohn. F. J. 19-19. Sedimentary Rocks. New York. Harper&: Brothers, pp. 526. Postgate, J. R. 1951. On the nutrition of Desulphoribrio desulphuricans. J. gen. l\.licrobiol., 5: 714-724. Rabinowitch. E. I. 1951. Photosynthesis, \'ol. II. Part I. New York, Interscience Publishers Inc. pp. , , 603-1208. ReYelle, R. and R. Fairbridge. 1957. Carbonates and carbon dioxide. Chapt. 10, Treatise on Mar. Ecology and Paleoecology. l\Iem. geol. Soc. _.\mer.. 61: 239-295. Rittenberg, S. C.. K. 0 . Emery. and W. L. Orr. 1955. Regeneration of nutrients in sediments of marine basins. Deep Sea Res., 3: 23-45. Senez, J. 1951. Etude comparative de la croissance de Sporovibrio desulfuricans sur pyruvate et sur lactate de soude. Ann. Inst. Pasteur, 80: 1-14. ----. 1953. Sur L'activit et la croissance des bacteries anaerobies sulfato-reductrices en cultures semi-autotrophes. Ann. Inst. Pasteur, 84: 595-605. Shepard, F. P. and D. G. Moore. 1955. Central Texas coast sedimentation: Characteristics of sedi­mentary environment, recent history and diagenesis. Bull. Amer. Assoc. Petrol. Geo!., 39: 1463­1593. Szent-Gyorgyi, A. 1957. Bioenergetics. New York, Academic Press, pp. 143. Wheatland, A. B. 1954. Factors affecting the formation of sulphides in a polluted estuary. J. Hyg., Camb., 52: 194-210. ZoBell, C. E. 1946a. Marine microbiology. Waltham, Mass., Chronica Botanica Co., pp. 240. ----. 1946b. Studies on redox potential of marine sediments. Bull. Amer. Assoc. Petrol. Geo!., 30: 477-513. Comparative Studies on the Metabolism of Marine Waters1 HowARD T. ODUM AND CHARLES M. HosKIN Institute of Marine Science The University of Texas Port Aransas, Texas Contents J'\TRODUCTION ---------------------------·-···­____ ____ ____________ ·--·------·----------------·-----Page 16 Acknowledgment METHODS -·---·---·--------------------····-··-------·---------------------------------------------------18 Instruction for Measurement of the Community Metabolism with the Diurnal Curve Method Diurnal Curve Methods in Estuaries Inverted Curves Diurnal Measurement from an Offshore Oil Drilling Tower Light and Dark Oxygen Bottle Methods in Estuaries Comparison of Diurnal Curve Methods with Bottle Measurements Bell Jars and Mud Bottles COMMUNITY METABOLISM ----------------37 Metabolic Rates in Texas Bays Seasonal Trends in the Grassy Laguna Madre and the Port Aransas Redfish Bay Diffusion Constants in Marine Bays with Suggestions for Computation of Pollution Capacity Heterotrophic Fertility and the l:se of Saturation Deficit to Classify Metabolism Daily Course of Efficiency of Photosynthesis SUMMARY -------------------------------­ 44 LITERATURE CITED ----·-------·-­ -----··-----------·-------···------------------------------------------· 45 Introduction Each day as the sun rises and retires the beautiful green bays like great creatures breathe in and out. By day photosynthetic production of food and oxygen by plants is plentiful, but day and night there is also a furious feasting. The animals, the consumer parts of plants, and the bacteria remove the food and oxygen previously created from the sunlight. On some days the production exceeds the respiratory consumption, and or· ganic food matter accumulates, but at other times respiration dominates so that the 1 These studies were aided by the Rockefeller Foundation. waters and their bottom oozes lose their store of energy. Just as the life in single or­ganisms is driven by the metabolism of the body cells, so many marine phenomena of theoretical or practical interest to man can be related to the composite metabolism of the environment. All through the year photosynthetic metabolism provides necessary chemical fuels for food chains, fisheries, sport of man, oyster reefs, plant beds, duck foods, dissolved oxygen for counteracting pollution, and biological activity that con­sumes and binds radioactive fallout to organic matter. Even the geochemical cycles are driven by the community metabolism which alternately binds and releases carbon­dioxide, nitrogen, phosphorus, and many trace elements. The organic matter and oil left in past ages in the marine sediments may be understood only by interpreting the metabolism of past seas. It is thus a first step in logical conservation as well as for scientific understanding to know the nature of the community metabolism. From metabolic rates may be estimated the magnitudes of dependent processes. Sensible management based on maintenance of fertility is a principal by-product of community studies in marine ecology. In this paper, metabolic methods are used for rapid study of whole water areas. Meas­urements of metabolic rates in contrasting waters indicate qualitative contrasts and a wide range of values in quantitative data. Such data can provide understanding of shallow marine bay systems anywhere and a basis for resolving conflicts for conserva­tion leadership. Diurnal curves of oxygen and carbon-dioxide (or pH) are numerous in the literature. Many recent efforts are being made to convert these data into metabolism figures. In addition to the papers cited in previous reviews (Oduin, 1956, 1957a), there are the following: in lakes: Jackson and McFadden (1954), Talling (1957), and Vinberg whose work in 1940 is quoted by Talling (1957); in marine waters: Sugiura (1953, 1956), Kohn and Helfrich (1955), and Ryther (1958); and in streams: McConnell (1958), Hoskin (1957), and Hornuff (1957). In this paper the diurnal curve method for measuring metabolism is widely applied in coastal waters with a number of tests of the methods. Acknowledgement We are grateful to the many persons who have helped with the laborious day-night sampling including Dr. W. McConnell, University of Arizona; Dr. Louis Kornicker, In­stitute of Marine Science; and Mr. William Guest and Mr. Richard Hoese, Texas Game and Fish Commission. Mr. Kilho Park and Mr. Herbert Bruce of A & M College of Texas cooperated iR the diurnal curve work concurrently with the pH studies reported elsewhere (Park, Hood, and Odum, 1958) and Mr. Thomas Hellier, Jr. aided with the Laguna Madre series. We are grateful for further testing of methods by the ocean­ography class, Duke Marine Laboratory, 1956 and the biological oceanography class at the Institute in 1957. Bell jar and tank experiments were supported by the National Science Foundation grant on Ecological Microcosms. We are grateful to Mr. M. Lerner for courtesy at Bimini. The Gulf Oil Company helped us in the oil tower work. Tanks were made available by Dr. A. Chestnut, Institute of Fisheries, University of North Carolina, Morehead City. These studies were begun at the Duke Marine Laboratory, Beaufort, North Carolina. Comparative Studies on the Metabolism of Marine Waters Methods For measurement of the overall metabolism, dissolved oxygen values representative of the free water were measured by means of the Winkler method by an observer in a small boat through a 24 hour period. Then as previously described (Odum, 1956) rates of diffusion, respiration, and gross photosynthesis were computed graphically. The light and dark oxygen bottle methods for measurement of plankton metabolism were also used at each station. In the Texas bays so far tested, the gentle drift of the water over flat bottoms tends to provide ideal conditions for adequate mixing for the diurnal curve method for measuring community metabolism. Since the marine estuaries and ponds are not entirely homogeneous, the reproduci­bility and variation of the measurements had to be determined. Replication of all the diurnal curves was not possible because of the labor, but duplicate curves from contrast­ing stations were obtained for Mesquite Bay (Table 2) July 22, 1957, and 3 separate stations in the upper Laguna Madre were run by W. McConnell on Jan. 22, 1958. The metabolic values in gm. Oxygen/ M2 / day for the three stations 1h M deep in the Laguna Madre are given below. Diffus:on constant is given in gm./M2 /hr. at 0 per cent satura­ tion (K). p R K Reds Camp, 5th post 2.1 2.7 0.3 Reds Camp, K. Park's station 2.5 2.6 0.3 Reds Camp, 1957 station 1.9 1.9 0.5 For the convenience of others planning to use the diurnal curve methods and in re­sponse to requests, instructions are given below with a sample calculation (Fig. 1). INSTRUCTIONS FOR MEASUREMENT OF COMMUNITY METABOLISM WITH THE DIURNAL CURVE METHOD 1. Locate an area of representative depth in which the history of the water moving into the area will be similar to the history of the water moving away. Circulation should be gentle as created with slowly drifting current and gentle wave action. Use fluorescein dye spread on the surface to determine the degree of vertical circulation. The dye will become vertically dispersed throughout the layer in 1-5 minutes in a suitable area. If the water is vertically stratified as indicated by temperature stratification, samples must be taken at several levels (Fig. 2). If it is a new area not previously tested, run several stations to determine if a single station is representative of the water mass (Fig. 3). 2. Sample and analyze with the Winkler method for dissolved oxygen every three hours for a duration of 24 hours. At each time of sampling, collect duplicates or tripli­cates, preferably separated by a minute of time. Express data as mg./ liter so that ready conversion to gm./ M2 can be made simply by multiplying by depth. At each sampling also obtain temperature, depth, and some estimate of current; collect two salinity samples during the period. Determine the average depth in the trajectory and during the period. 3. Plot oxygen and temperature curves as shown in Figure 1. From tables of oxygen saturation for given temperature and salinity determine the percent saturation and plot FIG. l._A typi~al ~iurnal curve and sample calculation for determination of gross photosynthesis ( P), commumty re~p1rat1on. ( R) , and diffusion constant (K). Redfish Bay, Port Aransas, Texas, July 18-19,1957. Calculations are included on the face of the graph. See text for explanations. REDFISH BAY JULY 18-19 1957 30"4• • 10 5 12 00 00 6 18 35 T •c 30 ONE SQUARE• 0 .5 Gt.yM3 DIFFUSION CONSTANT !PER VOLUME): : Q ttw-tE 0-LO SAMPLE DIFFUSION "1\ • 100 SM-~ =100 78_• 1.28 GM/%R. 0 CORRECTION AT S•IO~ CALCULATION: S-A. • C-0.68) Cl.28) • -0.87 NIGHT RESPIRATION• 1.3 G~Ma/ ' G1¥~HR. / HR•. ' : /DEPTH D1Fi:'us10N CONSTANT: K . • z/.. • (0.46M)(1.2eJ = o.59 GM/Mz, !PER AREA) / HR. AT S=I00% •·· METABOLISM PER DAY: .,· GROSS PHOTOSYNTHESIS: p • (56 SQUARES)(0,5~)(0.46M) =13 GM/Mt/ / DAY STIPPLED AREA/ ONE SQUARE) DE;.;9 R • ( -1,3 GM/M3-' > !24 HRS> C0.46Ml • 14GM/t/.1 'HR. RESPIRATION: ' DAY a saturation curve (Fig. 1) . See Steen (1958) for discussion of various tables available. Use millimeter graph paper for these graphs, since many subdivisions are necessary for counting squares. 4. From the oxygen graph, using a running two hour interval, determine the rate of oxygen change in mg./liter/hr.; plot the rate values just below the other graphs on the same horizontal time scale. A two hour computation interval provides some smoothing of the sharp corners of the oxygen curve eliminating artifacts that come from infrequent sampling. 5. Determine the diffusion constant from the rate of change graph (lower graph in Figure 1) with the use of two points, one in the predawn period (qm) and one in post sunset period ( q. ). Use care in choosing the points; do not take an irregularity or point known to be atypical due to sudden change of wind or other variable. The diffusion con­stant per volume is k. Little letters are used for volume; capital letters for measurements per area. In mg./l/ hr. at 0 per cent saturation k may be computed as follows: k=lOO qm ­ qe where qm is rate of change at a time before dawn Sm-Se in mg. /I/hr. q. is rate of change at a time after sunset in mg./l/hr. Sm is the percent saturation deficit before dawn at the time chosen for calculating qm. ( 78 percent in example in Fig. 1). S. is the percent saturation deficit after sun· set at the time chosen for calculating q •. ( 0 percent in example in Fig. 1). K, the area based diffusion constant, may be obtained by multiplying by the depth (Z) in meters. K in gm./ M2 / hr. at 0 percent saturation has been found to be: 0.1 to 1 in quiet waters, waters less than 1h M. deep, or stratified waters; 1 to 3 in bays with gentle circulation and wave action; and 3 or more in strong currents and large waves. In the example in Figure 1 sample calculations are given with result that k is 1.28 gm./M3/ hr. at 0 percent saturation ( 100 percent saturation deficit). At 0.46 meters depth K is 0.59 gm. /M2/hr. at 100 percent saturation deficit. 6. On the rate of change curve multiply the saturation deficit by k (volume diffusion constant) and add or subtract from the rate of change curve graphically so as to replace oxygen which diffused out or to retract the oxygen which diffused in. This is done in the example in Figure 1. One multiplication is given for a predawn point in Figure 1. The new curve includes community metabolism corrected for diffusion (dashed line in Fig. 1) ; it is the curve which might have resulted had there been no diffusion. 7. On the corrected curve determine the community respiration line by drawing a smoothed line through the average of the nighttime rate of change points. See example marked in Figure 1. Extrapolate this across the daytime period. In so doing one assumes nighttime respiration similar to daytime respiration which is probably not correct. It is possible to make this line in some different shape as soon as it is determined what the Fie. 2. Diurnal curve of oxygen for Stewart Farm Pond, Durham Co., North Carolina, March 29-30, 1956. Curves for two depths are drawn in the upper graph, but computations are based on an average of all depths in the second graph. In the lower graph the dashed line has been corrected for oxygen diffusion. STEWART POND MARCH 21·30, 1951 13 12 PPM o, II 10 I •./.:"i./ ...........'""'~­./;'--. L.4-----:--'\: ·~.IJ~ ~ • ~--J ~. \ 9 ~ • I 7 6 12 18 9 8 1 /·-0-· __.__ 1·"· . -· ""' 0 6 / LIGHT 120 110 'JC. SAT. 100 90 80 8 12 18 a 12 II NOON TIME 0, DAY IN HOURS Comparative Studies on the Metabolism of Marine Waters usual shape of daytime community respiration is. The average rate of respiration multi­plied by the depth and the number of hours (24) yields community respiration rate (R) in gm. oxygen/M2/day. See example of calculation in Figure 1. Values are best left in oxygen rather than converted fictitiously to carbon or glucose. For rough purposes grams of oxygen are almost the same as grams organic matter; further conversion is unrealistic unless the respiratory quotients and chemical nature of the community or­ganic matter are known in some detail. 8. The area between the respiration line and the daytime hump of the corrected rate of change curve is the gross community photosynthesis including simultaneous respira­tion of the community. This area is stippled in Figure 1. Determine the gm./m3 value of one square on the graph paper by multiplying the dimensions of the square. For ex­ample, in Figure 1 the square has dimensions 0.5 mg./ l/hr. in the vertical and 1 hr. on the horizontal time axis. The product, 0.5 mg./ l, is the metabol:c value of one square on the graph paper. The s~ippled area contains 56 such squares. Count the squares and multiply by the gm./M3 per square and depth in meters to obtain the gross production (P) in gm./M2/ day oxygen. In the example in Figure 1Pis13 gm./M2/hr. Both the respiration and the photosynthesis are much affected by the choice of the respiration line. The accuracy of the method is probably limited by this step. In some smooth curves (Fig. 1), there is little doubt and one has no choice in drawing the respiration line. In some graphs in which points are very irregular, sume subjectivity is introduced (ex­ample, Fig. 4), and results are held with less confidence, although the confidence inter­val can be estimated visually. It is probable that community respiration is actually varying as suggested by C02 curves (Park, Hood, and Odum, 1958). 9. If several different stations have been run, the oxygen curves should be superim­posed. If the differences are small, the data should be lumped and the combined curve used as the general curve for the water mass. If large differences are found, indicating heterogeneity in the communities, the results cannot be considered as accurate. In such cases the curve for one station does not represent metabolism at that station unless the water was stationary. The basic assumption in diurnal curve analysis (not considering the entirely different conceptual approach used for springs; Odum, 1956, 1957a) is that the water measured at a station has had a recent metabolic history characteristic of the water measured at the station during the previous sampling. If the water moving in has adifferent content of oxygen from the water which simultaneously leaves the station, results will be in great error. The underestimation of trajectory effect is the principal danger in the uncritical use of the diurnal curve method. For detailed work in a new area, the worker is obligated to present duplicate station results in order to establish the degree of uniformity of the body of water. In the diurnal curve method metabolic rates are computed on a volume of water basis and converted to an area basis by multi­plying by the average depth. Where possible an average depth should be determined with careful morphometric measurements. Such measurements for example were made in Prytherch's lagoon (Fig. 5) and Stewart Farm Pond (Fig. 2). In most of the data on Texas bays, however, the depth of the station has been used as a first approximation of the depth of the water mass measured. Even though Texas bays are remarkably flat Fie. 3. Simultaneous diurnal curves at three different stations in the upper Laguna Madre near Reds Fish Camp as obtained by William McConnell, January 21-22, 1958. See text for values of metabolic rate and diffusion constant. LAGUNA K. 9 8 7 9 8 7 MADRE JAN. 21-22, 1958 :30 °loo PARK's STATION 9 STATION USED IN 1957 8 00 06 12 18 Comparative Studies on the Metabolism of Marine Waters BAFFIN BAY JULY 26-27,1957; 57.3"Kt. 6 5 3 2 6 12 NOON 0.8 0.5 02 CHANGE 0 M'1 'l'HR -0.5 -0.8 00 6 12 18 00 Fie. 4. Example of a diurnal oxygen curve with somewhat irregular night respiration line. Baffin Bay, July 26-27, 1957, a turbid, hypersaline bay. The dashed line in the lower graph has heen cor­rected for diffusion. bottomed, the error in estimation of the depth of the water in its flow is likely to exceed errors in the diurnal curve measurements themselves. A few of the curves have irregular rate of change curves at night clearly associated with variations in wind velocity recorded by the observer in the field. A decrease in wind velocity, for example, just before dawn in Figure 6 decreased the diffusion coefficient. In general some smoothing of the night curve by averaging points is required where some basis for interpreting irregular points exists. LAGOON BEAUFORT. N. C. AUG. I 1956 · 35%0 8 CHANGE .5 0 .5 00 6 12 18 NOON FIG. 5. Example of a diurnal curve for a terminal bay area in a tidal marsh system, Prytherch's lagoon, Beaufort, N. C., Aug. 1, 1956. The dashed line has been corrected for diffusion. The separation of components of the diurnal curve such as given in this procedure and in variations previously described have been elaborately and formally expressed mathematically by Sugiura ( 1953, 1956). Although his results are similar, the math is not necessary for empirically established functions, which can be better understood with graphical work. DIURNAL CURVE METHOD IN ESTUARIES Experience shows that the diurnal curve methods cannot be applied in all estuarine water. A basic assumption of the free water methods is that analyses are made in water generally representative of the designated community area or water body. With the irregular flow systems usually found in estuaries, great care must be used in selecting suitable stations. Velocity is also important. A somewhat uniform diffusion constant must be assumed day and night. Since diffusion constants depend on current velocity, current variation introduces error. LAGUNA MADRE NOV, II, 1956 42. %0 O.GM 9 8 7 f-w1NO STOPPED -1.ooo.,.______,o,,..,s~---.,._...------,.1=----~oo Fie. 6. A diurnal cun·e in which a cessation of wind was noted before dawn accompanied by diminished diffusion and momentarily increased rate of oxygen decrease (see arrow) . Upper LagunaMadre, Red's Fish Camp, Nov. 11-12, 1956. To test for suitability of the diurnal cun·e in a bay, duplicate and simultaneous meas­urements should be made at several stations. If there is no great range of values, the data at each hour may be safely lumped and the computation applied to the whole curve. In Figure 3 are similar data from 3 stations in one bay area. When there is a slow drift over very fertile communities, dye markers may he applied and upstream-downstream measurements can be made as on reefs or the Florida Keys turtle grass (Odum, 1957b). Where there is a strong tidal system as at Beaufort, N. C., the method cannot he used in the open bays or in channels. The heterogeneity of current velocity causes large differ­ences in water mass, diffusion exchange with the atmosphere, and such large movements of water that no water mass may he followed with practicality. In such flows current markers mark streamlines rather than trajector~es. With this kind of mixing, trajectories of objects larger than colloidal size are a fiction as far as indicating water movement. Attempts to measure diurnal curves at the Institute of Fisheries of the University of North Carolina, Morehead City, in Bogue Sound yielded only saturation data for oxygen even though considerable metabolism was known to exist. Diffusion rates were very large. It was hoped that narrow uniform tidal canals were sufficiently uniform that a diurnal curve at one station would indicate a general metabolism. Such a curve is shown in Figure 7 for Core Creek, near Beaufort, N. C. Times of maximum tidal current are indi­cated. It is clear from the graph, however, that rates of diffusion were varying tremen­dously, being large during strong flow and diminishing during lesser flow. At times of strong current (arrows in Fig. 7), oxygen concentrations returned towards saturation. FortuQately, it is easy to recognize some complication when the diurnal curves have unusual shape. The diurnal curve can he used as an approximation however, in a tidal estuary if one selects a partially enclosed terminal body in which the water has minimal motion. A curve for such a body in Beaufort, N. C. is given in Figure 5. During the 24 hour period only part of the water moved gently out and gently hack. The extent to which the history of the water returning differed from the water leaving is the extent of the error. Fortunately for workers in the hays of Texas, tides are small (0.3-0.6 M/day) in the Gulf and mostly damped out in-the hays. Fresh water flows and winds dominate circulation. If one picks a 24 hour series in one wind regime, conditions are ideal with enough mixing to prevent stratification hut usually with not enough water current ve­locity to make diffusion corrections large. Since all of the Texas hays are only 1 to 12 ft. in depth, bottle measurements are of little use for total metabolism and the diurnal curve method is the only method now possible. See comparison of bottle and diurnal curve methods that follows. In the lagoon at Bimini, British West Indies the water enters with the tide from two sides. By choosing a station in the area where the flows meet, an area of minimal current was found for obtaining a diurnal curve (Fig. 8). The data from this shallow area represent the water that for the most part is being modified by the hard, algal covered, limestone substrates in the wide central area. INVERTED CURVES Among the situations which are easily recognized as unsuited for usual production and respiration measurements are the inverted curves in which oxygen is observed to in­crease during the first part of the night and to decrease during the day. One such curve was found on a dystrophic swamp river in coastal North Carolina (Fig. 9). Measure­ments with a thermistor during the day indicated temperature stratification in the top layer of the opaque brown water. A suggested interpretation of the inverted diurnal curve follows. The large respiration of the particulate and dissolved swamp materials far exceeded photosynthesis. A balance between respiration and diffusion from the air existed during the night hut was interrupted during the day by the temporary temper­ature stratification which cut down the diffusion constant so that oxygen removed by respiration could not he renewed as rapidly as after nightfall. A similar curve was found Comparative Studies on the Metabolism of Marine Waters CORE CREEK JUNE 21-22, 1956 8 7 ,---­ ·~:~-----..... --..--------------~: 6 • 5 6 12 18 / LIGHT \ 110 roo % 90 SAT. 80 70 6 12 18 0.3 \).2 0.1 02 CHANGE 0.0 PPM/HR. 0.1 0.2 0.3 0.4 6 12 18 NOON TIME OF DAY IN HOURS F1G. 7. A diurnal oxygen curve in a narrow tidal canal between two large bays, Core Creek, Beau· fort, N. C., June 21-22, 1956. Current velocities range 0-4 knots during the period. Metabolism can not be computed from this curve with accuracy. Arrows indicate times of maximum current velocity. BIMINI LAGOON JAN. 27 1957 CLEAR 9 02 MG L 8 , • , tlGH TIDE 7 ·-----..J 06 12 18 00 T °C ..-.... /.··. ~.. o.5 '· . . . ·' ,~. .. . . ' , . . . . . . \ RATE , . . " .. ' , . . .. , . * • .-:'"'. • I M9{_~R . . 0 ....-----~• • • .. • .. • • • ·.. • 4 • • • ' ----_-s-:..::.::...: ...:.., . .'...:.:·_;_· ..:...~·-=-· .~..:_~ ~-----­ -0.2 FIG. 8. Diurnal oxygen in a minimal current area in an area of convergence-divergence of Bimini Lagoon, Bimini, British West Indies, January 27, 1957. Data in cooperation with Dr. Louis Kornicker with courtesy of the Lerner Laboratory and the American Museum of Natural History. Substrate is an irregular limestone with algae, turtle grass, and benthic animals. by Laurie (1942) and by Hoskin in Singletary Lake, N.C.; another opaque. dystrophic body strongly colored with swamp seepages. DIURNAL MEASUREMENT FROM AN OFFSHORE OIL DRILLI'.'\G PuTFOR:\I It has been previously assumed that diffusion is so rapid that the diurnal curve methods are impractical in deep and relatively poor waters of the open seas. However, one can never eliminate diurnal changes in oxygen entirely because the rate of diffusion is proportional to the saturation deficit. As the deficit becomes small, so does the diffusion rate. Sugiura (1953, 1956) was successful in determining diurnal oxygen changes i11 open waters of Japan. NEWPORT RIVER JUNE 21-22, 1956 6 ..­ -.._----··-.. . ---­ ""-·­ 5 PPM 4 ·-~..-· ..---------­ 02 • 3 2 6 12 18 / LIGHT \. 0.3 0.2 0.1 o·2 CHANGE 0.0 PPM/HR. 0.1 0.2 6 12 18 NOON TIME OF DAY IN HOURS FIG. 9. Diurnal oxygen curve of inverted type in a brown dystrophic swamp river (200 ppm color), Newport River, Beaufort, N. C., June 21-22, 1956. Changing rates of diffusion and changing rates of respiration during the day produce a cun-e which cannot with present understanding be con· verted into metabolic figures. The dashed line in the upper curve represents correction from the Ohle test for interference. The possibility of determining metabolism of the Texas shelf waters from oil rig towers was investigated near Port Aransas. Accuracy was increased by obtaining a large number of duplicate samples for the Winkler analyses as plotted in Figure 10. There seems to be no doubt that a diurnal curve was observed during a gentle 1 M wave swell in 15 M of blue Gulf water. The strong coastal current running from the south during most of the period stopped in the period before 6 A.M., and the accumulations of fishes and the garbage of the rig may have affected metabolism in a local atypical manner. Even if this one time of day is ignored, a production value may be computed. The di­urnal changes in the lower part of the water column were greater than at the surface. The bottom occurs within the euphotic zone in the clear blue water and maximum me­tabolism may have been in lower levels. Diffusion from above may also explain the smaller daily amplitude of the surface data. The free water calculation indicated greater photosynthesis than in concurrent bottle experiments. There is a possibility that metabolic FIG. 10. Diurnal oxygen curves at 3 levels in blue Gulf of Mexico water beneath an oil drilling tower, July 14-15, 195i. LF SHELF PELICAN TENDER JULY 14-15 , 5 • •• • SURFACE 5 6 METERS • • 5 • 12 METERS • 5 MEAN 00 6 12 18 00 NOON estimates with bottles are too low even in shelf waers. A discussion of the use of bottles follows. LIGHT AND DARK OXYGEN BOTTLE METHODS IN ESTUARIES The light and dark bottle methods have now had wide use in limnology and ocean­ography, but some care with the details of their use has been found essential for obtain­ing reproducible results in estuarine waters as follows: Replicate oxygen samples in the free water normally indicate great heterogeneity of oxygen and populations. Thus plankton water used for filling start, light, and dark bot­tles must be homogenized in order to reduce variability. This is done by filling a 5 gallon carbuoy with bay water and swirling it for 5 minutes. Then this water is siphoned into 250 cc BOD (biochemical oxygen demand) bottles and filled in order: one start bottle, one light bottle, one dark bottle, etc. so that any bias in sequence of filling is dispersed uniformly among replications. Three replications of each category are considered necessary. Aluminum foil as a cover for black bottles is unsatisfactory because as crinkles de­velop, pinholes permit light to enter. Aluminum foil also reacts rapidly with salt water. Black, taped bottles are suitable with rubberized black cloth wrapped around and fastened with a heavy inner tube rubber band. When these precautions are taken, duplicate starting bottles are usually reproducible within 0.2 mg/l. After chemical fixation precaution in mixing the 250 cc bottles con­taining iodine is necessary prior to titration. It is believed that the distribution of iodine is affected by silt particles which settle to the bottom. In all oxygen analyses the direct Winkler process is usually adequate for non-polluted estuaries. The Ohle ( 1953) procedure was used for testing for Winkler interference. In all cases reported here interference was small (less than 0.3 mg./ l) and was ignored since the production and respiration measurements are based on diurnal differences. The actual value of the oxygen affects computations only in the calculation of the diffusion correction. Table 1 is a sample data sheet. A similar sheet was made for each bottle metabolic rate computed. In deeper waters 3 light bottles were placed at each of three depths. In shallower waters (1 M) bottles were placed at two depths. It is essential that light bottles be placed in the optimum light zone as well as in layers above or below. Optimum light is about 1 ,15 daylight. TABLE l Example of Data Sheet on Metabolism in Bottles. Data are given in mg./L (Gm./M3) . Laguna Madre, August l-2, 1957, depth 0.46 M U ght Light Slarl Dark lop bottom 5.90 4.89 6.92 7.05 5.90 4.73 6.96 6.98 5.99 4.70 6.90 7.40 5.93 4.Ti 6.93 7.14 '­ 7.03 R = (5.93-4.iil (0.46Ml =0.53gm./M2 /day Pn = 17.03 -5.93 1 ( 0.46 M) = 0.51 gm./M2/day (net yield of the bottle system) P, = P. +R = l.04 gm./1\12/ day (gross photosynthetic production of the plankton community) Comparative Studies on the Metabolism of Marine Waters COMPARISON oF DnncyAL CuRvE METHODS WITH BoTTLE MEASUREMENTS In Table 2 are reported the results of diurnal curve measurements and bottle measure­ments made simultaneously at the same station. Data from tanks are reported along with TABLE 2 Comparison Between Measurements of Metabolism made with Bottle Methods and Free Water Diurnal Curve Methods P and R Data are in gm./M2/day Oxygen. K is the Diffusion Constant in gm. 02/M2/hr. for 0 per cent Saturation as Computed from Diurnal Curve Data I. Systems without large bottom communities; plankton principally Fresh-water tank, stirred, pond water, Durham, N. C., May 25, 1956 ................... . Sea-water tank, open, Beaufort, N. C., June 20, 1956, stirred .................................. Sea-water tank, plastic cover, Beaufort, N. C., June 27, 1956 ................................... . Diatom bloom in marine creek, Ryther et al. (1958) ·· ·· ···-~---······· ········· ...... II. Systems with both plankton and bo.ttom mud components Stewart Farm Pond, Durham Co., N. C., .March 10, 1956, stratified ........................... . March 29, 1956, stratified ........................... . Prytherch's Lagoon, Beaufort, N. C., Aug. 1, 1956; data of Duke Oceanography class... . Baffin Bay, Riviera, Texas, July 26, 1957 ..... . Baffin Bay, Riviera, Texas, Dec. 23, 1957 ..... . Gulf shelf off Port Aransas, Pelican drilling tower, 15 M deep, July 14, 1957 ................ Tres Palacios Bay, June 17, 1957, salinity 2.4%o, turbid ................................... . Lavaca Bay, lift bridge, June 17, 1957, salinity 4.8%0, turbid .................................... Cedar Bayou, July 22, 1957, salinity 24.6%0, 1.5 M deep ....................................... . Mesquite Bay, July 22, 1957, salinity 15.5%0, Algal bottom, 1.2 M deep ........................... . Mesquite Bay, July 22, 1957, salinity 15.5%0, Shell bottom, 1.1 M deep ........................ . Copano Bay, May 18, 1957, Rt. 35 bridge, 1 M deep, salinity 13.1%0 ........................... . Bayside, salinity 29.9%0 .............................. Copano Bay, Aug. 8, 1957; salinity 11.8%0... . Cop;mo Bay, Oct. 20, 1957 ............................. . Aransas Bay, Rockport Pier, May 19, 1957, 1 M deep, salinity 21.0%0 ... Oct. 20, 1957, 1.3 M deep, salinity 18.7%o . Rockport Harbor, Aug. 5, 1957, salinity 34%0, after a bloom ....................... . Corpus Christi Bay, Ingleside, June 21, 1957, salinity 22.9%0 ................................... . Overflow tidal pool, South Jetty, Port Aransas, 0.15 M deep, July 1, 1957 ........... . 'Flood Pool, Ferry Road, Port Aransas 0.9 M deep .: ................................................. . III. Systems with dominant bottom plant and animal communities; plankton unimportant Laguna Madre, Texas, Mean, Diplanthera­ooze, 1957, Annual Curve (Figure 10) ..... . Redfish Bay, Port Aransas, Texas, Thalassia beds; Ransom Island, Mean of 5 days in all seasons ...................................................... Bimini Lagoon, British West Indies, Limestone Flats, 0.3-0.6 M depth, (Fig. 8) BotUe Method p R 1.50 0.83 2.40 0.73 1.17 1.58 4.8 1.12 1.26 1.63 0.79 6.25 2.76 1.90 7.91 3.0 4.0 2.0 9.4 4.3 1.1 1.53 1.50 2.5 2.1 0.85 0.95 0.75 0.86 1.86 2.3 0.26 0.44 0.59 1.1 2.3 1.5 0.96 0.24 2.4 1.01 0.70 0.29 0.12 0.20 0.36 0.31 Free Waler P R K 1.16 3.30 0.49 1.90 1.61 0.67 1.10 1.61 0.08 2.18 2.10 0.10 5.4 4.52 5.24 0.11 4.50 5.80 0.78 2.70 14.l 0.92 3.7 5.0 1.3 6.7 7.5 6.4 10.2 1.4 5.6 17.6 1.5 3.8 7.3 1.2 1.2 3.6 0.4 23.8 13.7 3.8 8.0 8.5 1.8 2.1 6.2 0.7 1.60 1.65 6.3 4.8 0.6 6.1 7.8 1.8 6.1 11.6 0.8 1.1 9.4 11.8 1.48 3.7 0.4 5.2 11.7 1.4 4.3 5.6 0.2-1.4 11.4 17.0 0.6-1.7 2.8 1.8 0.75 3-t Comparative Studies on the Metabolism of Marine Waters data from natural waters. All rates are reported in grams of oxygen per hour per square meter of community surface area. The bottles represent plankton metabolism only, whereas the diurnal free water meas­ments include effects due to the metabolism of the bottom and lateral surfaces. In the farm pond, the water was partially stratified, and oxygen at three levels was measured and multiplied by the pond volume at this depth based on a hypsographic curve of depth and area. In shallow bay waters steady stirring due to waves was observed and mixing was confirmed with fluorescein dye. Samples at several depths were similar in oxygen content indicating that mixing had been effective. In group I in Table 2 in freshly filled tanks where bottom communities are minimal, there is order of magnitude similarity of free water and bottle measurements. In Group II in Table 2 in the farm pond and shallow mud bottom bays, the respira­tion values are consistently much larger in the free water method probably because of the respiration of the mud surface and larger swimming animals included in the free water measurements. In environments of Group II where most of the photosynthetic production is planktonic (not benthic) it is not surprising that there is some similar order of magnitude in bottle and free water measurements of photosynthesis. It is clear from the data on the shallow mud bottom system:; that no adequate estimate of respiration or community BOD can possibly be obtained from bottle measurements. In this respect the bays and ponds resemble the streams where bottom respiration pre­dominates (Odum, 1956; Hoskin, 1957). Such conclusions must apply to any bottle measurements including radioactive carbon measurements. One is forced to suspect studies like that of San Diego Harbor based on bottles (Nusbaum and Miller, 1952). Only in deep water or bloom situations where the bottom role in respiration is relatively minor, can bottle measurements in shallow water indicate community metabolism. When bays are sufficiently shallow or clear enough so that appreciable light reaches the bottom, benthic plant communities develop lush communities. In such cases (Group III, Table 2) as Redfish Bay near Port Aransas, the bottle measurements of both photo­synthesis and respiration are completely unrepresentative of the whole community being one hundred times too small. Because bottom metabolism can be so important in bays 3 meters deep or less, discussions of nutrient cycles are incomplete if made on the basis of plankton alone as in Great Pond, Mass. !Hulbert, 1957). Although the bottle methods are not suitable in marine bays for total metabolism, they are apparently very useful in determining that part of the metabolism which is due to plankton. For this purpose, bottle measurements were continued in subsequent measurements in Texas Bays. Even where bottom metabolism is not a factor, free water measurements are likely to be superior to bottle measurements especially where diffusion corrections are small or known accurately. In bottles the normal turbulence is eliminated as well as the con­tinuous flux of water from regenerative parts of the system bringing nutrients. The constant level of a bottle suspended on a post or buoy does not provide the varying light conditions characteristic of nature. The rapid deterioration of metabolism in bottles has been shown by lchimura and Saijo 0958) with radio-carbon methods. However, other evidence ( Verduin, 19571 indicates there are disadvantages of very short time measurements before cells become somewhat adapted to new conditions. The rapid modifications of bottle communities have been much studied especially relative to the changing role of bacteria (Vaccaro and Ryther, 1954-). Whether the increased accuracy of a larger change in a longer period outweighs advantages of small population change!;' during shorter time is not known. Experience with lags and shocks in bell jar work (Odum, 1957a) as well as plant physiological work suggests that very short periods of a few hours may be imperfect. Most of the bottle measurements in this study were for 24. hour periods because of logistic requirements. BELL }ARs AND Mun BoTTLES Whereas the diurnal curve method provides estimates of metabolism of the plankton, nekton, and the bottom mud, the role of the bottom mud algae is determined wtih diffi­culty. Bell jars have been used previously over grass and reef substrates by Odum and Odum (1955), Odum (1957a), and Odum, McConnell, and Abbott (1958). As pre­viously discussed such measurements are very rough because of technical difficulties and because the rate of circulation of water is a controlling factor in benthic metabolism (Odum, and Hoskin, 1957). Some additional measurements of bell jar metabolism TABLE 3 Oxygen Metabolism Measurements in Shallow Texas Bays; Comparison of Bell Jar Methods with · Bottle and Free Water Methods P, gross photosynthesis; R, total respiration; P and R data are in gm./M2/day oxygen. K is the diffusion constant in gm. 0 2/M2/hr. for 0 per cent saturation as computed from diurnal curve data. Bottles Bell Jars Free Water p p p R R R K Laguna Madre, Reds camp, Nov. 11-12, 1956 .................... 2.1 2.1 5.1 3.2 1.4 Jan. 21-22, 1957 .................... 0.31 0.18 0.34 0.19 3.8 7.2 1.4 March 27, 1958. (by W. McConnell) ...................... 0.30 0.27 1.40 2.7 3.8 3.8 0.4 Redfish Bay, Ransom Island, . Jan. 31, 1957 ............................ 0.17 0.25 1.1 0.7 5.1 7.5 0.7 Baffin Bay, Riviera, Feb. 4, 1957 ----------------------------2.7 1.3 0.95 0.36 16.6 9.6 3.4 Bluegreen Algal Mat, 0.15 M deep, Portland, Texas, McConnell and Odum, May 26, 1958 -----------------··----·--1.3 1.0 2.2 1.9 0.3 over estuarine bottoms were made in Table 3 for the Texas bays. The bell jar measure­ments are consistently too low in both photosynthetic and respiratory metabolism as compared with the free water measurements. From evidence so far available, bell jars are good only for order of magnitudes and inferior where free water methods are avail­able. Another method involves measuring the metabolism of a mud surface which is allowed to develop in a 250 cc BOD bottle. A suspension of representative bottom mud is intro­duced into the bottle and a one inch layer is allowed to settle. Then the excess water is gently siphoned off and replaced with water from a carbuoy of known oxygen content. After 3 hours part of the water sample above the mud is gently siphoned off into a small oxygen bottle where Winkler analysis can be carried out. As indicated in Table 4 an estimate of the respiratory metabolism per area of mud can be obtained in this way. Tests for interference with Winkler analyses due to diffusing substances from the mud into: the free water indicate some effect. Thus such measurements are very rough, but they do indicate a large mud metabolism consistent with the high metabolic values ob­ :36 Comparative Studies on the Metabolism of Marine Waters TABLE 4 Dark Bottle Measurements in Estuaries; Comparison of the Respiration Rates of Stratified Mud Bottles with Plankton All data expressed as gm. oxygen/M2/day. Data from Oceanography Class, Duke University with Mary L. Sparlin!( and Edward McCoy PlankLon Dark Dark Bottles Mud Botlles Total Beaufort estuaries, N. C., 2-3 M deep. July 8, 1956 . 4.85 1.02 5.87a July 15, 1956 2.55 1.52 4.07a Aug. 20, 1956 0.55 2.64 3.19" Aug. 1, 1956 2.76 5.80b a Sum of plankton and mud boule estimates on an area of community basis. b Free water diurnal curve method used for total metabolism. 2-4 M REDFISH BAY 30-36 %0 o.e M SCIENCE FINE GRASS UPPER LAGUNA MADRE I Iii NANNOPLANKTON-CLAY Fie. 11. Map of marine waters in the vicinity of the Institute of Marine Science, Port Aransas Texas. Numbers are salinities during the period of measurement of this paper. tained with the diurnal curve measurement in the free water. Whereas neither the bell jars nor the mud bottles provide desirable accuracy, no other way to check the mud metabolism estimates of the diurnal curve method has yet been worked out. The sum of respiration of plankton and mud bottles is sufficiently large (Table 4) to confirm the important role of the bottom as indicated by the diurnal curve estimate of metabolism. Community Metabolism COMPARISON OF METABOLIC RATES IN TEXAS BAYS Measurements of gross production, community metabolism, plankton production, and plankton respiration were made in several seasons in the major Texas bays near Port Aransas as shown in Figure 11. The values for total metabolism computed from the diurnal graphs and from the dark-light bottle series are summarized in Table 2. The individual daily curves are not all reproduced, but each curve is based on 12 to 24 Winkler analyses. In most cases the corrected rate of change graphs have a hump shape like Figure 1, 2, 3, 4 and 6. In only two communities was a midday depression in total metabolism observed. These were communities with plankton bfooms (Fig. 5 and Rock­port Harbor, Aug. 5, 1957}. The shallow bays like Redfish Bay and the upper Laguna Madre are in general more productive than the deeper bays like Copano, Aransas, and Corpus Christi bays, (Figure 12}. The shallow bays in general maintain greater clarity with a large proportion of the metabolism on the bottom as indicated by the larger diurnal curve values in Table 2 relative to the simultaneous bottle measurements. Maximum production was 24 gm./M2 / day in a spring bloom in Copano Bay, but the highest sustained production was 7-18 gm./M2/day on the grass flats of Redfish Bay at a station where there is some influence of a treated sewage outfall. High as these figures are they are not as great as found in some clear, well circulated turtle grass at Long Key·, Florida with 34 gm./M2/day (Odum, 1957b}, at a coral reef at Eniwetok with 24 gm./M2/day, in Silver Springs in May with 30 gm./M2/day, or in polluted streams of · order of magnitude of 60 r;m./M2/day (See review Odum, 1956}. In the deeper and turbid bays, it is obvious that much of the available light is being absorbed by clay particles and is not reaching plant cells. The distribution of turbidity and turbulent stirring of the bottom oozes is a major factor in reducing potential productivity of Texas bays. There are possibilities for in­creasing productivity with procedures for clearing Texas bays. On any one day there may be a considerable excess of either production or respiration (Table 2 and Fig. 12} but there is no independent evidence yet to indicate that the bays are regions of export, storage, or organic matter import. At present the methods are not sufficiently accurate or the data replicated sufficiently in any one bay to determine if the community has an annual small net production or loss. Largest variations ( 1 to 24 gm./M2 /day range} from one time to another were ob­served in those bays with plankton populations like Copano Bay, Baffin Bay, and Aransas Bay. As might be expected photosynthesis of blooms is far more irregular than photo­synthesis by attached benthic plant communities. Even in Baffin Bay, benthic measure­ments equalled free water measurement on only one day (Fig. 12}. The salinities represented in Table 4 cover the tremendous range from fresh water to hypersaEne conditions of 79 parts per thousand. During this 1956-58 period salinity variation in single bays has been sharp and drastic. The general ability of the plant communities to maintain a considerable continuous primary productivity reflects the Comparative Studies on the Metabolism of Marine Waters 60 -----· 50 -----------· \ . / SALINITY %0 40 . ./~ 0100 30 ·-·~·----· 20 10 40 UPPER LAGUNA MADRE 30 GM1 2 M /DAY 20 10 5 .p (PLANKTON) J F M A M A S F M A M 1957 1958 MONTHS Fie. 12. For legend see next page. rapid capacity of such communities to survive or recover in such estuaries. Neither high nor low salinity provides a limit to basic plant photosynthesis. Some of the highest pro­duction values were found in the extreme salinities in which many consumers are ex· duded physiologically. Comparative Studies on the Metabolism of Marine Waters BAFFIN BAY, TEXAS, 1957 15 56-79 %0 METABOLISM 10 J REDFISH BAY, PORT ARANSAS, TEXAS, 1957 20 24-36 %0 10 5 p AND R OF BOTTL.ES J F M M J J A S 0 N D FIG. 12. Seasonal trend of metabolism in the shallow Upper Laguna Madre, Baffin Bay, and Red­fish Bay. P, gross primary production; R, community respiration. The lower curves are dark and light bottle measurements; the upper curves are the result of diurnal curve measurements. Comparative Studies on the Metabolism of Marine Waters SEASONAL TREND IN THE GRASSY UPPER LAGVNA MADRE AND PORT ARANSAS REDFISH BAy In order to gain some annual picture in one area, metabolism measurements were made over a year's period in a representative shallow water area of the upper Laguna Madre. Data on both total metabolism and plankton metabolism are plotted in Figure 12. As found by Odum, McConnell, and Abbott (1958 I with chlorophyll measurements, the producers include the plankton, the bottom algal ooze which often becomes sus­pended by wave action, and the fine bladed, sparse covering of Diplanthera wrightii. As the small values from bottles indicate, plankton metabolism is minor although small blooms occasionally occur as in l\ovember. 1957. !Fig. 121. Maximum production as well as maximum respiration occurred in late spring or early summer in both years. There is also a well known development of consumer popu­lations at this season due to the migration of larvae and rapid growth. Apparently these populations are closely geared to the gross community primary production although in any one day's measurement there is cons:derable range of the P /R ratio. In winter the migration of many of the dominant consumers such as mullet and shrimp into the Gulf for breeding is correlated with the minimum availability of pri· mary productiYe energy in the bays. Winter days are short and winter skies in Texas are markedly shaded by heavy cloud cowrs in many weeks. A study by Mr. Tom Hellier, Jr. is in progress in cooperation with the Texas Game and Fish Commission to determine the relat:onship of fish growth to the primary produc­tion in the Laguna. DIFFVSIOK CONSTANTS IK MARI:\E WATERS WITH S t:GGESTIO:\S FOR CoMPUTATIO:\ OF POLLUTION CAPACITY One of the by-products of the diurnal curve calculation is the gas transfer constant indicating the rate of oxygen diffusion into or out of the bay waters per unit saturation deficit gradient. The diffusion constants are in Table 4. There is a remarkable uniformity of diffusion constant with values between 0.3 and 3.8 gm./M2 /hr at 0 per cent satura­tion. The values below 1 were observed on nearly windless days in shallow bays where bottom friction was large and flow minimal. The values of 2 or above occurred in the deeper bays under windy conditions. These values are similar to constants previously re­ported for streams and other mixing waters !See review; Odum, 1956). The diffusion constant provides an estimate of the maximum excess of respiration over production which is permissible for a bay without conditions becoming anaerobic and producing fish kills. The diffusion of oxygen into a bay at the lethal oxygen content may be computed for a day by multiplying the diffusion constant per hour IKJ by the satura­tion deficit (SJ and by 24 hours. Oxygen added by photosynthesis is indicated by P. For a fish kill respiration IRJ must be greater than 24KS + P where Sis the saturation deficit considered lethal for desirable organisms. For example, in these bays K is usually about 1.0 gm./M2 /hr. at 0 per cent saturation or 24 gm./M2 /day. P is usually greater than 1.0 gm./M"/ day. If 10 per cent saturation 1S =90'/c J is lethal for fish, 24KS +P equals (24) (1.0) (0.9) + 1.0 or 22.6 gm. / M" oxygen added per day. For a fish kill Jl must exceed 23 gm./ M2/ day-a value well above usual values in Table 2. The 0 nlJ high rates of respiration were also accompanied by high rates of photosynthesis. Sucl computations indicate the maximum capacity of a bay to accept organ:c pollution that does not also stimulate photosynthesis. The capacity of Texas bay waters to reaerate by atmospheric diffusion is similar to that of moderate sized streams. It is markedly greater than the diffusion in enclosed stratified waters where diffusion is an order of magnitude less. The diurnal curve pro­vides means for practical management of oxygen budgets in Texas bays. HETEROTROPHIC FERTILITY AND THE UsE OF SATURATION DEFICIT TO CLASSIFY METABOLISM It is possible very quickly and without calculation to recognize environments that are receiving organic matter in large quantities and metabolizing it in excess of their own photosynthetic production. If the percent saturation remains mostly below 100 per cent during the 24 hours, the community has more respiration than photosynthesis. cent most of the day (Fig. 13) . If the percent saturation curve is above 100 per cent much of the time during 24 hours, the community is making an excess of organic matter over its respiration. The saturation deficit can thus be used to recognize any extensive organic matter import whether it is organic pollution or organic matter from natural sources. The type of metabolism domi­nated by imported organic matter was termed heterotrophic in a classification system previous I y presented (Odum, 1956) . However, small excesses of photosynthesis or respiration should not be considered as significant until some means for determining daytime respiration can be found. The very turbid waters in the low salinity bays of Texas are receiving organic matter from the land along with the silts. The respiratory metabolism of such bays as Tres Palacios, Aransas, and Mesquite (Table 2) exceeds photosynthesis. The percent satura· tion curve of Tres Palacios Bay on June 17, 1957, for example, remains below 100 per cent most of the day (Fig. 13). As Ohle (1956) has emphas'zed, communities of consumer organisms can be as abundant and a sy·stem just as fertile on organic Imports as on its own production. The hig'.1 values of respiration in some of the turbid Texas Bays suggest that there is at times as much consumer activity as in wme of the more productive bays. It has long been thought that organisms such as the oysters that dominate the turbid bays can survive e:ther on some kinds of algae or on some kinds of organic detritus. Menzel ( 1955) in his study of two species of oysters in the Texas bays near the Institute of Marine Science summarizes knowledge on oyster feeding without indicating the relative roles of live and dead foods in nature. The relative interchangeability of the two types of basic nutrition for consumers has yet to be adequately measured. DAILY COURSE OF EFFICIENCY OF PHOTOSYNTHESIS Diurnal rates of change provide an hour by hour chart of the photosynthesis of a community, which may be related to graphs of insolation in order to calculate efficiency. Considerable question has been raised as to the efficiency of photosynthesis at various times of the day (Pirson, 1957; Verduin, 1957). In Verduin's discussion of Silver Springs it was not recognized that oxygen curves in spring runs are the same as the oxygen rate of change curves (Odum, 1957a). Thus Verduin erroneously stated that morning efficiency was much greater than the afternoon efficiency in Silver Springs. Comparative Studies on the Metabolism of Marine Waters 8 • 7 ot PPM 6 5 6 12 18 L LIGHT \ 100 90 % SAT. 80 70 6 12 18 0.8 0.6 0.4 0.2 ot CHANGE o.o PPM/HR 0.2 0.4 0.6 0 .8 6 12 18 NOON TIME OF DAY IN HOURS Fie. 13. A diurnal oxygen curve from a bay which is receiving more organic matter than it is pro· ducin{!. Note the percent saturation curve below 100 percent throughout the period. TrespalaciosBay, June 17-18, 1957. Data were obtained with aid of William Guest, Texas Game and Fish Com· mission. In the lower graph the dashed line has been corrected for diffusion. Photosynthetic measurements are plotted as a function of light intensities in Figure 14 for one curve in Silver Springs and one typical curve for Redfish Bay. As might be ex· Fie. 14. Photosynthesis measurements of the whole community as a function of light intensity at rnrious hours of the day. The upper curve is plotted from a diurnal curve for Silver Springs (Odum,1957): The lower curve is from Redfish Bay, July 18, 1957. The diagonal straight lines are lines of constant efficiency for reference purposes. Comparative Studies on the Metabolism of Marine Waters 50 PERCENTAGE OF PHOTOSYNTHESIS AT NOON nls. THEORETICAL CONSIDERATIONS It is apparent that solar energy is the prime cause of diurnal change in pH either directly or indirectly. Both photosynthesis and respiration of marine organisms influence the carbonic acid equilibrium system in sea water. It therefore appears feasible to calcu­late the magnitude of both photosynthesis and respiration by observing pH change and other supporting field data, Wells (19221 measured carbon dioxide metabolism and related it to daily pH changes at Tortugas. Harvey (1955) has summarized various aspects of the correlation between the pH and carbonic acid sys~em in the sea, and Ryther (1956) has reviewed some of the problems with measurement of primary produc­tion by pH change. The buffer mechanism of sea water is quite complicated. However, the most sig­ nificant buffering action can be said to come from carbonic and boric acids and their salts which are present in the water. The remainder of the many weak acids in sea water may be considered to be insignificant. The magnitude of the contribution of boric acid and its salts to the buffer mechanism is approximately 5 per cent for oceanic water of a pH around 8.2 and a higher per cent for higher pH values. It is therefore necessary to consider boric acid when one computes the magnitude of primary produc­ tion in the sea by diurnal pH change. A brief review of the relationship between pH and carbonic acid system in the sea follows. From the first and second dissociation equations of carbonic acid the following re· lations are obtained: CHC03" pH = pK1 +log (1) CH co 2 3 Cco3-­ pH= pK2 +log (2) CHCOe" where capital C denotes molar concentration, and pK1 and pK2 denote the negative logarithm of the first and second dissociation constants of carbonic acid. Assuming pK1 and pK2 are constant, the pH changes observed are due to changes in the ratio of bicarbonate and carbonate ions, and/ or bicarbonate ion and carbonic acid. Correction for boric acid contribution on pH changes must be made. Therefore, any change of the relative concentration of carbonic acid components, H2C03, HC03-, and C03--, due to photosynthesis and respiration of aquatic organisms, tends to change the pH of the water. The dissociation constants of both carbonic and boric acids in sea water are given by Buch (1951) and Lyman (1957) as functions of temperature and chlorinity. Lyman's constants are used in this paper. His apparent first and second dissociation constants for carbonic acid are defined as: K:=-----­ (3) A(H2C00 + C02) Aw Cco3 -­ K~=-----­ (4) where A denotes active concentration. Solar energy also changes the temperature of the water thus affecting pH. Buch and Nyniis (1939) studied the relationship between temperature and pH in a closed system. They found that pH falls with rising temperature at a magnitude dependent upon the pH and salinity of the water. At a chlorinity of 19.5%0 , a temperature range of 20--30° C, and a pH of 8.2, the pH falls 0.01 unit per each 1° Crise in temperature. If one could assume environmental changes are restricted to temperature changes, one would expect the minimum pH to occur at the time of highest water temperature. Hence the minimum pH would be during daylight hours when the water temperature is max· imum, and maximum during night when the water temperature is lowest. This trend is opposite to the pH changes which one would expect from biological activity. If diffusion due to changes in solubility of molecular C02 is unimportant. a pH change caused by temperature change does not affect the total carbon dioxide content of the water. In computations of the concentration of carbon dioxide components in the en­vironment, temperature errors are minor since the constants used are corrected for temperature. In an environment in which calcium carbonate precipitation and dissolution are active, pH change due to biological C02 removal may be obscured unless a precise knowledge of the quantity of precipitate is obtainable since calcium carbonate affects buffering action. Another complication is the diffusion exchange of carbon dioxide across the air-water interface. The diffusional problem needs much more detailed investigation. STATIONS Four stations were selected for this study. The productivity of these stations was studied by Odum and Hoskin ( 1958) by means of the diurnal oxygen method. The locations of the stations are given in Figure 1. All stations were representative of large areas of the shallow bays, ranging from 1 foot to 6 feet in depth. Diurnal pH Variations in Texas Bays F1G. 1. Locations of stations occupied in diurnal pH measurements in Texas bays: Rockport Basin, 1: Red fish Bay station, 2; Laguna Madre station, 3; Baffin Bay station, 4. Redfish Bay was the shallowest of the bays studied (approximately 1 foot deep) and had the heaviest growth of benthic "grasses" (Thalassia and Ruppia). The "grasses" emerged at low tide. The water mass at the station was thought to be relatively stationary. The Laguna Madre station was midway between Pita Island and the Texas mainland in about 2 feet of water. A fairly heavy growth of Diplanthera "grass" existed. The Baffin Bay area was about 5 feet in depth. The water was turbid with a visibility of not over 2 feet. The chlorinity of the water exceeded 30%0 • Diurnal pH Variations in Texas Bays The Rockport station was in the Rockport Yacht Basin in water about 6 feet deep. During the investigation the water surfaces became intermittently covered w:th oil slicks. METHODS Measurements of pH were made with a Beckman Model G-S pH meter. The meter was enclosed, with desiccant, in a transparent polyethylene bag to exclude salt air. Both the glass and calomel electrodes were the general purpose extension type used with all lab­oratory models of Beckman Instruments. The electrodes were inserted through a No. 11 rubber stopper fitted to a glass container in wh'.ch the pH of water samples was measured. Standard buffer solution of pH 7.00 was kept at the water temperature by parfally im­mersing the container in the sea water, thus eliminating the temperature correction be­tween the water and buffer solution. Water samples were either taken by filling the bottles at the appropriate depth by hand or measurements were taken by inserting the electrodes directly into the water. In the former case the pH was measured as quickly as possible and without exposure to air. Although the Beckman Model G-S pH meter has been found sensitive to 0.003 pH units, with the possibility of greater field errors, the pH value was recorded to 0.01 unit. In order to obtain carbonate alkalinity of the water samples both titration alkalinity and total inorganic boron analyses were made on samples collected at three-hour inter­vals. Boron samples were frozen in dry ice until the analyses were made. The titration alkalinity was determined according to the method of Wattenberg (1933) using a pH of 6.0 at the end point. The total inorganic boron was measured by the method of Noakes (1959) . Results and Discussion The field data obtained at the four stations are presented in Figures 2 through 5 and summarized in Table 2. TABLE 2 Observed Field Data During the Investigation of Diurnal pH Variation in Texas Coastal Waters Titration Alkalinit"· Total In - Location Date (1957} pH Range t,,.pH Temp. Range (' C.) Cl%o (MilliequiYalent/l) orf_l'.anic Boron Range A•·ernge (Millimol/ l) Baffin Bay July 26-27 8.20--8.34 0.14 29.0-32.0 31.8 4.01-4.13 4.09 0.90 Aug. 15--16 7.94-8.35 0.41 29.0--32.0 36.l 4.34-4.39 4.37 1.10 Dec. 22-23 8.52-8.62 0.10 19.2-21.2 34.9 3.69-3.80 3.74 0.87 Redfish Bay July 18-19 8.23-8.91 0.68 28.0--35.0 16.6 2.45--2.65 2.53 0.37 Dec. 25--26 7.98-8.41 0.43 18.0--23.0 15.7 2.92-3.01 2.95 0.61 Laguna Madre Aug. 1-2 8.15--8.42 0.27 27.4-32.0 31.6 2.83-2.95 2.86 0.92 Dec. 28-29 8.21-8.49 0.28 14.4-17.5 17.6 2.82-2.88 2.85 0.40 Rockport Aug. 5-6 8.37-8.44 O.Q7 31.0--32.0 18.9 2.82-3.02 2.88 0.64 From the analyses it was observed that titration alkalinity and inorganic boron did not show a regular diurnal cycle but were about the same throughout the 24-hour period. Chlorinity differed quite markedly from station to station ranging from approximately 16%o at Redfish Bay to 36%0 at Baffin Bay. In comparison with other stations Redfish Bay showed the largest diurnal variation in Diurnal pH Variations in Texas Bays 12 II 10 9 8 7 o, 6 mg/ I 4 3 2 OOAM 35•c 25°C &90 8.00 SUMMER (JULY 18-19,57) 8.70 WINTER (DEC. 25-26, 57J SUMMER (JULY 18­19, 57) pH BAO A I I I I ~ \ 6AM NOON 6PM 12PM 8.30 J' /" I \ I I ~ I WINTER I \ 1 lDEC.25-26,57)\I I ,/'"',,, I:,; I \ \ / ....,,, ~----o.., 8.0 I -.! 6AM NOON 6PM 12PM 1s•c._____.______,_____,_____, OAM 6AM NOON 6PM 12PM Fie. 2. Diurnal oxygen, pH, and temperature rnriation at Redfish Bay, Texas. Solid lines, summer data for July 18-19, 1957. Dotted lines, winter data for Dec. 25-26, 1957. dissolved oxygen, temperature and pH (Fig. 2). The values of pH and temperature from this station in summer were higher and showed a greater diurnal range than in the winter. However, in the Laguna Madre a higher pH was observed in the winter than in the summer. Three sets of diurnal pH values were obtained in Baffin Bay (Fig. 4). Winter pH values were higher than the summer, but diurnal changes were less. The two summer stations showed a marked difference in the magnitude of diurnal pH change although the variation of oxygen and temperature were almost the same. The Rockport station (Fig. 5) was occupied after a dinoflagellate bloom. This station had the least diurnal variation in dissolved oxygen, pH and temperature of the locations studied. A maximum was observed at 1800 hrs. which was quite different from that at other stations. The amplitude of diurnal oxygen and pH was less in the deeper environ· ments where the effects of metabolism were dispersed in a greater volume of water. DIURNAL VARIATION OF TOTAL CARBON DIOXIDE Using the data for pH and carbonate alkalinity, the total carbon dioxide components, Ceo"--, CHco"-, CH~co,J and Cco2 can be calculated at any given time of the day. The equa­tions for this computation are summarized as follows: AH+ CHco -= Carbonate Alkalinity gm.-ion/ L (5) 3 2 K'2 +Aw Diurnal pH Variations in Texas Bays o, 7 mQ/I 12PM 8.50 otP..', ' "' / °""--..a... 8.40 I ' p ', pH "-, WINTER I ---..l._~C. 28-29,57) ,) --; p' ' ' SUMMER \ >~ • (AUG. 1·2,57) b'. 8.2 8.IOL-----'-----'-----'-----' OAM 6AM NOON 6PM 12PM 35'C 25'C 20"C IO'COAM 6AM NOON 6PM 12PM F1G. 3. Diurnal oxygen, pH, and temperature rnnat10n near Pita Island, Laguna Madre, Texas. Solid lines, summer data for Aug. 1-2, 1957. Dotted lines, winter data for Dec. 28-29, 1957. K'2 Ceo · · = Carbonate Alkalinity------gm.-ion L (6) 3 • 2 K'2 +Aw C(co + H col= ------gm.-Mole -'L (7) 2 2 3 K' 1 Let the total components of the carbon-dioxide system he repre;;;ented by L C02 : ~C02 = Cco3· + CHco-3+ c(C02 + H;C03\ Then: A + +K' ( l + ~~+ ) ~C02 = Carbonate Alkalinity H 2 1 gm.-:Mole L (8) 2 K 1 2 +Aw 54 Diurnal pH Variatwns in Texas Bays l ~ l (::! l (::! ~ ­z (/) :::> C> :::> ­:; ...... "'"' "' : "' ~ = CE .... ~ ~ ~ ~ = ~ ... .... =­"' 0 c;; = ;.. .... "' "' .. c\"go -. "' N - :E ,_ 0 (/) U 2.4 " ~ 2. "' " 0 u 2 .350 0:: "-' ()AM GAM l'()QN GPM 12PM ::> TIME UJ u -' 0 ::;; f"rc. 7. Diurnal computations for Redfah Bay, Dec. 25-26, 1957. See legend for Figure 6. 020 ...J :::; 016 0 ::t" 012 3 inn---­ oO OAM 6AM 6PM 12PM 224 z 0 ~ 160 !!i 140 ~ ~ "' c! u 0:: " ...J ::> u DAM 6AM NOON TIME 6PM 12PM UJ ...J 0 :E 40 FIG. 8. Diurnal computations for Laguna l\Iadre, Aug. 1-2, 1957. See legend for Figure 6. Diurnal pH Variations in Texas Bays TABLE 3 Comparison of Gross Primary Production and Respiration Estimations by O, and pH Measurements in Highly Productive Texas Coastal Bays Grnss Primary Production Respiralion Eslimation 0., Method CO, Method PQ 0, Method CO, Method RQ Location (Date (1957) mM/L/day mM/L/day o,;co, mM/L/day mM/L/day o,;co, Baffin Bay July 26-27 0.27 0.31 0.9 0.48 0.31 1.5 Aug. 15-16 0.24 0.90 0.3 0.30 0.91 0.3 Dec. 22-23 0.16 0.19 0.8 0.23 0.29 0.8 Redfish Bay July 18-19 0.88 0.92 1.0 0.96 1.00 1.0 Dec. 25-26 0.43 0.68 0.6 0.75 0.49 1.5 Laguna Madre Aug. 1-2 0.25 0.56 0.4 0.41 0.58 0.7 Dec. 28-29 0.17 0.27 0.6 0.15 0.25 0.6 Rockport Basin Aug. 5-6 0.13 0.12 1.0 0.24 0.12 2.0 0.38 ...J ' ...J 0.34 0 :i; :::; ...J ,. o" Q22 OAM 6PM t2PM z 2.58 0 ~ t80[ 2.54 ~ ...J ::::> 160 0 2.50 :i; ~ :::; ...J ,. ~ 0 u r! 2.42 u ~ ...J .. 2.38 ::::> u 2.34 "' cl :i; 60 OAM 6AM NOON 6PM 12PM Fie. 9. Diurnal computations for Laguna Madre, Dec. 28-29, 1957. See legend for Figure 6. Q06 ~ 0.04 ~ .0.02 ·. _ __ _____ _r.:11,_._,__ g o.oo Ffu"'··,·-._-*:-;-;-;---,,:;t,,7F' :::; ~-0.02 8-004 ~-Q06 3.200 :J 3.10 0 :i; :::; ...J ,. 3 0 u ... 2.900 OAM 6AM NOON 6PM 12PM Fie. 10. Diurnal computations for Baffin Bay, July 26-27, 1957. See legend for Figure 6. 010 0.08 0.150 ~ 006 0200~ 0100 ~ Q04 0.050 ~ 002 ::; 0 i QAM 6AM NOON 6PM 12PM QOO TIME 0-002 0 -0.04 .. <1-0.06 -0.10 t'­ 0 0 a: ...J "' a .., ...J 0 :::e FIG. 11. Diurnal computations for Baffin Bay, Aug. 15-16, 1957. See legend for Figure 6. .: O.o6 ~ 00 ...J ~ 3 :i rS-0.02 0 ~~ ::=:j .., O~AM~~--,,J6A~M~~~NO~ON=c-~--,6~P~M,---~=:!.~PM ~-004 TIME 2£00 ~ ...J ....5 2. :::e ::; ...J :E 2.400 :."""-_~/ !;;a: ~ t'­~ 140 120 10 80 6AM NOON TIME 6PM c! <.>.. OAM 6AM NOON TIME 6PM 12PM ~ ...J :> 0... ...J 0 :::e 60 4 20 FIG. 12. Diurnal computations for Baffin Bay, Dec. 22-23, 1957. See legend for Figure 6. and Smith (1958) show that many marine plants tend to utilize bicarbonate ion directly as a source of carbon in photosynthesis. EFFECT OF COLD FRONT AND HEAVY SHOWERS ON TOTAL CARBON DIOXIDE CONTENT At the winter station on Redfish Bay (Figs. 2 and 7) relatively heavy rainfall fell accompanied by a cold front between 1600 and 1900 Central Standard Time Decem­ber 25, 1957. During a period of 3 hours the pH of the water dropped abruptly from Diurnal pH Variations in Texas Bays _,022~ go.ta _, 0.14 _, ~ 0'0.10 OA~M~~-6~A~M~~~NOO""=N~~-6~PM'=c-~~1~2PM TIME i5 ii a:: :::> ~ 120 6AM NOON 6PM el~t: . ~· ~ Cr u a:: 10 80 60 .,.~rs .. 0 2.16 u H DAM I 6AM NOON 6PM 12PM u w_, 0 " 40 TIME fie;. 13. Diurnal computations for Rockport Basin, Aug. 5-6, 1957. See legend for Figure 6. 2.00 ~ _J '__J 0 2 __J __J ~ Fie;. 14. Change in the components of the carbon-dioxide system based on data from Redfish Bay,Texas, July 18-19, 1957. 8.41 to 8.03 and the temperature from 22.4° C to 20.4° C. These changes are more rapid than those normally observed due to a decrease in light intensity and appear to be caused by increased diffusion rates and/ or by addition of CO, rich fresh water through rainfall. The rain storm greatly increased surface agitation and mixing rates, resulting in increased carbon dioxide invasion of the bay waters. The over-all effect of the rain storm may be estimated from the data shown in Figure 15. The total CO, content of the water was increased by approximately 0.33 mM/ L in 3 hours during rainfall. The photosynthesis during this period was minimal because of 2.aoo ...J ' EXTRAPORATED 0 :;: RAIN EFFECT :J 2,600 ...J / / 8 N 2,500 / / i:..;i / / 2,400 RAINING 6AM NOON 6PM TIME Fie. 15. Estimation of the effect of rain by extrapolation of data from Figure 7. the low light intensity. Assuming the respiration had the similar tendency to that shown in Figures 6 and 9, it seems reasonable to draw a relatively straight line from 1600, the start of the rain, to 0600 next morning. From the extrapolation a correction value of approximately 0.08 mM C02 was obtained. The net change of C02 due to the storm was 0.25 mM/ L/ 3 hours or 0.08 mM/L/hr. It is interesting that similar changes have been observed with respect to dissolved molecular carbon dioxide in the Caribbean during the summer of 1958 (lbert and Hood, 1958). PHOTOSYNTHETIC AND RESPIRATORY QUOTIENTS Both photosynthetic (PQ=02/C02 by molecules and respiratory quotients (RQ= 0 2/C02 by molecules), based on the field data obtained, cover a wide range of values. PQ, from 0.3 to 1.0, and RQ, of 0.3 to 2.0, were observed (Table 3). Ryther (1956) reviewed and summarized the existing measurements of PQ in plank­ton reporting values which are rather close to unity, as expected for the synthesis of carbohydrates. Calculations of PQ can be made from the chemical compositions of aquatic plants (by the relative ratio to carbon, oxygen, and hydrogen). The calculated PQ values range from 1.09 to 1.48 with an average of about 1.20. When the plants use ammonia as their nitrogen source, the PQ is calculated to be about 1.10 and experiment­ally was found to be 1.06; with nitrate-nitrogen the PQ values calculated and measured were 1.45 and 1.47 respectively (Cramer and Myers, 1948). Contrary to the above expectations for single populations, the field data obtained in this study showed much lower PQ values. The reason for this anomaly is not clear. Low values were also found in some Florida springs by Odum (1957b). Apart from un­avoidable experimental errors in the fie!d a number of explanations are possible. The role of bacteria may significantly affect the PQ ratio since 107 bacteria/ ml. were found by Carl Oppenheimer in the waters of the bays studied. The bacteria may be consuming C02 without releasing oxygen. For example, photosynthetic anaerobic bacteria, grow­ing at the sunlit interface between mud and water, consume hydrogen sulfide and co~ during their photosynthetic activities and produce water instead of oxygen. There also may be diurnal changes in the kind of metabolism which may produce odd overall quo­tients. The correction of daytime photosynthesis with nighttime respiration may be less correct for one gas than for the other. Another possible source of the PQ discrepancy may be the alkalinity difference be­tween the shallow bay environments and oceanic waters; in some cases the chlorinity of the bay waters exceeded 3()%0 • In Figure 16 the correlation between chlorinity and :...1 ~ z 6.00 w ...J g ::::l 0 5.00 w /" :J / ...J Ba-L\ /-{ ~ >-. )/ 'aa-2 ~400 / ~ z Lo-2 / Bo-3 :J ru; z l&J Q 600 620 640 660 680 700 WAVE LENGTH M~ Fie. 1. Absorption spectra of liYing and decomposing aquatic plants. Beckman DU Spectropho­tometer at slit width 0.0.t mm. The peak optical density of each cun·e is adjusted to optical density 0.5 for comparison. The Chlorophyll "A" of Communities WATER SURFACE 0.01 CHLOROPHYLL "A" OOSE OOZE DEPTH OF EUPHOTIC Z 5 50 MICROGRAMS PER ML PERCENT OF SURFACE LIGHT Fie. 2. Vertical stratification of chlorophyll "A'' and light intensity (Dec. 29, 1957) in the shallow 'rnter and ooze of the upper Laguna Madre, Reds Fish Camp, Texas. TRANSMISSION IN SPARTINA--~ IN I CM TOP MUD 2 CHLOROPHYLL "A" MICROGRAMS PER ML 25 Fie. 3. \'ertical stratification of chlorophyll and light intensity in the marsh grass, Spartina sp. and the mud surface of Aransas Bay, Texas, March 21, 1958. Data on light intensity and chlorophyll "A" in the shallow, grassy, bay bottom of the upper Laguna Madre of Texas and in an area of sparse marsh grass near Port Aransas are given in Figures 2 and 3. The depth of the euphotic zone was measured as about 1 cm. deep into the bottom ooze. A submarine photometer was pressed into the ooze in order to determine the depth of penetration of effective light intensity. After collecting a core in a plast:c pipe pressed into the mud, the core was frozen and extruded from the pip~. Then one cm. sections of the top were cut off for chlorophyll extraction. There are Yery large amounts of non-functional chlorophyll below the euphotic zone in the mud ( r·g. 2 I. These deep layers are not assumed, a priori, as dead and incapable of photo­synthesis, but one can assume that if no appreciable light penetrates, there is no effective production. One must always consider the possibility of vertically migrating sediment algae, that migrate below the euphotic zone for part of the day (Aleem, 1950). Non living chlorophyll is a usual constituent of freshwater and marine sediments ( Vallen­tyne, 1955). Large amounts of non-living chlorophyll also occur below the euphotic zone in deep layers of marine and freshwater bodies. CHLOROPHYLL "A" DETERMINATION FROM OPTICAL DENSITY MEASUREMENTS AT ONE WAVE LENGTH The measurement of the optical density of chlorophyll in acetone extracts at three wavelengths is time consuming where large series are needed for meaningful averages. Comparisons of results based on measurements at 665 millicrons were made w:th results from measurements of three wavelengths. Widely different plant materials including blue-greens and higher plants were extracted and measured in the Beckman DU spectro­photometer. Each extract was measured at 3 wave lengths and the values computed as Chlorophyll "A" plotted against the optical density determined at 665 millimicrons (Fig. 4). The close alignment of points indicates that the 3 wave length procedure is unneces- DoI 0 0 F .... L ::E z 0 c. "' ::E 14.3 tO :r 0 \::•n 1-. "' 0 It'. < :r u a: 5 0.5 10 OPTICAL DENSITY 663-667 a.u.t ~... BECKMAN DU SPECTROPHOTOMETER Fie. 4. Chlorophyll "A" concentration determined by the Richards-Thompson method using read­ings at 3 wavelengths as a function of optical density at one wavelength (665 mill:microns). Beckman DU Spectrophotomete", slit width, 0.04 mµ.. U, Uirn faschlla ; L, Lynbya sp.; G , Geplwrtia mitche'ae: 0, Oscillatoria sp. ; Dr, fresh Diplanthera wrightii blades; D,, decomposed Diplanthera blades: S, decomposed sedimentary material containing blue green algal mats. sary considering tht> other greater sources of error. Where optical densities are read with a 1 cm. optical light path in a narrow band pass spectrophotometer, the Richards and Thompson Chlorophyll "A" procedure is abbreviated in equation (2) as follows: Chlorophyll A in mg./ l = 14.3 cl,.,.,, (2) CHLOROPHYLL "A" DETERJ\!IN.-\TIOX ON A PORTABLE SPECTROPHOTOMETER WITH BROAD BAND PASS The Bausch and Lomb "Spectronic 20" spectrophotometer has been found useful for dett>rminations of chlorophyll because of portability, reliability, and relatively low eost ($240) permitting semi-field use. The width of the spectral band pass is given by the manufacturer as 20 millimicrons. Thus when the wave length is set on 665 milli­microns, light between 65:) and 673 millimicrons is passing into the extract tubes. Thus the range of absorption is broad enough that the amount of Chlorophyll "B" present can affect Chlorophyll "A" determinations. Chlorophyll "B" has an absorption peak at about 645 millimicrons not far from the 653 millimicron light. A series of measurements was made on both the Beckman narrow band pass instru­ment and on the broad band pass Bausch and Lomb instrument. In Figure 5 the Chlorophyll "A" figures with the Richards-Thompson method are plotted against the optical density readings at the single 665 millimicron setting on the Bausch and Lomb instrument. Data for a red, green and blue-green alga differ considerably. Readings on broad band pass instruments differ from narrow band pass measurements even when there is but one pigment present. The average per cent absorption over the 20 millimi­cron range must be on the average less than the per cent absorption in a 5 millimicron band pass located on the sharp peak of absorption. Thus when the Bausch and Lomb instrument is used for field use it is desirable to convert optical density measurements into Chlorophyll "A" values with curves like those in Figure 5. A calibration graph determined for higher plants should not, for {·xample, be used to compute Chlorophyll "A" values in extracts from red algae. In the chlorophyll procedure of Hogetsu and Ichimura (1954) chlorophyll is con­verted into phaeophytin with acid. In the chlorophyll procedure of Creitz and Richards (1955) magnesium carbonate is used to prevent the decomposition of chlorophyll into substances like phaeophytin. In these considerations a shift or possible shift of the red peak of chlorophyll is involved. If a wide band spectrophotometer is used all of the compounds with peaks in this area are measured, and the conditions of handling may not be so crucial. For field use in a wide variety of types of plant material, simplicity in chemical procedure is essential. Where the largest errors are likely to be ones of sampling, rapid methods which permit replications are desirable. Since the roles of Chlorophyll "B" and "C" in photosynthesis are not adequately known, it is not possible to state whether or not plants with much Chlorophyll "B" or "C" will have a different assimilation number relative to Chlorophyll "A" in compari­son to plants without these pigments. Assimilation numbers in this paper are computed relative to Cholorphyll "A" only. NOTES ON EXTRACTION The extraction of Chlorophyll "A" from widely divergent materials for rcological purposrs, requirrs a variety of extraction methods. Large, leafy plant itssues are passed through a meat grinder, slurrird in water, aliquotrd, partially dried, and extracted with CHLOROPK'fLL A RICHARDS­THOMPSON METHOD 5 ~ L. 0.5 OPTICAL DENSI TY B a L METHOD 653-673 Ill.I Fie. 5. Chlorophyll "A". determined with the Richards and Thompson method on the Beckman DU spectrophotometer as a function of optical density at 663 millimicrons and 1.1i' cm. light path on a Bausch and Lomb "spectronic 20" spectrophotometer. Um fasciata, a green alga: Gephortia mitchelae,a red alga: and L)·ngbya major, a blue-green alga. 90 per cent acetone. Discs from rain forest leaves were punched, bruised. placed in 10 cc. acetone (90'·() in vials, and extracted completely in 24 hours in a refrigerator without further grinding. Some plankton samples on millipore filters extract readily without mechanical damage to cells t Creitz and Richards. 1955). Other algae require grinding in mortar and pestle with clean sand. Corals and calcareous algae were cut with hacksaw and ground in a mortar. Rocks in streams were extracted by immersion in pans of ac::-tone accompanied by brushing by McConnell I 1958) and in class sampling work at Duke in 1956. Freeze drying of leaves was effective where presen·ation was necessary and where greatest accuracy was desirable. Some algae are sufficiently bruised in a Waring blendor for subsequent extraction. Others resist extraction. Species of Prasiola, Cfra, Cladophora, Tlzalassia, Diplantlzera, Phormidiam and Enteromorpha were especially resistant. With larger tissues, observa­tion of the residual color may be used to ascertain if most of the extraction has taken place. With tiny plankton cells on a millipore filter, the lack of extraction may be readily overlooked. In one extraction invoking a species of Clzlorella, only a small fraction of chlorophyll was removed from the filter residue until grinding was done with sand. Extraction is made with 90 per cent acetone. If 100 per cent acetone is used to dis­solve millipore filters, a part of the filter goes into solution which precipitates as a milky turbidity later when the solution is adjusted with water to 90 per cent acetone. Ideally The Chlorophyll "A" of Communities final extracts are in 90 per cent acetone. Final stages of extraction are usually carried on for 24 hours in a refrigerator in the dark. VALIDITY OF R1ctt..\RDS A:"D Ttto~Pso:-;'s PRoCEDt:RE FOR 90 PER CE:"T AcETO:SE In deyeJoping the chlorophyll assay method 1equation 1 in this paper) Richards with Thompson I 1952 J cite the work of :'.\facKinney I 19411 and Zscheile, Comar, and Mac­Kinney I 1942 I and adopt extinction coefficients for use in 90 per cent acetone solutions. Richards with Thompson 119521 prepared Chlorophyll "A" from Macrocystis with chromatographic methods and used the coefficients of the preYious authors for deter­mining the grams of chlorophyll represented by the extraction. MacKinney made meas­urements at 663 millimicrons whereas Richards used his new curve to determine ah­sorbancy at 665 millimicron~. In actuality. the extinction coefficients reported by Zscheile and co-workers and :\IacKinney are for 80 per cent acetone solutions of pig­ments. The effects of changes of soh-ent on the spectrophotometric properties of plant pigments ha,·e been widely discussed elsewhere 1Franck and Loomis, 1950; Rahino­witch. 19511. A series of experiments was run by \'\"alter Abbott to test the effect on absorbance of a change of soh-ent from 80 per cent to 90 per cent with pigments from terrestrial plants. (This was not a test of extractability, but a test of the basis of Richards and Thompson's equation. -, Small portions of fresh leaves from a rnriety of plants were extracted with 10 ml. of neutral. redistilled reagent grade acetone. Extraction was carried out in the dark at 5° C. When extraction was completed, two -1 ml. aliquots were withdrawn. With a micro­burette, these were adjusted to 80 per cent and 90 per cent acetone contents by addition of distilled water. Correction for nonideal beha,·ior of acetone water mixtures was made from an empirically determined chart of rnlume relationships. Samples were then cooled and stored briefly in the dark until spectra could be scanned on the Beckman DK recording spectrophotometer. Data from these measurements are in Table ]. TABLE 1 Ah5orhance Yalu~ (Optical Den5ity in 1.0 cm. Light Path I Species. 663 mu 80<( 90<( 645 mu au<;< 90~ 630 mu 80% 90% Catalpa bignoniodes 0.600 0.6-19 0.26-t 0.279 0.170 0.166 Quercus phellos O..J..H 0.568 0.-tl-t 0.581 0.199 0.258 0.219 0.268 0.113 0.145 0.124 0.147 Rhus glabra 0.-110 0.66-t 0.-13-t 0.718 0.116 0.303 0.189 o. 332 0.100 0.167 0.103 0.189 Celtis occidentalis 0.538 0.952 0.56-t 0.956 0.2.J..I 0.421 0252 O..J..17 0.140 0.248 0.140 0.228 [°/mus americana 0.286 0.2.:.2 0.302 0..356 0.131 0.133 0.135 0.15-t 0.082 0.084 0.082 0.088 Quercus r:irginiuna 0.890 0.972 0.333 0.3.J..I 0.189 0.188 l uniperus m.exicanus 0.366 0.502 0.16-t 0207 0.085 0.117 Cereis canadensis 0.511 0.558 0.175 0.209 0.082 0.101 Cnidentified 5eedling 0.568 0.581 0.227 0.226 0.110 0.107 The Chlorophyll "A" of Communities Accompanying the change of solvent from 90 per cent acetone is a 6 per cent lower­ing of absorbance at 663 millimicrons and at 645 millimicrons. Data taken by the Rich­ards and Thompson method should be corrected by multiplying absorbances at 663 millimicrons and 645 millimicrons by 0.94. Equation (2) above thus modified becomes equation (3) below: Chlorophyll "A" in mg/ l of 90% acetone = 13.4 d665 ( 3) With the Bausch and Lomb broad band pass instrument, however, the composition of the extract is not so important since part of the solvent effect is a change in wave length of the peak of the spectrum. McConnell (1958) found relatively minor effect of changing water composition. Steady State Energetics of Input-Output Systems in Which the Input Energy Arrives as Packets of Fixed Energy Content; The Cannon-Ball Catcher In order to better understand the energetic coupling of biochemical machines to the steady inflow of photons of light, it is convenient to describe the cannon-ball catcher (Fig. 6). Although the machine is an analogy, its energy aspects are thought to be similar to those of primary production systems. It is possible to erect hypotheses which account for the behavior of input-output systems containing chlorophyll with respect to light, temperature, nutrients, and other factors. That the behavior of chlorophyll adap­tation seems logical when compared with a cannon-ball catcher is at least an aid to the memory. In the discussion that follows the behavior of chlorophyll is related to mech­anisms of self regulation believed to exist in plants at a higher level of organization than single reactions of photochemistry and photosynthesis. The cannon-ball catcher machine is a coupled input-output system like those described in a previous paper (Odum and Pinkerton, 1955) . Energy arriving in the form of cannon-balls is caught by cups on a turning wheel of the cannon-ball catcher. The input energy system thus receives a steady flow of energy in discontinuous similar packages. After catching a ball, a receptor to be used again must be revolved through a cyclic process requiring a discrete time. The input system is geared by a pulley system and a fly wheel to a machine for doing work. Therefore, a discontinuous energy input drives an output that may have a more regular power distribution with time. A review of the possible kinds of biochemical reactions by which such input-output systems might function is given by Strehler (1958). There are several similarities between a cannon-ball catcher and a photosynthetic machine. A plant receives discontinuous photons just as the catcher receives cannon­balls, and is geared to a chemical machine that produces organic, synthetic work. The number of cups per machine is comparable to the concentration of chlorophyll per chloroplast. A cyclic biochemical process is necessary for maintaining chlorophyll "A" in receptive state (see flashing light research: Rabinowitch, 1956: 1433) just as the cups on the wheel must be revolved for reuse. As discussed previously (Odum and Pinkerton, 1955) maximum power output is gained with a particular and optimum force adjustment between input and output systems. It is likely that plants contain mechanisms for maintaining optimum adjustments, and these are established by natural selection. Consider a situation with a steady influx of cannon balls. The steady torque developed The Chlorophyll "A" of Communities by the impinging balls can be varied by changing the number of cups on the catcher wheel. With more cups more balls are caught. In this way the input drive can be adapted for a given output load so as to provide the maximum power. The rest of the cannon-balls can pass uncaught to be accepted by another catcher machine below. Similarly in photosynthesis one can imagine that the number of receptive chlorophyll molecules provided can be controlled in order to maintain the ideal photochemical drive to the respiratory system. Changes in light or in factors affecting respiratory system rates are likely to be accompanied by changes in chlorophyll in order to maintain the adjustment within the chloroplast. Such adjustment prohibits acceptance of too many or too few photons. Conservation of photons not only maintains proper load ratios in the immediate biochemical system but contributes to the maximum output of cells just below. ~ 19 CANNON-BALLS c \ LLI z 0 N 0 l­ o ~ Q. :::> LLI '> FIG. 6. Three cannon-ball catcher machines in vertical sequence. The upper machine is adapted !o the area of great flux of cannon-balls because relatively few cups are protruding; the lower machine JS adapted to the area of few cannon-balls with a large number of cups in position. According to the theory of energy adaptation, outlined in the text, successfully surviving photo­synthetic systems adapted in nature have mechanisms to change number of cups in order to adapt' to the rate of energy influx and so as to develop maximum power output of the entire population throughout the euphotic zone. The Chlorophyll "A" of Communities The total yield of the whole euphotic zone is thus higher. Maximum yields are possible if a multi-layered reception system is active. If the theory of optimum input-output adjustment of chlorophyll is correct, then the chlorophyll fluorescence as an input energy flow meter (Katz, 1949) is likely to be maintained in fairly constant ratio to photosynthesis under adapted circumstances. A correlation of the fluorescence of live chloroplasts and photosynthesis does exist (Rabi­nowitch, 1951: 819; Brugger, 1957). The fluorescence of living chlorophyll systems may have some analogous similarity to the loss of energy that occurs when a cannon-ball strikes the cups of the catcher machine. Both phenomena are likely to be accentuated when the cycles of the receptive material l cups or chlorophyll) are being retarded and are not dispersing energy at optimum speed. Katz (1949) found increased fluorescence with artificially retarded photosynthesis. Discussion of chlorophyll, light, and other factors follow with reference to energetics and input-output systems. The purpose of the cannon-ball picturization is to make clear the hypothesis that photosynthetic systems 'n order to maintain maximum power output can regulate thf' input power by regulating the chlorophyll. Summary of Ecological Factors Affecting Chlorophyll and Assimilation Number Reports from the literature on the spatial and temporal patterns of chlorophyll m natural communities are not uniform. Nor do the data from the experimental litera­ture at first suggest simple relationships between environmental factors, chlorophyll, and assimilation. In the following paragraphs an attempt is made to summarize and ac­count for the adaptive behavior of systems containing chlorophyll. LIGHT INTENSITY AND THE CHLOROPHYLL CONCENTRATIONS IN A Part OF A COMMUNITY Consider only plants growing under conditions of illumination to which they are adapted for effective competition. Such adaptation may be in the laboratory or outdoor situations. In adapted condition the chlorophyll of single leaves, cells. and other com­munity parts tends to be diminished as the chloroplasts adapt to increasing light intensity (See various authors cited by Rabinowitch, 1956: 1261). However the effectiveness of adapted chlorophyll for photosynthesis (i.e., assimilation number) in­creases with light intensity. (See examples cited by Rabinowitch, 1957: 12621. With less chlorophyll per cell a larger part of the incident light passes through to the next layer of plants so that the total energy absorbed for photosynthesis in the top layer may be less. Consequently single cells or leaves may have lower, equal, or greater total photosynthes'.s in brightest light. It may be a misnomer to refer to diminished output of naturally adapted cells in bright light as saturation in any sense of malfunction. The effect of adaptation to intense light is to increase the assimilation number of the top cells in spite of decreasing chlorophyll concentration. In a yellow variety of Sambucas Willstatter and Stoll ( 1918) measured an assimilation number of 84 gm./gm./ hr. In contrast, shaded cells of a community possess higher chlorophyll concentrations and lower assimilation numbers. At light intensities for maximum photosynthesis of single cells, assimilation numbers of 7.4 gm./ gm./hr. (range, 4 ­ 12 gm./ gm./ hr.) are found (Ryther and Yentsch, 1957). These relationships of assimilation number, chlorophyll, and light have been shown by many authors. For example, Phillips and Myers (1954) presented a graph for Chlorella which shows diminishing chlorophyll per cell with adaptation to greater light intensity. lchimura (1958) reports data on lake plankton in field conditions with ris­ing assimilation numbers correlated with increasing light intensity for three seasons. Assimilation numbers were about 1 gm./ gm./ hr. with 3 per cent light rising to 7 gm./ gm./ hr. at 60 per cent light. The adaptation found for chlorophyll and assimilation in bright light may be com­pared to the cannonball-catcher in a zone of numerous flying cannonballs. In order that the same torque be exerted upon the output machinery, it becomes necessary to diminish the number of cups on the wheel. An overloaded system runs too fast for most effective use of the energy. By reducing the number of cups the top machine works at optimum and the uncaught balls pass below to the next cannonball machine for effective use there. The decrease in number of cups in this situation may be compared to decrease in chlorophyll in bright light. With plenty of cannonballs for the available cups, there is a large work output per cup just as there is a large assimilation per unit chlorophyll. The large chlorophyll content of cells adapted to the shade may be compared to the numerous cups on the cannonball-catcher. Many cups are needed to maintain enough power output for survival. With few balls received the output work per cup is small just as the assimilation per unit chlorophyll is small in the shade. Efficiency is large in a slowly turning wheel because there is a greater potential energy difference main· tained between incoming and acceptor energy states. The efficiency of the slowly turning wheel may be compared to the high efficiency that occurs with low rates of assimilation of plants adapted and operating in the shade (Nihei, Sasa, Miyachi, Suzuki and Tamiyo, 1954). Gessner (1949) adapted lake plankton for five days in light and darkconditions. The plankton initially had an assimilation number of 1.4 gm./gm./hr. At the end of the experiment the shaded plankton had an assimilation number of 0.1 gm./gm./hr. whereas the plankton in bright light had an assimilation number of 5.8 gm./gm./hr. Marshall ( 1956) showed relative increase of marine plankton chlorophyll in bottles in shade relative to bottles at the surface. Manning and Juday (1941) found higher assimilation numbers in hypolimnetic plankton. Sargent (1940) showed twice as much chlorophyll in shade adapted Chlorella as in light adapted algae, although the light adapted algae had the greatest yields. He reviewed the work of five earlier authors with similar results. In Sargent's work the chloroplasts in the sun had less chlorophyll present. Gessner (1955) described thinner chloroplasts in Spirogyra in light adapted plants. Movements of the chlorophyll and chloroplasts so as to transmit more light under bright light con· ditions are summarized by Rabinowitch (1951 :681). LIGHT INTENSITY AND CHLOROPHYLL IN THE WHOLE EuPHOTic ZoNE OF A COMMUNITY The total chlorophyll which develops under a square meter tends to absorb most of the · available light. A thick zone of chlorophyll may develop which is light adapted at the The Chlorophyll "A" of Communities top and shade adapted below. The greater the light intens~ty is at the top of the com­munity the greater the total amount of chlorophyll which develops within the euphotic zone. For a whole lake lchimura (1958) found increasing assimilation numbers as chlorophy11 increased. The chlorophyll per unit area thus increases with light intensity under adapted con­dition even though the chlorophyll per cell may decrease in the upper layers. An ap­parent paradox results. Communities adapted to bright light may seem pale from above and yet may contain more chlorophyll. The greenest appearing communities are not necessarily the most photosynthetic or the greenest in extraction. Pale appearing com­munities, however, do not necessarily have additional chlorophyll below, for there are other limiting factors such as nutrients. The effectiveness of the total chlorophyll of the whole euphot:c zone increases with incident light intensity. Increase of light intensity causes bright adapted cells with high assimilation numbers to develop on top of and in addition to the cells present under lesser Eght intensities. Thus the overall assimilation number increases with incident light. Since in whole communities both chlorophyll and assimilation number increase with light intensity, total photosynthesis increases with light intensity even under conditions of full daylight. The whole zone therefore in photosynthetic response to bright light differs from the behavior of single cells, leaves, or small plants. Parts of communit:es have optimum light intensities for maximum photosynthesis. See numerous examples in Rabinowitch ( 1951 :964) and Gessner (1955). Whole naturally adapted communities have maximum photosynthes's at maximum light intensity. Examples of communities (whole euphotic zone) without maxima in their photosynthetic curves (per area) with light are: mass algal culture (Burlew, 1953: 17); Silver Springs (Odum, 1957); pine tree (Kramer, 1958: 163); aquatic plants (Gessner, 1955: 86); marine waters (Mar­shall and Orr, 1929; Jenkins, 1937; Ryther, 1956b); Cabomba (See Rabinowitch, 1951: 968); lakes (lchimura, 1958); Lake Erie plankton in seasonal comparisons (Verduin. 1956); and surface algae (Manning, Julay, and Wolf, 1938). Takeda and Maruta (1956) found light saturation in individual rice plants but not in communities of rice plants. In plant cells in which there is an optimum light intens' ty for maximum photosyn­thesis, a midday depression in production may occur (Rabinowitch, 1951 :873). In an adapted community with a full depth of chlorophyll, midday depression is not expected. Any tendency for saturat'on of surface cells is more than compensated for by additional photosynthesis of the deeper cells receiving light only during the brightness of midday. Because shaded communities are more efficient (not more productive) than light adapted zones, assimilation and assimilation numbers do not diminish on cloudy days as much as the light intensity. Steeman-Nielsen (1954), Verduin (1956), and Edmondsen (1956) have reported smaller deviation in community photosynthesis than expected from changes in light intensity. Community efficiency increases as light intensity di­m'nishes (Wassink et al., 1953; Odum and Hoskins, 1958). FOUR TYPES OF CHLOROPHYLL ADAPTATIONS TO LIGHT BY PRODUCER SYSTEMS The general principles concerning chlorophyll states and light in plant communities can be summarized by considering whole natural communities and laboratory experi­ments as belonging to one of four types pictured in Figure 7. The difficulty encountered The Chlorophyll "A" of Communities INTENSITY EUPHOTIC ZONE t COMMUNITY SHADED MIXING ALL . BRIGHTTYPES TOTAL AREA CHLOROPHYLL "A" Go/M' 0.4-!.0 0.001-0,5 0.02-1.0 O.Ol-0.80 COMMUNITY ASSIMILATION RATIO 0.4-4.0 0.1-1 1-10 8-40 G%t.tifR EFFICIENCY OF VISIBLE LIGHT ABSORBED "4 1-a 10-30 PH OTOSYNTHESIS GM/M/DAY 1-so O.Ol-110 0.11-10 GROSS Fie. 7. Four types of chlorophyll adaptations to light by producer systems. The concentrations of chlorophyll are indicated diagrammatically by the density of dots in the circles. Estimates of chloro· phyll, assimilation number, efficiency, and community photosynthesis are derived from Table 2 and from the literature. in recognizing light adapted patterns in nature is attributed to the interplay of other factors on the vertical distribution of chlorophyll. 1. Stratified community. There are light and shade adapted portions. Chlorophyll per cell increases downward. All of the incident Fght is effectively used. Assimilation number decreases downward but the overall assimilation number of the community is high. The overall efficiency of photosynthesis is also high. The total chlorophyll per area is large but may not appear so from the top view. Such communities are the most productive of any yet measured. Forests and benth'.c communities are examples of this type. Some stratified, quiet waters develop vertical stratifications of planktonic algae with similar patterns. 2. Shaded Community. A community developing and becoming adapted to low light intensity may be considered to be the same as the lower portion of a stratified community in full daylight. In other words, the shaded community lacks the light adapted parts. It is relatively high in chlorophyll concentration but has less than the stratified type although it may appear much greener. With lower light intensities efficiency is greater but with so little light the total photosynthesis is less. The excess chlorophyll needed to gather light in shaded conditions is not in effective continual use and the assimilation numbers are small. Such communities may occur in winter seasons, in caves, and in most laboratory work where light intensities are rarely as bright as full daylight. Many mis­leading ideas have developed concerning nature from work with shade adapted com­munities. Many popular writers overestimated the possible yields of algae by consider­ing as representative the high efficiencies found at low light intensity. 3. Mixing Community. In turbulent upper waters of lakes and oceans at some seasons, the principal producers are moved up and down and are subjected to alternating con­ditions of shade and light. Cells move from full daylight to half daylight in a matter of a minute or two where mixing is effective. (See discussion of the diffusion of oxygen (Odum and Hoskins, 1958). Waters in the upper zones of aquatic environments also re­ceive flashing light due to wave effects at the surface in diffracting and reflecting light. Especially in the sea such cells cannot become adapted to either shade or br'ght con­ditions. It is possible that plankton cells of this type are adapted to intermediate light. As might be expected, such cells have intermediate chlorophyll concentrations and assimilation numbers (Table 2 and Fig. 7). If the euphotic zone is spread over any ap­preciable depth, the water absorbs a large proportion of the potentially usable light as Steeman-Nielsen (1957) has discussed. Such communities may thus receive less total light, develop less total chlorophyll per area, and yield less total photosynthesis. Ryther (1956a) was unable to find much light and shade adaptation in Dunaliella in contrast to the adaptation in other species found by many workers. Ryther found assimilation numbers of only 1-5 gm./gm./ hr. in contrast with higher numbers in other data for light adapted cells. However, no adaptation was attempted above 1500 ft. candles whereas daylight is usually greater than 10,000 ft. candles. Whether usual plankton algae are capable of becoming Fght and shade adapted cannot yet be generalized. In a rain forest in the intermediate depths, light is rather uniform due to the scattering of light by the leaves (unpublished data of Odum, Abbott, and Selander). Such forest leaves may also be adapted to intermediate light intensities. As the leaves are rustled by the breeze on bright days, flecks of br:ght sunlight momentarily illuminate the lower layers. Thus forest leaves on a sunny day are exposed to considerable variation of inci­dent light intensity, and adaptation to one light intensity may be difficult. Steeman­Nielsen (1957) presents a discussion of light and chlorophyll in forests and waters. 4. Thin culture with all bright light. In many laboratory cultures especially those with lights inside the containers, there is no part of the system where there is shade. Such cultures, which are never allowed to get so dense as to produce their own shade, may be v:sualized as the upper part of a full stratified community. Unfortunately, gen­eralizations have been made about such cultures without realization of their specialized nature. One may overestimate potential yields if one extrapolates with the high assimila­tion numbers of light adapted cultures. A br:ght adapted community has low concentrations of chlorophyll per volume of cells and has very high assimilation numbers. Efficiency of use of the light absorbed is low at high light intensity. Total area based photosynthesis may be higher than in the shade adapted community but is less than in the two types of communities that repre­sent full euphot'.c zones. Such cultures are particularly misleading about chlorophyll since the assimilation numbers are much higher than in full depth communities (20 gm./gm./ hr.; Ryther, 1956a). Such communities occur in nature where only thin vege­tation can develop sometimes on surfaces of rocks, water, or in new colonization. Agri­ culture is often of th'.s type. Sunflowers have high assimilation numbers (8-16 gm./gm./ hr.; Willstatter and Stoll, 1918). LIMITL'G MATERIAL REQUIREMENTS AND CHLOROPHYLL Consider next plants growing with adequate light but sub-minimal rates of supply of Lhe materials needed for photosynthesis and growth. The literature of plant physiology (Hill and Whittingham, 1955, for example) and plant ecology (Harvey, 1955, for ex­ ample) are full of cases in which limiting factors are demonstrated. Chlorophyll con­centrations diminish when plants adapt to maintain minimal existence in limiting environments. Physiological studies indicate that subminimal amounts of different limiting factors including iron, nitrogen, potassium, magnesium, and water induce lowered chlorophyll as summarized by Rabinowitch ( 1945: 427; 1956: 1267). The operation of nutrients in stimulating increase of chlorophyll in whole community systems also has been frequently noted. For example S. Conover ( 1956: 85) showed the effects of nutrients on chlorophyll of plankton populations in bottles, and Edmondson (1955) showed increased chlorophyll with fertilization in a tank. No attempt is made to separate instances where the increased chlorophyll is within cells already existing from those where new cells form. If the output machinery is limited by some factor, a reduction of input torque is nec­essary to maintain the proper load ratio and to allow energy to pass to lower levels of the community where limiting conditions are possibly equal or less restrictive. The rate at which energy is entering may be diminished by the elimination of some of the re­ ceptors in each machine. In the machine the decrease in receptors is an adaptation to the mechanical requirements of the input-output gearing. The disappearance of chloro­phy11 in environments unfavorable for photosynthesis need not be described as patho­ logical in plants surviving in such conditions, for the diminished chlorophyll may be an adaptation of input to the externally limited rates of output for purposes of economy. Just as the output per cup remains high so in these limitations the assimilation per unit chlorophyll remains high. The high assimilation numbers of yellowing autumn leaves (Willstatter and Stoll, 1918) may be from a diminish'.ng supply of raw materials at the leaf site. Many studies have related the higher chlorophyll content of generally rich eutrophic waters to high nutrient levels (Deevey, 1940; lchimura, 1956). However, some of these studies accentuate the lesser chlorophyll of oligotrophic waters by expressing data on a volume of water basis. Thus clear water has a deeper euphotic zone and less chlorophyll per volume even when chlorophyll per area is the same. Such regional comparisons should be made on an area of community basis. The nutrient factor is affected by temperature. At high temperatures there is a rapid regeneration rate in respiratory systems of the community so that a steady supply of nutrients is not as important. TEMPERATURE, TISSUE SIZE, AND CHLOROPHYLL There are two factors which affect chemical reaction rates and as a result affect the chlorophyll content of communities. First consider temperature. <\lthough the initial photochemical events of photosynthesis are physical in nature and little subject to temperature, most of the photosynthetic machinery is biochemical, The Chlorophyll "A" of Communities metabolic, and made of reactions which follow laws of chemical reaction rate. As the temperature rises, the rates of reaction and recycling of materials increase. Materials are released by respiration more rapidly also. At high temperatures, therefore, less ma­ teials are needed for the same flux of energy because materials are reused more often. In photosynthesis chlorophyll is connected to an action and recovery cycle that is chemical in nature, and its synthesis is chemical. Consequently, at higher temperatures less chlorophyll may be needed at any instant because of the greater possible turnover rate. Even though the chlorophyll functions in a temperature independent photochemical action, the amount of chlorophyll is regulated by a temperature dependent system in order to maintain optimum input-output adjustment. Assimilation numbers may be larger (Rabinowitch; 1956: 1272). Willstatter and Stoll (1918) found slight increase in assimilation number from 25° C. to 30° C. Myers and Kratz (1955) found less chlorophyll developed at higher temperatures in Anacystis. With less materials to main­ tain, cells may be smaller for the same job at high temperatures. Next, consider tissue size as a factor. It is now known in plants that in photosynthesis as well as in respiration smaller cells have greater turnover (Margalef, 1957; H. T. Odum, 1956; E. P. Odum, Kuenzler, and Blunt, 1958). The metabolism per gram inversely varies approximately as the 113 power of the linear thickness of the tissue. It has been postulated that diffusion proper­ties are responsible for the greater metabolic rate of smaller cells because the power of diffiusion per gram to carry materials into and out of the cell varies as the 113 power of the thickness. In any case the smaller the plankton cell or the thinner the leaf tissue, the more rapid is the possible metabolism per cell. In small cells with more chemical respira­tory turnover per gram, more chlorophyll per gram (or per volume) is necessary to maintain proper input torque than in large cells, but as in temperature effects, it may again be postulated that conditions permitting rapid turnover will similarly permit the cells to maintain higher assimilation numbers and less chlorophyll for the same photo­synthetic flux. The respiratory output rate by this hypothesis controls the adjustment of chlorophyll for proper input-output stress. If these postulations are correct, higher assimilation numbers may be expected in smaller cells and at higher temperatures. In terms of input-output systems any modifica­tions of the circulating machinery which permits a more rapid turnover, will permit fewer receptors to catch the same number of energy units. In Table 2 are presented data on chlorophyll in communities. There is a tendency for higher assimilation numbers and smaller chlorophyll contents per area in plankton com­munities than in communities made up of plants with thicker tissues. In Chlorella smaller individuals have more chlorophyll per gram and greater assimilation numbers (Tamiya et al., 1953). In the Laguna Madre of Texas (Table 3) assimilation numbers are higher in the plankton than in the benthic plants beneath. CHLOROPHYLL IN PARTIALLY HETEROTROPHIC PLANTS In completely heterotrophic plants chlorophyll may be absent, but in some partially heterotrophic organisms there may be chlorophyll based nutrition as well as hetero­trophic based energy intake. At the base of a growing aquatic plant, blades in the tissues recently formed from meristematic growth, and in Euglena cultures placed in the dark, there are cells which may have photosynthetic capacity when placed in the light at a later time, but which are living off imports of organic food. Such tissues apparently have TABLE 2 Chlorophyll and Assimilation Number in Communities PLANKTON COMMUNITIES (Not including bottom plants) Fertilized tank, marine, 1.5 M deep (Edmondson and Edmondson, 1947) . ·············--··· English Channel, 70 M deep (Jenkins, 1955! Summer and winter Spring blooms ····-·-···-·-··· Long Island Sound (Riley, 1956 ; Conover, 1956) Phosphorescent Bay, Lajas, Puerto Rico, 3.7 M deep January 23, 1957 . . .. ··--······· --· --··-····-·-···-······--· ·---·---···--··­Baffin Bay, Texas, 1-2 M deep, turbid hypersaline North Sea Fladen Ground, May bloom (Steele, 1956) Wesslingsee, Bavaria, 10 M deep (Gessner, 19491 Pool, 30 cm. deep (Gessner, 1949) ··-·--·-··-·····-····-. Lake Suwa, Japan, :\1 deep (lchimura, 19541 Summer Winter -······· Wisconsin Lakes, '.\[ deep (Manning and Juday, 1941) Dinoflagellate hloom, New Haven Conn. marine (Conover, 19541 ................................. Lunzer Untersee (Gessner, 1949 ) Lake Washington, Washington (Edmondson, Anderson and Peterson, 19561 _ .............. . Diatom bloom, l\foriches bay, 1 M (Ryther et al., 1958) .. Sewage Pond, Kadoka, S. D. (Bartsch and Allum, 1957) euphotic zone 20 cm. _ Corpus Christi Bay, Texas, depth 4 m. August 9 and 22, 1957 _ ...... ................ . Marine pool, stagnant, filled from flood tide, Port Aransas, Texas, Aug. 6, 1957, 1 m. deep . .. Gulf, Port Aransas, Texas, 30 M deep, January, 1958 .. REEF MATERIALS AND CALCAREOUS ALGAE Coral Reef, Eniwetok <.Odum and Odum, 1955) Whole Reef . Anemone sheet, Zoanthus pulchelus, Lajas, Puerto Rico.. Reef corals and calcareous algae, Lajas, Puerto Rico Montastrea sp. ... .. ................ . Porites asteroides _ ................ . Dendrogyra cylindricus . Diploria clivosa Penicillus capitatus ball, from center .... Halimeda opuntia clump . ALGAL CULTURES Chlorella film , dense (Myers, 1954 I Chlorophyll est'mated from biomass Sewage culture (Ludwig, et al, 1951) ................................ . Dunaliella culture 'ect". To obtain wet weight estimates of the standing crop of each of the domi1iant species, plants were removed from square meter quadrats with one hundred percent cover. The <>xcess water was removed from the plants by draining and air drying for one hour. The quadrat sample was then weighed. For each species the weight of plant~ at 100 percent cover was multiplied by percentage cover of the species to obtain an actual standing crop estimate for the quadrat. The 26 species studied in detail represented over seven-eighths of the total benthic crop· of plant material present in the estuary at any season. Hence, the instantaneous biomass for all species measured was considered a reasonable estimate of the apparent net plant production for each month of the survey. Biotic surveys were run every week, but individual quadrats were visited once every two weeks. Water analyses were made monthly during 1952-53, but every fortnight during 1954. Detailed analyses were based on the 1954 data, while those for the previous year were used to interpret trends in the environment during the period of observations on the biota. Limitations of time prevented detailed synoptic analysis of physical, chemical and biological changes in 1952-53. In January and April of 1953 storm conditions prevented observations on all transects. However, the partial data are used. AH data were transcribed to McBee "Keysort" punch cards. MEASURING PHYSICAL AND CHEMICAL FACTORS Salinity.-ln 1952-53 salinity measurements were made with the aid of a precision· calibrated specific-gravity hydrometer. Salinity was determined from specific gravity readings by using the average of three readings per sample. Glass and temperature corrections were applied, and chlorinity and salinity values obtained from Knudsen's tables (Knudsen-Oxner, 1946). Samples taken with a water sampler were placed in thrice-flushed 100 ml. glass· stoppered reagent bottles, sealed with petroleum jelly and inverted in storage. Measure· ments with the hydrometer were made the following day when all samples had reached room temperature. In 1954 all salinity samples were titrametrically determined by the Knudsen silver nitrate method (Knudsen-Oxner, 1946) and the values expressed in parts per thousand (%0) . Water T emperatures.-Water temperatures were recorded in both years with the aid of a -10° to +50° Centigrade (C.), partial immersion, guarded thermometer graduated in tenths of a degree. Both surface and bottom water temperatures were taken with the aid of a water sampler. Direct sunlight and exposure to cold winds were avoided during the readings. Density.-Density of the estuarine water was obtained by converting salinity and temperature data at atmospheric pressure to in situ density using Knudsen's tables (Knudsen-Oxner, 1946) . The results were expressed in Sigma·t Units (at) obtained from density figures by moving the decimal three places to the right and dropping the initial 1: thus density 1.0260 gm./ml. = a, 26.0. Dissolved Oxygen.-The standard Winkler method (Amer. Pub. Health Assoc. et al., 1946, 1955) was used to determine the partial pressure of dissolved oxygen in sea water. Examination of results indicated amounts of ferrous and nitrate salts were not great enough to affect the accuracy of the method. The results were expressed in milliliters of dissolved oxygen per liter (ml./L.). pH Measurements.-Measurements of pH were made within three hours after sampling with a Beckman Model G pH meter. Total and Inorganic Phosphorus.-lnorganic phosphorus was determined by a modifi· cation of the Deniges colorimetric, stannous chloride method (Truog and Meyer, 1929; Amer. Pub. Health Assoc. et al., 1955) . The modifications were: (1) use of low· phosphate sea water for the preparation of standards to eliminate salt error and, (2). the use of a sea-going colorimeter, called the electric eye photometer (Ford, 1949). Total phosphorus was determined by a method after H. W. Harvey (1948) and the results expressed in microgram atoms of phosphorus per liter (µ,g·at P / L) . Nitrate and Nitrite.-Nitrite was determined according to a method first described by Ilosvay and later modified by Weston (1905). Nitrate was determined colorimetrically Seasonal Growth of Benthic Marine Plants using a strychnine reduction method of H. W. Harvey ( 1948) . In both instances, the results are expressed in microgram atoms of nitrogen per liter (µ.g·at.N / L.). Titratwn Alkalinity.-The method used is described in the American Public Health Association water analysis manual (1946, 1955). Both phenolphthalein and methyl orange alkalinity were determined and the results expressed in hydrogen ion equivalents. llluminatwn.-Photometric measurements using a Leitz camera photometer were made seasonally to determine the minimum and maximum transparencies which could be expected in the estuarine waters. Data kindly provided by the Epply Co. of Newport, R. I., measured with an Epply pyroheliometer, gave total radiation in Langleys or gram calories per square centimeter per day ( gm.·cal./ cm. 2 /day) . Current Measurements.-Both velocities and directions of the surface and subsurface tidal currents of Great Pond were measured and the values plotted for analysis. For determining surface currents, a standard 60 cm. meteorological balloon was attached directly to a galvanized sheet iron cross. For mid-depth current estimates the cross was suspended from the balloon by a 1 meter length stainless steel wire. The balloon was inflated just enough to buoy up the cross but not enough to cause appreciable wind drag. By releasing a number of current crosses from a given line and keeping track of their positions with time, the direction and rate for each could be obtained. In addition, a line of a half dozen current crosses were released and followed to obtain an estimate of the overall ebb and flood tide circulation for interpretative purposes. One analysis was made during a period of moderate southwest prevailing wind direction and another during a period of a fresh north-northwest wind. Water Sampl.er.-The sampler consisted of four 100 ml. glass-stoppered bottles strapped to a wooden platform and separated by a wooden cross. A centered, two-foot­long upright member supported a 400 ml. plastic overflow chamber stoppered at both ends. The platform was weighted with a keel of six pounds of lead. The four reagent bottles were fitted with rubber stoppers with a pair of holes in each to accommodate the plastic tubing which led from the first through the last bottle in a series and finally into an overflow cylinder. The inlet tubing extended to the bottom and the outlet tubing was inserted just into the top of each bottle. This arrangement permitted a flow of water to move in succession through each of the four bottles and, finally, to fill the overflow cylinder. Normally a thin-walled, rubber air-escape tube could be attached to the top of the overflow cylinder and brought above the surface of the water. A hose clamp could then be closed on the free end and the air-filled sample bottles stoppered and lowered in a rack to any desired level in the water column without filling from unwanted levels. Upon opening the hose clamp water pressure would force the air from the bottles, fill them with sea water and in turn fill the overflow cylinder. In shallow water or at the surface insufficient water pressure is developed to operate the apparatus. To speed up the process, a 4 meter thick-walled rubber tube was attached to the top of the overflow cylinder and terminated with a bicycle pump with inverted pump washer. This arrange­ment provided enough suction without creating cavitation to fill the sampler in 20 seconds at any depth. The first bottle in the series received the greatest degree of flushing and its contents were used for oxygen determination. The second sample was used for pH and titration alkalinity, the third for phosphorus, nitrate and nitrite, and the fourth for salinities. The water in the overflow chamber was suitable for temperature measurements since flushing tended to bring the chamber to water temperature. During the 1954 survey, A Seasonal Growth of Benthic Marine Plants weighted four liter glass pail, which could be filled at any level by pulling its stopper by an attached cord, was used to obtain water samples for all analyses except dissolved oxygen, pH, alkalinity and temperature. Samples for these were collected using the sampler described previously. Data from Other Sources.-Precipitation, air temperatures and pressures as recorded at Nantucket Island, Massachusetts, were obtained from the U. S. Weather Bureau's climatological summaries, both national and regional. Rochford's ( 1952) estuarine classification, based largely on salinity and bottom type, was used to characterize Great Pond. The zones, according to the system, are indicated in Table 1 and Figure 2. Great Pond estuary falls into Rochford's classification TABLE 1 Hydrographic Zones in Great Pond Based Upon Differences in Salinity and Bottom Sediments Salinity %o Zone Seclion~ Stations Bottom Types Surface Bottom Neri tic none 1 coarse sand 28-31 30--32 .Marine I 4 coarse sand 28-31 29-31 Tidal II, III 4A,6,6A,6B,7,7 A,7B,8,8A,8B firm sandy clayey silt 23-30 26-31 Transition IV 9,9A,9B soft silty sandy clay 12-30 20--31 Gradient Fresh V,VI VII 10,ll,12,12A,13,13A 12E,13C soft silty sandy clay ooze firm silty sand 1-28 0 ...J ~ z2 1­Ul 0'-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~....1 :r: J F M A M J J A s 0 N 0 1911!1-19112 Fie. 3. Seasonal fluctuations in the total standing crop of benthic marine plants in Great Pond in l952-53. Data are given in kilograms wet weight per square meter. TABLE 2 Check List and Monthly Distribution of Benthic Plants in Great Pond, Mass. (after Taylor, 1957) T =Trace Present; D = Dormant; G = Growth Indicated; M = Period of Maximum Standing Crop; ? = Uncertain. Section of Seclion o( the Es.luat~' lhe-Estuar~· Month~ which had which had maximum minimum Sep I. Oct. :"o,·. Dec. Jan. Feb. Mar. Apr. May June Jul~-Aug. f!:tOWlh growth AQUATIC PHANEROGAMS Ruppia maritima L. G D D D D D D G G G M G 3-5 2 Zostera marina L. G G D D D D G G G G M G 1-4 5 XANTHOPHYTA Vaucheria thuretii Wor. M D D D D D G G M M M M 2-4 5 V. compacta V. minuta Blum et Con. G G D D D D D D D D D D G G G G M M M M M M M M 2-4 2-4 5 5 V. coronata Nordst. G D D D D D G G M M M M 2-4 5 V. arcassonensis Dang. V. intermedia Nordst. G G D D D D D D D D D D G G G G M M M M M M M M 2-4 2-4 5 5 CHLOROPHYTA Ulothrix implexa Kiitz. G G G 6 Enteromorpha clathrata (Roth) J. Ag. G G M M M G M G M G 2-4 5,6 E. compressa (L.) Grev. ... G G M M M 4,5 2,3 E. intestinalis (L.) Link M G G G G G M M M G G G 3,4 2,5 E. linza (L.) J. Ag. M M G G G G M G M G M G 1,2 3 E. plumosa Kiitz. M G G M G M G G G 3,4 5,6 E. torta (.Mert.) Reinb. M G G M G G 2,3 4 TABLE 2-Continued Check List and Monthly Distribution of Benthic Plants in Great Pond, Mass. (after Taylor, 1957) T =Trace Present; D =Dormant; G =Growth Indicated; M =Period of Maximum Standing Crop; ? = Uncertain. Section of Section of the Esluary the Estuary Months which had which had maximum minimum Sept. Oct. Nov. Dec. Jan . Feb. Mar. Apr. May June July Aug. growlh growth Monostroma oxyspermum (Kiitz.) Doty G M G G G G G M 5,6 Ulva lactura L. G G G M G G M M M M G G 1-3 4,5 U lva lactuca var. latissima (L.) De Cand. M M G G G G G G M M M G 4,5 2,3 Ulva lactuca var. rigida (C. Ag.) Le Jolis M M G G G G G G M M M G Chaetomorpha aerea (Dillw.) Kiitz. G G G G G G G M M M G G 1 C. linum (Miill.) Kiitz. G G G G M M G G 5 3 Cladophora gracilis (Griff. ex Harv.)Kiitz. F. tenuis Far!. M M M G G G G M G G 3,4 2,5,6 C. refracta ======-~~~~~~~.>.....1 St I lo phoro rhliolft 1 Mono.• lromo oiy1p1 rm11m Ecloc orp 111 co t1hr•oidt1 U IY a l oc t11c o ,_, •.• Fie. 4. Standing crops of some benthic plants dominant in the spring and fall periods in 1952-53. Data are given as grams wet weight per square meter. Seasonal, Growth of Benthic Marine Plants growth was generally very limited and only certain dormant persistent forms carried these species through the warm summer and fall months. In midwinter, usually by Janu­ary, Bangia fuscopurpurea and Porphyra leucosticta made their appearance and flour­ished for a few months. By May only an occasional plant remained. The spring vegetation in Great Pond appeared in March but most of the plant species were waning or gone by May or June. The list of species included ephemeral, short­growing-season species such as Acrothrix Novae-angliae, Chorda filum, Desmarestia viridis, Dictyosiphon foeniculaceus var. americanus, Stilophora rhizoides, Striaria attenuata, and Nemalion multifidum, as well as various species of Cladophora, Entero­morpha, Monostroma, and Ulva. The greens usually reappeared or showed another in­crease in abundance in the fall period. The sea grasses, Zostera marina and Ruppia mari­tima showed initial growth in mid and late spring but did not reach their peak growth until summer. The summer plant populations included the sea grasses which began initial growth in late spring. The most common algae species among these were Enteromorpha compressa, Ulva lactuca and its two varieties, Agardhiella tenera, Gracilaria verrucosa, Ceramium rubrum var. proliferum, Hypnea musciformis, Spy­ridia filamentosa, and a number of species of Dasya, Ceramium, Champia, Lomentaria and Polysiphonia. In the fall, there was a reappearance and renewed growth of some species which grew in the spring. The list includes mostly greens such as the various species of Enteromorpha, Ulva, Cladophora, and Monostroma, and the brown Stilophora rhizoides. Between September and November no really new additional species appeared in Great MONTHS JFMAM J JA SO N D J FMAM JJ AS O ND J FMA MJJ A SON D Elllt ro ll'lo rpho c lolhro to (ocotyp1 I) R11ppio mor it i m o Zo thra ma rin a .,.: 40 " ~ 30 w "0 2 ~ c •• 0 Polytiphonio d1nudoto Grocilorio .4.9ordlllollo tentro ... ···~···.. 2000 . 4 00 :;· ~. Hypn to m111ci form i 1 Ent o ro in orpho l n t o 11inoll 1 Pol y1 iphon io novoo-on9 lio e o.oe Fu;. 5. Standing crops of some dominant species appearing in spring and summer, early spring through fall, and late spring through fall. Data are given in grams wet weight per square meter. Pond. The winter plant communities generally appeared year after year, in late October along the coast and in early November in the estuaries. For further details on the seasonal histories of benthic plants in Great Pond, see Table 2. The reader should be cautioned that these data represent measurements on the quadrats and do not include the occasional appearance in other parts of the estuary of other species. These seasonal histories are considered to be characteristic of an estuary rather than the open coastline. COMPARISON OF SEASONAL GROWTH IN GREAT POND WITH GROWTH REGIMES JN OTHER LATITUDES Great Pond, at Latitude 41 ° N., as indicated in this study, has two major groups of plants each with growth at different seasons. Polar and tropical seas each have one dominant group with one growing season. Seas of intermediate latitudes are believed to be populated by more or less mixtures of warm and cold climate species. A brief review of studies concerning seasonal growth of marine plant communities from polar seas to the tropics shows the intermediate nature of southern New England plant com· munities. In the Arctic, above Latitude 60° N., visual estimates of benthic marine plant growth suggest that there is only one seasonal growth period. Apparently growth only occurs during the short summer interval. Yet, reproduction may occur at other times of the year in some species. A review of these studies may be fom.d in Fritsch ( 1945). MONTHS J FM AMJ J A SONO J FMAMJ J AS ONO Efllerotnorpho llnzo Ul•O h1cluco war, r I ;id P11ncfario p lofltotlneo Scyto•iPhon lom•ntorio Ee toe orprn 1il ic 1,t1011n Fie. 6. Standing crops of scme benth'c plants seasonally dominant in the fall and winter period in 1952-53. Data are given in grams wet weight per square meter. MAP OF GREAT POND, FALMOUTH SHOWING AR EA S PO PULATED tell - BY MA RINE PH ANEROGAMS ,_,_¥~ i l ~ v I !\.' c y 4 R D s 0 u N 0 F1G. 7. Map of Great Pond, Falmouth, Mass., showing areas populated by marine phanerogams. In boreal waters, Macfarlane (1952) reported a predominance of spring growth for the important commercial kelps and in Irish moss along the coast of Nova Scotia. These quantitative data are for open coastal water rather than protected or estuarine situations. Bell ( 1927) and Bell and Macfarlane ( 1933) provided visual estimates of a less promi­nent summer-growth group. No information appears to be available on winter growth for the area. Bermuda, at Latitude 32° N., has really only one major seasonal growth group and :>ossibly two minor ones (Bernatowicz, 1950, l 952a, l 952b). Only three winter species, ~allithamnion corymbosum. Scytosiphon lomentaria, and Sphacelaria tribuloides, grow in those waters. Six annuals with short growth seasons appear in the spring but the rnlk of the vegetation persists the year around. The maximum growth period for most ,pecies is in May and June. Bernatowicz's data are for quiet, protected habitats and ience may be readily compar2d with the Great Pond da·a. Results of studies in the 11editerranean by Feldmann ( 1937) and Funk ( 1927, 1954) suggest a similar seasonal . ;rowth asp~ct to that reported from Bermuda. According to an early report by Svedelius (1906) thr tropics may have only one 11ajor s2asonal growth period. Presumably growth is maximum in winter and early ~ Jring at Ceylon. LOCAL DISTRIBUTJO'.'J I'.'< THE ESTUARY Standing crop totals for the commonest species (Figs. 4, 5, 6, 7; Table 2) clearly i1 1dicated that in Great Pond the bottom of the entire major basin (transects 6 through 9 sections II-IV, Fig. 2) supported the largest accumulation of benthic marine plants. L1 comparison, the intertidal areas of both basin and riprap supported a relatively n T • I I • 6 5 r I 4 2 •a I QUAD, A B C D E F G H A B C D E F G H A B C D E F G H A B C 0 E F G H STILOPHORA RHIZOIDES (EHRHJJ.AG. FIG. 8. Seasonal distribution and percent cover of Stilophora rhizoides in Great Pond in 1952-53. The graph shows transects 1 through 13 including transects in both sloughs. T indicates a trace. A dot indicates absence of plants. A number indicates the percentage cover of plants where 1= 10 per­cent. Relative length of each transect is suggested by the length of each line of quadrats. Each dot on the riprap transects 1, IA, and 2 represents four 25-centimeter-square quadrats. The number of quadrats is the sum of dots and numbers. Species population isopleths indicate the plant's density in the estuary at four selected periods of the year. Isopleths are drawn at intervals: 15%, 30'7<, 45%, 60%, 75%, and 90%. found at the upper end of the major basin in section IV were scarce, dying and heavily epiphytized. A fair abundance of somewhat healthier plants was found toward the lower end of the basin lsection II, transect 6, Fig. 2) _ During August and September indi­vidual plants were rarely seen in the estuary and none within the measured quadrats. The few occasional plants found in late summer were in fairly saline habitats near the mouth of the estuary (section II, transect 6, Fig. 2) . A similar history was found for Cracilaria verrucosa as shown in Figure 9. CHANGES FROM ONE YEAR TO THE NEXT JN SPECIES CoMPOSITIOX AND ABU'.'IOANCE Since the present quantitative vegetation survey was begun in July of 1952 and terminated in August 1953, direct comparisons between 1952 and 1953 were possible for the months of June, July and August. Spot checks on some quadrats were made in 1954 and some cursory inspections at several seasons conducted in 1955 and 1956. Some species were either missing or had altered their positions in the economy of the estuary. Conspicuous by its absence was Punctaria plantaginea in the winters of 1954 and 1955. Not even an occasional plant appeared in or near the entrance of Great Pond. The species Punctaria latifolia was not nearly as abundant in 1954 and 1955 as the previous winter, 1952-53_ Some Zostera marina beds failed to produce leaves in 1954-55 becom­ing less abundant. The sea grass, Ruppia maritima, showed a distinct increase in SEPTEMBER 1952 NOVEMBER ' 1952 MARCH 1953 JULY 1953 TRAN.,.. 13 : ~22~. I T T T 12a 12 II T T 10 .a·jj):•· ~·ldT T I T I 4 2 I a T • T 1 QUAO. A B C 0 E F G H A B C 0 E F G H A B C 0 E F G H A B C 0 E F G H GRACI LARI A VERRUCOSA(HUD~)PAPEN~ Fu;. 9. Seasonal distribution and percentage cover of Gracilaria verrucosa in Great Pond in 1952­ 53. Isopleth interval is 15 percent. Legend is the same as for Figure 8. 1954-55 especially in areas occupied largely by Zostera in 1952-53. The species ac­cumulating the largest biomass in Great Pond was Gracilaria verrucosa. This form was abundant in 1952, less abundant in 1953 and was further reduced in 1955. The alga, Gelidium crinale, absent for several years from the Woods Hole-Falmouth area, accord­ing to Taylor (personal communication), had appeared as a conspicuous but narrow intertidal zone 3 cm. wide on the jetty. By the winter and spring of 1955, very little of this population remained. In 1956, Gelidium seemed to be reclaiming ground lost in 1955 but diminished after the more severe winter of 1957. The species Bostrychia rivu· laris was first observed in July 1952. By the summer of 1956 no plants could be found anywhere in the estuary. OvERSEASON PERSISTENCE OF SUMMER AND WINTER SPECIES An opportunity to study the winter persistence of some species was afforded during the seasonal growth survey. Branched basal portions of the previous season's thalli of Gracilaria verrucosa, which had persisted through the winter months, were observed to regenerate new, branched systems the following spring. During the winter period no intact, branched, basal portions of this species were observed along the exposed coast where holdfasts are commonly found. Dormant, branched portions of thalli may only occur in quiet, protected waters such as Great Pond. In Agardhiella tenera, branched portions of the thallus likewise persisted over winter and in late spring produced new growth. In addition, small germlings of Agardhiella were found in late spring at the same time that regeneration was detected in the branched thalli of the previous growing season. The juvenile plants were seen only out· side the estuary on pebbles and shells in the shoal water of Vineyard Sound. Apparently Seasonal Growth of Benthic Marine Plants Agardhiella over-winters as holdfast discs in coastal water and vegetatively in south shore tidal estuaries. Solitary plants of Nemalion multifidum, an early summer annual, were observed in mid-winter attached to boulders below the intertidal zone. In contrast the early summer habitat was in the upper intertidal zone. Maximum luxuriance of this species occurred in late June. Unattached, branched portions of Hypnea musciformis were found in May apparently dormant, but viable, suggesting that these may regenerate after the winter. Some individuals of Ceramium rubrum passed the winter months mature, full-grown, and actively reproducing. The size and number of individuals were usually reduced suggesting very limited vegetative growth. SEASONAL PERIODICITY IN REPRODUCTION Some observations of a general nature were made on the occurrence of reproduction in the seasonal histories of some species of benthic plants. Generally, conspicuous re­productive structures such as those developed in the Rhodophyta and Phaeophyta are present throughout the period of maximum luxuriance and continue into the wan· ing phase. In Polysiphonia novae-angliae, P. denudata, Champia pravula and Agard­hiella tenera, both the gametophyte and the sporophyte were present during the season of active growth. An increase in population usually was paralleled by an increase in the number of individuals bearing reproductive structures. A correlation between growth and reproduction for some species was also found in Bermuda by Bernatowicz (1952a). Smith (1947) found a correlation between spring tides and sporulation in Ulva lactuca. Evidence for similar periodic sporulation was obtained in Great Pond. Measurements were made on juvenile plant populations of Ceramium rubrum, Petalonia fascia, Sargas­sum filipendula and Scytosiphon lomentaria. In the study of Scytosiphon, for example, the juvenile plants were measured on four 25-centimeter-square plots on different, closely associated, rock surfaces at about the same intertidal exposure. The plants were very uniform in height. Within each plot less than one third of the height measurements exceeded the standard error for all plots. This evidence suggests that nearly all p!ants in each plot germinated at about the same time. Over a period of 20 days. the mean height of juvenile plants in each plot increased uniformly as shown in Table 3. Older plants TABLE 3 The Mean Height of Juvenile Plants of Scytosiphon Lomentaria in Four Plots on a Jetty at Falmouth, Mass. (Height in centimeters) Plot number 1 2 3 4 March 3 2.8 2.3 1.9 2.0 March 8 3.4 3.5 3.3 3.4 March 14 5.2 5.7 5.1 5.3 March 21 8.1 8.4 8.0 8.2 m all plots were significantly taller than the juveniles. Possibly the sexually mature plants in the population had responded to some rhythmic factor in the environment. The study of the seasonal growth of benthic marine plants in Great Pond has pro­vided measurements of a series of biological events. These events may be characteristic of other quiet water lagoons or estuaries in southern New England. The environmental factors causally related to these growth events are considered in the following chapter. 120 Seasonal Growth of Benthic Marine Plants Effects of Environmental Factors on the Seasonal Growthof Benthic Plants Those environmental influences which normally occur year after year may be sepa­rated from those which occur intermittently, irregularly, or rarely. Some of the regularinfluences found to be important are insolation, air and water temperatures, chemicalproperties of the water mass, tidal movements, the run-off waters from fresh-waterstreams, and the seasonal weather patterns of the region.The irregular types of environmental influence include insolation in prolonged peri­ods of clear skies, severe low temperatures, icing, unusual numbers of severe frontalstorms, tropical hurricanes, prolonged emersion and immersion due to offshore or on­shore winds, pollution by man, wholesale wasting diseases due to plant parasites (Dexter, 1953, Taylor, 1933) and unusually heavy grazing by a large year-class of some herbiv­orous animal (Lodge, 194-8; Odum, 1953). During the course of the present survey, anevaluation of both regular and irregular factors was made. LIGHT, TEMPERATURE, A::\"D SEASONAL GROWTH The separate effects of li ght and temperature on plant growth may be provisionallyseparated in a field study since the annual temperature and light curves (Figs. 10, 11) JUL'!' AUCUST IOTUU[ll OCToau 'f)(C( MICll . MO'ICMIU s?~f-----------------~----------------1 r ~ '~.... ,'. g_ ,.. r----'t-1-------------------------------"'---l~~~~--... ~ ; RAINFALL l9!U: . ,3 _....__..___._,___.__ l.~~ _...'-"-...._-------..L.J...._..-_........._____.___~'--'-~~...LJo ~ Fie. IO. Weekly average of total solar rad:at'on at Newport, Rhode Island, 1952-53. The bar graphat the bottom of the figure indicates rainfall in centimeters. show minima, maxima, rises and falls at different times. It is fully appreciated that inter­pretations of results are limited due to the many interacting variables on growth. Theseresults reported here are thus tentative. Further experimental studies are needed to con­firm the suggested relationships. Overall benthic plant production in Great Pond during 1952-53 indicates that thelargest standing crop developed between June and September with the maximum inAugust (Fig. 3). The minimum plant biomass occurred in January and February. Thesemaxima and minima are associated with the max:ma and minima in temperature andinsolation (Figs. 10, 11) suggesting that these two factors play the leading role in theseasonal growth of benthic plants in Great Pond. Seasonal Growth of Benthic Marine Plants J.uu.un 'ElltUAllY MAill:CM ....ltlL JUM[ Sl:f"TEMIU OCTOIElt NDVEMl[llt 0[CUtHlt ~ 1oi--~~~~~~~~--,.c-,._,,_~~~~~~~~~~~~~~~~~~~'<-...,._~~-1 ! FIG. 11. Seasonal changes in the bottom water temperature in 1952-53 in the "neritic" zone (Figure 2) adjacent to Great Pond. Late winter and early spring species including some short-growing-season ephemerals, (Fig. 6, Tables 2, 4) attained maximum population density in April and May and showed a marked reduction by June_ Presumably these plants were adjusted for opti­mum net phytosynthesis at low temperatures and illumination_ Plant species with maxi­mum growth in the spring period were of two types. One showed irregular growth (see section on nutrients) through the season, i.e., Enteromorpha compressa (Fig. 4), while the other maintained a steady increase in growth, i.e. Stilophora rhizoides (Fig. 4). The latter type, growing in late spring (Table 5), is believed to be more dependent on light t!-ian temperature. Temperature effects are undoubtedly present. A more rapid weekly increase in light as compared with temperature supports this opinion (See Figs. 10, 11; compare the slopes of curves for April-June). A different group of sp~cies attained maximum growth rates during June and early July (Fig. 5, Table 6) during the peak illumination period (Fig. 10). Other plants (Fig. 5, Table 7) grew best during the temperature maximum (Fig. 11) in August. These data suggest one group of species may be adapted for optimum net photosynthesis at maximal light and submaximal temperature while the latter group is adapted to maximal TABLE 4 Standing Crops of Some Dominant Benthic Plants Occurring in Late Winter and Spring, 1952-53 (in Grams Wet Weight per Square Meter) Species March April May June Cladophora re/racta 1.4 132.0 346.0 84.2 Enteromorpha clathrata 134.8 0.1 161.0 + Enteromorpha intestinalis 17.9 12.9 228.0 30.8 Enteromorpha linza 4.3 0.0 6.9 0.0 M onostroma oxyspermum 0.0 2.2 1.9 0.5 Ectocarpus con/ervoides 6.4 ? 62.0 2.4 Gifjordia mitchellae 0.0 ? 7.5 O.l Ectocarpus siliculosus 49.5 93.9 49.5 8.9 Punctaria lati/olia 14.0 5.4 3.9 O.l Scytosiphon lomentaria 6.4 3.7 4.9 0.5 Ulva lactuca 48.2 22.6 98.7 77.2 Seasonal Growth of Benthic Marine Plants temperature and submaximal light for this latitude. It should be pointed out that the few species listed in Table 6 are responsible for the largest fraction of the plant cover for the mid-summer period. A small group of species, including plants with warm climate affinities, grew in Great Pond over September and into early October of 1952 (Tables 2, 8).The influence of light may have dominated other factors in its effect on growth. The fall off in light intensity was rather rapid in September and October (Fig. 10), but bottom water temperatures tended to fall more slowly (Fig. 11). There was a general reduction in standing crop of most species in the estuary in October (Figs. 3, 4, 5; Tables 2, 8) possibly associated, TABLE 5 Standing Crops of Some Dominant Benthic Plants Occurring in the Late Spring Period, 1953 (in Grams Wet Weight per Square Meter) Species April May June July August Cladophora gracilis f. tenuis 5.7 98.5 98.0 30.3 0.2 Enteromorpha intestinalis 12.9 228.0 30.8 33.2 13.1 Enteromorpha Linza 0.0 6.9 0.0 0.0 0.0 Enteromorpha plumosa 3.8 16.7 16.3 4.6 3.0 Stilophora rhizoides 12.8 222.0 398.0 253.0 0.6 Ulva lactuca var. latissima 13.7 652.0 274.0 191.4 44.3 TABLE 6 Standing Crops of Some Dominant Benthic Plants Occurring in the Mid.Summer Period, 1952-53 (in Grams Wet Weight per Square Meter) Species April May June July August Seplember Agardhiella tenera 350.0 124.0 406.0 464.0 989.0 785.0 Enteromorpha compressa 1.9 10.2 0.6 9.9 0.0 0.0 Graci/aria verrucosa 1,262.0 733.0 1,310.0 2,310.0 2,980.0 1,452.0 Ruppia maritima + 35.4 34.0 161.0 96.9 98.5 Zostera marina + 208.0 170.0 293.0 188.0 136.0 TABLE 7 Standing Crops of Some Dominant Benthic Plants Occurring in the Late·Summer Period, 1952-53 (in Grams Wet Weight per Square Meter) Species June July August September October Enteromorpha intestinalis 30.80 33.20 13.10 107.00 6.74 Enteromorpha plumosa 16.30 4.57 3.02 47.30 2.38 Hypnea musci/ormis 0.00 0.00 0.35 6.91 0.00 Polysiphonia denudata 6.95 3.35 60.00 118.00 1.60 Polysiphonia novae-angliae 0.01 0.00 0.001 0.10 0.001 TABLE 8 Standing Crops of Some Dominant Benthic Plants Occurring in the Autumn Period (in Grams Wet Weight per Square Meter) Species August September October November December Cladophora gracilis f. tenuis 0.2 0.6 77.7 134.5 0.0 Clcdophora refracta 0.0 22.7 41.1 29.3 0.0 Ectocarpus confervoides ? 9.3 30.6 29.7 ? Ulva lactuca 0.6 0.6 1.9 1.4 11.2 Monostroma oxyspermum 0.0 0.0 0.1 0.1 0.1 Seasonal Growth of Benthic Marine Plants in part, with the decrease in light intensity. During the last two weeks of October bot­tom water temperatures dropped below 18° as the rate of cooling increased. Decreases in both temperature and light at this time may have accelerated the degradation of warm season plant populations. A few plants, which grew in the spring, including Cladophora gracilis F. tenuis, C. refracta, Enteromorpha clathrata, E. intestinalis, E. plumosa and Monostroma oxysper­mum (Figs. 4, 6; Tables 5, 8) showed renewed growth in the fall. Light intensities were similar in spring and fall during the periods of growth. Yet, water temperatures were several degrees higher in the fall for the same light intensity. Does this mean that light was the determining factor? Phytoproduction was much lower in the fall for these species populations. There is a possible explanation. In the spring light and tempera­ture as well as photosynthesis and respiration were increasing. In the fall solar radiation and metabolism were decreasing. In November, the species shown in Table 9 appeared to be dominant and by Decem­ ber had carpeted the estuary. Many of these cold season annuals had appeared along the open coast a week or two earlier. The difference in time of initial growth between TABLE 9 Standing Crops of Some Dominant Benthic Plants Occurring in the Winter Period (in Grams Wet Weight per Square l\Ieter) Species :\onm1ber December Januar~· February March Apdl Ectocarpus siliculosus ? 152.000 44.600 42.600 49.500 93.900 Punctaria lati/olia ? 2.460 1.800 2.480 14.000 5.450 Punctaria plantaginea 0.184 0.390 0.265 0.265 0.604 1.050 Scytosiphon lomentaria 0.145 4.000 3.040 2.940 6.440 3.680 estuary and coastal communities in 1952 may possibly be explained on the basis of the presence of higher temperatures (Fig. 12) within the estuary than along the coastal areas. By early November temperatures in both protected and exposed communities were below 18° C. The light intensity associated with early NoYember (Fig. 10), the time of initial growth of these winter plant communities, was the same as that in March and April. Water temperatures in the spring period, however, were 5° C. lower (Fig. 11 )_ Maximum growth was exhibited by these species in March and April (Fig. 6, Tables 2, 4). These data seem to indicate that the role of light was possibly more critical as related to growth than water temperature. The only species which grew best during the winter temperature minimum in Janu­ary and February of 1953, were Bangia fuscopurpurea and Porphyra leucosticta (Table 2). These plants grow well at or near freezing temperatures. The plants appeared while illumination increased from the winter minimum of less than 100 gm. cal./cm.2 /day on January 10th, 1953, to 300 gm. cal./cm.2/ day by mid February. Here is another ex­ample in which light intensity may have been more important than water temperatures. Many of the winter species present in Great Pond grow in summer in polar and boreal seas (See review by Fritsch, 1945). It is suggested from the above results that a study of boreal and polar benthic plant growth with respect to light and temperature may show the critical role of illumination at high latitudes. NITRATE, PHOSPHATE AND SEASONAL GROWTH From physiological studies l Gessner, 1955) it is known that the amounts of nitrate and phosphate available can regulate plant growth during periods of nutrient scarcity. JANUARY FEBRUARY MARCH APRIL MAY JUNE JULY A UOUST SEPTEMBER OCTOBER NOVEMBER 0 EC EMBER STA. 12 II IO 88 • 0 O A T A 6. 5 6. 9 JO.I ~:: 14.~ \ ::~ \ ( \ \ I I I I 17. 6 17.6 17.6 17.2 C~) '"· 1@ 18. 2 I I I I1 I I I I 88 I @ @ I 0 I ' ® 78 SB 4 ~~:::~::: ~\~ 17.8 20.7 20. 6 23,1 22.1 22,1 I I I I I I ( 11.0 \'·· 18, 0 17. 7 I f I I I I ( I I I I 10,6 \ I 13 ••7 I I I I I 8A G • 0 0 A T A I 7A I I I I I 10.2 \ \ 4 f I \ I ' ' I I 10.2 Fie. 12. Temperature in the bottom water of Great Pond, Falmouth in 1954 in degrees centigrade (C 0 ). In this and all subsequent frequency polygon projections of Great Pond data, two graphs are shown separated hy a douhle line. The lower graph represents asection along the western side of the estuary including station 1 in Vineyard Sound and station 13 in Perch Pond Slough. The upper graph represents a section on the eastern side of the estuarine basin also including station 1 in the sound as well as stations 10 through 12 in the Coonamessett Slough. The two graphs providea means of comparing the more swiftly moving tidal water of the eastern half of the basin with the more sluggish water of the western half. Seasonal Growth of Benthic Marine Plants Multicellular plants store large concentrations of nutrients in their extensive tissues (Vinogradov, 1953). In estuarine communities the available nutrients are in part sup­plied by regeneration from dead plants. Evidence that nitrogen and phosphorus may limit growth in Cape estuaries is provided by a fertilization experiment (Pratt, 1950) conducted in Megansett estuary. Enriched fertilizer was dumped into the small embayment. Large excesses of nitrogen and phos­phorus yielded blooms of phytoplankton, and masses of benthic plant growth fouled the shallow basin. Great Pond is like Megansett estuary with a similar geologic history. In Great Pond there are low concentrations of nitrate (N0 3-N) and inorganic phos­phate (P04-P) during the spring and fall periods (Fig. 13). One period of maximum JAMl.iUY HllfllUUV MUCH JU ME AU&U$ T S EPT.ENIElt OCTOIEll NOVEN8Elfll DE CEtUE lt TOTAL PHOSPHORUS ·INORGANIC PHOSPHATE J -_/\_ .·~ .:o.s ~----------......_ I. 01$SOLYED OXYGEN~ ,,..... --..... --·,.!> ........' J • ... ....... ··~5 APPROXIMATE OXYGEN~ ..... UTILIZATION .,., ........· ····........ ................... ..JUI NI.TRUE NITROGEN .--~-­ i .. // ·-v,,....--·~ ,,'' z ______........... , , r.o :o.s ,,'' •. .__.........__.__.__.__~_.___.__._~_...___.____,'--_.___.._...___.__...__.___._ _,_~-'--.........--' Fie. 13. Seasonal changes in total phosphorus, inorganic phosphate phosphorus, nitrate and dis­solved oxygen in the bottom water of the "transition" zone (Figure 2J of Great Pond in 1954. nutrient salt regeneration in situ within the estuary occurs in August, and minor ones occur in late fall and in the spring (Fig. 13). The spread and reduction in abundance of many benthic plant species seems to follow changes of phosphorus in spring and nitrogen in the fall. Data for the seasonal fluctuations in the nitrogen cycle in Great Pond (Fig. 13, 14, 15, 16) furnish some evidence that nitrate may have been low enough in the bottom water to affect the growth of benthic plants in October. Measurements of nitrogen in October indicate that less than 0.4 µ.g-at. N/ L were present in those sections (II-IV, Fig. 2; Table 2) of Great Pond where phytoproduction was high in other seasons. Concentra­tions of nitrogen and phosphorus were considered low when amounts were near or at limiting levels for unicellular algae under laboratory conditions (Ketchum 1939). Dur­ing October a number of species, many of them dominants, exhibited drastic reductions in standing crops (Figs. 3, 4, 5; Tables 2, 8). Many plants terminated growth. The pos­sibility that light and temperature play an important role here was discussed earlier. No new additions to the benthic vegetation were observed in October. Only two species Cladophora gracilis f. tenuis and Ulva lactuca var. rigida increased in standing crop. There was a delay in the appearance of winter species within the estuary (Fig. 6, STN IZ II 10 9B BB 7B 6B 4 13 9A BA 74 6A 4 I JAN UARY FEBRUARY MARCH' APRIL MAY J UNE JU LY AUGUST SEPTEMBE R OCTO BER NOVEMBER DECEMBER 0 A T A " 0 0 A T A " 0 F1G. 14. The seasonal <'hunµ;es in nitrate in the bottom water 0£ Greut Pond in 1954, expressed in minoµ;rnm ntoms N/ L. Tiu~ two µ;rnphs shown nre frequency polygons, des<'rihed in detail in the leµ;end [01· Fiµ;nre 12. Seasonal Growth of Benthic Marine Plants APjlt ll WAY JUHE JUL'!' ..... ------­ .-­ <:.::__-· --::_::_:-~ -·--:____.---=---.---­ ____./~-"N­ ...... ·-------­~I~~~~· '-=====~~~~_j~: ;;i..__ ',..,._......... /--~~~=====­ 0 ,_cc z -~~-­ -·---­ --·-------­ _._-............ ----------------··--:-----·- Fu;. 15. Seasonal distribution of nitrite (NO_, I and nitrate (:'iO. I in the bottom water of Great Pond by hydrographic sections (Figure 2.1 in 195-1, expressed in microgram atoms N/ L. Table 4) mentioned earlier. All these responses may haYe been related to a nitrogen deficiency. The seasonal changes in the phosphorus content for Great Pond are giwn in Figures 13, 17, 18, 19. The relatively low concentrations in spring and fall with possible limiting amounts of phosphorus in April and June suggest phosphorus as well as nitrogen may have limited growth in Great Pond. Phosphorus concentrations are charted in Figures 17, 18, and 19. Diminished growth of benthic vegetation may be observed in Figures-!, 5, 6; Tables -1-, 5. Species present in the spring first appeared in March, sustained diminished growth rates in ApriL but increased again in May. In April P04-P levels were less than 0.3 }-tg-at P Lin the bottom water of the usually productiYe sections of the estuary (fig. 2, sections II-1\-1 _Another period of reduced standing crops for many of these and some additional species occurred in June. At this same time P04-P concentrations had decreased to low rnlues. ~ot all species in the estuary were equally affected by the low supply of nutrient salts. Different species have different minimum nutrient requirements (see reYiew by Ketchum, 195-1-1. Unfortunately, there are few published data on the minimum nutrient requirements of attached marine plants (Feldmann, 1951). During the winter months in 1952-53, when nutrient salts were in highest concentra­ tion, the six species of Vaucheria found in Great Pond marshes were in fruit I Table 21. When these species were still showing good vegetative growth in late summer to falL nutrients were possibly near limiting levels_ Whether the nutrient cycles in the marsh areas are similar to those in the basin is not known. The largest supply of inorganic phosphate and nitrate t Figs. 13 through 191 occurred in the estuary within hydrographic sections II and IV (Fig. 2). These sections were also the scene of the highest plant production per unit area in the entire estuary t Fig. 9_ Table 2). The sea grasses, Ruppia maritima and Zostera marina tfig. 7). which repre­sent an important part of the plant biomass in summer. are most abundant in these sections of the major basin of Great Pond. Since sections II and IV were the scene of marked tidal water exchanges (see current velocities, Fig. 20), more nutrients per unit JANUARY FEBRUARY MARCH APRIL MAY JUNE JU LY AUGUST SEPTEMBER OCTOBER HOV EMBER DECEMBER STN . I I I I 12 I , ' ,' Q. 18 0.14 0.1, 0.35 o.o e I , I , I CE I I 0 .11 II I (___ ~ I , 0.1• 0,1-40.lt 0.14 0 .12 10 ~ I 0,17 (50• I / I .2 01 <0 I 0 . 11 Q03 . 0 .14 0 .11 I 9B N 0 0 A T A. I 0.22 ~' I I ~r I / , 8B , ,/ , .o ,, 0 06 7B 0" 0.08 O.t3 8B ----_ e--,, 4 0 .04 0, 11 0 09 0 26 0 .18 13 '\ -o~_,..,,,,,. 0.16 0 ,12 0,16 I I 8A H O 0 A T A Q.1 4 / .,..... ,.,,. 0 23 7A / ~0.09 .--' 0 .12 F~: 6A 4 // _, -----~­ 0 .06 0.08 0 II 0 ,05 -----&" Q.0 4 0,01 / 0 .41 I FIG. 16. The seasonal changes in nitrite in the bottom waler of Great Pond in 1954, expressed in microgram atoms N/L. The two graphs are frequency polygons,described in detail in the legend for Figure 12. Seasonal Growth of Benthic Marine Plants ,,...,,.. ....-.... v ---·-----·---...---------.------------·------·-------·--~---------~ ~.. I' •.o·----.. IV _,':"~..-..""~-·-­ "" ---'~.-----------'IE• P0.. -1' SI-~~~~~~~~~~~===-~~~~~~~~~~~~~~~~~~~~--;~ "' , I J ~~~~~~~~~~~~~~~~~==-~~~~~~~~~~~~~~~~___,g >­ ., I! 16AJ ____ • __ ... --•---·---·---------­ 06 t:I -----•-----·-....... ··---·­ FIG. li'. Seasonal distribution of inorganic phosphate (PO,-PI and total phosphorus <:::Pi in the bottom water by hydrographic sections (Figure 2 I in Great Pond in 195~. expres."t'd 'n m:crogram atoms P/ L. of time due to currents may have aided plant production in these areas (Gessner. 1952). Thus the recycling of organic materials in Great Pond may occur primarily in situ possibly augmented by contributions from Vineyard Sound and terrestrial run-off. Hul­burt (1956) did not include macroscopic benethic plants in his explanations for seasonal phosphorus changes in Great Pond. The nutrient limitations may help to explain \1·hy plant growth in Great Pond is limited and irregular. DISSOLYED 0XYGE:\" .-1.:\"D GROWTH The effects of oxygen deficiency on benthic plant growth could not be observed in Great Pond since no anaerobic conditions were observed in the bottom water at anv sea­son. Oxygen data for 1952-53 and 1954 (Figs. 13, 21, 22) substantiate this conclusion. Ketchum (personal communication I did measure an oxygen deficit 0.0 ml. O~/L in Great Pond immediately following a hea\·y rain storm. Such short term depletions may not seriously affect plant growth in an estuary with good transport and mixing. Reduced oxygen in the bottom water in some seasons was correlated with high growth rates and the regeneration cycles of nutrients. PH ..\'ID BE:\"THIC PL.UT GROWTH The pH of the bottom water (Figs. 22, 23) may have had some direct influence on benthic plant growth especially in brackish water. Maximum growth occurred in the "tidal" zone (Fig. 2) of Great Pond where there was only a limited annual range of pH between 7.8 and 8.2. pH may affect plants by controlling the carbon dioxide state arnil­able for photosynthesis. The brackish waters exhibited pH extremes from 1.0 to SA.. During the summer months growth rates (Fig. 23) were highest (fig. 3) during periods of lower pH (7.8 usually). One explanation is that overall attrition and respiration may have exceeded photosynthesis during peak growth periods causing the pH to drop. It is also possible that the regeneration of nutrients during such periods may have stimulated growth in particular plants. ...... w 0 JANUARY FEBRUARY MARCH APRIL MAY JUN E JULY AUG US T SEPTEMBER OCTOBER NO VEMB ER DE CEMBER STN " 16 II 0 9 0 7 19 12 18 12 I 6 1,1 II .. 0' 0'·~ 18 16 10 11 I CE 0 I.I 1.4 10 0' 0 7 I 0 I 0 I> 12 I> I 6 I.I 10 I " Vi 0 6 16 '0 I 3 1,4 10 98 N 0 0 A T A ~ ~ c "" 3 "' ;::, 88 ·~ \' 12 2) a. ,,....---..._,_ 11 10 08 o.e 10 I 8 10 0 9 1.0 .... ;.:v c 78 "' ~ I 1': 12 10 0.9 0' 0,9 0 9 12 11 14 I 2 0 6 1.7 1.4 66 s. 4 1.2 c ~ ....... 16 1.4 09 0' 0. 1.2 12 0 6 10 0 ' l:tl ;::, "" 13 1.2 0' 1/ I 6 0.7 S­ 0---;::;· I 14 I 6 1.0 9A 1.0 I.I 1.3 I I" I.I 1. 4 ...... I ' / ""' ·~)··( " / ~ / .... , ,/ ~· BA / "" N 0 0 A T A I.I 12 10 0 7 13 1,1 12 1.5 7A ~ ~',o ,.---/ I ~ 0 0 I 10 0 • 0. 0 • 12 1.0 I 4 I 3 1.0 6A l.O "' ~ 12 4 1.6 0. 0 • I 2 12 12 I' 0 ' FIG. lll. The seasonal dis1ribution of total phosphorns in the bottom water of Creal Pond in 1954, expressed in mit·ro~run1 uton1s P/L. The two Al'llphs shown are frequency polygons, described in detail in the legend for Figure 12. JANUARY FEBRUAR Y MARCH APRIL MAY JUNE JULY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER STN 12 0 30 Q.14 0 . 14 ~0. 1 4 IC£ I I 0 .24 o . 31 0.13 0 . 11 ~· 0 . 14 0 . 14 10 0 .29 o. 13 I 0. 19 _,.f5----;:.--­--'"6-­ 9B • 0 D A T A o . .s o 0.1 5 88 a.is 0 .21 Q.16 7B Q. 21 68 4 Q.27 I _.... _...,."' 0 .21 0 .33 13 -==-~---&--------­ 0 .4 9 9A 8A • 0 0 A T A 7A o.za 0 .23 4 o .27 0.7 9 Fie. 19. The seasonal distribution of inorganic phosphate in the bottom water of Great Pond in 1954, expre$sed in Microgram atoms P / L. The two graphs are frequency polygons, described in detail in the legend for Figure 12. MAP OF GREAT POND, FALMOUTH WITH CURRENT TRACKS ON EBB AND FLOOD TIDES UNDER CONDI ­TIONS OF r .< fvAILING SOUTH ­ WEST OR NORTH NORTH WEST WI NOS ~· s LEGEND: MA XI MUM CURRENT VELOCITIES OVER I KNOT _..-­OVER Q. S KNOT ,,...­LESS THAN 0.5 KNOT ;r 0.3 KM. 4 R D s 0 u N D FIG. 20. Map of Great Pond, Falmouth with current tracks on ebb and flood tides under conditions of prevailing southwest or north northwest winds. Maximum current velocities are indicated by arrows. JANUARY FEBRUARY MARCH APRIL MAY JUNE JU LY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER STN 12 7 9 \ 6 .7 / / / I I / I I /I I I / I I II • ' ..,.) 6. --0 /, ,,/ ,/,/ l,,r·c;8 I 10 6 . 6 6 .8 ,'/' I // I ,' 8 2 98 N O· D A T A 5 . 0 \ I I: I I\ 8 O.._ ...., BB 78 68 4 6 .2 6 . I $. I 6.1 5 .4 5 . • ' ' '. 0 I I G{ 16.4 I I I 16. I \ ~ ,~-,'-, \ '...~-) '\ \ 7 . I \ \ '\ I I 13 • . 7 5 . • 6 9 B.I I 9A BA 7A N 0 0 A T A 5. 2 ::[t~~'\, 7 . • 7 . 1 F1G. 21. The seasonal distribution of dissoh·t>d oxyµ:t>n in the bottom water of Creat Pond in 1954, expressed in milliliters per liter. The two graphs are frequency polygons, dt>snihed in deta'.I in the lt>g:end for Fiµ:ure 12. 7. 4 7 .0 b.3 5. 5 5.3 5. 2 5. 2 5.1 6.1 6.3 6.2 ', \ 7.2 I \ 6A ·· ··, ·~ I I 6 'I 4 I ' I I I \ 7 .. 6 .9 6. 2 6, I I I 5 .I • . 6 5. 1 6.1 6.0 6.1 \ \ I '·' I Seasonal Growth of Benthic Marine Plants TITRATION ALKALINITY v IU l HYDR£~4EN IONS v Ul8l , DISSOLVED OXYGEN ------...-..!.!,.:i .. .I 0•"·~· ~ ~ ------·--­19"<>-,,o;;-----il~-.,---r--.... /, .r~ .,, v 19 0 98) ---....___ '·---·-,,/' .......'":-,l ',, _/ i ............ l ............._______.,/ • F1G. 22. Seasonal changes in titration alkalinity in 1954 and pH and dissolved oxygen in the bottom water of the "transition'" zone of Great Pond in 1952-53 and 1954. PHYSI CAL FACTORS A.XO THE DISTRIBCTIO:\ OF BE:\THIC PLA'.\'TS Wave Action: When intense storms, occurring in the period from October through December, sweep over the estuary, waves have been observed to reach the bottom and to leave ripple marks. Poorly anchored benthic plants such as Agardhiella tenera, Gracil­aria verrucosa and Stilophora rhizoides were transported a considerable distance as ob· served on several occasions. The occurrence of severe northeasterly storms has probably been largely responsible for the dense accumulation of algal beds within the arcs of the sand spits of the western shore. These arcs are also the scene of substantial changes in sedimentation during such storms. The net movement of the poorly anchored species is mostly across the long axis of the estuary. Especially after a period of high winds, moribund plants of Gracilaria and Agardhiella were found in the sloughs. These plants, which presumably came from the major basin, differed from those which normally grow in the sloughs. Slough inhabiting plants have distinct modifications, such as short internodes and bushy habits. Storms frequently separate the dying elements in the plant communities of the basin and collect them into windrows on the beach. This action tends to provide space for the younger, healthier plants in the habitat and at the same time aids in the recycling of nutrient salts. Current: Paths taken by the flooding and ebbing tidal currents in Great Pond pass over areas abundantly populated by attached juvenile plants and by more fully de· veloped algae attached to cobble. In Figure 24 are shown small areas where deposits of coarse sand cobble coincide with the tracks of tidal current (Fig. 20) near shore, and in narrow channels. Three explanations are offered for the abundance of the plants in the areas with currents. Active tidal currents probably transport a large number of viable spores over these areas from both neritic water and the estuary. Second, due to the constant movement of tidal water, more nutrient salts per unit of time are probably supplied to the plants in the path of the currents (Odum and Hoskin, 1957). Third, coarse sediments for attachment generally are present where water movements are JANUARY FEBRUARY MARCH APRIL MAY JUNE JU LY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER STN 1, \ 7., IZ 8 09 80\ 11 • 00 , 70 T 10 7 .80 8 00 8 .03 7 . 91 / _ 9B M 0 0 A T l 8.00 7. 90 e.oe \,__8 .10 ~.) "~ ' " 8B 7B 8 .09 8. 10 8.05 8.0 8 6B 8 .22 8 . 10 8.12 8 .06 8 . 09 i'" 0~9~A(9' 8 .12 4 u \~ 806 8. 17 8 .13 8 .tO g 13 7.80 730 8 .2.2 8 .10 9A 8 .07 7 .90 7 .90 7 .99 8 .12 8 .02 80 8A M 0 0 A T A. 7A •o 8 .10 7 .90 7 .86 8 .17 8 .16 e .20 '·' e.o3 8 .28 91 0 6A , 90 7 " 7 ., 8 09 8 16 8 .18 8 12 • 19 • 10 4 / 8 01 1 eo e 05 7 90 8 00 '06 8 17 8 13 910 /B 00 Fie. 2.'3. The seasonal distribution of the pH of the bottom water of Great Pond in 1954. The two graphs are frequency polygons described in detail in the legendfor Figure 12. Seasonal Growth of Benthic Marine Plants rapid. Such coarse sediments in Great Pond include large sand grains, pebble, cobble and boulder. These coarse materials probably were deposited by glacial outwash streams and are being eroded from the shoreline of the estuary. They are now very suitable for plant attachment. Sedimentation: Butcher's data (Fig. 24) provide evidence for the development of sandy slopes strewn with pebble and shell on both shores, whereas the bottom is silt· covered. There is believed to be a correlation between the abundance of juvenile plants on the sandy shoals of the shoreline and the poorly anchored, well developed plants found on the basin floor. Poorly anchored, well developed plants growing in the major basin were nearly always attached to a sizeable sand grain, pebble or shell fragment. The cycle of events suggested by these findings starts with the attachment of spores and growth of juvenile plants on the coarse sediments of the sandy shoals. As these plants grow to some height, they are transported by wave action into deeper water where they accumulate in large masses. Eventually the algae die, after on~ or more seasons, and are cast upon the beach by wave action to decay. The decomposed plant debris releases nutrients for new growth. There is little doubt that this mechanism is quite necessary for the development of large accumulations of benthic plants on the basin floor. Poor conditions for attachment, and rapid burial, are often brought on by the sedi· mentation of fine-grained materials in a "gradient" zone (Rochford, 1952) . The fine silt enters the system in the run-off water, becomes flocculated by an electrolytic proc· ess and settles rapidly to the bottom of the sloughs. The rate of deposition is usually determined by the amount of salt present (Rochford, 1952 .1. The soft, yielding sediments of the two sloughs (Fig. 24) are incapable of supporting attached plants. Thus, the survival of benthic plants on the slough bottom is improbable due to lack of a stable sub. stratum. This is one reason that the floors of the two sloughs remain essentially barren of benthic plant life. Only the immediate banks and sandy shoreline slopes of the sloughs support attached algae and Ruppia beds. Salinity: The effects of salinity on plants are variously related to other variables in a tidal estuary. Local distribution in estuarine water may be controlled by salinity and temperature (Luther 195la, 195lb; Doty and Newhouse, 1954). Large changes in salin· ity require rather severe osmotic regulation in an organism (Sverdrup et al., 1942). Certain benthic marine plants such as Chondrus crispus, Corallina officinalis, Gelidium crinale, Nemalion multifidum, Petalonia fascia, Polysiphonia novae-angliae, Puncwria plantaginea, Ulva lactuca var. rigida and others were never observed within the "tidal" or "gradient" zones of Great Pond (Figs. 2, 25) but established themselves only on the riprap along the open coast. In contrast, Agardhiella tenera, Ectocarpus confervoides, E. siLiculosus, Gracilaria verrucosa, Polysiphonia denudata, Porphyra leucosticia, Punc· taria latifolia, Scytosiphon lomentaria, Spyridia filamentosa, Stilophora rhizoides and others (Table 2) established themselves in the major basin of Great Pond and a few plants grew in the brackish water of the sloughs. In the colder months species including Ectocarpus siliculosus, Punctaria latifolia and Stilophora rhizoides were found in brackish water areas from which they were conspicuously absent in the spring. These plants were much less tolerant of low salinities at the relatively high water temperatures obtained in mid to late spring. Doty and Newhouse (1954) reported similar plant dis· tributions in low salinity water when comparing summer and fall periods in the Oyster River. Due to a relatively uniform rainfall the year around (Table 10) seasonal (Fig. Seasonal Growth of Benthic Marine Plants 1f MAP OF GREAT POND, FALMOUTH SHOWING THE SED I MENTATION FEATURES OF THE ESTUARY I !: 00l(IJLOCCtA..AHTI ' ""~'"""'"" ------o:;~ ~~~"o!:Z!c':zc COAR$£ UNOAHD Ollt'1•NICOClll1$ 111i.o cn.occU1.»1n '"' ou~k~•""'w ""' ---04; IRACI( OOICJ.lilC 002[ "NO S.t.110 SAllO,.NOO'IGANICO(llllS ',:::~llLACI( Ol'IGA~C OOZE I SllCLL (J LOCCULANT) l.2 111 OllC Y OllGAH K; OOZE ANO $Hlll COAllS(S.o.NO? CHAll080T10"'1 IH[\.l.,tOAllUUNO .U lll ~ ""'"DITTO, · l"l..U$~G..NICOUllll& ------­ .'"""""""""'~:=.· .. f;fl[Y lll UO(OltG...NIC) s w ' Ol't[YWVD,SHEl.LS :: ,,..'""..~o,.,. 91.ACI( lllUO COMAll lCI :-~~-­~~~:[~':.!~":::~~) 2 0 Ill , SANO All OliHCLL 0.4,W l l .lCk OllG•NIC WUO l~ lll ANO rLAHT OCIR!S SAHQ,IHCLL, rcHLC I.I Ill ~ \ Fie. 24. Map of sediments of Great Pond based on unpublished data by Butcher. 138 Seasonal Growth of Benthic Marine Plants 0 z ~ 5 (/) r:s l­ a: ·W Q. (/) l­ a: rt ?; to (/) er w I-15 ~ ~ ~ I-10 0 ID Li.'.. 0 5 >­ t::: z :::i 0 ~ -........ ___..............._ __ _ ~ -.__.,_ __________.,.______.... _.-..... ......... """, -----, ',', ....................... "" ~ ~ ---~ .__i_,. __ ____1..., _ ,,.1 __v~_._, 1 1 \ s E c T 0 N s \ \ \ ' ,......-' .__..,.. ' 1\ '',,,_-z/ \ --~ \ -~--\ ' -..... liJ \ ~ '~i ~ ', ~ ~ ~ \ ..-------------·~------~,--­ [ \ \ z 0 N II \ \ \ \ ~"---..----~11~1 _,_,1v_... I\ I I \ I \ ::c \ ~ t:f -.1~ \ \ I I \ I 1­ I I I I \ \ \ I I 4 SA SB 7A 78 • t.\ 18 IO 13 II 13A 12 12E HYDROGRAPHIC STATIONS IN ESTUARY Fie. 25. The annual means and extremes of salinity of bottom water in Great Pond, Falmouth along the long axis of the estuary in 1952-53. Data are given in parts per thousand (0/00) . The zones (after Rochford, 1952) and sections are those indicated in Figure 2. The anomalous extremes of de· creased salinity at stations 6A and 6B are due to mixing of bottom water with surface water over a shoal sand bar at the seaward end of the basin (see Fig. 2) . TABLE 10 Monthly Precipitation in Centimeters for Nantucket Island, Massachusetts 1952 1953 1954 January 16.0 20.9 13.0 February 20.5 13.7 6.0 March 11.6 15.8 6.7 April 6.6 21.4 9.8 May 8.2 4.7 9.2 June 1.7 0.5 1.8 July 0.4 14.9 2.9 August 14.9 8.0 12.2 September 1.9 11.5 12.7 October 5.1 11.1 3.3 November 5.2 19.9 11.4 December 12.1 10.0 16.2 Annual 104.2 152.4 105.2 26) and daily ranges (Barlow, 1956) in salinity were nearly the same in Great Pond. Temperatures (Fig. 12), however, varied widely with the season. Where the dilution of sea water becomes too great, the normal growth of a species may not continue (Legendre, 1921). The sloughs are the last outpost for a few invading marine species of benthic algae and marine phanerogams and there is only a small total plant standing crop at all seasons (Fig. 7, Table 2). Considerable growths of Entero· morpha clathrata, E. compressa, E. intestinalis, Monostroma oxyspermum and Ulva lactuca var. latissima were found, but the two species Enteromorpha linza and E. minima, JANUARY FEBRUARY MARCH APRIL MAY JUNE JULY AUGUST SEPTEMBER OCTOBER NOVEMBER DECEMBER STN 12 29.2 Z•9 ' \ , "' ' ' \ , /,­I .:~~19jl9'I 29 0 \I 11 .7 , /@ \ r,;/. /,27 o\~<10 IO /_.._.__._'·' 26.9 ICE II 29 .0 28.• 25.6 266...____"\.. 199 199 293 ~ 274 17 . 1 281 ICE 10 29 .2 29.4 29. 7 2.9.4 21 .:uo.1 2.7. 8 29.4 26.9 28.9 --­ 26 . 8 98 N 0 0 A T A 29 . 2 29.3 29.8 25.1 \ 27.3 29.9 32.2 29.6 2 8 . 2 27 .6 ' -----~-------....------­29.4 ' ' ' 88 ',,-@._____ __......­ 7B 27.8 28.4 27.0 29. 3 29.7 27.8 26.6 30.2 U .4 30.8 30.7 &B 30.8 30.0 30.3 _30.4 29.2 30.6 30.7 31.2 31.5 SI. 2 4 30.6 30.9 31. 4 31.0 30.8 31.1 .30.7 30.• 30. I 30.8 31.2 31.5 ~25.2 23.5 23.1 23 .5 26.6 21 .7 I 22 .8 27.4 13 ... I ...____®--------·;,~-­ .... 30.8 21.2 211.0 19.0 21.6 29.6 zt.t IA --------®----------­ • 0 0 A T A 7A 30.3 28. 7 30.t 30.4 &A 30.0 24.8 .Sl .2 31.1 4 30 .6 30.9 31.4 31.0 30.8 31.1 30.7 30.I 30. I 30.8 31.2 31.4 31 .5 FIG. 26. The seasonal patterns of salinity in parts per thousand in the bottom water of Great Pond in 1954. For explanation of the coordinates see legend of Figure 12. were never found in brackish water. Reduced salinities were characteristic of the "tran­sition" zone (Figs. 25, 26; Table 2) yet the area supported a large portion of the total plant material in the estuary. Brackish water in the sloughs apparently greatly reduced growth and caused abnormal ties in some species of marine plants such as Ceramium rubrum, Gracilaria verrucosa, Punctaria latifolia, Scytosiphon lomentaria and Spyridia filamentosa. There is evidence that some species of benthic marine plants remain sterile if they inhabit brackish water (Levring, 1940; Hayren, 1940, and others). In those sections of Great Pond where the mean salinity of the bottom water was below 26100, sterility was common among the benthic algae. Probably the spores which arise on attached plants in brackish water come from plants in normal sea water habitats. Among the Phaeophy­ to no unilocular reproductive organs were observed in Punctaria latifolia or Scytosi­ phon lomentaria. The alga Ectocarpus siliculosus, appeared to be one of the few species in the division Phaeophyta which bore both unilocular and plurilocular reproductive structures in brackish water habitats. Nothing is known concerning the viability of the spores which this species produces in low salinity water. The identity of specimens of Ectocarpus siliculosus found in the "gradient" zone may be open to question. The gametangia were more attenuated and elongated than in the typical form. Although plants of Agardhiella tenera, Ceramium rubrum var. proliferum and Gracilaria verrucosa grew in the "transition" and "gradient" zones (Figs. 2, 26), all specimens examined were sterile. The species, Stilophora rhizoides, could be sepa­rated into two morphological types in the "tidal" and "transitional" zones and the "gradient" zone respectively. The latter was smaller and finer branched. Both forms were induced to liberate swarmers in the laboratory suggesting that the species is cap· able of reproducing in brackish water. Fruiting periodicity was studied in six species of marine Vaucheria (Blum and Con­over, 1953). Fruiting occurred in several species from January through May soon after the occurrence of spring tides. Thus fruiting appeared to be related to prolonged im­ mersion in water of much higher salinity than normal for the habitat. Water Density : Whereas there seemed to be no clear-cut correlation between tem­perature or salinity alone and the seasonal occurrence of some benthic marine plants, there was a correlation between density and plant growth. Some plant species grew in brackish water habitats in winter but were absent from these areas by midspring and summer. Species which followed such a pattern included Ectocarpus siliculosus, Graci­laria verrucosa (Figs. 5, 6) , Punctaria Latifolia, Scytosiphon lomentaria and Stiloplwra rhizoides (Figs. 4, 6). These plants grew in brackish water as long as the water density (Fig. 27) was above 20 sigma-t units (1.020 gm./ cc.) , but disappeared from the sloughs when the density fell below 20 sigma-t units. In the major basin, where growth was not affected, the density ranged from 20.6 in winter to 24.5 sigma-I units in summer. In the sloughs the density ranged over the year from 12.0 in summer to 22.5 simga-t units in winter. From these results it would seem that density is correlated with growth in some seasons more than either temperature or salinity alone. Exactly why there is a correla­tion of density and growth is not known. Boyle and Doty (1949) believed that toler­ances of marine algae to diluted sea water may decrease with increases in temperature. Since density is a function of temperature and salinity, the apparent correlation may be a matter of tolerance to low salinities at low temperatures. JANUAIH FEBRUARY OCTOBER N OVEMB~R DECEMBER STN 12 II 10 ;:..._;_;:_-._:::::-~~~-_ IJJ ~ !;) 98 N 0 D AT A "" c ;::s 88 ~ ~ / / 2 4,3 .... 78 c ~ /@ / / 24 ;;:. 61 / / 24.1 c 4 -tl:l ~ ;::s ;;:. 13 14. 0 ~ i:;· 9A ~ ~ !;) / .... / / ;:;· 8A / ~ 22 NO 0 A T A 7A ~ 24 .1 "" ~ 6A 22. 4 , z 4.5 I I 4 I 22 .• 22.1 24 .1 Fie. 27. Water density in sil.(ma-t units in bottom water of Great Pond in 1954. The two graphs shown are frequency polygons, described in detail in the legend for Fii:i:nr1> 12. 142 BIOTIC FACTORS AND SEASONAL GROWTH Animal grazing apparently begins or takes place immediately after there is an initia­tion of phytoplankton growth (H. W. Harvey, 1955). With benthic marine plants in Great Pond, there is a period in early growth when grazing and epiphytism appear to he slight. Immature plants of Agardhiella tenera, Ceramium rubrum, Gracilaria verrucosa and Polysiphonia novae-angliae were studied periodically to observe the relationship be­ tween growth and attack by other organisms. Very little grazing by the snails, Bittium alternatum or Nassa obsoleta, was witnessed. Only slight coverage by algal epiphytes was noted. On well developed plants about three weeks old many evidences of grazing were found, and epiphytes were noticeable with the unaided eye. Epiphytism and graz­ ing were marked when the species populations had attained or passed their peak produc­ tion levels. Summer populations under conditions of high illumination and temperature were those most severely affected, but Polysiphonia novae-angliae showed attack by epiphytes and grazing early in its growth period. During the time of most active growth, attached plants in normal sea water did not seem to be seriously reduced by animal grazers. Ob­ servations on some marine plant species in both normal sea water and brackish water indicated that many more epiphytes had developed, and grazing by gastropods was far more intense on plants growing in more brackish water habitats than in normal sea water. Apparently, plants, whose optimum growth processes take place in normal sea water, are less able to resist attack and predation during early growth in a brackish en· vironment. Although morphological differences provide adequate grounds for establishing va­ rieties in algae, other evidence may also be used. Seasonal growth data on varieties of Ulva lactuca provided evidence for a separation. In Great Pond Ulva lactuca var. latissima and Ulva lactuca var. rigida (Figs. 4, 6) exhibited dissimilar seasonal growth patterns. STORMS, EXTREME TEMPERATURES AND lcING Winds and Tropical Hurricanes: During the entire 1952-53 survey period water temperatures were mild (Fig. 11) and no violent storms (Fig. 10) were recorded. Thin sheets of ice formed on the sloughs only for a brief two-week period in February. According to the small deviation from the 62-year norm for the area {Table 10) rainfall in the survey years was fairly normal except in January and April, 1953. The illumina­tion received in the summer of 1952 was somewhat above average. Measurements of standing crops of the dominant benthic marine plants during 1952-53 were believed to be representative for tidal estuaries on the Cape, according to sample surveys of several other estuaries along the southern coast of the peninsula. From June 1952 to August 1954 average climatic conditions prevailed in New England. On August 31, 1954, a tropical hurricane swept over southern New England, bringing over 180 km. per hour wind velocity and 13 cm. of rainfall to the south Cape region. This storm was followed by a second one a week later. A series of salinity measurements were made in Great Pond to observe what had taken place (Table 11). Following the first hurricane, "Carol," the salinity of waters in Vineyard Sound increased only about 1%0 at the seaward end of the major basin. Then due to the heavy rainfall accompanying TABLE 11 Salinity in Parts per Thousand in Great Pond Before and After Hurricanes "Carol" and "Edna", September, 1954 (Figure 2) Slaliom Le,·el July 23 Aug. ll Aug. 31 Sept. 8 Sept. 11 Sept. 16 Sept. 27 6A surface 30.6 27.1 Hurricane 31.0 Hurricane 30.5 31.0 bottom 30.7 28.4 "Carol" 31.0 "Edna" 30.7 31.0 9A surface 19.7 20.5 16.4 26.l 28.6 bottom 29.7 25.3 30.0 26.l 28.1 10 surface 28.6 8.5 4.7 23.2 B.O bottom 29.4 26.9 27.0 28.9 30.1 surface 30.8 29.9 31.2 30.5 31.0 bottom 30.8 30.1 31.2 30.8 31.2 the two storms, salinities decreased in the surface waters of the sloughs as the river dis­charge rate nearly doubled the normal amount for that season. These changes in salinity had little effect on the benthic vegetation according to a cursory examination. Precipi­ tation in hurricanes had a greater effect on salinity than high storm tides. At any rate, such changes are quite temporary. Wave action on the riprap disarranged the boulders, stripped most of the larger plants off the rocks and removed large amounts of the basin-inhabiting algae from Great Pond. In the summer of 1956 quadrat studies were made on transect 6 (section II, Fig. 2) to determine the fate of benthic plant populations. To all appearances the former luxuri­ance of the benthic vegetation had been restored. Extreme Low Temperatures and Icing: In 1955, ice completely covered Great Pond to a depth of 7 cm. and persisted from January 20 through February 12. In 1956 icing conditions were even more severe. An ice sheet, which had a maximum thickness of 12 cm., remained for six weeks over the estuary. Inspection of the riprap at the mouth of the estuary showed ice-scouring had removed large quantities of Chondrus crispus, Corallina officinalis and Fucus spiralis but had done little damage to populations of Fucus vesiculosus. The latter occupies a sublittoral position on the riprap. Ice-scouring affected the attached plants on most Falmouth jetties during this period also, as large sheets of fringe-ice lined the shore. For the first time in nine years {records of the Woods Hole Oceanographic Institution), Buzzards Bay was covered by large sheets of ice which broke off during the thaw as large chunks passed through Woods Hole and entered Vine­ yard Sound. The effect of low water temperatures and subsequent icing during this period also may have been responsible for the near extinction of Gelidium crinale. This species occupied a habitat in the intertidal belt where ice-scouring was severe. At this time the alga, Bostrychia rivularis, disappeared from section IV and could not be found anywhere in the estuary through the summer of 1956. The alga, Punctaria plantaginea, so abun­dant in the winter of 1952-53 did not appear along the Falmouth coastline in 1954 or the following winters through 1957. No explanation could be given for disappearance of a species that is found characteristically in arctic waters in a sublittoral habitat little affected by icing (Kjellman, 1883). SIMILARITIES AND DIFFERENCES IN SUCCESSIVE YEARS A question arises in a two-year study such as the present one in which the biotic survey was made one year and the hydrological survey the second year. Can one logically relate the biological data for 1952-53 to oceanographic data for 1954? Since some properties JANU ARY FEBRUARY SEPTEMBER OC TOBER NOVE WBER OEC(\18£111 STATIONS TOTAL PH OS PH ORUS ,1.,~1.10.• '-AI) 0 ,6 0.1 ;;;:v •.• ·' •.• \0 J:· '·''·' " =;-;j'.~ " ~ INORGA NIC o,•• o,n 0.1 3 0.11 .. -o,z o,z , o.iz Fie. 28. Total phosphorus and inorganic phosphate in the surface water of Great Pond in 1954,expressed in microgram·atoms P/1. The two graphs shown are frequency polygons. Stations were located at the center in 1951 and along the eastern side of the pond in 1954. Refer to Figure 12. SEPT£ 1i1 8£A OCTO BE A N OY E M BER 0£C£ \,1 8£R •.• \.'.V •.• ··~· ;_, . ·•.._ ,' TOTAL PHOSPHORUS .. '·' ~·· '·' ('~~--::~---­ @ o.' \~&> 1,0 t.• o. ~ ZJ) u 1.0 -----... _ Fie. 29. Total phosphorus and inorganic phosphate in the surface water of Great Pond in 1950 (data after Hulbert), expressed in m'crogiam atoms P / L. Compare these data with those of Figure28 for 1954. The legend is the same as for Figure 28. Seasonal Growth of Benthic Marine Plants were measured in both years, a partial answer can be given. In Figure 22 dissolved oxygen and pH for 1952-53 and 1954 have been plotted. The same trends are present. There is also a close parallel between the 1950 and 1954 seasonal distribution of surface phosphorus. Data for 1950 are from Hulbert; Figures 28 and 29. Apparently the regu­lar factors represented by the patterns of oxygen, pH, and phosphorus are repeated with similar amplitude from year to year. The plant populations, however, are not exactly similar (See section on changes from year to year). It may be inferred that irregular factors and unusual departures are responsible for the changes in seasonal growth and reproduction of the plants from one year to the next. LITER.ATURE CITED American Public Health Association and American Water Works Association. 1946. Standard Methods for the Examination of Water and Sewage. 9th Ed. Lancaster, Pa., Lancaster Press,286 pp. American Public Health Association, American Water Works Association and Federal Sewage In­dustrial Wastes Association. 1955. Standard Methods for the Examination of Water, Sewage, and Industrial Wastes. 10th Ed. Baltimore, Md., Waverly Press, Ind., 522 pp. Barlow, J. P. 1956. The hydrography of Great Pond, a tidal estuary. J. Mar. Re8., 15(3): 193-203. Bell, H.P. 1927. Seasonal disappearance of certain marine algae. Trans. N. S. Inst. Sci., 17(1): l-5. Bell, H. P. and C. MacFarlane. 1933. The marine algae of the maritime provinces of Canada, II. AStudy of Their Ecology. Canad. J. Res., 9: 280-293. Bernatowicz, Albert J. 1950. Seasonal aspects of the Bermuda algal flora. Pap. Mich. Acad. Sci., 36: 3-8. ----. l952a. Seasonal changes in the marine algal flora of Bermuda. Univ. :\Iich., Ann Arbor,Mich. Dissertation Series Pub!., 186 pp. ----. 1952b. Marine monocotyledonous plant8 of Bermuda. Bull. Mar. Sci. Gulf & Caribb.,2(1) : 338-345. Black, W. A. B. 1950. The seasonal variation in weight and chemical compo8ition of the common British Laminariaceae. J. Mar. Biol. Ass. U. K., 29: 45-72. Blinks, L. R. 1955. Photosynthesis and productivity of littoral marine algae. J. l\Iar. Res., 14(4):363-373. Blum, John L. and John T. Conover. 1953. New or noteworthy Vaucheriae from New England salt marshes. Biol. Bull., 105(3): 395-401. Boyle. M. and M. S. Doty. 1949. The tolerance of stenohaline forms to diluted sea water. Biol. Bull., 97: 232. Boysen-Jensen, B. 1941. Studies concerning the organic matter of sea bottoms. Rept. Danish Biol. Sta., 22: 5-39. Braun-Blanquet, J. 1932. Plant Sociology. New York, McGraw Hill Book Co., 439 pp. Dexter, R. W. 1953. Recession of eelgrass at Cape Ann, Mass., Ecology, 13(1): 229-231. Doty, M. S. and J. Newhouse. 1954. The distribution of marine algae into estuarine waters. Amer. J. Bot., 41(6): 508-515. Ehrke, Gerhard. 1929. Die Einwirkung der Temperatur und des Lichtes auf die Atmung und Assimi­lation der Meeresalgen. Planta, 9: 631--638. ----. 1931. Uber die Wirkung der Temperatur und des Lichtes auf die Atmiing und Assimila­tion einiger Meeresund Siisswasser Algen. l'lanta, 13: 221-310. Fassett, N. C. 1928. The vegetation of the estuaries of North Eastern North America. Proc. Bost. Soc. Nat. Hist., 39(3): 73-130. Feldmann, Jean, 1937. Recherches sur la vegetation marine de la mediterranee. La Cote des Alberes. Rev. Algol., lO: 1-339. ----. 1951. Ecology of marine algae, in Manual of Phycology. Waltham, Mass., Chronica Botanica Co.; pp. 313-334. Fogg, G. E. 1953. The Metabolism of Algae. London, England, Methuen & Co. Ltd., ix+ 149 pp. Ford, Wm. L. 1949. Seagoing photoelectric colorimeter. Analyt. Chem., 22: 1431-1435. 146 Seasonal Growth of Benthic Marine Plants Fritsch, F. E. 1906. Problems in aquatic biology with special reference to the study of algal periodic­ ity. New Phytol., 5(7): 149-167. ----. 1945. The Structure and Reproduction of the Algae. Cambridge, England, CambridgeUniv. Press, Vol. II. xiv+ 939 pp. Funk, G. 1927. Die Algenvegetation des Golfs von Neapel nach Neueren Okologischen Untersuch­ungen. Pubbl. Staz. zoo!. Napoli, 7(Suppl.): 1-507. ----. 1954. Beitrage zur Kenntnis der Meeresalgen von Neapel. Pubbl. Staz. zoo!. Napoli, 25 (Suppl.): 1-178. Gessner, Fritz. 1952. Der Druck in Seiner Bedeutung fur das Wachstum submesser Wasserpfianzen. Planta., 40: 391-402. ----.. 1955. Hydrobotanik, die Physiologischen Grundlagen der Pfianzenverbreitung im Was­ ser. I. Energiehaushalt. Berlin, Germany, V.E.B. Deutscher Verlag der Wissenschaften, 517 pp.Gunter, G. 1957. Temperature. Mem . geol. Soc. Amer., 67. Treatise on Marine Ecology and Paleo·ecology., 1: 159-184. Harvey, H. W. 1948. The estimation of phosphate and total phosphorus in sea waters. J. Mar. Biol.Assoc. U.K., 27(21: 3.37-359. ----. 1955. The Chemistry and Fertility of Sea Waters. Cambridge, England, Univ. Press, vii+ 224 pp. Harvey, W. H. 1851. Nereis Boreali-Americana I. Melanospermeae. Smithson. Contr. Know!. Pt. I, 3(4): 1-67. Hasegawa, Y. and E. Fukuhara. 1956. On the seasonal and local variations in the weight of fronds. Bull. Hokkaido Reg. Fish. Res. Lab., 12: 29-39. Hayren, E. 1940. Uber die Meersalgen der Insel Hogland im Finischen Meerbusen. Acta, phytogeogr. suec., 13: 50-62. Hedgpeth, J. W. 1957. Marine Biogeography. Mem. geol. Soc. Amer. 67. Treatise Marine Ecologyand Paleontology, 1: 359-382. Hoffmann, C. 1929. Die Atmung der Meeresalgen und ihre Beziehung zum Salzgehald. Jb. wiss. Bot., 71: 214-268. Holmes, R. W. 1957. Solar Radiation, Submarine Daylight, and Photosynthesis. Mem. geol. Soc.Amer. 67. Treatise on Marine Ecology and Paleoecology, 1: 109-128.Hulburt, E. M. 1956. Distribution of phosphorus in Great Pond, Massachusetts. J. Mar. Res., 15(3):181-192. Institute of Seaweed Research. 1953. Annual Report, 1953. Ketchum, B. H. 1939. The development and restoration of deficiencies in the phosphorus and nitrogen composition of unicellular plants. J. cell. comp. Physiol., 13: 373-381.----.1954. Mineral nutrition of phytoplankton. Annu. Rev. Pl. Physiol., 5: 55-74.Kjellman, F. R. 1883. Algae of the Arctic Sea. K. svenska Vetensk. Akad. Hand!., N. F., 20(5):1-349. Klugh, A. B. and J. R. Martin. 1927. The growth-rate of certain marine algae in relation to depth of submergence. Ecology, 8: 221-231. Knudsen, Martin (trans. by M. Oxner) . 1946. The Determination of chlorinity by the Knudsenmethod (hydrographic tables by M. Knudsen). Woods Hole Oceanographic Institution, WoodsHole, Mass., 63 pp. Knight, Margery and Mary W. Parke. 1931. Manx algae. Liverpool. Mar. Biol. Comm. Mem., 30: 1-155.----. 1950. A biological study of Fucus vesiculosis L. and F. serratus L. J. Mar. Biol. Assoc. U.K., 29 : 439-514. Kylin, Harald. 1907. Studien iiber die Algen flora der Schwedischen Westkiiste. Akad. Afh. Upsala, 288 pp. Legendre, R. 1921. Influence de la salinite de l'eau de mer sur !'assimilation chlorophyllienne desAlgues. C. R. Soc. Biol., Paris, 85: 222-224. Levring, Thor. 1940. Studien iiber die Algenvegetation von Belkinge Siidschweden. Lund, Sweden, Akad. Abhandl., 178 pp. Lodge, S. M. 1948. Algal growth in the absence of Patella on an experimental strip of foreshore,Port St. Mary, Isle of Man. Proc. Lpool. biol. Soc., 56: 78--83.Luther, Hans. 1951. Verbreitung und Okologie der hotheren Wasserpflanzen im Brackwasser der Ekenaesgegend in Suedfinnland. Tei! I. Acta bot. fenn., 49: 1-232; 50: 1-370.Macfarlane, C. 1952. A Survey of certain seaweeds of commercial importance in southwest NovaScotia. Canad. J. Bot., 30: 78--97. Odum, E. P. 1953. Fundamentals of Ecology. Philadelphia, W. B. Saunders Co., 384 pp. Seasonal Growth of Benthic Marine Plants Odum, H. T. and C. M. Hoskins. 1957. Metabolism of a laboratory stream microcosm. Puhl. Inst. Mar. Sci. Univ. Tex., 4(2): 115-133. Petersen, C. G. J. 1912. Uber Menge und Jahresproduktion der Bentospflanzen an den westuropais· chen Kiisten. Int. Rev. Hydrobiol, 5: 47-52. Pratt, D. M. 1950. Experiments in the fertilization of a salt water pond. I. \lar. Res., 8(1): 36-59. Printz, H. 1939. Uber die Kohlensaure Assimilation der Meeresalgen in ver<;chiedenen Tiefen. Skr. norske Vidensk. Akad., 5: 1-272. ----. 1950. Seasonal growth and production of dry matter in Ascophyllum nodosum . Arh. norkse Vidensk. Akad., 4: 1-15. Rees, T. K. 1934. Algal migration on the Gower coast. Proc. Swansea sci. Fld. Nat. Soc., 1: 235-239. ----. 1935. The l\larine Algae of Lough lne. J. Ecol., 23: 69-133. Rochford, D. I. 1952. Studies in Australian estuarine hydrology. I. Introductory and comparath·e features. Aust. J. Mar. Fresh. Res.,2(1): 1-116. Sargent, M. and L. Lantrip. 1952. Photosynthesis, growth and translocation in giant kelp. Amer. J. Bot., 39: 99-107. Setchell, W. A. 1915. The law of temperature connected with the distribution of the marine algae. Ann. M. bot. Gan., 2: 287-305. ----. 1917. Geographical distribution of the marine algae. Science, 45 (11571 : 197-204. ----. 1920a. The temperature interval in the geographical distribution of marine algae. Ibid. 42 (1339): 187-190. ----. 1920b. Stenothermy and zone invasion. Arner. Nat., 44: 385-397. ----. 1932. Macrocystis and its holdfasts. Univ. Calif. Puhl. Bot., 16: 445--492. Smith, G. M. 1947. On the reproduction of some Pacific coast species of Ulra. Amer. I. Bot., 34(2): 80--87. Svedelius, Nils. 1906. Uber die Algenvegetation eines ceylonischen Korallenriffes. in Bot. Stud. till F. R. Kjellman. Upsala, pp. 184-220. Sverdrup, H. W., M. W. Johnson, and R. H. Fleming. 1942. The Oceans, Their Physics, Chemistry and General Biology. New York, Prentice-Hall, Inc., x + 1087 pp. Taylor, W.R. 1933. Epidemic among zostera colonies. Rhodora, 35: 186. ----. 1957. Marine algae of the northeastern coast of North America. 2nd. Ed. Univ. ~lich. Sci. Ser., Ann Arbor, Mich., Univ. Mich. Press, viii + 509 pp. Truog, E., and A. H. Meyer. 1929. Improvements in the Deniges colorimetric method for phosphorus and arsenic. lndustr. Engng. Chern. (Anal. l, 1: 136-139. Vinogradov, A. P. (Trans. by J. Efron and J. K. Setlow). 1953. The Elementary Chemical Com­position of Marine Organisms. New Haven, Conn., Mern. Sears Foundation \lar. Res. No. 2, xiv+ 647 pp. Walker, F. T. 1952. Sublittoral seaweed surveys: Dunbar to East Castle, East Scotland. J. Ecol., 40: 74-83. Weston, R. S. 1905. Notes on the deterrninat'on of nitrogen as n:trites in waters. J. :1.mer. Chern. Soc., 27: 281-282. Artificially Formed Mud Balls Loms S. KoRNICKER, CARL H. OPPENHEIMER, AND JOHN T. CONOVER Institute oJ111arine Science The Un:versity of Texas Port Aransas, Texas Armored mud balls were described by Pettijohn (1949, p. 146) as large, subspherical balls of clay that are coated or armored with fine gravel. Bell (1940) showed that gravel­armored mud bails originated as clay chunks which had been eroded from rapidly under­ cut stream banks and which had acquired a gravel armor upon rolling down stream. Richter (1926) illustrated mud balls armored with shells. Shell-armored mud balls (Fig. 1) are fairly common on the beaches along the Laguna Frc. l. A typical shell-armored mud ball from beach on Aransas Bay, Texas. Mud ball is 8 cm. long. Origin is questionable. Shells on surface have been identified by W. Rice as follows: Cerithium vari­able. Modulus modulus, Bittium varium, Melanella sp.. Nassarius vibex, Polinices duplicatus, Ceri· thidea pliculosa, Odostomia sp. (broken) , Cardita floridana, 2 unidentified gastropods. A fish vertebra is also exposed. Photographed specimen collected by C. M. Hoskin. Madre, Texas. They are usually oval shaped with a long axis averaging about 8 cm. (-3 Phi) . These balls were observed to roll to and fro on the beach with the long axis remaining parallel to the water's edge. While migrating up and down the beach with the swash and tides, the mud balls pick up shells which adhere to their surface. During the construction and subsequent deepening of the lntracoastal waterway, which extends the length of the Laguna Madre, sand and clay dredged from the channel were deposited on either side. The dredged material forms small islands, called spoil Artificially Formed Mud Balls l.anks. and flat lying strata on the shores of the Laguna Madre. Dredged material along many parts of the Laguna Madre is co\·ered by sand dunes and vegetation which mask the artificial origin of the deposit. Waves created by wind and large ship» which use the intracoastal waterway break against the dredge-deposited sediment to form beach cliffs. Exposed in many of these cliffs are mud balls similar to those found on Laguna Madre beaches. The unexposed surface of the mud balls in these cliffs are not as round as the surface projecting from the cliff indicating that the continuous lapping of waves against the cliff tends to further round the projecting side of the mud ball. It is postulated that the waves eventually undercut the cliff and free the mud balls, which are then further rounded by rolling to and fro on the beach. Accumulations of mud balls at the base of several cliffs lend support to this hypothesis. Rounding and elongation on the beach probably is due partly to abrasion and partly to water erosion. The authors Yisited a dredge in operation in order to determine if the mud balls found in the dredged material were derived from sediment containing mud balls or were formed during the dredging operation. The dredge operated by laterally boring into the sediment with a steel screw several feet in diameter. The removed material was sluiced through a pipe about 16 inches in diameter and several hundred feet long. Extra lengths of pipe were added as either the place of dredging changed or the dumping area became filled. At the time of the visit, the dredge was removing an embankment whose top was just a few inches above water level. The sediment being removed seemed homogenous with no evidence of the presence of mud balls. The sediment ejected from the pipe end, how· ever, contained abundant mud balls (Fig. 2). During the boring operation the sediment F1G. 2. l\lud ball accumulation at end of dredge pipe-line. :\Iud balls awrage about 10 cm. in length. Top background is stream of water from pipe whose rnd is not discernable in upper left corner of picture. Artificially Formed Mud Balls was broken into chunks which became more or less rounded by abrasion and water action as they tumbled through the pipe. A few of the mud balls leaving the pipe had shells adhering to their outer surface, hut most were barren of shells. The shells could have been picked up by the mud halls as they travelled through the pipe, but they also may have been encased in the sediment and merely exposed during the mud ball fomation. A dredge of the type observed is probably capable of forming mud balls from any sediment it is able to cut into, providing the sediment contains sufficient clay, which seems to act as a binder. An artificial mud ball was analyzed and found to contain about 80 per cent clay. The clay was greenish, and according to Dr. Edward C. Jonas (written communication, 1958) , who examined it by means of X-ray, it had a micaceous crystal­ line structure. Evidently many mud balls found on beaches of the Laguna Madre and Texas bays are formed artificially by dredges. These mud balls are eroded from undercut spoil bank cliffs. Some collect a shell armor when rolling to and fro on a beach. Rolling tends to elongate the mud balls. The long axis usually parallels the water edge and is perpendic­ular to current or wave action. W. Armstrong Price and T. D. Cook have observed mud balls formed by erosion of naturally deposited clay cropping out along the Laguna Madre (personal communica· tion, 1958). The balls were most abundant immediately after a storm, and, according to Cook, naturally formed mud balls in large numbers probably are formed only during periods of rapid erosion, although, any clay shoreline can provide mud balls . Dredged waterways and canals form a network along the coastal areas of the United States. Mud balls found in coastal areas, therefore, may have formed in the dredging process. The writers were not able to identify with assurance isolated mud balls found on beaches as having been formed by natural or artificial processes. A large number of mud halls found in place in unlithified layered sediments near dredged channels may be a clue that the deposit is probably spoil material. McKee ( 1958, p. 1735) interpreted rounded clay balls and cross bedding observed in strata on an island situated in the lagoon adjacent to the Port Aransas causeway, near a major channel, as indicative of strong current action in this part of the lagoon. The proximity of the channel suggests that the clay balls and cross bedding may not have been formed naturally, but are the result of dredging. Mapping of spoil areas where artificial mud balls occur as well as those areas where clay cliffs of Recent or Pleistocene age are exposed and natural mud balls may be ex· pected, would be helpful for determining the origin of mud balls found in these areas. LITERATURE CITED Bell, H. S. 1940. Armored mud balls-their origin, properties and role in sedimentation. J. Geol. 48: 1-31. McKee, Edwin D. 1958. Primary structures in some recent sediments. Bull. Amer. Assoc. Petrol. Geol., 41: 1704--1747. Pettijohn, F. J. 1949. Sedimentary Rocks. New York, Harper and Brothers, 1-526 pp. Richter, R. 1926. Flachseebeobachtungen zur Paleontologie und Geologie. XV-XVI, Senckenbergiana, 8: 297-315. Notes on Blanquilla Reef, the Most Northerly Coral Formation in the Western Gulf of Mexico DONALD R. MOORE Gulf Coast Research laboratory, Ocean Springs, Mississippi Introduction Published accounts of coral reefs in the western Gulf of Mexico are rare, and little is known about this region except for a paper by Heilprin (1891) on the reef at Vera Cruz. Joubin (1912) published a map of the coral reefs of the Gulf based on Heilprin's work and on various unpublished sources. Smith (1954) published a map showing the coral reefs in the Gulf and also listed the corals found at Vera Cruz by Heilprin. The present account is the first addition in nearly half a century to the knowledge of coral reefs in the western Gulf. Blanquilla Reef is a rather small coral reef located near the tip of Caho Rojo in the western Gulf of Mexico. It lies offshore from Laguna de Tamiahua and about sixty miles from Tampico. Territorially it is part of the State of Vera Cruz. It is of more than ordinary interest. because it is apparently the northernmost surface reef in the western Gulf. There are rocky substrates further north which do not have reefs. Thus Caho Rojo forms the dividing line between two faunal provinces; tropical to the south, and sub­tropical or warm temperature to the north. Although currents are poorly known in this part of the Gulf, it is apparent that a branch of warm, clear oceanic water permits the growth of reef corals in this area. The author and H. H. Hildebrand visited Blanquilla Reef April 26 and 27, 1955. The trip from the mainland was made in a small skiff with an outboard motor, which held little besides its passengers. The author contented himself with a small collection of invertebrates, and with some observations on the structure of the reef. These observa­tions were made while wading in shallow water, and swimming and diving in deeper water. Observations Blanquilla Reef lies approximately three miles southeast of Caho Rojo (21°33'N.­97017'W.). The depth is about fourteen fathoms between the reef and the mainland, and fairly deep water is found very close to the lee side of the reef. A rough estimate of the size of the reef would be about eight hundred yards long in a north-south di­rection, and five or six hundred yards wide. On the highest point, the Mexican govern­ment has erected a concrete monument with an automatic light as an aid to navigation. The surrounding area is awash at low tide and is paved with loose coral cobbles. There is an extensive shallow water area around the high portion which gradually deepens in all directions toward the reef edge. This edge is poorly defined on the windward (east) side, but there is a small cliff with about a ten foot drop on the west side. Notes on Blanquilla Reef Live coral was rare on the seaward side, but became more common along the west side and was flourishing along the edge of the cliff. Small "heads" of Siderastrea siderea and Diploria clivosa were living in shallow water not far from the monument. Acropora palmata was fairly common in slightly deeper water close to the western edge. A. cer· vicornis was also found in this area, but in much smaller quantity than A. palmata. Montastrea annularis formed large massive colonies along the cliff edge, and was quite common. M. cavernosa was uncommon, and all colonies examined were found in deep water beyond the cliff. A few small Porites furcata were living in the Acropora zone. A very flourishing growth of coral heads in deep water below the cliff was made up mainly of two species of Diploria (strigosa and labyrinthijorms?). This reef differs considerably in structure from the reefs off the Florida Keys. It faces a long reach of open water, over a thousand miles in the northeastern sector. Storm waves from this direction build up to near maximum power and velocity. As a result, the seaward side is almost a flat expanse of bare coral rock with a very gradual slope toward the open sea. It is covered by a pink encrusting alga, and the burrows of a sea urchin. The Florida reefs do not have such a wide reach of open water and are pro­ tected somewhat except in hurricane seas, by the proximity of the Bahama Islands and their associated shallow banks. The seaward side of such reefs has considerable relief and quite a bit of live coral, and tends to slope down at a fairly steep angle. The land­ward side of Blanquilla Reef is also quite different from Florida reefs. The bottom descends to about ten fathoms within about a hundred yards of the western edge, while the Florida reefs usually have sand bottom from ten to thirty feet deep on the landward side. Differences in reef structure are apparently attributable to different physical condi­tions between the two areas. Most of the invertebrates were small and easily overlooked, compared to the massive colonial corals. However, a conspicuous constituent of the fauna, the sea urchin Echino­metra lucunter, was very numerous on the seaward sloping portion of the reef. All were living in burrows which they had excavated by scraping algae from the rock. This burrowing habit enables the urchin to live on the exposed seaward side despite the action of the larger waves. Dr. Voss of the Miami Marine Laboratory indicates that Echinometra lucunter forms a distinct zone in semi-protected shores of the Florida Keys and the Bahamas. Another large sea urchin, Diadema antillarum, was found under rocks or along the edge of the cliff on the west side, but it was not common. Other echinoids were Eucidaris tribuloides found under rocks, Tripneustes esculentus scattered in deep water on the west side, and a starfish, Linckia guildingi, on rocks. Brittle stars were ob­served, but none were identified with certainty. A small armoured holothurian probably Thyonepsolus braziliensis, a shallow water West Indian species, was found when a piece of coral rock was closely examined in the spring of 1958. Some traces of color remained. It was apparently bright red when alive, and was evidently overlooked due to its simi­larity· in appearance to Homotrema rubra. Alcyonarians were scattered and rather scarce. Only three species were observed living on the reef. They were Antillogorgia acerosa, Plexaura flexuosa, and Plexaurella dichotoma. The remaining animals identified were all mollusks except for one unusual protozoan, Homotrema rubra. This is a bright red encrusting foraminiferan found attached to the Notes on Blanquilla. Reef underside of rubble on reefs. It is large for a protozoan and conspicuous, for its color contrasts with the light colored rock. Although common on reefs in the West Indian region, it is not confined to shallow water. and may be found liYing to depths of forty fathoms or more off the Alabama coast I personal observation I. The Mollusca identified were also common West Indian species found on coral reefs. The gastropods Thais deltoidea and Astra~a sp. were li\·ing out in the open on coral rocks, but other species such as Drupa nodulosa, Cerithium literatum, and Pisania pusio were only found under rocks. Pelecypods included the widespread species Arcopsis adamsi, I sognomon radiata, Lima seabra., and Lithophaga aristata. Some gastropods common on Florida reefs that were not found on Blanquilla include Astraea ca.elata, Conus regius, Cypraea. cinerea. and Mitra barbadensis. Corallipphila abbrei·iata is usually found liYing on the coraL Acropora palmata, but none could be found on this reef. Strombus gigas, one of the largest of the West Indian gastropods. was also absent, but a small specimen of the giant horse conch. Pleuroploca gigantea, was collected. The fauna of Blanquilla Reef appears scanty when compared with that of the Florida reefs. However, this reef is a small one and is lacking a number of ecological zones in which corals and other invertebrates are found living in the Florida Keys. A number of corals in the Florida area occur usualh· in shallow water behind the main line of reefs. Lobos Island, a few miles south of Blanquilla Reef, may have a larger and more varied fauna, and should prow a fruitful field for inwstigation. Another factor to be con­sidered is that this list of fauna is based on a cursory examination and hurried collection. The close relationship of the Blanquilla Reef fauna with the rest of the West Indies is without question. Almost all of the species listed are also found on the Florida reefs and much of the West Indian region. :\early all of the species under consideration haw pelagic larvae and establish themseh-es where,·er natural conditions are favorable. Al­though conditions are apparently not suitable for the best dewlopment of a coral reef off Caho Rojo, the minimum requirements are met for at least some tropical species. List of Species Identified :\Iany of the larger animals were identified in the field but most of the smaller organ­isms were collected and identified later. In the case of corals of which identity was doubtfuL an attempt was made to collect a small colony for more accurate identification in the laboratory. In the following list an asterisk ind:cates field determinations only. The author alone is responsible for the names in the list. except the Alcyonarians. which were identified by F. M. Bayer. The specimens collected. except the Alcyonarians and a holothurian. were deposited with the i1n-ertebrate collection at the Institute of Marine Science, Port Aransas, Texas. The _.\Jcyonarians were sent to the l'nited States :\ational Museum. PROTOZOA Foraminifera Homotrema rubra Lamarck Notes on Blanquilla Reef COELENTERATA Corals Acropora palmata Lamarck Acropora cen;icomis Lamarck Siderastrea siderea Ellis &Solander Diploria clivosa Ellis & Solander Dipwria strigosa Dana Diploria labyrinthiformis Linne? * Montastrea annularis Ellis & Solander Montastrea cavernosa Linne* Astrangia brasiliensis Vaughn? .llycetophyllia lamarckana Edwards &Haime? Porites furcata Lamarck * Alcyonarians (Identified by F. M. Bayer I Gastropods Pelecypods Antillogorgia acerosa Pallas Plexaura flexuosa Lamouroux Plexaurella dichotoma Esper MOLLl"SCA Mitra nodulosa Gmelin Nitidella ocellata Gmelin Charania variegata Lamarck Columbella mercatoria Linne Ocenebra aberrans C. B. Adams Thais deltoidea Lamarck Engina turbinella Kiener Acmaea jamaicensis Gmelin Cerithium literatum Born Astraea sp. Conus mus Hwass T egula semigranosa A. Adams Drupa nodulosa C. B. Adams Pisania pusio Linne Leucozonia nassa Gmelin Bulla occidentalis A. Adams Aplpia dactylomela Rang* Arcopagia fausta Pultney lsognomon radiata Anton Arcopsis adamsi E. A. Smith Arca umbonata Lamarck Barbatia candida Helbling Lima scabra Born Lithophaga aristata Dillwyn ECHINODERMATA Echinoids Diadema antillarum Philippi * Echinometra lucunter Linne * Eucidaris tribuloides Lamarck * Asteroids Linckia guildingi Gray* Holothurians Thyonepsolus braziliensis Theel? LITERATURE CITED Heilprin, A. 1890. The corals and coral reef of the western waters of the Gulf of Mexico. Proc. Acad for this rep::>rt were collected by M. N. Bramlette, of the Scripps Institution of Oceanography, from a phosphate mine at Gafsa, Tunisia, and from Ocean Island in the Gilb::-rt group in the S::>uth Pacific Ocean. Bramlette (1955) states .. .. (I ·.. \~· 'I·, I .FIG. 1. P~otoi;nic.rograph.of a Thin Section of Phosphate Rock from Tunisia Showing Fecal Pellets w: th a Rad:olan an ? Inclusion ( X250) . Evidence for Fossil Bacteria in Phosphate Rocks that "the phosphate rock from Tunisia is typical of much of these Korth African deposits which form widespread beds in the early Tertiary of the region. The formation of the phosphate rock must have been during the time of deposition of the associated strata, some 60-70 million years ago." The phosphate rocks consist of small round bodies with little interstitial cementing material. Thin sections were made by the usual grinding procedure. The sections were examined by bright field and phase m:croseopy. The round and oval bodies (Fig. I), haYing an average diameter of 150 microns, are assumed to be ooids (individual oolites) and phosphatized fecal pellets from some small organism. Many of the pellets contain broken fragments of diatoms, rad:olarians and foraminifera. Some contain unbroken organisms such as the radiolarian seen in figure]. Moore t1939) describes some of the fecal pellets found in marine de;)Os:ts. Fu;. 2. Photomicrographs by Phase :'.\licroscope of Fecal Pellets in Thin Sections of Pho$phate Rock from (a) Tunisia and (bl Ocean Island. The elongate dark bodies I.arrow I are a$$Umed to be Fossil Bacteria. Figure 2a aboYe. Evidence for Fossil Bacteria in Phosphate Rocks FIGURE 2b. Modern fecal pellets contain many bacteria and therefore, fossil fecal pellets are not an unlikely source of fossil bacteria. Pettijohn (1949) states that small granules and oolites are always present in phosphorite sediments. In some phosphorites, where phos­ phorus has been precipitated, a fossil fragment serves as a nucleus thus producing a ra­dial structure. Pettijohn describes the formation of ooids by the preciptation of phospho­rus which sometimes initially cements small particles of organic matter and fragments of organisms. If this is true, then bacteria must certainly be present and associated with the organic matter and particles. It is difficult to distinguish between fecal pellets and ooids and, therefore, the forms described in this paper will be referred to as pellets even though the author has assumed them to be fecal pellets. Microsopic examination of thin sections of Tunisian rock were made. At magnifications of 1000 many small uniform rod shaped objects of a diameter of 0.5 to 1 microns were Evidence for Fossil Bacteria in Phosphate Rocks seen which resemble bacteria. The length of the forms varied from 1 to 3 microns, the ends were rounded and chains of two were present. Figure 2a illustrates the size and shape of the small forms. It is difficult to obtain good photomicrographs of the thin sec­tions because of depth of field limitations. Therefore, the elongate objects seen in the photomicrographs in figure 2 represent only a few of the many found in the sections of rock. It is presumed that the forms represent bacteria which were incorporated in and, or, were decomposing the organic material of the pellets at the time of sediment deposi­tion. Bacteria must have been present during or before phosphatization. Cultural tech­niques were not used to determine whether the bacteria were still alive because it would be difficult to determine if bacterial growth was due to bacterial contamination in the interstitial spaces or the fossilized forms. The abundant bacteria-like forms were found primarily inside the pellets associated with other inclusions. It is unlikely that the forms are due to contamination during the sample preparation because of their abundance and the fact that pellets or oolites have a rather hard impermeable outer shell which would discourage bacterial entry. However, it is possible that bacteria could enter the rock formation when the rocks were more permeable than at present. Some bacteria were seen in the spaces between the pellets and could possibly be contaminants of a later geological date or from the laboratory treat· ment. Bramlette (personal communication, 1955) reports that, "the sample of phosphate rock from Ocean Island came from one of the open pit mines of the higher central part of the island. The blanket of ore is due to phosphatization of an original calcareous reef formation." Thin sections of this rock examined at 1000 diameters had many of the bacteria-like forms which are shown in figure 2b. Ht it is assumed that these are fos· silized bacteria, they" were presumably present during the time of the reef formation and probably not earlier than the later Tertiary or Pleistocene. This is a note of academic interest to be added to the growing evidence for fos· silized bacteria, the existence of which is difficult to prove. However, the recent work by Abelson (1956), which demonstrates the presence of amino acids in ancient shells, certainly suggests that some constituents of protoplasm might persist during geological time and it seems reasonable to assume that bacteria would be represented. As new tech­niques of organic analysis are developed it might be possible to identify the protoplasm of small organisms such as bacteria in ancient sediments and rocks. LITERATURE CITED Abelson, P.H. 1956. Paleobiochemistry. Sci. Amer., 195: 83-92. Barton, H. M., and D. J. Jones. 1948. Electron microfossils. Science, 108: 745-746. Moodie, R. L. 1920. Thread moulds and bacteria in the Devonian. Science, 51: 14-15. Moore, H. B. 1939. Recent Marine Sediments. Tulsa, Okla., Amer. Assoc. Pet. Geol., 736 pp. Pettijohn, F. J. 1949. Sedimentary Rocks. New York, Harper and Brothers, pp. 526 Renault, B. 1895. Sur quelques Micrococcus du Stephanien, Terrain Houiller Superieur. C. R. Acad. Sci., Paris, 120: 217-220. ZoBell, C. E. 1957. Annotated bibliography on paleoecology of bacteria. Treatise on marine ecology and paleoecology. Mem. Geol. Soc. Amer., 67 (2) : 693--098. A Bacterium Causing Tail Rot in the Norwegian Codfish CARL H. 0PPENHEIMER1 Institute of Marine Science, Port Aransas, Texas During the spring of 1953, while at the l"niversity of Oslo Biological Station a study was undertaken to determine the causative agent of the so-called "Tail Rot" disease of the Norwegian codfish Gadus callarias. A similar disease in the sea herring, Clupea harengus, is caused by a gram negative pleomorphic rod which has not been character­ized (Sindermann and Rosenfield, 1954). A partial review of other marine fish diseases of bacterial origin is given by Oppenheimer and Kesteven ( 1953) . In Oslofjord near Oslo live cod are kept in submerged wooden cages in the fish markets and tail rot is quite prevalent in these fishes. The waler temperature in the fjord varied between 5° and 9° C. The infection usually started on the margins of the fins or tail and produced a characteristic whitish color (Fig. 1) . The infection pro­gressed inward from the edge of the fin or tail and soon involved the integument and muscle tissues. In advanced stages of the disease only the spines of the tail or fins re- Fie. 1. Typical appearance of "Tail Rot" disea5e of the Norwegian codfish. 1 Fulbright Fellow, 1952-53. A Bacterium Causing Tail Rot in the Norwegian Codfish mained, bleeding was noted and other areas of the integument developed lesions. Con­ current examination of the blood of the fish in advanced stages of the disease demon­strated motile rod-shaped bacteria suggesting that the infection was systemic. The fish usually died within 48 hours after lesions appeared and the muscle became infected. Experimental Methods Material from the lesions of the integument and tail regions of diseased living fish was examined by phase microscopy and by cultural methods. Large numbers of motile rods and cocci were present. Material from the lesions was infective when inoculated into the fleshy base of the tail of healthy fishes. The material from the lesions was streaked on petri plates containing sea water nutrient agar (Medium 2216, ZoBell, 1946) enriched with codfish extract. The plates were incubated at 15° to 18° C.; i.e. approxi­mately 10° C. warmer than the aquarium water. The bacterial colonies which appeared on the plates after incubation were purified and tested for pathogenicity by the inocu­lation of a dilute suspension of the bacteria into healthy fishes. One microorganism was isolated in pure culture which produced the characteristic tail rot when inoculated into the skin at the base of the tail of healthy fish. A small lesion soon appeared at the site of inoculation and the infection spread to the tip of the tail and then progressed as the disease did in nature. The tail rot was marked after 24 hours when lesions typical of the advanced disease appeared and death occurred in approximately three days. A microorganism identical with the original inoculated culture was reisolated from the les:on of the experimental fish. The reisolated culture produced the typical disease. The characteristics of the pathogenic bacterium were determined by "Standard Methods Procedures" except that 75 percent sea water was substituted for distilled water. Litmus milk was made with distilled water since sea water reacted with the medium. Incubation was at 16° C. Description Short rods: 1 x 2 microns occurring singly, in pairs and in short chains. Motile with single polar flagellum. Gram Negative. Agar plate: Round with entire edge, moist, glistening, creamy white color, slightly opaque. Gelatin stab: Crateriform liquefaction at 7 days. Broth: Turbid, no pellicle. Potato: No growth. Litmus milk: No growth. Starch: Not hydrolyzed. Ammonia produced from nitrates. lndole not formed. Distilled water nutrient broth: No growth. Acid from maltose and saccharose. No acid or gas from glucose, lactose, mannitol, salicin or glycerol. Sensitivity tests with antibiotics indicated that growth of the organism was inhibited in vivo and in vitro by penicillin, streptomycin, chloromycetin, neomycin and poly­myxin. A Bacterium Causing Tail Rot in the Norwegian Codfish Conclusions The above description places the bacterium in the genus Pseudomonas according to Bergey's Manual of Determinative Bacteriology, 6th Ed. The description applies to no other described organism. The culture was lost during transit back to the USA and therefore other cultural characteristics were not determined. The above description is not adequate to name a new species, but may be useful to other microbiologists working with "tail rot" diseases of fishes. LITERATURE CITED Oppenheimer, C. H., and G. L. Kesteven 1953. Diseases as a factor in natural mortality of marine fish. Fish. Bull. F.A.0., 6: 1-8. Sindermann, C. and A. Rosenfield 1954. Diseases of fishes of the Western North Atlantic. I. Diseases of the sea herring Clupea haungus. Bull. Dept. Sea & Shore Fisheries, Maine, Res. Bull., 18: 1-23. ZoBell, C.. F.. 1946. Marine Microbiology. Waltham, Mass., Chronica Botanica Co., 240 pp. An Incidence of Pink Oysters in Galveston Bay, Texas RICHARD Y. MoRITA1 Department of Biology, University of Houston, Houston, Texas During the middle of December, 1955 an incidence of pink oysters was brought to the attention of the investigator by Mr. G. J. Brittain. Engineer Assistant, Division of Sani­tary Engineering, State Department of Health, Harris County Health Unit). A pink discoloration was noted in the juice of shucked oysters taken from Redfish Reef, Galveston Bay. These shucked oysters were obtained from many of the oyster houses around Seabrook, and had been placed in pint, quart, and gallon containers. Ac­cording to the oyster house operators, this condition was confined to the oysters taken from Redfish Reef. Normally oyster juice has a colorless translucent appearance. However, pink discolora­tion of the oyster renders them unsightly and unfit for food (Tanner, 1944). The occur­rence of pink oysters has been attributed to the presence of yeasts (Tanner, 1944) and to the presence of carotenoids, from the food of the oysters, which have been concentrated in the tissue and which bleed from the tissue when the oyster is shucked (Adam, 1949). Since microorganisms may cause the pink discoloration, a bacteriological examination was performed on the oyster juice. Slides were made from the oyster juice to determine whether or not one morphological type of organism was predominant. Slides that were Gram stained revealed a large num­ber of Gram negative, short, straight rods, but no yeast cells were demostrated. Yeast cells were not demonstrated on slides stained with methylene blue. Routine plate counts were made from fresh pooled oyster juice, obtained from oys· ters from Redfish Reef, which had been standing in the oyster houses for about four days. This freshly prepared oyster juice was collected as aseptically as possible. It was used in preference to oyster juice from previously shucked oysters that were already in containers to obviate gross contamination by the shuckers. Oyster juice collected with this method had a slight pink discoloration. Employing a sea water medium, 2216E (ZoBell, 1946), peptone (Difeo), 5 g.; yeast extract (Difeo), 1 g.; ferric phosphate, 0.1 g.; aged sea water, 1000 ml.; pH 7.5, a plate count of 180 x 108 bacteria per ml. was obtained, of which 130 x 108 bacteria produced a dark pink pigment. A plate count of 30 x 107 bacterial per ml. was obtained from the oyster juice, using nutrient agar pep­tone (Difeo), 5 g.; beef extract (Difeo), 3 g.; distilled water, 1000 ml.; pH 7.0, of which 21 x 107 bacteria per ml. produced a dark pink color. Incubation of the plates was performed at room temperature. Pure cultures were isolated from the oyster juice. Microorganisms that produced a dark pink or reddish color were subjected to routine bacteriological tests to determine, 1 Present address: Department of Bacteriology, University of Nebraska, Lincoln, Nebraska. An Incidence of Pink Oysters in Galveston Bay, Texas if possible, the genus and species. All of the isolated microorganisms gave the following results: Gram stain : Gram negative, short, straight rods (24 hour culture) . Spore stain (Wirtz ) : No spores ( 48 and 72 hour cultures) . Agar colonies: Granular, circular, thin, white becoming dark pink in 3 or 4 days. Broth: Turbid, turning dark pink in 3 or 4 days. lndole not produced. Nitrites from nitrates. Odor of trimethylamine. Aerobic and facultative anaerobic growth. Litmus milk: Acid, soft curd, dark pink growth. Pigment: Not readily soluble in water, soluble in alcohol and ether. Glucose broth: No gas. Acetylmethylcarbinol is produced. Gelatin: No pellicle. From the above data the organisms would be classified as Serratia marcescens. According to Bergey's Manual of Determinative Bacteriology (Breed, et al., 1957) S erratia marcescens is found in water, soil, milk, foods, silk worms, and other insects. Other species of Serratia have been implicated in the reddening of codfish (Tanner, 1944). Since Serratia marcescens is normally found in soil and water, the utensils and water used in the oyster houses were checked for Serratia marcescens. Results were nega· tive and, therefore, ruled out the possibility of gross contamination of oysters from the oyster houses. Oysters from other sections of Galveston Bay were obtained from the oyster houses. Freshly pooled oyster juice was prepared and checked for the presence of Serratia (as evidence by a pink or red colony) by the plate count method. Medium 2216E was used for checking the presence of Serratia colonies. A plate count of 35 x 106 bacteria per ml. was obtained from the oyster juice. Plates made from a 1:1,000,000 dilution of this oyster juice had no red or pink colonies. The oyster shuckers apparently did not notice a discoloration when shucking oysters. The discoloration was more evident after the shucked oysters had been allowed to stand. Evidentl y the time interval served as an incubation period for the growth of Serratia. Difficulty of securing proper samples, as well as arriving on the scene late, prevented the investigator from making a more complete bacteriological investigation. It is suggested from the data that the occurrence of pink oysters resulted from an in· fection of Serratia marcescens. LITERATURE CITED Adams, R. J. 1949. Red Oysters. Frosted Food Field and Food Merchandising, 9: 11. Breed, R. S., et al., 1957. Bergey's Manual of Determinative Bacteriology, 7th ed. Baltimore, Md., Williams & Wilkins Co., 1094 pp. Tanner, F. W. 1944. Microbiology of Foods. Champaign, III., Garrard Press, 1196 pp.ZoBell, C. E. 1946. Marine Microbiology. Waltham, Mass., Chronica Botanica Co., 240 pp. The Drop-Net Quadrat, a New Population Sampling Device1 THOMAS R. HELLIER, JR. Institute of Marine Science The University of Texas Port Aransas, Texas Determining the biomass of fish in a prescribed area is one of the most difficult problems in studies of fish productivity. Many methods have been proposed for this purpose, such as tagging and clipping, determinations of catch per unit effort, elec­ . trical sampling, and in small areas draining or poisoning. Although these methods have often been used and modified for specific areas and studies, size selection is usually a problem (Carlander, 1956). A survey of the fisheries literature indicates that no method is applicable to all situations. The above methods, their advan­tages and their limitations, are described by Haskell (1940) , Smith (1940), Krum­holz (1943), Ricker (1943, 194.Sa, 194-8b, 1956), Shetter and Leonard (1943), Rounsefell and Kask (1946) , Carlander ( 1948, 1956, 1958) , Delury ( 1951, 1958) • Meyer-Waarden (1957), and many others. None of these methods seemed completely suitable for studies in broad, shallow marine bays. This paper describes a new method being used for a fisheries population study in the large Laguna Madre of Texas. General biological features of this shallow hyper­saline bay were recently described by Simmons (1957). It is hoped that this new method, the drop-net quadrat, will eliminate much of the selectivity inherent in more standard methods . The author acknowledges the services and generous contribution of ideas from members of the staffs of the Institute of Marine Science and the Marine division of the Texas Game and Fish Commission in Rockport. A somewhat similar trap was used in a farm pond in a class project at A and M Col­lege by Mr. William B. Wilson in 1950. The trap covered 225 square feet and operated with sliding panels (personal communication) . Description of Method The drop-net quadrat method of population sampling consists of surrounding a large quadrat (100 ft. by 100 ft.) with a small mesh net (Fig. 1). This net is sup­ported by a steel cable suspended approximately 3 feet above the surface of the water from eight piling. The piling may be easily placed or removed by a % inch jet of water supplied at 20 to 30 pounds pressure per square inch. A heavy chain 1 These studies were aided by the Texas Game and Fish Commission as part of the project: Produc­tivity of the Striped Mullet in the Laguna Madre of Texas, a study of herbivores in a bay system, Howard T. Odum and Thomas R. Hellier, Jr. 166 The Drop-Net Quadrat, A /\"eu· Population Sampling Det"ice /ROPES HOLDING PART OF NET IN ELEVATED POSITION WATER__LINE TRIGGER PILING--~ ROPES REL EASED Fie. 1. The Drop-net Quadrat. ( 1 pound per foot) is used for a lead 1ine on the net. A line is secured to the chainat each piling. These lines are passed through pulleys located on crossarms at thetop of the piling, and the net is hoisted clear of the water. After the lines are securedto a trigger mechanism the net is ready to drop. The area is left undisturbed forseveral hours after the net has been rigged. The trigger is then released, freeing the:Jines, and the net drops into place. This method effectively and rapidly isolates thequadrat. The quadrat is then seined repeatedly with a 120 ft. bag seine until all ofthe fish are remoYed. The Drop-Net Quadrat, A New Population Sampling Dei;ice Discussion The drop-net quadrat is at present adapted for a relatively shallow homogeneous area of Texas bays, but might also be used for quantitatiYe collection of other estu­arine and coastal bay areas_ Initial trials of the method have been conducted with netting of two mesh sizes on a 14 acre quadrat, and on a small quadrat (1/16 of an acre) with small mesh netting l % inch stretch) . The results of the first trials are presented in Table l. Even though these early trials TABLE 1 Yield of Fishes from Drop-net Quadrat in Pounds per Acre of Preserved weight 1h aero quadrat,8 1/ 16 acre quadrat,b ~-4 acre quadral, <' 3 inch slretched rne~h %. inch stretched me~h ~~ inch stretched me~h No. of Trials 6 6 6 Average yield, per trial 25.5 (2.9g/l\I2) 54.6 (6.lg/M2) 121.3 06.6g/ W) Yield of highest single trial 73 (8.2g/.M2) 145.3 (16.3g/l\.P) 176.2 (19.8g/M2) Yield of lowest single trial 3.2 (0.4g/l\i2) 5.4 (0.6g/l\12) 64.5 (7.2g/ l\F> a Principal fish species taken : Galeichthys Jeli.$, .llugil cephalu.s, Cynoscion nebulosus, ,\licroposon und.ulatus, Pogonias cronis, Sciaenops oceUata. b Principal fish species takeo: Anchoa mitchilli, A. hepserus, .llenidia beryllina, Mugil cephal.us, Leiostomu..s xanthurus, .fficropogon undulatus, Lasodon rhomoides. ~ Principal fish species taken include all of abon. were necessarily selective, because the large mesh in one set of trials captured only the largest fish and the small area of the quadrat in the other selected against the larger fish, the data obtained indicates that a 14 acre trap with % inch stretched mesh netting is adequate to take all size groups. There are now several traps located in representative areas of the Laguna Madre which are being tripped at regular intervals. These are the basis for continuous estimation of the biomass, not only of fishes, but of some inverte­brates also_ The initial data from these traps ( 14 acre; % inch stretched mesh I also are presented in Table l. LITERATCRE CITED Carlander, K. D. 1948. Some precautions in estimating fish populations. Progr. Fish Cult., 10 (3): 135-137, ----. 1956. Appraisal of methods of fish population study-Part 1. Fish growth rate studies: techniques and role in surveys and management. Trans. 21st N. Amer. Wild!. Conf., 262-274. ----. 1958. Some simple mathematical models as aids in interpreting the effect of fishing_ Iowa St. Coll. J. Sci., 32 (3) : 395-418. DeLury, D. B. 1951. On planning of experiments for the estimation of fish populations. J. Fish. Res. Bd Can., 8 (4): 281-387. ----. 1958. The estimation of population size by a marking and recapture procedure. J. Fish. Res. Bd Can., 15 (1): 19-25. Haskell, D. C. 1940. An electrical method of collecting fish. Trans. Amer. Fish. Soc., 69 (1939) : 210-215. Krumholz, L. A. 1943. A check on the fin-clipping method for estimating fish populations. Pap. :\lich. Acad. Sci., 29 (1943): 281-291, 2 figs. Meyer-Waarden, P. F. 1957. Electrical fishing. Fish. Study, F.A.0., No. 7, ii pp. Ricker, W. E. 1943. Creel census, population estimates and rate of exploitation of game fish in Shoe Lake, Indiana. Im·est. Ind. Lakes, 2 09421: 215-253. The Drop-Net Quadral, A New Population Sampling Device -----1948a. Methods of estimating vital statistics of fish populations. Ind. Univ. Pub!., Ser. 15, 101 pp. ----. 1948b. Computation of fish production. Bull. Bingham oceanogr. Coll., 11 (4): 1-283. ----. 1956. Symposium: Uses of marking animals in ecological studies. The marking of fish. Ecology,37 (4). Rounsefell, G. A., and J. L. Kask. 1946. How to mark fish. Trans. Amer. Fish. Soc., 73 0 943): 320-363, 4 figs. Shetter, D. S., and J. W. Leonard. 1943. A population study of a limited area in a Michigan trout stream, September, 1940. Trans. Amer. Fish. Soc., 72 ( 1942) : 35-51, 4 figs. Simmons, E. G. 1957. Ecological survey of the upper Laguna Madre of Texas. Pub!. Inst. Mar. Sci. Univ. Tex. 4 (2): 156-200. Smith, M. W. 1940. Copper sulfate and rotenone as fish poisons. Trans. Amer. Fish. Soc., 69 (1939): 141-157. Fluctuations in the Relative Abundance of Sport Fishes as Indicated by the Catch at Port Aransas, Texas 1952-1956 VICTOR G. SPRINGER1 AND JACQUES PIRSON Institute of Marine Science The University of Texas Introduction Port Aransas is a small coastal town located at the mouth of Aransas Pass inlet on Mustang Island, which is adjacent to the Texas mainland (Fig. 1). One of its primary industries is sport fishing and many of the inhabitants derive their livelihood directly or indirectly from it. Unfortunately quantitative data on sports fishing are scarce every­where and especially in the western Gulf. This report of records gathered at Port Aransas provides a view of the annual cycle. The Port Aransas Tourist Bureau was established in 1951. One of its functions is to keep the public aware of the varieties and numbers of fishes being caught in the area. This function has been served by the dissemination of a weekly mimeographed bulletin summarizing the catches of sport fishermen as total numbers of fishes caught of each species. The sport fishes reported at Port Aransas are those caught at the jetties of the Lydia Ann Channel or in the open Gulf from piers and charter boats. Information regard­ing catches is supplied to the Bureau by local establishments integrally associated with the area's sport fishing. In recent years these sources of information have been primarily Mathews Sportsman's Headquarters, the Island Charter Service, Fisherman's Wharf, and Pappy and Mammy's; all are located within Port Aransas. Compilation of the catches and publication were originally under the supervision of Mrs. G. L. Webb but are presently supervised by Mrs. Norma Skinner. The catch, which is reported, is only a part of the total catch at Port Aransas. What percent is reported is not known. ACKNOWLEDGMENTS Sincere appreciation is due Mrs. Norma Skinner for her patience and alacrity in providing us with information necessary to the completion of this study. We wish also to thank the numerous residents of Port Aransas who offered us pertinent information on our subject. Dr. H. T. Odum called our attention to the possibility of the study and offered suggestions concerning the manuscript. 1 Present address: Florida State Board of Conservation, Marine Laboratory, Maritime Base, Baybor0< Harbor, St. Petersburg, Florida. Fluctuations in the Relative Abundance of Sport Fishes 97° 301 w + Gulf of Mexico 27° 301 N + Frc. 1. Port Aransas, Texas, and vicinity. PROBLEMS IN ESTIMATING THE RELIABILITY OF THE DATA Evaluation of the completeness with which catch reports are made shows there has been considerable variation in the methods used. Some of the participating organizations estimate that their reports are as much as 95 percent complete, while others admittedly state their reports contain only the best two or three catches each week out of some hundreds. Some omit certain fishes from their reports (i.e. catfishes) . Report sheets are missing for sixty-five of the two hundred and sixty-one (or about 25 percent) weekly periods covered by this study. However, all but one of the months during the period are represented by at least one report. The most serious deficiencies occur for the following dates: June 25-July 15, 1952; August 27-September 16, 1952; December 10, 1952-January 1, 1953; April 29-May 28, 1953; September 15-0ctober 5, 1954; December 8, 1954-January 5, 1955; March 31-April 27, 1955; and May 5-May 30, 1955. No reports are issued the last two or three weeks of December in compliance with Post Office Department requests that bulk mail be withheld during the holiday mail rush. No uniform method of dating the reports has been maintained over the years. Some­times part of the last week of the year is included with the first week of the next year; Fluctuations in the Relatii·e Abundance of Sport Fishes other times recording starts with the first week of the year. Frequently report periods are three or four days only; occasionally there is a gap of two or three days between the dates covered by the reports. Dates often overlap (i.e. January 1 to 7 followed by January 7 to 14). In most of these instances the catches for the overlapping days would be split between the two reports, probably because the information was received too late to be included at the time of the first publication. Graphs As the reports for certain weeks of certain months are not strictly comparable from year to year, it was necessary to de,·ise a means of graphing which would most accu­rately portray the tendencies of the fishery. For the purpose of this study the policy has been to recognize the first week of the year as the first week which contains reports only for that year, and the last week of the year as the last week which contains reports for that year whether or not the first few days of the next year are included. The same method applies to weeks that include days from two months. The reports are published approximately fifty times a year and despite the fact that the weeks are not exactly com­parable, they are listed on the graphs in accordance with their numerical sequence, one year below the next. The graphs report the numbers of fish caught for each period. If the scale is too low for the peak months, the number is recorded to the side of the point on the graph. Total catch of each species for each year is also reported on each graph. Index of Fishing Intensity The only index of fishing intensity which could be de,·ised is a relati,·e one. It is based upon a tally of the cars (Fig. 2) paying tolls to and from Port Aransas via the Port I:,::>~ l:iJ.:> \,j..,~ 1:;);,;0 1956 J FM Alill J J AS 0 If 0 J FM AM J J AS 0 ND J F Ill AM J J A 0 ND J FM AM J J A F Ill A M J • s IPJO's Totol: 11'9,!89 UK>,854 t90,°'8 155,906 165,908 1Mu: 3.02 2.>3 3.21 4.37 3.42 I !\ I ~ I \ I I A (\ "'.. '-I ....I \ I I I \_ Fie. 2. Number of vehicles paying tolls to and from Port Aransas. Index equals number of times. highest traffic month exceeds the lowest. Aransas Causeway. This tally is maintained by the Nueces County Road District. No distinction is made between the number of cars entering or leaving Port Aransas, but presumably by halving the numbers for any date ~to account for round trips) , one would approximate the number of individual cars making trips for that date. The index has a number of intrinsic faults. It is necessary to assume that the porportion of fishermen to non-fishermen remains constant throughout the year. This is obviously Fluctuations in the Relative Abundance of Sport Fishes incorrect sinct-the ratio of fishermen would be considerably higher in the winter when sunbathing and swimming activities are almost non-existent. In addition, it assumes that a constant proportion of the occupants of the vehicles will proceed to and remain in the Port Aransas area_ It is well known that some fishermen concentrate their activities on the causeway itself and others on the nearly twenty miles of beach between Port Aransas and Corpus Christi. A final difficulty is encountered when one considers the years 1955 and 1956. On November 9, 1954, the Padre Island Causeway opened a paved road from Corpus Christi to Port Aransas. This shunted many vehicles away from the Aransas Pass Causeway. which until that time was the major access to Mustang Island. The decrease in number of automobiles crossing to Port Aransas in 1955 and 1956 has been attributed by causeway authorities to bad fishing. This may be true as the propor­tionate change in the number of vehicles is not sufficient to account for the propor­tionately fewer fishes caught during 1955 and 1956. During the years covered in this study the number of vehicles using the Padre Island Causeway has steadily increased. There is no way of knowing if the number of cars entering Port Aransas via the Padre Island Causeway has increased at the same rate. The graph of fishing intensity is reported as thousands of automobiles. The total num­ber of cars for the year is given at the top of each year's record. Below this is listed an index which represents the number of times the heaviest traffic month of the year exceeds the least traffic month. The largest index ( 4.37, July and December, 1955) is not suf­ficient to account for differences in the peaks for the various species. Under the circum­stances it was decided not to correct each month's catches using a monthly index based on the most active month, but to report the raw data unchanged and interpret it directly. We believe that the annual recurrences of largest catch peaks, bimodalities, etc_ as seen on the graphs, are actual indications of abundance. Fishing Tournaments Every year during the second week in July, and for the past three years during the second week in October, fishing tournaments have been held in Port Aransas. These include private and charterboats. The only manifestation of these periods of increased fishing intensity. as indicated by the graphs, are found in the catches of sailfish and per­haps tarpon. Other than for these two species it is not felt that increased fishing during the tournaments influenced the catches of the other forms reported here. Small Craft Warnings A list of dates during which small craft warnings were raised nearby, at Rockport, Texas, are given below. These warnings affect offshore catches for the periods of their duration. However, since most offshore fishing occurs during the summer and most small craft warnings occur during the winter, it is doubtful that the charterboat catches re­ported here would be affected. In more violent periods jetty fishing is also affected by bad weather, and correlations can sometimes be made between periods of bad catches on the graphs and periods of bad weather as indicated in the list. 1952. January 9, 10, 22, 23, 27, 28; February 1-6, 13-16, 20, 24-26, 29; March 3, 9, 10, 12, 14, 15, 17, 18, 21, 22, 27, 28; April 1-4, 9, 10, 12, 22, 23; May 9, 17-19, 23-25, 27, 28; June 5; (Data missing for July, August, September and October) November 2, 9, 10, 18, 22-26; December 4, 9, 2L 22. 30. 31. Fluctuatwns in the Relative Abundance of Sport Fishes 173 1953. January 1, 2, 8, 9, 11, 12, 14-17, 20, 22-24, 27, 28, 31; February 1, 2-6, 8-11, 14-17, 19, 20; :\larch 1--1, 12-14, 21-23, 29-31; April 5, 6, 12, 15-18, 23--25, 28-30; :\lay 4, 10, 12, 13, 18, 19: lune, July, August, September: none; October 26; NoYember 7, 8, 19, 22, 24, 25: December 2,3,5,6,8,9, 11-14,21-23,27-30. 1954. January 8-16, 20-22, 30, 31; February 6, 7, 10, 11, 15, 16, 18-20, 23, 25-28; :\larch 1-3, 11-13, 17-19, 31; April 1, 7, 8, 15, 16,29,30; :\lay 1,2, 11, 12: June25, 26; ]uly2,3 ; August: none; September 10-12, 16, 17, 21, 22, 30; October 1, 3--7, 14, 15, 22, 23, 26-30: November 2-4, 10-15: 17, 18, 23, 24, 26-28; December 4-6, 8, 9, 11-17, 27-29. 1955. January 5, 6, 12-15, 17, 18, 20, 21, 23. 24, 27. 28; February 3--5. 9-11, 17-21, 26, 27; :\larch 5. 6, 16, 17. 20, 22, 25-27, 31: _.\pril 1, 3, 4, 6, 7, 12, 13, 23: :\lay 7, 8, 11, 12, 25. 26, 28, 29: lune 8, 9, 21, 22: July 13, 14. 31; August 1: September 4-7, 11-14, 17-19, 28-29: October 6, 7, 12-14, 16, 17, 23, 24, 28, 29; No,·ember 2, 3, 5, 8, IL 12, 15, 16, 22, 23, 27. 28: December 3, 4, 8, 11, 12. 14-16. 30. 1956. January 7, 17, 18, 29, 30: February 1-13, 17, 18, 25. 27; :\larch 3, 6-8. 10-13. 21, 22, 27, 28; _.\pril 2, 3, 5-7, 9. 10. 14, 15 ; :\lay 2. 3. 15. 31; June 1: July 26. 27; _.\ugust: none: September 6, 7, 10-12, 22, 23, 30: October 18, 20, 21, 25, 26, 30; NoYember: data missing; December 8,9, 12. 13. 17. 18.22.23. Discussion of the Species The Tourist Bureau reports list the fishes by their common names. Most of these names are ascribable to particular species. others are not [i.e. shark, flounder. bonito (includes EuthJnnus alletteralus and Sa rd a sarda L whiting I includes JIenticirrus littoralis and M. americanus), grouper. trout I may include Cynoscion arenarius, C. nothus, and occa· sionally C. nebulosus)]. For recognizable fish both the common and scientific names are included in this paper; the unidentified fish are not discussed. Tarpon-Mega/ops atlantica I Fig. 31 J F M A M J J A S 1952 Tola1:301 40 1953 Talal: 285 40 ;· ."" /\.../'--·-~.----·­ 1954 0• Tolol: 2045 i 'O ::> ,/\.v->..._,_. ..,.,,·,~ [ 1955 Total:3'S ~ \/'·/'V\_ A. 1956 Totol: 32!! Month FIG. 3. Tarpon. 174 Fluctuations in the Relative Abundance of Sport Fishes Some of the game fishes which are included in the reports are not discussed here be­cause they are caught or reported only sporadically. These fishes include white and blue marlin, barracuda, amberjack, junefish, wahoo, ladyfish, spadefish, and gafftop catfish. Tarpon are primarily caught in the summer and fall. The graph indicates that there is a bimodal peak each year: one in June and one in October. Mr. Terry Leary, of the Texas Game and Fish Commission, informs me that the tarpon are in the bays during the summer, and Gunter ( 1945) reports two specimens taken in late July and late Sep­tember from Capano Bay. Peaks would be expected for Gulf catches when the tarpon are entering and leaving the bays. Whether or not tarpon enter or leave the bays to spawn is not known. It may be that the peaks are artifacts associated with the fishing tourna­ ments, but if such were the case, no recognizable peaks would be found during 1952 and 1953 when October tournaments were not held. From November or December through March there are no records of catches. Pre­sumably the tarpon are in the warmer offshore waters. Farrington (1949) notes that the tarpon arrive off Port Aransas in April and remain until November, and lists the best fishing months as May through October. Sailfish-/stiophorus americanus (Fig. 4) J F M A M J J A c; 0 N n .. 1952 To,al: 102 30 /\,,__ 1953 To!al : 68 . /~ ·-_.__...-'------' ·­ . 1954 i""­ Total: 154 30 ----"! V\_., 1955 Total: 120 30 ./ \ ·~ ~-........___/ ~----~­ 1956 Total: 102 30 ___/ "'-....---.,.., ·---­ Month Fie. 4. Sailfish. Sailfish are caught only from late May through September. Baughman (1954a) states that sailfish appear off the Freeport-Galveston area as early as May 2. Peak months are usually July and sometimes August. The peaks seem to be associated with the July fishing tournaments, although they may be the result of spawning and associated concentrations. Gehringer (1956) reports larval sailfish in the western Gulf from June, and spawning possibly occurs through August (Baughman 194la). Commercial fishermen and charterboat men have informed us that sailfish are often seen in schools at the snapper banks, approximately forty miles offshore, during the wintertime. Summer fishing for sailfish takes place as close as six miles offshore. It is Fluctuatwns in the Rela!ive Abundance of Sport Fishes the opinion of the charterboat men that these fish could be caught year round, but that no one has made an effort to do so, most likely because of the increased expense of traveling further out to sea and because of the less predictable winter weather. It is quite possible that there is a potential winter sport fishery in the area which could, with proper advertising, result in increased revenue for the area. Spanish Mackerel-Scomberomorus maculatus \fig. 5) F I I Ii ii I 1952 M I I A M ···~>A'' I I J I I I I J I I I A S ·~~1\''' I 0 I I I I N i I D I I Totol: 22,340 1953 /\ /~; Totol: 23,!591 /\ io,ooo---11 Totol: 42,220 · 1954 11\ I ., \._ 1955 Totol : 13,203 /~ 1956 Totol : 2!5,292 / ·-· ;\ Month Fie. 5. Spani~h l\lackerel. Spanish mackerel are most frequently caught during the months from March through October. The times of peak months rnry, but there appears to be a bimodality with the greatest catches occurring in March and April and again in July, August and September. Gunter I 1945) states that some mackerel \about 11 percent of his catches) enter the bays in the summer, and it is possible that the phenomenon here may be the same as indicated for the tarpon. Gunter also noted that the greatest commercial catches were made on the Texas coast during August. He states that the only month during which mackerel were not taken was February. Although not graphed here, mackerel were being caught during the week of February 19 to 26, 1957. 176 Fluctuations in the Relative Abundance of Sport Fishes LaMonte (1951) reports that this species is common off Florida from November through March; the months when this species is least common off the Texas coast. This might indicate a migration occurring from one side of the Gulf of Mexico to the other. Kingfish-Scomberomorus cavalla (Fig. 6) J F M A 1952 1!500 Total: 3923 1953 Total:9841 .._/'·,/\/'--· .. 0,. 1954 :0 Total: 11,2'7 £ ..,>I /,/'-v·V\ .... 1900 Total: 11,598 1955 _./'-.... /~t- Total: 7803 ._____, 1500 1956 .-/!\ Month FIG. 6. Kingfish. The kingfish is taken primarily in the summertime. It is possible that this also is an example of lack of fishing intensity during the winter months. These fish, like the sailfish and dolphin, are caught only from the charterboats and private boats. Rivas ( 1951) reports that this species is most common around south Florida during the winter months, which may indicate a migration around the Gulf similar to that suggested for the Spanish mackerel. Dolphin are caught almost exclusively during the summer months, showing very sharp peaks during the month of August, although the 1956 peak occurred in May (this has not been repeated in 1957). Sharp peaks such as these may indicate migrations, wherein fish are present in large numbers for short periods of time while they pass a point. Baughman ( l 94lb) reported a large run of dolphin which passed by Freeport during the last week of May, 1939. LaMonte ( 1951) states that dolphin are, in general, present from June through October on the eastern seaboard. She also states that young are plentiful around Beaufort in late summer. We have collected several young from drifting seaweed (Sargassum) in the Lydia Ann Channel during July, 1957, and Pew (1957) reports collecting young from Aransas Bay during the months of April, May, August, and October (ranging from 32 to 55 mm.). The common jack is caught from February or March through October and sometimes November. Peak months are variable, occurring in March, May, June, and October. Gunter ( 1945) collected most of his specimens from the bays during August. All these were young of less than 60 mm. which may indicate that the bays serve as nurseries for this species. Fluctuations in the Relatii·e Abundance of Sport Fishes Dolphin-Coryphaena hippurus l Fig. Tl J F M A M J J A S 0 N D 1952 /\_ Total: 1516 1953 . -·-/\ / Total: 2162 1954 Total: 1607 ;\ -· ---· .... 1955 Tatal: 1168 IOOO /"'\_ ...... ­ -· -·--· 1956 Total: 2809 Month Fie. 7. Dolphin. Jack Crernlle-Caranx hippos I Fig. 8) J F M A M J J A 0 N 1952 Totol: 8~6 !\ '--­ 1953 T-­ /\_ 0• ...:t ~ 1954 :;;; ToCwl : 1377 E A._ _•.?-­ 1955 i'v ToCwl: 793 1956 Total: 1151 .,,./\ Month FIG. 8. Jack CreYalle. Pompano are caught all year long, but major catches occur during :\larch. _.\pril. and May. These months are probably jm•t prior to the breeding period as Gunter 119.J.5) reports young as small as 13 mm. from June, and states that fish as small as 23 mm. were taken from June through December. He also notes that this species is taken all year by commercial fishermen on the coast, but to a lesser extent during the winter. Simmons (1957) reports that large schools of pompano were present in Laguna :\Iadre during the months of September and October. 178 Fluctuations in the Relative Abundance of Sport Fishes J 200 F 1952 Totol: 1531 M Pompano-Trachinotus carolinus (Fig. 9) A M J J A \ .\I·-._.__ ....._ 0 N D 1953 Total: 19~ /--\·-­.............. 1955 Total: 1368 /\ ,,._.____ 1956 Total: 1110 - Month FrG. 9. Pompano. Bluefish-Pomatomus saltatrix (Fig. 10) J F M A M J J A s 0 N D 1952 Totol: 72S _._........__/""[\·-­ . 1953 Totot: 273 1954 Total: 1957 _/-...... 1955 Total: 96 -___....._ 1956 Totol: IOI Month FIG. 10. Bluefish. Bluefish are caught primarily from March through September, and appear to be entirely absent during December and January. The peak months are July and August. Catches are variable from year to year, but during 1955 and 1956 they fell off markedly. Gunter (1945) noted an absence of this species during the years 1936-1942, exclusive of 1937-1938, during which time 395 pounds were reported from the Texas commercial landings. Fluctuations in the Relative Abundance of Sport Fishes Bigelow and Schroeder ( 1953) give interesting data on the bluefish off the north­eastern coast of the United States, but state that little is known about Gulf of Mexico populations. Ripe individuals are taken off Chesapeake Bay during late spring and summer, and catches are low or absent during the winter in that area. However, some commercial catches are made off Key West, Florida during the winter. J F 100 1952 Tolol: &40 1953 100 Total: 573 1954 100 Totol: 699 1955 IOO Totol: 297 1956 IOO Total: 360 J F 1952 Toto!: 2 to 1953 Total: 27 to 1954 0• ,, Total : 40 :2 to > :;; = 1955 Totol:H to 1956 Tolol : l420 Cobia-Rachycentron canadus (Fig. 11) M A M J J A s 0 N 450 ! '\ -------·-----... --·-·--/'·, __,/·,__._._.__.___......._,,,,.......-...... ...........-... /'.-... ----·------~/----·-·--­ - ·----·-·- ...........~·---·---·--·· ......•-·-·------· ·-·­ Month Fie. 11. Cobia. Snook-Centropomus undecimalis (Fig. 12) M A M J J A 0 N D ~­------·~ ....._.______ Month Frc. 12. Snook . Fluctuations in the Relative Abundance of Sport Fishes Cobia are caught from February or March through October or November. Only during April, 1952, did there appear to be a recognizable peak. Although not fished for specificall y, cobia are esteemed as a food fish. Snook are caught infrequently at Port Aransas. The best months appear to be from mid-July until late September. Nothing is known of the spawning habits of the snook in this region. Mangrove Snapper-lutjanus griseus (Fig. 13) J F M A M J J A s 0 N D 1952 Total: 676 I/\. 1953 Total: 2485 250 -. Jvv \ _._ -·--------·-·--·/ !!!. 0 1954 ::s Total: 124 'O ·;;: :0 -= ·-· 1955 Total : 556 250 1956 Total: 781 2 Month Fie. 13. Mangrove Snapper. Mangrove snapper are caught throughout the year. The catches are quite variable from year to year, but the peak month is most often found in October and is sharply defined. Information concerning the breeding habits of this species is not available. Spotted sea trout are caught throughout the year, but the best months are May through July, and secondarily, October and November. Catches appeared to be very low during 1956, but this may be the result of tabulation. Catches were frequently reported simply as "trout" during the latter part of 1956 and were not graphed, but they may have been partly ascribable to this species. Gunter (1945) found peak catches for this species during the winter in Copano and Fluctuations in the Relatit·e Abundance of Sport Fish es 181 Spotted Sea Trout---CJnoscion nebulosus IFig. U l J F M A M J J A s 0 1500 1952 Total; ll,3U ..___,,,,...-.i'---·-/ -·-'--./· .............. ___.,,,.,........._.___ l!ICO 1953 Totol: 22,3~ 1954 Toto l : ~.911 1955 Totol: II, 133 /\ 1----------~~==~/'-_-..._'_:-;L_o/_-'..___::::•-,.,,-=---'' =-----·-'/-,....__,_/""~­ ~ 1956 Totol: 3509 Month Fie. H. Spotted Sea Trout. Aransas bays. Ripe fish were found from April through l\ovember. Simmons \1957) found small trout common in Laguna Madre in April. May. and October. According to Pearson 11928) spotted trout spawn in tht> bays from early spring through Octobt>r. with the height of the season during April and :\lay. Drum are caught throughout the year. The best catches art> usually made from Octobt>r through February, and the highest peak occurs in :\oyember of each year. Gunter I 1945 l found October through April to be the pt>ak months for the yt>ars 19-H and 1942. He also pointed out that the fall and winter months provide the greatest catches of this spt>cies in tht> commercial fisht>ry of the Tt>Xas coast. Pearson I 1928) states that from February until May the drum are spawning. Large migrations take place from the bays to the Gulf. He also states that a secondary spawn­ing sometimes occurs from July through :\o,·ember. Sport catches for 1953 and 1954 would indicate such a secondary migration into the Gulf. Although croaker are caught during most of the year, there is no month which can compare with the October-l\ovember pt>aks. Pearson 11928) reports that ripe adults migrate from the bays to the Gulf during Septembt>r and October; spawning takes place from October to February. Suttkus 119551 found that croakers migrate out of Lake Pontchartrain, Louisiana, during September. Octobt>r. and Xovember. associated with the decrease in water temperature. First young of the spawning were caught in the lake during :\"o,·ember. It was found that most of the fish did not t>xceed two yrnrs in age (also Pearson, op. cit., for Texas) . On the basis of the abow information the single peak period at Port Aransas might be explained by the migration into the Gulf of the pre­viously spawned fish which had migrated into the bays as small indi\-iduals, too small to be caught. Texas Game and Fish Commission yearly reports show that croaker are frequently absent from the commercial fishery I, which is in the Gulf) during the spring and summer. 182 Fluctuations in the Relative Abundance of Sport Fishes Drum-Pogonias cromis (Fig. 15) J F M A M A 1952 Total: 1974 500 __......'-,,,,,,............._ /'--....... 1953 Total: 4321 0~-·-......... ....... ·-· ~-­ .!!? 1954 0 Total : 4-0tl5 ::> "O :~ "O ..!: i'· . \ -.....-~""' \j\~. ......._. ......... 1955 Total: 1741 500 .....\. ............ .....,,,..,,,,,. ---·---...}\ ' 1956 Total: 1181 500 Month FIG. 15. Drum. Croaker-Micropogon undulatus (Fig. 16) J F loll A loll J J A s 0 N 0 1000 1952 Total: 2275 1000 1953 Total: 3533 !11 0 ::> "O 1954 ;2E 1000 Total: 7312 "O ..5 ----­1000 1955 Total : 3290 1956 Total : 2119 -·--· Month Fie. 16. Croaker. Fluctuations in the Relative Abundance of Sport Fishes and also that the peak month is most commonly October. In collections from Texas bays, Gunter 11945) reported the largest catches of croaker during the spring and summer months. Redfish (Channel Bass )-Sciaenops ocellata (Fig. 17) F M A M J J A s 0 N D 1952 Toto! : 6964 1000 . -·-·-·....___ /'-)\ 1953 Total : 4567 1000 /\~ ;\ -"\ /\_\ _.-·............., . ·-.--· _/ ./ ·­ 1954 /\ . Total : 9&44 -· /'---/, _/'-."--·/ V\__ 1955 Total: 3311 1000 ___.,,,....'--_·-'-.-·'-· - ·­ 1956 Total : 1140 1000 . ---·--------.............· Month Frc. 17. Redfish. Redfish are caught throughout the year, but the peak months usually occur in the fall when the adults are leaving the bays to spawn in the Gulf {Pearson, 1928). Sometimes, however, large catches occur in July and August. A notable decline was observed during 1955 and 1956. Sheepshead are caught all year, but the best months are from October to February. These findings agree with Farrington's (1949 I report and also with tabulations of the commercial catch for the Texas coast, as reported in the yearly reports of the Game and Fish Commission. An additional high for the month of July, 1956, is reported by the Game and Fish Commission for the commercial fishery, but this does not show on the Port Aransas records. Conclusions and Summary The Port Aransas sport fishery is an open Gulf and channel mouth fishery, as opposed to numerous bay sport fisheries in the area. Open Gulf fishing takes place from charter­boats and piers, and the channel mouth fishing takes place from the jetties. The types of Fluctuations in the Relative Abundance of Sport Fishes Sheepshead-Archosargus probatoceplzalus \Fig. 18) J F M A M J J A s 0 N D 1952 7llO Totol: 2441 .~ /-...._ - -·-­ /'v·'· 1953 7SC Totel:5304 /'-._..-/~\. ............._. /-·.,,--­ 1954 7'< Toto!: 3516 .------./'·...... /\_ -· --· /----.../ 7!1C .~j\_ Totol: 8S08 \ • ·/\._. -­- - - -­ ,,.~.......... 1956 1SO Total: 4849 ~----·,·-./\ - - -­-- r·-·-· \ Month FIG. 18. Sheepshead. fishes caught from the charterboats (sailfish, kingfish, dolphin, cobia, etc.) differ from those caught from the jetties and piers ( redfish, mangrove snapper, drum, croaker, sheepshead, etc.) ; howewr, tarpon are taken from all three places. Fishing from charter· boats takes place almost exclusively during the summer although it appears that the fishes caught may be present in the area all year. Fishing from the piers and jetties occurs all year long with certain fishes pro\·iding peak catches during various times of the year; however. most of the species recorded here provide fall or winter peaks for fishing. On the basis of life history studies in the literature (particularly for the sciaenids) the Port Aransas jetty and pier fishing is dependent upon the fact that certain species leaw the bays and enter the Gulf to spawn. They exit at Port Aransas through the only channel I Lydia Ann) for approximately sixty miles to the north and one hundred thirty miles to the south. In view of the recent interest in the cutting of fish passes to the north and south of Port Aransas. it should be pointed out here that there is a possibility such passes will diminish the catches of the migrating bay species at Port Aransas. This effect could be accomplished by diwrting the fishes which would ordinarily exit at Port Aransas. The cutting of passes may also diminish total fish caught if sports fishing does not take place at points in the vicinity of the Gulf openings of the passes. Whether or not the opening of passes will allow the bay species to increase in number in the future and offset the diwrting effect is a matter of less predictable nature. There ha;; been a considerable drop in the number of fish caught for several of the species discus;;ed herein during the years 1955 and 1956, as compared with 1952-1954. The causes for the decreases are manifold. and we do not feel the information available is ;;ufficient to explain them. Fluctuations in the Relative Abundance of Sport Fishes LITERATURE CITED Baughman, J. L. 194-la. Notes on the sailfish, l stiophorus americanus (Lacepede), in the western Gulf of Mexico. Copeia, 1941 (1) : 33-37. -----194-lb. On a heavy run of dolphin, Coryphaena hippurus, off the Texas coast. Ibid. 1941 (2): 117. Bigelow, H. B., and W. C. Schroeder. 1953. Fishes of the Gulf of Maine. Fish. Bull., U.S., 74: 577 pp. Farrington, S. K., Jr. 1949. Fishing the Atlantic. Coward-McCann, Inc., New York. 312 pp. Gehringer, J. W. 1956. Observations on the development of the Atlantic sailfish, lstiophorus ameri­canus (Cuvier), with notes on an unidentified species of istiophorid. Fish. Bull., U.S., llO: 139-171. Gunter, G. 1945. Studies on marine fishes of Texas. Puhl. Inst. Mar. Sci. Univ. Tex. 1 (1) : 1-190. LaMonte, F. 1951. 1951, a preliminary survey of marine angling in North Carolina. In Survey of Marine fisheries of North Carolina. Univ. North Carolina Press, pp. 251-286. Pearson, J. C. 1928. Natural history and conservation of redfish and other commercial sciaenids of the Texas Coast. Bull. U.S. Bur. Fish., 44: 129-214. Pew, P. 1957. Occurrence of young dolphin, Coryphaena hippurus, in a Texas bay. Copeia, 4: 300. Rivas, L. R. 1951. A preliminary review of the western north Atlantic fishes of the family Scombridae. Bull. Mar. Sci. Gulf & Carib. 1 (3) : 209-230. Simmons, E. G. 1957. An ecological survey of the upper Laguna Madre of Texas. Puhl. Inst. Mar. Sci. Univ. Texas, 4 (2) : 157-202. Suttkus, R. D. 1955. Seasonal movements and growth of the Atlantic croaker (Micropogon undulatus)along the east Louisiana coast. Proc. Gulf & Carib. Fish. Inst., 7 (1954) : 151-158. Population Studies of the Shallow Water Fishes of an Outer Beach in South Texas GORDON GUNTER Gulf Coast Research Laboratory Ocean Springs, _1/ississippi Introduction The vertebrate life of the beach enYironment is little known. Pearse, Humm, and Wharton !1942) reported the fishes and invertebrates taken in 68 seine hauls on inside and outside beaches near Beaufort, :\orth Carolina. Gunter !1945) reported on 35 minnow seine and 23 beach seine hauls made monthly over a year and a half on the beach of Mustang Island, Texas. These are the only recent reports of seining studies of outside beaches from :\orth America. Warfel and Merriman !1944) made a study of an inside beach on the east side of the harbor of :\ew Haven, Connecticut. Several earlier reports of fishes collected in Atlantic waters, but compiled for other purposes, are cited by Warfel and Merriman. In this study hauls along the Gulf beach of Mustang Island were made during separate periods, totaling 10 months, in three different years. The Institute of Marine Science of The Cniversity of Texas was headquarters for the work. Mr. David Kramer, who first began this work with the writer, collected most of the data. I am indebted to him and in part to Mrs. Kramer, who also assisted. LOCALITY Mustang Island is one of the barrier islands lying between the bays and the Gulf, which are typical along the Texas coast. It lies east of Corpus Christi Bay and is part of the western shore of the Gulf of Mexico. This island lies in the dry subhumid zone of Thornthwaite (cf. Collier and Hedgpeth, 1950). It is 16 miles long and runs almost due north and south, bearing a little to the east. The island and the general locality have been figured and described by Gunter 11945) and Collier and Hedgpeth !1950) and is also described in current geological papers. A map of this area is provided as Fig­ure 1 by Hoese (19581 in this volume. The beach itself has very gradual slope back to the sand dunes and is made up of fine sand. The offshore slope is gradual and approximately six sand bars lie parallel to each other to a distance of a quarter mile offshore. In 194-7, the southern end of the island was separated from Padre Island by a shallow pass, Corpus Christi Pass, into Corpus Christi Bay. However. Aransas Pass at the north end of the island is maintained at an artificial depth as ship channel, and the nearby passes "-ill not stay open. Corpus Christi Pass closed in May, 1949, joining Padre and :\1ustang Islands. This abolished collection station number one, which was located inside Corpus Christi Pass. The :'.\1ustang Island beach differs from those on the northern Gulf coast in two major respects. Large ri\-ers do not occur nearby and there ic.; usually very little driftwood on the beach. Secondly, it is part of a very windy coast, the average wind speed at Corpus Christi being 11 m.p.h. Therefore, the surf is usually rough and the inshore water is often roiled and muddy. It is remarkable that so many delicate, young and juvenile fishes, as are found there in warmer months, can live in this surf. METHOD The seine used for collecting was fifty feet long and four feet deep and made of mesh measuring one quarter of an inch stretched. After the work was finished the meshes measured three to an inch, but whether stretching was slow or took place after the preliminary use is not known. Eight stations were established. The first was a little less than one half mile inside Corpus Christi Pass, on the Mustang Island side, with the second at the mouth of the pass. The third was one mile north of the pass on the outer beach. Stations four, five and six were equally separated at about five-mile intervals going northward, the last within one mile of Aransas Pass. Station seven was at the angle made by the south jetty of this pass and the beach. The last station was a little more than one half mile inside the mouth of Aransas Pass. They were numbered one to eight starting at the inner Corpus Christi Pass station and ending at the inner Aransas Pass station. At each station the seine was carried as far out as possible, then spread and pulled ashore. At the two pass stations, where the water was deeper close to shore, sweeps par­allel to the shore were made. The fishes captured were identified and measured in millimeters in the field. Air and water temperatures were taken by a mercury thermometer reading in tenths of a degree centigrade. The salinity at each station was determined by titration. Results and Discussion TEMPERATURE AND SALINITY In Table 1 the water and air temperatures at each station are indicated. They show rather consistently the lag of water temperatures behind air temperatures when the seasons are changing. The temperatures for August, 1949 are close to the maximum for shallow open waters in this region, but the recorded temperatures for the cooler months do not approach the minimum since collections were not made in January through March. In Table 2 the salinities at each station are given. The period 1948 to 1953 was very dry due to a long drought (cf. Parker, 1955) , but the salinities ran higher than those given by Gunter (1945) for 1941 and lower than those of 1942. Probably this was due to the effects of local rains. The lowest salinities followed spring rains in April and May, 1949 and the highest came in August of the same year. THE FISHES In all 10,633 fishes were caught in 144 hauls, an average of 73.8 fish per haul. There were 44 species, including six identified only as to genus. Twenty-four specimens, which were mostly small puffers, Sphoeroides, and scombroid fishes were listed as unidentified. All the fishes taken in this work were small species or the young of larger species. Almost all of them ranged from 20 to 99 mm. in total length, although the total range was 14 to Population Studies of the Shallow Water Fishes TABLE 1 The Water and Air Temperatures in Degrees Centrigrade Are Given for Each Station. The Air Temperatures Are in Parentheses. Stal ion 1947 Nov. 9 20.5(22.7) 21.2122.8) -----------------­Nov. 29 20.5(21.5) 19.01 . . ) 18.8( ... ) 19.0( ) 18.0( ... ) 18.0( ..... ) 18.0( .. ) 16.5( .. Dec.6 22.3(23.5) 23.0120.6) 20.8(22.0) 20.5122.6) 20.4(21.8) 20.8(21.8) 21.4(21.1) 23.2(21.6) Dec. 31 22.3(21.1) 19.5(19.5) 19.5(19.8) 19.5(20.0) 19.3(20.2) 19.0(20.1) 19.0(20.0) 19.0(21.0) 1948 Sept. 19 28.0(28.8) 28.5129.5) 28.3(29.5) 28.0127.8) 28.8(27.0) 28.5(28.0) 29.1(28.1) 29.5(28.0) Sept. 26 22.7(22.7) 22.5(21.5) 24.2(22.2) 25.2123.0) 25.3(23.2) 25.5(23.0) 26.2(23.0) 22.0(23.2) Oct.10 26.3(27.5) 26.5(28.0l 26.5(28.8) 26.8128.0l 26.8(28.0) 27.0(27.5) 27.0(27.5) 28.5(27.8) Oct. 31 24.2(25.4) 24.8(24.5) 24.6(25.5) 25.0f25.5l 24.7(26.2) 25.1(25.6) 25.5(25.5l 28.0(28.5) Nov. 7 19.8(24.0l 23.0<24.0) 22.2(22.8) 21.5<22.0l 22.0(22.1) 23.5(22.0) 23.0(22.51 --­Dec. 12 22.0(22.0) 19.0(21.3) 19.2(21.3) 18.8(20.5) 19.0(20.5) 19.0(21.8) 19.0(21.2) 20.0(20.0) 1949 Apr. 10 23.8(25.2) 22.2(24.2) 23.2(26.0l 22.8126.l) 23.0(26.4) 23.0(26.3) 22.3(26.0l 25.8(21.0) Apr. 24 28.1(26.2) 23.7(25.1) 22.4(26.6) 24.8(24.0) 23.9(26.0) 24.2(25.0) 23.8(24.2\ 26.0(24.6) May8 26.4(23.8) 21.8<28.8) 27.0(26.8\ 27.U26.8l 26.0(26.8) 27.0(27.0) 27.2(27.8\ 30.5(28.8) May22 27.0(28.4) 26.8(27.5\ 26.8(27.7) 27.4(28.2) 27.8(27.5) 28.0(28.5) 28.0(28.0) 29.8(30.0) .Tune 11 ---28.9128.2\ 29.0(28.8) 29.0<28.2) 29.0(28.1) 29.0(28.5) 29.0(28.51 --­June 24 ---29.5(30.0l 30.0(29.8) 29.9130.0) 30.0(29.8) 30.2(29.4) 30.1(29.8) 31.5(32.0) July9 ---29.0(30.0) 29.1(30.1) 29.3130.l) 29.3130.0) 30.0(30.0) 30.1(30.0\ 31.5(30.5) Aug. 7 ---27 .5(29.0) 27.8(28.5\ 28.0f28.5l 28.0(29.0l 28.0(29.0) 28.8(28.8) 29.8(31.0) Aug. 20 ---31.4(29.9) 32.0(29.0) 32.0128.9) 32.0(28.8) 32.0(29.1) 32.1(29.1) TABLE 2 The Salinity in Parts Per Thousand Are Given for Each Station. The Determinations Were !\lade by Titration. Stal ion 1947 Nov. 9 32.5 32.5 32.5 32.5 31.9 31.9 31.6 31.5 Nov. 29 29.2 29.1 29.4 29.2 29.3 29.0 29.2 25.8 Dec. 6 27.7 28.3 28.3 28.3 28.4 28.0 27.8 24.4 Dec. 31 31.5 31.7 31.7 31.6 31.8 31.8 31.8 31.8 1948 Sept. 19 29.6 29.8 29.3 29.3 29.2 29.2 28.7 28.7 Sept. 26 28.5 28.4 28.0 28.0 28.0 27.9 27.9 28.l Oct. 10 31.9 31.5 31.6 31.9 31.8 31.5 31.5 30.9 Oct. 31 31.5 31.1 31.2 31.0 31.0 31.0 30.9 31.0 Nov. 7 38.6 32.3 32.3 32.3 32.1 32.0 31.7 Dec. 12 31.9 31.6 31.7 31.7 31.4 31.6 31.9 32.2 1949 Apr. 10 29.3 32.3 32.3 32.2 31.9 32.2 31.4 29.8 Apr. 24 27.7 28.1 28.3 28.2 28.0 27.6 26.5 24.7 May8 33.2 23.l 23.1 22.5 22.3 21.7 21.6 21.9 May22 35.9 27.6 27.8 May 29 32.9 32.7 32.9 32.3 31.9 June 11 33.2 32.8 32.2 31.9 31.6 31.6 June 24 34.7 34.6 34.3 34.l 34.00 33.9 33.6 July9 35.2 34.9 35.3 35.3 35.l 34.3 34.9 Aug. 17 31.7 31.6 31.5 31.6 31.7 31.5 31.2 Aug. 20 35.9 35.6 35.7 35.7 35.6 35.7 177 mm. Only eleven fish were as long or longer than 100 mm. These were: Elops saurus, Strongylura timucu, Oligoplites saurus, Polydactylus octonemus, Mugil cephalus 2, M. curema 2, Trachinotus carolinus 2, and M enidia beryllina. The last specimen was 114 mm. long, which is huge for that species. The species and the numbers caught are listed in decreasing order in Table 3. It is divided into the three periods of the work, as shown in Tables 1 and 2. The pass and TABLE 3 The Catch by Species at the Pass Stations and the Beach Stations Are Shown Separately for Each of the Three Periods of Work, With Totals and Average Catches. Nov.·Dec., 1947 Beach Pass Sepl.·Dec., 1948 Beach Pa ss April-Aug ., 1949 Beach Pass TOTALS Beach Pass Beach and Pass Trachinotus carolinus Harengula pensacolae Mugil curema Menidia beryllina 73 0 4 5 1 0 15 146 170 1,481 3 80 313 64 276 2,959 679 468 58 11 362 590 244 3,202 2,160 475 143 12 675 669 666 3,214 2,835 1,144 809 Fundulus similis 2 56 8 222 2 153 12 431 443 Anchoa mitchilli 3 4 208 5 171 0 382 9 391 Micropogon undulatus 0 0 0 2 238 0 238 2 240 Brevoortia sp. 0 0 0 0 190 0 190 0 190 Menticirrhus littoralis 23 0 47 0 119 0 189 0 189 Anchoa hepsetus 0 0 125 5 58 0 183 5 188 Cynoscion nothus 0 0 0 0 159 0 159 0 159 Cyprinodon variegatus 0 1 0 110 1 45 1 156 157 Caranx Latus 0 0 0 0 117 4 117 4 121 Eucinostomus sp. 0 0 0 0 0 104 0 104 104 Stephanolepis hispidus 0 0 0 0 97 0 97 0 97 Polydactylus octonemus 0 0 0 0 95 0 95 0 95 Sciaenops ocellata 0 0 0 0 29 16 29 16 45 Syngnathus sp. 0 0 5 0 31 1 36 1 37 Anchoa sp. 0 0 2 0 30 0 32 0 32 Gobionellus boleosoma 0 0 0 0 0 27 0 27 27 Menticirrhus americanus 1 0 13 1 0 0 14 1 15 Trachinotus glaucus 0 0 0 0 14 0 14 0 14 Strongylura sp. 0 0 3 5 2 1 5 6 11 Others 6 7 19 13 29 2 54 23 76 Totals 117 230 2,164 1,016 5,546 1,560 7,827 2,806 10,633 Number of Hauls 24 8 36 11 55 10 115 29 144 Average per Haul 4.9 28.8 60.1 92.4 100.8 156.0 68.l 96.8 73.8 outside beach stations are also separated. At the two pass stations 2,806 fishes were caught. These are listed separately because certain fishes which are found in inside waters are not found in the outside beach waters. All of the gobies, Gobionellus boleo­soma, and all Cyprinodon variegatus but one, were taken in the passes. Similarly, almost all Fundulus similis were taken inside. It was noted before (Gunter, 1945) that this species is not found on the outside beach, except for rare strays next to the mouths of the passes. The great majority of the silversides, Menidia beryllina, were taken in the pass. Gunter (op. cit., Table 24) has shown that this species and the Cyprinodontes are pri­marily bay fishes and that their restriction to such waters is not caused by salinity limits. Eleven species, caught mostly on the beach, made up 0.1 per cent of the catch. Five species, Trachinotus carolinus, Harengula pensacolae, Mugil curema., Anchoa mitchilli and Micropogon undulatus, made up 82.4 per cent of the beach catch. The last three are scattered widely in the bays, too, but the first two are more abundant on the outer beach. The little pompano feed to a large extent on young Harengula, and are commonly found in the summer with their stomachs distended with them. The average catch was consistently greater within the passes than on the beach. It appears that small fishes are more abundant in protected waters than on the open shore. The 1947 catches may be called late fall catches. An average of 10.9 specimens were caught per haul. The average beach haul was quite low, 4.9 fish per haul. In six hauls nothing was caught and several hauls yielded only one fish. On the beach the most abundant fishes were the pompano, Trachinotus carolinus, and the Gulf whiting, Menti­ cirrhus littoralis. Menidia beryllina and Fundulus similis were the most abundant species in the passes. The 1948 catches may be taken as late summer and fall catches. Forty-seven drags yielded 67.6 fish per haul. The silver sardine, Harengula pensacolae, the two anchovies, Anchoa mitchilli and A. hepsetus, and the pompano, T. carolinus, were the most abun­dant fishes on the beach, while Harengula pensacolae, Menidia beryllina, Fundulus similis and Cyprinodon variegatus were the most numerous in the passes. The 1949 catches may be seasonally classified as spring and summer catches. An average of 109.3 fish per haul were caught from 65 hauls. On the beach the pompano, T. carolinus, the white mullet, Mugil curema, and the sardine, Harengula pensacolae, were dominant. Mugil curema, Harengula pensacolae, Menidia beryllina, and Fundulus similis were most abundant at the pass stations. Ninety-six specimens of Stephanolepis hispidus, the filefish, were taken on the Gulf beach from April to August, 1949. They have not been reported there before. They ranged from 31 to 44 mm. long and were caught where the water salinities ranged from 21.7 to 34.8 per mille. In addition to the more abundant fishes listed in Table 3 the following species were taken occasionally: Strongylura sp., Eucinostomus sp., Mugil cephalus, Astroscopus y-graecum, Oligoplites saurus, Eucinostomus gula, Trachinotus falcatus, Eucinostomus argenteus, Membras sp., Chloroscombrus chrysurus, Strongylura timucu, Strongylura notata, Lobotes surinamensis, Anchoa lyolepis, Hyporhamphus unifasciatus, Elops sau­ rus, Lagodon rhomboides, Pomatomus saltatrix, Abudefduf saxatilis, Selene vomer, Caranx sp., Alutera scripta, Sphoeroides sp. CHARACTERISTICS OF THE DOMINANT SPECIES POPULATIONS The two chief beach fishes are the common pompano, Trachinotus carolinus, and the sardine, Harengula pensacolae. Both have offshore populations of larger individuals, but the young are restricted to the Gulf beach. The sardine is a plankton feeder and the young pompano seem to subsist largely upon them during their early life in the surf. These two fishes made up 69.3 per cent of the total beach catch. The mullet, Mugil curema, the anchovies, Anchoa mitchilli and A. hepsetus, and the small croaker, Micro­pogon undulatus, were common at times. These findings correspond to the work previ­ously reported (Gunter, 1945) except that Polydactylus octonemus was found to be more abundant than M. undulatus. At least three species of small pompano and possibly four were taken on the Mustang Island beach. One of these, which was called Trachinotus argenteus in the original notes, is here synonymized with T. carolinus, following Ginsburg (1952). It was separated at sizes up to 82 mm. but specimens are no longer available and the case is uncertain. Another species, at a size of 23 mm., was dusky or blackish, had reddish orange dorsal spines and a reddish orange eye. It seemed to be T. falcatus. It was taken in June, 1949. A third species, at a size of 23 mm., was reddish orange on the lower caudal lobe, the anal fin edge and large spine. It was called T. goodei, which Ginsburg synonymizes with T. glaucus, and probably belonged to the latter species. Table 4 shows the size limits and crude modes of the pompano caught. It confirms the previous statement (Gunter, 1945) that the very young are recruited into the population over an extended period during the warmer months, indicating a long spawning season for the pompano on the Gulf coast. Population Studies of the Shallow Water Fishes TABLE 4 Length-frequency Data on the Pompano, Trachinotus carolinus Caught on the Gulf Beach of :'.\lustang Island. Years Months Size Limits mm. Crude Mode ~umber of Specimens 1947 1948 1949 November December September October November April .'\Iav Jun'e July August 20-123 83-142 21-79 19­84 27-74 l+-52 15-83 13-83 16-81 l+-99 Indeterminable Indeterminable 33 30 33 17 23 50 l8 23 38 2 26 43 39 68 856 242 104 202 TABLE 5 Length-frequency Data on the Sardine, Harengula pensacolae, Caught on the Gulf Beach. Years Months Size Limits mm. Crude Mode :\"umbe-r of Specimens 1948 September 23-81 38 and 63 864 October 29-67 43 and 53 258 November 28-39 2 1949 April 25-32 28 41 May 27 l June 23-52 28 366 July 25-67 28 30 August 29-93 47 141 Table 5 gives size limits and crude modes for the silver sardine, Harengula pensacolae. The data indicate a relatively long spawning season. Gunter {1945) has previously noted evidences of two spawning peaks during the summer and a low point of abundance in July. The same thing is indicated but not proven by these data. Similarly the peak of abundance was found to be in the fall. SEASONAL AND L.\TITL"Dll\"AL COMPARISONS There is a seasonal succession on the Gulf beach, which can be partially characterized as the result of this and the previous work I Gunter. 1945). In most years the beach is practically unpopulated in winter except for a few bay fishes next to the passes. In some years this depopulation shows up in December, but in others it comes as early as Octo­ber. Besides the diminution in numbers or absence of many species in the cooler months there is a slow change in the complex of a few dominant species, which is doubtless different each year. Table 6, lists the species in order of their abundance for the seasons. The complexes for the different years are different, but the pompano and sardine are generally present in considerable numbers except during the cooler months. Such seasonal successions are quite different from those on a plot of grassland or marine fouling organisms on an underwater object. In the latter, what happens or what is present at a given season depends somewhat on what was present before. In fish populations, which grow as individuals and migrate about. the connections with their successors are much more tenuous and there is no proof that the midwinter population would not be the same if the summer population had been altogether different. Pearse, Humm and Wharton 0942) found that the silverside, Menidia menidia, the anchovy, Anchoa mitchilli, and the pompano, Trachinotus carolinus, were the most Population Studies of the Shallow Water Fishes TABLE 6 A Comparison of the Most Abundant Species on the Gulf Beach as Found by Gunter 0945) in 1941-42 and the Present Work 1947-49, at the Different Seasons. The Fishes are Listed in Order of their Abundance. Gunter (1945) Presenl dala Number Spring Anchoa mitchilli Trachinotus carolinus 1,076 Mugil cephalus Mugil curema 420 Polydactylus octonemus Micropogon undulatus 232 Summer Polydactylus octonemus Trachinotus carolinus 905 H arengula pensacolae Harengula pensacolae 605 Trachinotus carolinus Brevoortia sp. 155 Fall A nchoa mitchilli Harengula pensacolae 1,481 Harengula pensacolae Trachinotus carolinus 246 Trachinotus carolinus Anchoa hepsetus 132 Winter M ugil cephalus Menidia beryllina 61 Anchoa mitchilli Fundulus similis IO abundant fishes on an outside beach at Beaufort, North Carolina, during the summer of 1939. The next summer the most abundant species were M. menidia, Fundulus majalis, and T. carolinus. The two stations where these hauls were made are close to the broad inlet into Beaufort Harbor, which may account for the presence of M. menidia and F. majalis. Nevertheless, the association seems to be much the same. The species are the same or are cognates of those commonly found on the Texas coast. Warfel and Merriman (1944) worked an inside New England beach. Among the six most abundant fishes were Menidia menidia, Fundulus majalis and Brevoortia tyran· nus. The New England populations of beach fishes are more similar to those of North Carolina than to those of Texas, but several of the abundant species are cognates of those common on Texas shores. It appears that over such great distances as from South Texas to southern New England the populations of small beach fishes are similar. Summary 1. A study of fish populations was made in Aransas and Corpus Christi passes and the outer beach of Mustang Island in South Texas. Eight stations were sampled by small. mesh seine hauls eighteen times in three different years, 144 hauls being made. Water temperatures and salinities were taken at each station. 2. The fishes taken were all small and only 11 were greater than 100 mm. in length. The total catch, representing 44 species, was 10,633 fishes. The most abundant species were: Trachinotus carolinus, Harengula pensacolae, Mugil curema, Menidia beryllina, Fundulus similis and Anchoa mitchilli. Five species, T. carolinus, H. pensacolae, M. curema, A. mitchilli and Micropogon undulatus, made up 82.5 per cent of the beach catch. The last three are common bay fishes, but the first two are abundant only on the beach and in outside waters. 3. Fishes were most abundant on the beach in the spring and summer, less abundant in late summer and fall, and quite scarce in late fall and early winter. The Gulf beach is primarily a habitat of young pompano, T. carolinus, and the sardine, H. pensacolae; these two fishes made up 69.3 per cent of the catch. Comparisons made with seine hauls of earlier years showed some change of the populations based upon fluctuations of a few species. 4. The dominant species of the beach were the same species or species cognate to Population Studies of the Shallow Water Fishes those of North Carolina beaches, and they were all cognates of those reported from New England beaches. 5. All Gobionellus boleosoma, nearly all Cyprinodon variegatus and Fundulus similis, and the great majority of Menidia beryllina, were taken in the passes rather than on the outside beach. More fishes were caught per haul at the stations inside the passes than on the open beach, but the species were somewhat different. Four species, H. pensacolae, Mugil curema, M. beryllina, and F. similis made up 87.0 per cent of the pass catch. 6. The work confirms previous findings that young pompano are recruited into the population over an extended period during warm months. It also supports reports of a double spawning peak for H. pensacolae. Small specimens of the filefish, Stephanolepis hispidus, are reported from the beach for the first time. LITERATURE CITED Collier, Albert W. and Joel W. Hedgpeth. 1950. An introduction to the hydrography of the tidal waters of Texas. Pub!. Inst. Mar. Sci. Univ. Tex., 1(2): 121-194. Ginsburg, Isaac. 1952. Fishes of the family Carangidae of the northern Gulf of Mexico and three re· lated species. Pub!. Inst. Mar. Sci. Univ. Tex., 1 (2): 43-117. Gunter, Gordon. 1945. Studies on marine fishes of Texas. Pub!. Inst. Mar. Sci. Univ. Tex., 1(1): 1-190. Hoese, H. D. 1958. A partially annotated check list of the marine fishes of Texas. Pub!. Inst. Mar. Sci. Univ. Tex., 5: 312-352. Pearse, A. S., H. J. Humm and G. W. Wharton. 1942. Ecology of sand beaches at Beaufort, North Carolina. Ecol. Monogr., 12: 135-140. Parker, Robert H. 1955. Changes in the invertebrate fauna, apparently attributable to salinity changes,in the bays of central Texas. J. Paleont., 29: 193-211. Warfel, H. E. and Daniel Merriman. 1944. Studies on the marine resources of southern New England. I. An analysis of the fish population of the shore zone. Bull. Bingham oceanogr. Coll., 9: 1-91. Ecology and Taxonomy of Recent Marine Ostracodes in the Bimini Area, Great Bahama Bank Loms S. KoRNICKER Institute of .llarine Science, The Cniversity of Texas, Port Aransas, Texas Contents ABSTRACT ------------------------------------------------------------------------------_________ ----________________Page 194 Il\TRODUCTI0 l\ --------------­ ----------------------------------------------------------------------------------------------194 Acknowledgment M ETHODS ------------------------------------------------------------------------197 GEl\ERAL PATTERl\S OF OsTRACODE D1sTRrnuno:-; ---------------------------------------------------199 E:-;vrnoNME;\TAL FACTORS AFFECTil\G OsTRACODE D1sTRIBCTION --------------------------202 Influence of Substrate on Ostracode Distribution Influence of Salinity and Temperature on Ostracode Distribution Influence of Water Color and Photoaxis on Ostracode D:stribution Influence of Dissolved Oxygen and pH on Ostracode Distribution Influence of Organic Carbon and Organic Detritus on Ostracode Distribution Influence of Current Velocities on Ostracode Distribution SYSTEMATICS Al\D O c ccRRE!\CES OF OsTRACODE SPECIES --------------------------------------------229 Su :\1 :\er5 indicate the to:al number ofspecimens per sample I 10 cc. I. Ecology and Taxonomy of Recent Marine Ostracodes ~ 0 2 Km SCALE D I -5 D 6-40 Fie;. 6. Areal distribution of living Podocopa and Platycopa. Numbers indicate the total number of specimens per sample (10 cc.). Environmental Factors Affecting Ostracode Distribution INFLUENCE OF SUBSTRATE ON OsTRACODE D1sTRIBUTION The sediments in the Bimini area have been subdivided by Mr. Purdy and the writer into six facies: 1 chalky facies (I), pitted facies (II), unpitted facies (Ill), glazed facies (IV) , faecal pellet facies (V), and oolite facies (VI). Figure 11 represents the areal dis· tribution of these facies. These facies are distinguishable visually under a binocular microscope and have peculiar physical and chemical properties. Characteristic prop· erties of the sediments of facies I, II, IV and VI are presented in Table 1. 1 Facies is here used to define a geographical area coYered by sediments distinguishable on thebasis of physical, chemical, or biological properties from contemporary sediments covering adjacent or nearby areas. Ecology and Taxonomy of Recent Marine Ostracodes 203 TABLE l. Chalky Pitted Glazed Oolite facies facies facies facies I II IV VI Dry Bulk Density, gm/cc* 1.0 1.45 1.45 1.6 Organic Carbon, '}"o dry wt. 0.7 0.5 0.5 0.4 Median Diameter, mm.** 0.28 0.26 0.28 0.26 Sortin~ Index, (So) * * 1.80 l.35 1.26 1.20 • Dry bulk density was obtained by weighing a known volume (10 cc) of dried sediment. **Data supplied by Edward G. Purdy. For definition of So see Krumbein and Pettijohn (1938). A description of each facies follows: Chalky fades (I) : The sediments within North Bimini harbor and North Sound, with the exception of the northern end of the latter region, consist of bioclastics which often appear chalky and are mixed with extremely fine-grained particles of calcium carbonate (mostly aragonite). These sediments do not pack well, and the dry bulk density is ' ' ' 0 I I 0 0 j( I I I I I I 0 14 0 6 0 14 0 22 0 3 I 0 t I I I I I I I I I I I I I : 2 0 25 0 6 0 3 0 r 0 13 0 6 0 3 0 0 I 0 ' 0 ! I I 6 0 6 I 0 8 0 14 I ' ' ' /I 9 o•0 0 4 0 0 5 0 0 0 I 4 ' o',1i10 19 5i Oo 0 I 0 3 0 I I I I I I I I 17 0 0 0 3 I 0 7 0 0 0 7 0 2 0 29 0 3 01 0 2 0 3 0 15 0 3 0 3 0 38 0 cP56 55 ~Oo 8 44 I 0 21, <{) ,420 1 8 5156 I ~,p::?.C§. •t/?.' ' t 21 12 0 26 0 0 7 8 21 0 ~~o 49 I b o 21 0 90 0 •· 0 5 0 24 0 I 0 I 0 0 0 4 0 7 0 Fie. 7. Areal distribution of dead Podocopa and Platycopa. Numbers indicate the total number of specimens per 10-gram sample Oarger than 124 M fraction) . Ecology and Taxonomy of Recent Marine Ostracodes 1 D 0 -2 2 Km 0 3 -25 SCALE CEJ 2 6-200 [£J 201-600 FIG. 8. Areal distribution of total dead Podocopa and Platycopa. Numbers indicate the total number of speeimens per 10-gram sample (larger than 124 .'\I fraction J. very low I about 1.0 gms./ cc.) . Because of the presence of the fine material, the sorting index (SO) is relatively high (about 1.8). The region covered by this facies includes extensive areas partially exposed at low tide, tidal channels traversed by fairly rapid currents, thickly vegetated sand as well as sand virtually barren of plants, and rock bottom areas covered by a few millimeters of sand (Figs. 12-15). Organic detritus forms a fi lm over the sediment in practically all of this region, and the organic carbon content of the sediment fa higher than in the sediments on the Banks or outer platform. The abundance of living ostracodes within the area covered by this facies varies. Myodocopa, Platycopa, and Podocopa are common over most of the region. The abun· dance of dead shells of Podocopa and Platycopa is unusually high. This is thought to be the result of a high rate of ostracode accumulation in comparison to the rates of accumu­lation of preservable remains of other organisms. Pitted facies (II): Bioclastic sediments which ha\·e been subjected to intense pitting. possibly by algae, are situated south of South Bimini and extend into the deeper waters west of Turtle Rocks. These sediments form a thin laYer from a few millimeters to a meter or so in thickness on top of the submarine platform from which the Bimini Islands I I I I I I I I I 5 0 I 0 0 I I 0 I I I I I I 0 0 0 0 2 I I I I 0 0 0 0 0 I I I 2 I I 0 I I I I I 0 0 0 0 I I I I I I I I I I 0 0 2 4 c 0 0 ( 8 0 43 8 0 828 I 0 ') c c 0 18 SCALE 2 5:38 0 0 fflo 8 0 6 v 2 !· 0 0 6 60 0 Fu;. 9. :\ real distribution of dead Podocopa and Plat~·copa. :\umber:; indicate the number ol :;pec:i­mens per 10-gram sample (smaller than 12-! :\I fraction I. 206 Ecology and Taxonomy of Recent Marine Ostracodes and numerous cays project. The sediments of this facies butt against and blend with oolitic sediments which delimit this facies on the east. Although sand, often supporting profuse vegetation, usually of Thalassia, is accumu­ lating in the protected near-land areas, a large part of the region covered by pitted bio­ clastics is rock bottom containing small pockets of sediment or at most a film of sarid a few millimeters thick. Where a rock bottom is present, a varied algal flora, often dominated by the genus Sargassum, combine with sea whips, corals, and sponges to form a community rich in numbers and diversity (Figs. 16--21). 1 0-5 0 2Km SCA LE !--'>.---1 6 . 100 ~ JOO PLUS Fie. 10. Areal distribution of total dead Podocopa and Platycopa. Numbers indicate the number ofspecimens per 10-gram sample (smaller than 124 M fraction). Living ostraode abundance in the area covered by this facies is variable. Living ostracodes are common over most of the area but extremely abundant east of South Bimini and in a narrow strip just south of Turtle Rocks. The reason for this abundance is not known. The distribution of the dead ostracodes here is similar to that of the living ostracodes. Unpitted facies (III): The sediments within Cavelle Pond are well preserved bio­clastics and do not exhibit intensive pitting. Cavelle Pond supports a profuse vegetation. The western segment of the Pond con­tains Thalassia as the dominant flora, whereas Laurencia is dominant in the eastern part. The passageway which connects the two segments is covered in many places by only 15 cm. of water at low tide. Living ostracodes are common in this pond. In general, the same species living here are living in other marine environments in the Bimini vicinity. Dead ostracodes are ex­tremely abundant and consist predominantly of brackish water forms not living in Cavelle Pond at present. These shells are probably relicts from the t:me Cavelle Pond SCALE Fu;. 11. Sediment facies map. I, chalky facies; II, pitted facies; Ill, unpitted facies; IV, glazed • facies; V, faecal pellet facies; VI, oolite facies. 208 Ecology and Taxonomy of Recent Marine Ostracodes F1G. 12. Intertidal zone, North Bimini Harbor in front of Lerner Marine Laboratory. Mounds inforeground are formed by the worm Arenicola. was separated from the sea and contained brackish water (see section on "Salinity").Glazed facies (IV): Bioclastics which are rounded and appear glazed form the beach sands and sediment along the western shores of North and South Bimini. Aggregates ofthe glazed grains are common where the glazed bioclastics blend with oolite sandsaround the northern end of North Bimini. The shifting sands which are parallel to the coast line support little vegetation, al·though isolated patches of Thalassia and Chondria tenuissima (Goodenough and Wood­ward) appear to be thriving on this substrate. Rock-bottom areas off the coast arecovered by only a film of sand. These support abundant and diverse plant and animallife (Figs. 22, 23). Living and dead ostracodes are rare in the area covered by this facies. The rock-bot·tom areas seem to contain more ostracodes than the surrounding sand areas. The rarityof ostracodes in this region is considered to be the result of agitation of the sand bywaves. Ostracodes are, for the most part, burrowing animals, and many species may notbe adapted for existence in shifting sand. Faecal pellet facies (V) : Faecal pellets of the gastropod Batillaria minima wereabundant in the sediment in the extreme northern end of North Sound (Kornicker and Ecology and Taxonomy of Recent Marine Ostracodes 209 Purdy, 1957). Although the lime-secreting organisms (ostracodes, Foraminifera, pelecy­pods, gastropods) form but a small percentage of the total sediment in the upper fewcentimeters, their proportion materially increases in the sediment below the surface dueto poor preservation of the faecal pellets. Myodocopa are absent from the area covered by this facies. This is thought to be the F1G. 13. Holothurian and algae on rock bottom, North Bimini Harbor. Water depth is about 1 meter. Ecology and Taxonomy of Recent Marine Ostracodes FIG. 14. "Coral patch'', 1orth Bimini Harbor. Water depth is about 1 meter. result of the high and variable salinities encountered in this region. It was not possible to evaluate the role of sediment type, which actually might be quite important. Living Podocopa and Platycopa are fairly common. Dead ostracodes are extremely abundant. The abundance of dead ostracodes is attributed to the relative scarcity of other organ­isms with calcareous parts living in this area. Some podocopid species which are present in abundance in the sediment are restricted to this region; other species abundant else­where are absent here. This distribution is also considered the result of high and variable alinities (see section on "Salinity"). Oolite facies (VI) : The term oolite sand is loosely used in this paper to include calcium carbonate coated faecal pellets as well as spherical grains with well developed concentric structures. Two oolite bores are a minor part of the oolite sands in the Bimini area. Level oolite sand areas are commonly coated with organic detritus which makes them darker than the oolite bores. They also support more vegetation than the bores. Identification of these sands as oolite requires close examination of the sediment. They are not distinguishable from bioclastic sands on aerial photographs as are the bores. Sed:ments belonging to the oolite facies are characterized by being extremely well sorted; the diameter of most grains lies within the 124-248 micron range. The dry (and Ecology and Taxonomy of Recent Marine Ostracodes wet) bulk density of oolite sand is higher than that of other sands in the vicinity of Bimini (Figs. 24, 25). The organic carbon content of oolite sand is lower than in other sediments in this region. However, a film of organic detritus is not uncommon on the sediment. Living and dead ostracodes are virtually absent from the area covered by oolitic sands. The influence of the oolite sands on the distribution of ostracodes is especially notable in the submarine area around the eastern end of South Bimini. In this region ostracodes are abundant in the pitted and chalky bioclastic sediments but are absent or extremely rare in the adjacent oolitic sands. The reason for the rarity of ostracodes in the oolite sands is not definitely known. Vegetation is sparse in this region, and the organic carbon content of these sands is lower than in the other sediments in the Bimini vicinity, so it is possible that food supply may be the limiting factor. Organic detritus is present on much of the oolite sediment, and the vegetation, although sparse, possibly occurs in sufficient quantities to support more ostracodes than are present in these sands. In the Bimini vicinity, except for the two bores shown on the map insert, which probably are shifting, the sand has been somewhat stabilized by sparse vegetation; ripple marks and other evidence of shifting were not observed. Therefore, in this area Fir.. lS. F.,,hinoi 70 c. 0 w er N = 14 ... 50 lO 10 1675 1.87, I HS ORY BULK DENS I TY GJl:A WS ,_CF' CC F1G. 24. Comparison of dry bulk density of sediment from '·A·· chalky facies. ··ff· glazed facies, an·d ..c · oolite facies. is by far the dominant vegetation and mostly grows either in small patches or covers larger areas where it is usually sparse. Before the investigation the writer thought that large variations in ostracode distribu­tion and abundance would be found among the different bottom type~. Therefore, sampling was conducted so that each bottom type was sampled in all areas. Statistical consideration of the data did not produce significant differences with the exception of the platform area west of North B:m' ni, where both the liYing and dead ostracode abundance varied directly with the estimated areal coverage of the sampling locality by flora and fauna (P less than 5). The flora and fauna were principally confined to the rock bottom and practically absent from the wave-agitated sand areas (Fig. 27). The writer has no explanation for not finding larger variations in ostracode content on different bottom types in the other areas. Perhaps it indicates an abundance of food in all areas. Detailed sampling may bring out subtle differences masked by the present sampling procedure. INFLUENCE OF SALINITY AND TEMPERATURE ON OsTRACODE DISTRIBCTIO~ Salinity.-Typical salinity values (in parts per thousand) reported by Turekian (1957) for the Bimini area are listed in Table 2. These salinities were obtained during the spring mostly under dry conditions. After a heavy rain the regions with limited inter­course with the open sea experience marked lowering of salinity. For example, the salinity just below Mosquito Point in North Bimini harbor decreased from about 39 parts per thousand to 30.8 following a rain (Turekian, 1957). The salinity fluctuation of the Florida Straits and Great Bahama Bank water in the vicinity of Bimini was very small. In North Bimini harbor the fluctuation in salinity was somewhat greater, but in all three regions the salinity remained within the 30 to 40 218 Ecology and Taxonomy of Recent Marine Ostracodes ,' I'IIIIIIIIII I 1.40 I' 0 ,,,I 145 I I I I' 165 !' 0 ' l 50 0 ,' 160 0 ' ' I 1.25 I 0 , 1.40 ! 0 I 65 1.45/ o, 0 I I l I I I I' I 1.60 I 0 1.6 0 I 160 150 1. 65 0 ' 160 0 Ql.3 0 0.8 5 1.00 1. 60 0 I 2Km SC ALE b'' I 20 I 65 0 ;. 0 I 55 0 F1G. 25. Dry bulk density of sediment in vicinity of Bimini (grams per cc.). parts per thousand range considered by Hedgpeth (1951) to include normal marinewater. Fluctuation within this range generally has relatively little effect on the distri­bution of fauna (Dahl, 1956). Ostracode communities which inhabited these areas in theBimini region contained essentially the same species, and variation in ostracode abun­dance does not seem attributable to these small differences in salinity. The movement of water in and out of North Sound during change of tides is restrictedbecause of its geographical position at the end of the harbor and because of lateral landprojections and a rock sill athwart the current at its lower end. In June the water during periods of low rainfall became highly saline (more than 40 parts per thousand), andafter a considerable amount of rain the salinity dropped to about 31-35 parts per thousand or lower (Turekian, 1957). The steep salinity gradient encountered here orpossibly the high variability of the salinity seemed to have considerable effect on thefauna! and floral communities inhabiting the region. Echinoderms were absent, and Ecology and Taxonomy of Recent Marine Ostracodes 0 10 20 ~~ o:> 30 40 ~ GLAZED BIOCLASTICS 100 % 50 L---'==--~~~~~---=::::!'..~~~---'"=LJO 6 LIVING OSTRACOOES per IOcc 4 L-~~~~~~~~~~~~~"=~o 140 100 DEAD OSTRACODES Pt r 10 Q 60 20 FrG. 27. Platform Area West of North Bimini. A. Cross-section showing distribution of the sediment. B. Estimated areal subsurface coYerage by flora and fauna. C. Number of living ostracodes per 10 cc. sample. D. Number of dead ostracodes per 10-gram sample. TABLE 2 Salinity of Water in Bimini Area Salinity Florida Straits (near Bimini l 35.85-35.92 Great Bahama Bank (near Bimini I 37.50 North Bimini Harbor 36.10-39.40 North Sound 40.00-46.50 Cavelle Pond, South Bimini (1 sample I 31.52 corals as well as most sponges were restricted to the lower end where normal marine salinit'.es occurred (Fig. 28-29 I. The ostracode population also experienced a change in this region (Fig. 30). The genus Bairdia which was common in all other environments in the Bimini area did not occur in the upper part of North Sound. Loxoconcha dorsotuberculata (Brady) also was restricted to the lower end. Loxoconcha Levis (Brady) was collected only in the upper part of North Sound. Another species (Hemicythere sp.) was also restricted to the northern end. The last two species were not collected alive. Hemicytere sp., how­ever, often contained appendages which indicates that the specimens died recently. The Myodocopa also seemed restricted to the less saline !'\Outhern part. The sediment in the northern part of the Sound contained an abundance of empty ostracode carapaces. These could not be accounted for by an abundance of li,·ing ostra­codes. An explanation for this -favored by the writer is that the absence of other sedi­ment-forming organisms such as corals and the alga Halimeda permitted the empty ostracode carapaces to form a large part of the sediment. Cavelle Pond is a small pond on South Bimini (Fig. 1). It is almost bisected by a 220 Ecology and Taxonomy of Recent Marine Ostracodes 0 I 2 Km SCALE 1 (\ BAIRDIA ~ {seve ra l species) V \ Im 68~~8WMfcuLATA!B..,,1 mIJ HEMtCYTHE RE ton• 1peti11l ~MYODOCOPA PODOCOPA El LOXOCONC HA LEVIS {Br ody) ~PLATYCOPA ~GASTROPODS PELECYPODS WORM MOUNDS ~BATILLARIA SPONGE S ~CHONDRILL A NU GULA CORAL ~SEVE RAL S PE CIES FORAMINI F ERA Fie. 28. Faunal ranges in North Sound. north-south trending mangrove spur. The two halves of the pond are connected at thenorthern end by a shallow stream which has a rock bottom on which corals grow. Each section of the pond has a sediment substrate which supports a luxuriant crop of vege· tation. The only salinity determination available for this pond is 31.5 parts per thousand,obtained from the center of the western section of the pond (Turekian, 1957). Inter­change of water during tidal change is quite rapid, and it is probable that the waters inat least the western section are normally marine, although it is not unlikely that aftera heavy rain the salinity falls below 30 parts per thousand.The dominant ostracode genus present in the sediment is Cyprideis, a known brackishwater form (Brady, Crosskey, and Robertson, 1874; Sohn, 1951; Swain, 1955-). Alsoabundant in the sediment are valves of the pelecypod Anomalocardia cuneimeris whichis a recognized estuarine (poikilohaline) species (Hedgpeth, 1953, p. 178). Neitherof these forms which were abundant in the sediment were found alive.In Cavelle Pond many species of dead ostracodes in the sediment are not living in the Ecology and Taxonomy oj Recent ,llarine Ostracodes I 2Km SCAL E B BATOP HO RA ACETABU LA RIA D THALASS IA CYMODOCEA GRASS LAU RENCIA HALIMEDA PENICILLUS RHI POCEPHALUS FIG. 29. Floral range;: in '.\orth Sound. pond at the present time. Marine forms are li,·ing in the pond, whereas estaurine forms compose a large part of the sediment. An examination of the older maps of the region rewaled that Cawlle Pond was once closed off from the sea. The earliest map re· ,-iewed showing a connection between Cawlle Pond and the sea was the :\arnl Air Pilot. 1940, H. 0., No. 194-. united States l\ayy Department. Hydrographic Office. p. 198. Inhabitants of Bimini recall that the pass connecting Cavelle Pond with the sea was excavated in an attempt to reduce the number of mosquitoes breeding in the brackish water of the pond. Although living ostracodes belonging to the suborder Myodocopa were abundant around Bimini, their remains were absent from the sediment probably because of poor preservation of the carapace. which usually soften after death of the animal. With the exception of the North Sound and Cavelle Pond areas, species belonging to the sub­ 222 Ecology and Taxonomy of Recent Marine Ostracodes A ~ ROCK F:"•:•A BIOCLASTICS f:::.:.J FAECAL PELLETS I00 °1. ESTIMATED AREAL COVERAGE 3 by 50 L_~~~~~~~-=::::::::~====~======::::::_~~~~__JO 6 LIVING OSTRACODES 3per IOcc L.::::::::::::::=::=:=:::=::::::=:::==::=:T:ro:v:••:•:•=A= )==~======~~==========~'.__~~~~~__:~j O 500 D DEAD OSTRACODES per 10 9 (Traverse A} ~================~======~=====-~~~~~o Boirdia (several species) Loxoconcho dorsotuberculoto (Bro dyl E Hem1cythere (one species) RANGE OF TYPIC AL SPECIES {Traverse A) Loxochoncho levis {Brody) 137 81 119 204 187 216 217 219 F G H ISOHALINES SALINITY %0 SAMPLE LOCALITIES Boirdio (several species)6/3/55 0 2Km Loxoconcho dorsotuberculoto (Brody) SCALE Hemicythere (one species) Loxochoncho levis' (8rody) Fie. 30. l\orth Sound. A. Cross section showing distribution of sediment. B. Estimated co,·erage of subsurface by flora and fauna. C. Number of living ostracodes obtained in 10 cc. sample. D. Number of dead ostracodes obtained in 10 gram samples. E. Range of typical species along Tra,·erse A (see part G for location of Traverse A). F. Salinity isohalines in l\orth Sound. Data partly from Turekian ( 19571 . G. Sample locality map. H. ~lap showing distribution of typical species in North Sound. orders Podocopa and Platycopa represented by many empty carapaces in the sedimentwere also collected aliw during the periods in which this study was made. T emperature.-Daily temperature variations in the Bimini Yicinity were only a few Ecology and Taxonomy of Recent Marine Ostracodes degrees (Turekian, 1957), and under normal conditions this difference is not likely to have any effect on ostracode distribution. Occasionally, however. unusually cold winter temperatures have been encountered. For example, Dr. Louis A. Krumholtz (personal communication, 1956) recorded a temperature of 14° C. in North Bimini harbor on January 12, 1956. At this time fish were killed by the cold water of the harbor. The effect of this low temperature on the ostracodes of the region is not known, but it is pos­sible that the absence of living Hemicythere sp. from North Sound was a result of the low temperature of the previous winter. The carapaces of this species, which in many instances still contained appendages, were abundant in the sediment. I:wu.T:~cE OF WATER CoLOR A'ID PttoTOTAXIS O'I OsTRACODE DISTRIBUTIO'I The waters of North Bimini harbor often were greenish-yellow when viewed from a distance. This was especially noticeable during ebb tide where the harbor water entered the blue waters of the Gulf Stream. Yellow color may be due to a trace of organic com­pounds which absorbs blue and near ultra,·iolet light (Harvey, 1955) and may be de­rived from mangroves or marine vegetation. The color. as determined by a Taylor Color Comparator (Table 3), which probably did not differentiate subtle color variation, did not seem to have any influence on ostracode distribution. TABLE 3 Water Color and pH Location Water Color* pH* Northern end North Sound 15 8.0 Center North Sound 0-5 8.2 Big :'lfangrove Cay Southern end North Sound 0-5 0-5 8.2 8.2 Center North Bimini Harbor 0-5 8.l Tokas Cay Southern end North Bimini Harbor 0-5 0-5 8.2 8.1 West side Cavelle Pond 0-5 8.2 East side Cavelle Pond 0-5 8.2 South of Ca,·elle Pond 0-5 8.2 *All samples were obtained between the hours of 6:00 A.)!. and ):00 P.)l. during June, 195<>. In shallow water near land areas a considerable number of particles which reflect light upward are usually in suspension. The amount of reflected light is also affected by the substrate; open sand reflects more light than either grass or unewn rock. The amount of light reaching the bottom probably has an influence on ostracode distribution. Ostra­codes seem to be either negatively or positively attracted to light (see Kesling, 1951). The response to light of specimens of Myodocopa collected by dragging a net through a Thalassia patch in North Bimini harbor was observed in a petri dish for one hour and 35 minutes. The net response of this population to light was positiw I Fig. 31). INFLUENCE oF DissoLVED OxYGEN AND PH ON OsTRACODE DISTRIBUTION Dissolwd oxygen in the waters of Cavelle Pond and North Bimini harbor was de­termined using the Winkler procedure (Table 41. The waters were found to be suf­ficiently aerated, and it is improbable that ostracode distribution was affected directly by oxygen. ' ®: @'o @" 70 30 815 30 27 0 ® 2 miR. '''"" + ®+8 ®', 10 4 0 13 0 18 30 0 ®®I 21 ITIJ/1. t!) 111 111. 40 "'" "· Fie . 31. Response of Myodocopa to light. Petri dish was rotated before each observation. Light source was microscope lamp. Sample of ostracodes was obtained from Thalassia patch in North Bimini harbor. Community contained Sarsiella truncana (13 ), S. carinata (11), S . gigacantha (5), S. punctata (3), S. capillaris (3), Philomedes multichelata (1) , P. paucichelata (8J, Rutiderma dinochelata (1) , Asteropteron monambon (2). TABLE 4 Dissoh·ed Oxygen Content of Waters North Bimini Harbor* Oxygen ml/L Water Temperature c. Time Air Temperature c. Date 3.69 23.5 10 :30 22.9 24.4 11/ 10/ 56 3.93 24.l 11:30 24.9 23.4 4.64 25.7 1 :30 24.7 5.25 4:45 24.5 CaYelle Pond, South Bimini** Oxygen ml/L Water Temperature c. Time Air Temperature c. Date 5.03 5.16 25.2 2 :00 P. ~I. 2:15 11/ 6/ 56 4.99 24.9 2:20 25.1 5.30 2:30 5.38 24.7 2:45 * Location: east of Lerner Marine Laboratory, about 22 meters from shore, water about 1 meter deep. High tide was at 2 :30 P .~!. ** Location: about 30 meters within the entrance of Cavelle Pond, water about 1 meter deep.High tide was at 12 :00 ~!. The pH variation in the Bimini vicinity was small and did not appear to be a factor directly influencing ostracode distribution (Table· 3). This conclusion coincides with observations which were made in other areas by Tressler and Smith (1948) and Benson (1955). l '.\'FLUE'.'iC E OF 0RGA'.'i!C CARBON AND 0RGA'.'i!C DETRITUS ON 0STRACODE D1STRIBUTION Organic detritus.-A gelatin-like film containing fragments of shell debris and plants, un=cellular algae, and diatoms (Fig. 32) formed a tan-colored coating on the submarine sediments in the vicinity of Bimini (Fig. 33). A simple experiment in which water from North Bimini harbor was pumped into a settling tank demonstrated that this material was also suspended in the water (Fig. 34). Organic detritus is considered a basic food for benthic fauna (Dexter, 1944). Yonge (1953) questions the actual food value of detritus and states, "... possibly its major Ecology and Taxonomy of Recent Marine Ostracodes Sponge spicule Diatom (green tint) Fu;. 32. Organic detritus (sketched under high magnification). 0 o, 0 0 0 0 0 0 0 2 km 0 SCAL E 0 0 I o 8 I ' I Fu;. 33. Distribution of sample localities coated with organic detritus. Ecology and Taxonomy of Recent Marine Ostracodes ,,' Fie. 34. Photograph of sand on bottom of small tank through which 3,000 gallons of water were passed at the rate of 2/ 3 gallon per minute. In the top part of the photograph the organic detritus which settled from the water has been removed to expose clean sand. The contrasting dark color in the lower part of the photograph is the result of organic detritus accumulation. importance may be as a culture medium for Protozoa and bacteria." Cannon (1934) concludes, after studying the feeding mechanism of Cypridina antarctica and members of the subfamily Philomedinae, that these ostracodes probably feed on minute detritus. Klugh (1927) found that several fresh water species ingested both algae and detritus but did better on algae. Hoff (1942) states, "The foods of few species (of ostracodes) have been studied in detail, although a study of food habits might have some bearing on problems of dis· tribution." Very little has been added to knowledge concerning the food habits of ostra· codes since Hoff made the statement quoted above. It seems probable, however, that at least some of the ostracodes in the Bimini vicinity feed on organic detritus. The presence or absence of organic detritus seemed dependent to a large extent on current action. In shallow areas where the sand was in almost continuous action, organic detritus was rare and probably had been removed by currents. Ostracodes were usually rare in these sands, but this was probably because agitated sand is an unfavorable habitat rather than because of the lack of detritus. Extensive areas of oolitic sand which were almost barren of ostracodes were coated with organic detritus. Ecology and Taxonomy of Recent Marine Ostracodes Organic Carbon.-Sediment from creeks flowing through mangrove swamps on East Bimini were found to be relatively high in organic carbon ( 1.39 to 2.90 percent) (Fig. 35). Many of these sediments were dark in color, emitted an H2S odor, and contained fragments of mangrove leaves. Ostracodes were rare in these sediments. Sediment within North Bimini harbor and most of North Sound contained the next highest organic carbon content (0.44 to LOO percent; average 0.74 percent). The lowest organic carbon content was encountered in the sediments on the outer side of the islands (0.39 to 0.61 percent; average 0.49 percent). Within small areas, such as North Bimini harbor, the organic carbon content of the sediment was uniform, even though the amount of visible vegetation varied at the sample stations. Perhaps this is the reason for finding in this study about the same number of ostracodes living in open, nonshifting sand areas as in neighboring vege­ tated areas. INFLUENCE OF CURRENT VELOCITIES ON 0STRACODE DISTRIBUTION The highest current velocity in the area caused by tides was encountered at the entrance of North Bimini harbor, where the velocity reached 1.2 meters/ second during ebb tide. This dropped to 0.16 meters/second outside the entrance. Within the harbor the highest current velocities occurred within the main channel. A measurement of 0.3 meters/ second was made in this channel about 1.5 kilometers from Entrance Point. In the northern part of the harbor the current velocities were very low (0.01 meters/ second). South of South Bimini, except close to shore where the crescent shape of the southern shore line created a low current velocity area, the current velocities approached those encountered in the main channel of the harbor. The highest velocity measured in this area was 0.5 meters/second, which occurred immediately above the crest of an oolite ridge. The restriction in the path of water flow created by this ridge caused an increase in current velocity. Velocities in about the same range as those south of South Bimini were obtained north of North Rock. Velocity measurements taken immediately to the east of Bimini, on the Great Bahama Bank, indicated low velocities in this area (maxi­mum velocity recorded here was 0.18 meters/second). A current velocity measurement of 0.2 meters/ second was obtained west of North Bimini, between Paradise Point and Stepping Rocks. In this area waves were of greater importance as current producers than were the tides. Submarine ripple marks as well as the presence of winter berms on the beaches testified to the sediment-moving ability of the waves along the western shore of North Bimini. Ostracodes were rare in the shifting sand area west of North Bimini and were absent in the channel at Entrance Point, where high water velocities probably disturbed the sediment. In the area between Turtle Rocks and South Bimini the current seems to keep the rock swept clean of sediment, and ostracodes were rare. From these observa­tions the writer concludes that current velocities sufficiently high to stir the sediment may be a major factor in inhibiting the colonization of an area by ostracodes. Most of the submarine environment included in this study was in shallow water less than seven meters deep. I I I I I I t I I I I I •I 0 .61 I I I I I I I I I I I I I • I 0.50 I I I I I I I I I I I I • • I 0 .39 I I 0.55 I I I I I I I I 0 .72 I 0.44 I • I I • • 0.50 I 0.55 I I • I I I •I 0.39 I I I Q I I I I I I I 0 2KmI SCALE I I i I I I II gI I I ~ IIII I ' . I I I I I I Fie. 35. Organic carbon content of the sediment (percent dry weight). Ecology and Taxonomy oj Recent Jlarine Ostracodes Systematics and Occurrences of Ostracode Species Descriptions of Bimini ostracodes follow. These include diagnostic morphologic fea­tures only and are not complete descriptions. Morphological features are shown in detail in illustrations. When sexual differences are striking. or are important for identi­fication, both sexes have been figured and described. Variations in taxonomically im­portant appendages are mentioned in the diagnoses and usually are figured . App' ndages (most frequently the furca) of juwniles haw been figured as an aid to identification. Numbers given to spec:mens are based on sample or station numbers used in the present investigation. Dissected specimens haw been mounted in glycerin and protected with paraffin-sealed coyer-glasses. Holotypes are dried shells. or. if these were not aYail­able, shells preserved in alcohol. Holotypes haYe been deposited with the Columbia Ilnivf rsity Geology Department. Shells for measurement were selected at random from ,·arious growth stages. The drawings and photographs of ostracode morphology, Figs. -16--39. are collected at the back in an appendix for the com·enience of taxonomists. Suborder ~Iyodocopa Family Cypridinidae Subfamily Cypridininae Genus Cypridina l\lilne-Edwards Cypridina squamosa subspecies lerneri Kornicker. new subspecies trigs. -17 lA-B: 48, A-D: -19. A-El Diagnos's: The shell is elliptical in lateral Yiew, with protruding postero-Yentral corner. The antenna! sinus is fairly deep and the rostrum prominent. A cluster of muscle scars occurs near the center of the shell. Maximum shell height is posterior to the middle. The adult furca bears eight claws: the fourth claw is smaller than the second and fifth; the second claw is attached to the lamina without a demarcation line at its base. The second antenna has basal spines which are longer on the distal joints. The second­ary branch of the second antenna bears two long and two short bristles on the proximal joint and one long bristle on the end of the distal joint. The sewnth limb has 23 to 29 lateral setae. Comparisons: The shell of C. squamosa lerneri is wry similar to that of C. squamosa squamosa Mueller. C. squamosa lerneri is smaller than C. squmnosa. A graYid female of C. squamosa lerneri (specimen number 2-11-31 measured 2.0-1 mm. whereas the female of C. squamosa reported from the Bay of :\aples by Mueller was 3.3 mm. in length. C. squamosa squamosa bears six to Sfien claws on the furca. whereas the adults of C. squamosa lemeri ha,·e eight. Remarks: It is possible that two female ostracodes from the Dry Tortugas identified by Tressler 11949, p. 3351 as C. squamosa belong in the subspecies C. squamosa lerneri. Tressler's specimens haw only seven furcal claws but, as these specimens were only 1.71 mm. in length, they may haw been immature. The present study shows that im­mature specimens ha,·e fewer furcal claws I Fig. -18. A-Cl. Ecology and Taxonomy of Recent Marine Ostracodes Shell measurements (in mm.): Specimen number Length Width Height 212-1 ( holotype) 1.31 0.55 0.79 247-3* 2.04 1.40 656F 1.42 0.94 P-1 0.71 0.4.S 246-7 1.18 0.80 93 1.58 0.62 1.03 127-1 1.29 0.87 127-2 0.81 0.54 127-3 1.26 0.84 127-4 1.26 0.81 127-5 0.71 0.45 127-6 0.74 0.48 93-2 0.84 0.54 170 1.19 0.68 5306M 0.71 0.48 * Gravide female. Material: Twenty-four specimens were obtained from bottom samples and one (a male) from a surface tow. Five specimens. including a male and a mature female, were dissected. The holotype is specimen number 212-1 (an immature specimen) which is illustrated in Fig. 47, la-b. Carapace form was the same in mature and immature specimens. Occurrence: Cypridina squamosa lerneri was most abundant east of South Bimini (Fig. 36). Specimens were collected in waters having temperatures of about 29.2°C. and salinities about 37.5 parts per thousand. Substrate was calcareous sand, and usu­ally algae and Thalassia were growing in the area. One male was obtained in a sudace plankton tow in North Bimini harbor. Subfamily Philomedinae Genus Philomedes Lilljeborg Philomedes multichelata Kornicker, new species (Figs. 46, 3A-B; 50, A-E; 51, A-D) Diagnosis: The shell is elliptical in lateral view, with a shallow but distinct anterior sinus. The surface of the valves contains closely spaced pits and muscle scars. The furca is long, narrow, and triangular in outline, with five strong and nine weak claws. The first antenna bears, on the penultimate joint, a sensory bristle which has numerous setae on the outer edge of its flattened base. The second antenna bears non­spinose, natatory setae; basal spines are absent. The mandible has, on the ultimate joint. two strong claws, one small claw, and two bristles. The exopodite of the mandible bears two bristles. The seventh limb has about five setae and terminates in a claw-like process. The female is not known. Comparisons: Philomedes multichelata is closely related to Philomedes oblonga Ecology and Taxonomy of Recent Marine Ostracodes 0 0 I 0 0 • 0 0 80 0 0 0 0 0 0 0 0 •20 0 0 0 0 •I 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 8 0 •2 ;. I 0 0 • 0 00 0 0 Fie. 36. Distribution of Cypridina squamosa lemeri Kornicker, new subspecies. Species was absent from samples shown as empty circles. Small circles represent trawl samples; large circles representspot samples. (Juday) Mueller, from which it differs in having 14 furcal claws, whereas, P. oblonga has only 10. Shell measurements (in mm.) : Snecimen m1rnher Length Width Height 686Z-6 (holotype) 1.04 0.57 0.60 248 0.75 0.3 0.51 245 0.91 0.5 0.58 686Z-l 0.99 0.41 0.55 686Z-3 0.96 0.52 686Z-4 0.98 0.53 238 1.04 0.66 (approx.) Ecology and Taxonomy of Recent Marine Ostracodes Material: Four specimens were obtained from bottom samples and several hundred from a single night surface tow. Four specimens (all males) were dissected. The holo­type is specimen number 686Z-6, Fig. 46, 3a-b. Occurrence: Two individuals were collected from the bottom southeast of South Bimini and one from the lagoon (Fig. 37) . Several hundred specimens were collected around a night light suspended from the end of the Lerner Marine Laboratory dock in North Bimini harbor. The water at the time of collection had temperatures of 28.3° to 29.2°C. and salinities of 36.l to 37.5 parts per thousand. The bottom was calcareous sand. 0 0 0 0 0 0 t 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 2 0 0 ~ o., 8 0 I 0 (jJ I 0 0 I (() I I ct Jf:l?C§ I I 0 0 . r:tx? 0 0 ~ 8 0 0 0 ;. 0 © Philomedes. mult1chelata 0 0 00 " <;_yclaberis biminiensis 00 0 0 FIG. 37. Distribution of Rutiderma polychelata Kornicker, new species; Pseudophilomedes /erulanaKornicker, new species; Philomedes multichelata Kornicker, new species; Cycloberis biminiensis Kornicker, new species. Species was absent from samples shown as empty circles. Small circles repre· sent trawl samples; large circles represent spot samples. Ecology and Taxonomy of Recent Marine Ostracodes Phiwmedes lomae Juday ~Figs. 46, 7 A-B; 52, A-E; 53, A-D; 86, A, E, I) Philomedes lomae Juday, 1907, Univ. Calif. Puhl. Zoo!., Vol. 3, l\o. 9, P. 14-1 , PL 19, Figs. 1-6 Diagnosis: A process projects posteriorly from each side of the carapace, commencing at about the center. The antenna! sinus is shallow but distinct. The surface of the valves is pitted. The dorsal margin of the living ostracode is orange-red which fades rapidly after death, being no longer visible after a few months. The furca contains six cla\\·s. The distal end of the fourth joint of the first antenna bears a sensory bristle which contains one thick and many fine setae. The secondary branch of the second antenna consists of three joints; the third joint is reflexed upon the second. A small sensory appendage which is ciliated on its outer extremity is on the distal end of the first joint of the mandibular polp. The terminal joint of the mandible contains one claw and three slender bristles. The female is not known. Shell measurements tin mm. I : Specimen number Length Width Height ll5-Cl 1.24 OA4 0.75 215 1.18 0.34 0.73 92 0.89 0.33 Material: Nine specimens were collected from bottom samples; two of these (both males I were dissected. Occurrence: Philomedes lomae seemed to be uniformly distributed in the Bimini area ~Fig. 381. Specimens were collected in waters haYing salinities of 36 to 42 parts per thousand, and temperatures of about 29°C. The species was obtained at water depths ranging from one to 20 meters. The bottom in the collecting area was calcareous sand. Philomedes paucichelata Kornicker. new species tFigs. 46. 4A-B; 54. A-E; 55. A-C; 87. B. E. H.) Diagnosis: The shell is elongate_ with a prominent rostrum and a posteroventral pro­jection, and is not strongly calcified. The surface is coarsely punctate and contains irregular horizontal ripples as well as light depressions above the central part of the shell. The fifth leg terminates in a large quadrate tooth. The furca bears six claws; thf first, second, and fourth claws are strong; the third, fifth. and sixth claws are weak. Natatory bristles on the second antenna are spined. The seventh leg bears four terminal and two lateral setae. The liYing ostracode has a characteristic orange-colored body and eyes. The male is unknown. Comparisons: This species may be distinguished from Philomedes .flexilia Brady and Philomedes sculpta Brady by the fewer number of claws on the furca. The shape of the shell is remarkably similar to that of Streptoleberis crenulata Brady but. as the appendage structure of the latter species is unknown, the relationship of these two forms remains in doubt. 234 Ecology and Taxonomy of Recent Marine Ostracodes 0 0 0 00 0 0 0 0 0 0 ., 0 0 00 0 0 00 0 0 0 00 Oo 0 00 0 0 0 0 0 I I I 0 0 0 0 I 0 0 0 0 o Oo 6 0 0 0 0 cP I 0 ~ <{) 0 0 O ~C§ ~J 8 0 0 0 8 0 0 .. 0 I• 0 0 00 0 0 0 FIG. 38. Distribution of Philomedes lomae Juday. Species was absent from samples shown as empty circles. Small circles represent trawl samples; large circles represent spot samples. Shell measurements (in mm.): Specimen number Length Width Height 287-1 (holotype) 0.83 0.35 0.46 62* 0.8 0.48 288 0.87 0.35 0.50 156-5 0.80 0.4 * Gravid female. Material: Eighty-one specimens were collected from bottom samples; two of these (both female) were dissected, and, in addition, individual appendages were examined from several other specimens. The holotype is specimen number 287-1, illustrated on Fig. 46, 4a-b. Occurrence: Philomedes paucichelata was a common species in the Bimini area (Fig. Ecology and Taxonomy of Recent Marine Ostracodes 235 39). It was found in water from one to about 20 meters in depth. Specimens were col·lected in waters having temperatures of about 29°C. and salinities of approximately 38parts per thousand. The bottom in the collecting areas was calcareous sand. ' '' 0 I ' 0 I II 0 0 f 0 0 III 0 I I I I III 0 0 0 0 I 0 0 II 0 II 0 0 II 0I • 2 0 /I' 0 0 0 I' 0 0 I /' 0 0 0 /I/ 0 o/ I 0 Poo o I III 0 I' 0 0 0 I I 0 I I 0 0 I () I I I 0 I I 0 0 I I'I I I I' • 2 ! Oe ~7 I 0 0 II 0 ~I I 0 I I <{) 3 O~C§ 0 0 8 I Cf:O 0 io o1~ 8 SC ALE I 0 0 I 0 ;. 0 0 00 00 0 0 Fie. 39. Distribution of Philomedes pau cichelata Kornicker, new species. Species was absent fromsamples shown as empty circles. Small circles represent trawl samples; large circles represent spot samples. Genus Pseudophilomedes Mueller This genus is characterized by having an elongate tooth on the fifth limb. Pseudophilomedes f erulana Kornicker, new species (Figs. 46, lA-B, 2A-B; 56, A-D) Diagnosis: The carapace of this species is elliptical in lateral view, with a small ros­ trum in the upper part of the anterior margin and a short postero-ventral projection.The caudal furca contains a total of eight claws. The claw arrangement is typical of Ecology and Taxonomy of Recent Marine Ostracodes the genus: two large claws on the distal end, followed by several smaller claws; the fourth claw is larger than the third, fifth, sixth, seventh, and eighth claws. The maxilla bears on its end a blunt, elongate process. The remaining appendages are typical of the genus. The seventh limb contains six terminal and four lateral setae. The male is unknown. Comparisons: Three species of this genus have been described previously: Pseudophil­omedes /oveolata G. W. Mueller; Pseudophilomedes inflata (Brady and Norman); and Pseudophilomedes angulata Mueller. P. ferulana may be distinguished from these forms by having eight furcal claws, whereas P. foveolata and P. angulata have only six and P. inflata has 10. P. f erulana is also distinguished from the other species in having a long blunt process on the last joint of the maxilla. This feature may be of subgeneric value. Shell measurements (in mm.) : Specimen number Length Width Height 165-1 (holotype) 1.15 0.79 93-1 0.63 0.40 118B-3 ?0.98 Material: Five specimens were collected from bottom samples; two of these (both females) were dissected. The holotype is specimen number 165-1, illustrated on Fig. 46, 2a-b. Occurrence: Three specimens were obtained east of South Bimini, and one west of Tur· tie Rocks (Fig. 37). Water temperature in collecting area was about 29°C., and the salinity 37 parts per thousand. Water depth was about six meters. Family Rutidermatidae Genus Rutiderma Brady and Norman This genus is characterized by having a strong chela on the mandible. Subgenus Rutiderma Kornicker, new subgenus Diagnosis: Caudal laminae have few (six) claws, which include three or four strong claws followed by two or three weak "claws" with hairs at their base. Comparisons: Three species, Rutiderma compressa Brady and Norman, Rutiderma rostrata Juday, and Rutiderma dinochelata Kornicker, new species, form the subgenus Rutidenna which is based on the furca having few claws which decrease in length pos­teriorly. The subgenus Alternochelata is based on the included species (one at present) having a furca bearing numerous claws, with the strong claws alternating with weak claws and spines. Rutiderma (Rutiderma) dinochelata Kornicker, new species Figs. 46, 8A-B; 57, A-F; 58, A-D; 86, B, F, J) Diagnosis: The shell is oval in lateral view with a truncate posterior. A shallow anten· nal sinus does not have an overhanging rostrum. The shell surface is coarsely punctate and ornamented with numerous riblets; two prominent longitudinal ribs run almost the complete length of the shell. The caudal furca bears three strong claws which show distinct demarcation lines at their bases, followed by two weak claws. The remaining appendages are typical of the genus. The seventh limb bears six terminal and four lateral setae. The male is unknown. Comparisons: Rutiderma dinochelata differs from Rutiderma compressa Brady and Norman in the distribution of the strong and weak claws on the furca. R. dinochelata differs from R. rostrata Juday in not having an overhanging rostrum. Shell measurements (in mm.): Specimen number Length Width Height 57-1 (holotype) 1.22 0.45 0.87 91-1 1.18 0.45 0.8 247-10 Ll4 0.79 Material: Seventy-nine specimens were collected from bottom samples. Two speci­ mens were dissected (both females), and.individual appendages were examined from several additional specimens. The holotype {s specimen number 57-1, illustrated on Fig. 46, 8a-b. Occurrence: Rutiderma dinochelata was common in the Bimini area (Fig. 40). Speci­mens were obtained from waters which ranged in depth from one to 20 meters, and in salinity from 31 to 42 parts per thousand. Water temperature was about 29°C. Alternochelata Kornicker, new subgenus Type species: Rutiderma ( Alternochelata) polychelata Kornicker, new species Diagnosis: Caudal laminae with many claws; weak claws and spines alternate with strong claws. Comparisons: The subgenus Alternochelata differs from the subgenus Rutiderma in having more claws on the furcal laminae and in having these claws arranged so that weak claws and spines alternate with strong claws. Rutiderma ( Alternochelata) polychelata Kornicker, new species (Figs. 46, 6A-B; 59, A-E; 86, C, G) Diagnosis: The shell is oval in lateral view, with a narrow antennal sinus and over­hanging beak. The postero-ventral corner is compressed, forming a small but distinct projection. The surface of the shell is smooth and without ribs or other ornamentation. The furca bears 10 claws and spines; two long distal claws are followed by a small an­nulated spine; then a short claw is followed by a weak claw, which is succeeded by a larger claw and four weak claws. The second antenna bears a secondary appendage which seems to have two joints. The proximal joint of the secondary appendage bears several short spines (five or six), whereas the distal joint bears a long spine which has secondary setae near its middle. The seventh limb contains six terminal and four lateral setae. The first and second antenna, maxilla, and the fifth and sixth limbs are typical of the genus. Comparisons: Rutiderma ( Alternochelata) polychelata differs from all previously described species of this genus in possessing a greater total number of claws ( 10) on the furca and also in that these claws are arranged so that strong and weak claws alternate. Occurrence: Two specimens were collected east of South Bimini and eight were ob­tained in North Bimini harbor (Fig. 37). Water depth in collecting areas ranged from Ecology and Taxonomy of Recent Marine Ostracodes 0 0 0 0 0 0 t 0 0 0 . 0 0 3 0 0 0 0 0 0 ·2 0 0 0 0 0 1' 0 Oo • 0 0 0 •I 0 0 0 0 0 0 0 0 0 2 I0 0 •• I 0 I 0 ce2 0 0 <{> '~ o~csl I 19!-0° 0 0 , ~ 0 0 SCAL E 0 0 0 cj1 2 ;, 0 0 0 •5 0 0 0 0 F1G. 40. Distribution of Rutiderma dinochelata Kornicker, new species. Species was absent from samples shown as empty circles. Small circles represent trawl samples; large circles represent spotsamples. one to five meters. Temperatures were about 29° C., and salinities were about 37 parts pe1· thousand. The bottom consisted of calcareous sand. Shell measurements (in mm.): Specimen number Length Width Height 144-1 1.03 0.49 0.73 llOF-2 (holotype) 1.36 0.55 0.98 Material : Ten specimens were collected from bottom samples; one specimen (sex not known) was dissected. The furca of a second specimen was examined and found identi· cal to that of first specimen. The holotype is specimen number llOF-2, illustrated on Figure 86, C, G. Ecology and Taxonomy of Recent Marine Ostracodes Family Asteropidae Genus Asteropina Strand Asteropina mulleri l Skogsberg) (Figs. 60, A-F; 61. A-F; 87, A, D, G) Cylindroleberis teres G. W. Mueller. 1894, Fauna Neapel, Monogr. Vol. 21, P. 220, PL IV, Figs. 13, 30, 43; Pl. V, Figs. 15, 24, 25; Pl. VIII, Fig. 5. Asterope mulleri Skogsberg, 1920, l"ppsala Cniversitat, Zoo!. Bidr. l"ppsala, Suppl. Ed. L PP. 483-491, Fig. LXXXIX; Klie, 1940, Kieler Meeresforshungen, Band 3, PP. 409-411, Figs. 7-10. Diagnosis: The shell is more or less pear shaped, with its greatest height and width behind the middle, but elliptical shells are occasionally encountered with the greatest height and width in the middle. Live shells have characteristic brown markings on the inner surface of the shell which are visible from the outside. These markings fade after death. Shell measurements lin mm.): Specimen number CP5-l* Length 1.26 Width 0.49 Height 0.90 CP5-3 CP5-2* CPAC-1 1.3 1.26 1.34 0.57 0.87 0.89 0.93 * GraYid female. Remarks: This species is described in considerable detail by Skogsberg (1920). Skogsberg ( 1920) and Mueller ( 1894) describe this species as having six claws on the furca. The adults examined by the author (five females) had either five or six furcal claws. One individual contained six on the left lamina and five on the right. The Bimini specimens are slightly smaller than Skogsberg's or Mueller's forms and much smaller than specimens collected by Klie ( 1940). Material: Fifty-three specimens were collected in the Bimini area from bottom samples. Seven specimens were dissected; of these three were gravid females, and the sex of the remainder was not determined. Occurrence: Asteropina mulleri was widely distributed in the Bimini area (Fig. 41). It occurred in waters having salinities ranging from about 31 to 39 parts per thousand and temperatures around 29°C. Individuals belonging to this species were adept at ag­glutinating debris in order to form burrows in which they resided. Occasionally two in­dividuals occupied the same burrow. Mueller t,1894, p. 14) noted burrow formation by other members of this genus. Distribution: Also reported from the English Channel, the Mediterranean Sea, and the coast of German Southwest Africa. Asteropina setisparsa Kornicker. new species (Figs. 46, 9A-B; 62, A-D; 63, A-D; 64, A-E; 86, L-P) Diagnosis: The carapace is elongate with almost parallel dorsal and ventral edges. 240 Ecology and Taxonomy of Recent Marine Ostracodes 0 0 0 ;( I 0 0 0 0 ~ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ~ Oo 8 0 0 cP 0 0 I ~ I 0 0 8 0 0 0 SC ALE 0 0 0 0 Fie. 41. Distribution of Asteropina mulleri (Skogsberg) and Actinoseta chelisparsa Kornicker, new species. Species was absent from samples shown as empty circles. Small circles represent trawl samples; large circles represent spot samples. The postero-ventral corner projects slightly in some individuals; in others the posterior end is evenly rounded. In dorsal view the carapace is slightly wider behind than in front. A narrow slit, the antennal sinus, is present in the lower half of the anterior part of the shell. The furca contains Eeven to eight claws. The anterior five claws are sturdier than the last two or three. The seventh limb contains six lateral and six terminal setae. The antero­ventral corner of the sixth limb bears one thick bristle and one or two slender bristles; the ventral edge contains, at most, one bristle and many fine hairs. Two large black eyes are visible through the shell. The male is unknown. Comparisons: The shell of this species is similar to that of Asterope mariae Baird. A. setisparsa differs from A. mariae in having fewer claws on the furca and in the absence of numerous bristles along the ventral edge of the sixth limb. A. setisparsa is also smaller than A. mariae. Ecology and Tax-0nomy of Recent Marine Ostracodes Shell measurements t,in mm.) : Specimen number Length Width Height 48---6 1.45 0.53 0.69 2.t-0-3* (holotype) 1.54 0.75 127G 1.5 0.57 0.78 122D-3"" 1.52 0.77 1-llK 1.22 0.47 0.63 * Gra,·id female. :\laterial: Thirty-seven specimens were collected from bottom samples. Three speci­mens tall female l were dissected. The holotype is specimen number 246-3 illustrated on Fig. 86. L-P. Occurrence: Asteropina setisparsa was abundant east of South Bimini and rare in other areas (Fig. 42 l. East of South Bimini where this species reached its greatest abun­dance. the depth of water from which this species was collected ranged from one to five meters and had a temperature of about 29"C. Salinity at this locality was about 37.5 parts per thousand. The bottom was calcareous sand and rock. Patches of Thalassia and the algae laurencia grew in this area. but .4 . setisparsa was not restricted to them . .4steropina extrachelata Kornicker. new species t Figs. 65. A-E; 66. A-E; 87, C. F, I) Diagnosis: The shell is om! in lateral outline and has a smooth surface; however, small pits may be observed under high magnification. The antenna! sinus is a narrow slit just below the center of the anterior edge. The furca bears fiw claws followed by four annulated bristles. The penultimate joint of the mandible has a "claw" at about the middle of the anterior edge. Along the wntral edge the sixth limb bears approximately 24 bristles. Two bristles are on the anterior edge. and a total of fi,·e bristles are on the antero-,·entral corner. Six lateral and six terminal setae are on the sewnth limb. Two lateral eyes are present as well as the gill­like organs typical of the genus. Comparisons: The shell is similar to that of Asterope abyssicola Sars. .4steropina ex­trachelata differs from .4. abJSsicola in having fewer setae on the sewnth leg. The furca of A. extrachelata differs from that of .4. abJSsicola. as well as Asterope mariae Baird, in that the last four digits are distinctly annulated bristles. whereas the other two species bear weakly dewloped claws. Only one individual of this species was collected. The '·claw" on the anterior edge of the penultimate joint of the mandible differentiates this species from all other species of this genus known to the author. For this reason a new species was erected with this minimum of indi,·iduals. Shell measurements (in mm.): Specimen number Length Width Height 118-1" t holotype) 1.88 1.01 • 1.ra,·id female. Ecology and Taxonomy of Recent Marine Ostracodes 0 0 0 0 0 0 J( l 0 I 0 4 0 0 0 0 0 0 0 0 •2 0 0 0 0 ' ' 0 0 0 ' ' I I ' Oo o' 0 0 /o:, Pooo, 0 0 I I,, I I 0 I • 0 0 I , 0 I I I 0 0 0 I I 0 I 0 0 0 cf) 0 0 I 0 Q) 0 sf!?°& 0 0 ' a:2 8 0 0 ~ 8 0 $C ALE 0 0 ;. 0 0 0 0 0 00 0 0 Fie. 42. Distribution of Asteropina setisparsa Kornicker, new species. Species absent from samples shown as empty circles. Small circles represent trawl samples ; large circles represent spot samples. Remarks: The individual examined was a female. It contained rosette-like markings on the shell similar to those described by Tressler (1949) on Cyclasterope sphaerica. These markings have been observed by the author on some individuals of other species and seem to be due to the decomposition of the soft parts as they are not present on the living animal. A diagram of one of the markings on A. extrachelata is presented on Fig. 41, 8c. Material: One specimen (a female) was collected in a bottom sample. The holotype is specimen number 118-1, illustrated on Fig. 87, C, F, I. Appendages of this specimen are figured on Figs. 65, A-E; and 66, A-E. Occurrence: The only specimen of Asteropina extrachelata collected came from the eastern section of Cavelle Pond, South Bimini. The sediment there gave off an odor of hydrogen sulfide and supported an abundant growth of the algae Laurencia. Ecology and Taxonomy of Recent Marine Ostracodes Genus Cycloberis Skogsberg Cydoberis biminiensis Kornicker, new species (Figs. 67, A-D; 68, A-F; 85, A-E) Diagnosis: The shell is oval in lateral outline and only slightly longer than high. An antennal sinus is approximately in the middle of the anterior edge. A distinct muscle scar of rosette form is visible near the center of the shell. Lateral eyes are situated above the muscle scars. The furca bears four claws followed by three or four bristles. Between the third and fourth claw is an additional small bristle. The seventh limb bears 23 to 32 setae. Six protuberances are on the anterior edge of the penultimate joint of the first antenna. Otherwise, this limb, as well as the frontal organ, is similar to that of Cycloberis ameri­cana lMueller). The male is unknown. Comparison: Many species of Cycloberis have a carapace of similar shape. The num­ber and arrangement of claws and bristles on the furca are used for differentiating species. Cycloberis brevis (Mueller) is the only species known, other than C. biminiensis, whose furca bears four claws; all other known species bear only three claws. C. bimini­ensis differs from C. brevis in the presence of a small bristle between the third and fourth furcal claws. C. brevis has five to seven bristles following the fourth claw, whereas C. biminiensis has only three or four. Shell measurements (in mm.): Specimen number Length Width Height 177-1 (holotype) 1.8 0.74 1.4 177-2 1.5 1.1 156-1 1.8 0.9 1.3 Material: Seven specimens were collected from bottom samples. Of these specimens three (?females) were dissected, and the caudal furca of two additional specimens examined. The holotype is specimen number 177-1, illustrated on Figs. 85, A-B; 67, A-C, and 68, A-B, D-E. Occurrence: Cycloberis biminiensis was collected in waters ranging in depth from three to 10 meters with temperatures around 20° C. (Fig. 37). The salinity was about 37 parts per thousand. Actinoseta Kornicker, new genus Type species: Actinoseta chelisparsa Kornicker, new species Diagnosis: The shell is strongly calcified, densely pitted, and appears imbricate in dorsal view. In the type species the shell is almost oval in lateral view and contains a shallow antennal sinus. The natatory bristles of the second antenna bear spines on the distal half; basal spines are not present. The posterior appendage of the second antenna bears numerous spines on the first and second joints. The seventh limb has abundant pinnately arranged setae. The furca bears three (always?) sturdy, medium-length claws with fine hairs at the base. Following these claws are a few weak non-annulated setae. Comparisons: Actinoseta is closely related to Cyclasterope Skogsberg and Asterop· teron Skogsberg and possesses some of the diagnostic morphological features of both genera. Actinoseta differs from Cyclasterope in not having powerful basal spines on the second antenna and in having only a few furcal bristles. It differs from Asteropteron in having spines on the natatory setae on the second antenna and in possessing numerous setae on the seventh limb. Actinoseta chelisparsa Kornicker, new genus, new species ffigs. 46, lOA-B; 43, A-L; 69, A-F; 70, A-I; 89, H-J, P, Q) Diagnosis: The shell is almost oval in lateral view, with a shallow antennal sinus and small rostrum. The surface of the shell is densely pitted and appears imbricate in dorsal Yiew. Three or four faint protuberances are present in the postero-dorsal corner paral­leling the margin. These are not always apparent and are absent on some specimens. The furca bears three strong claws. At the base of the second and third claws are numerous fine hairs. Occasionally these are also found between the first and second claws. In the posterior corner of the £urea are three short, slender bristles. The seventh limb contains about 46 setae which are on about the distal 22 segments; two setae are attached to each segment. The st>cond joint of the posterior appendage of the second an­tenna bears eight lateral setae. The male is not known. Considerable change in append· age morphology takes place during ontogeny I Fig. 43). Shell measurements I in mm.) : Sj)ecimen number Length Width Height 60-1 1.68 0.92 1.19 CP52-19 0.95 0.73 CP52-16 0.83 0.63 CP52-6* 2.47 1.95 CP38-2* (holotype) 2.42 1.29 1.92 5216B 2.0 1.55 CP52-24 0.61 0.46 CP38-7 0.83 0.71 246-2 0.99 0.55 0.78 * Gravid female. Material: Thirty-two specimens were collected from bottom samples. Six specimens (all females) were dissected for comparison. Many others were partially dissected dur­ing identification. The holotype is specimen number CP38-2, illustrated on Figs. 43, A-L ; 69, A-F; 70, B-D, F-1; 89, I, Q. Occurrence: Actinoseta chelisparsa was found in waters three to 15 meters in depth FIG. 43 Ontogenetic stages in female of Actinoseta chelisparsa. A, D, G, J. Young instar: A, carapace; D, Furca; G, First antenna; J, Posterior appendage of second antenna. Specimen number CP52-16. B, E, H, K. Immature ins tar: B, carapace; E, Furca; H, First antenna; K, Posterior appendage of second antenna. Specimen number CP52-19. C. F, I, L. Mature female : C, carapace; F, Furca; I, First antenna; L, Posterior appendage of second antenna. Specimen number CP38-2. Figures with simil ar magnification: A, B, C; D, E, F; G, H, I; J, K, L. Ecology and Taxonomy of Recent Jlarine Ostracodes 245 A 8 c D E !/ ~\ j J K Ecology and Taxonomy of Recent Marine Ostracodes (Fig. 41). Salinity was about 37 parts per thousand and temperature about 29°C. This species was common in Cavelle Pond where salinities as low as 31.5 parts per thousand were recorded. Genus Asteropteron Skogsberg Asteropteron monambon Kornicker, new species (Figs. 46, llA-B; 71, A-G; 72, A-D; 86, D, H, K) Diagnosis: The shell is elliptical in lateral view, with a shallow antennal sinus slightly below the middle of the ventral margin. The width of the shell increases gradually from the anterior margin to a point about one-fourth of the shell length from the posterior margin and then decreases rapidly so that in dorsal view the carapace resembles the head of an arrow. A raised border parallels the shell's outer edge except at the postero· dorsal corner, where the border forms two or three nodes. These are more distinct on some individuals than on others. A more or less horizontal slightly raised ridge extends from the anterior border, at a point directly behind the antenna! sinus, to the posterior portion of the carapace, dying out before reaching the posterior border. In the center of the shell this ridge expands and encloses about 15 reticulations. The surface of the shell is punctate. The furca is elongate and bears three main claws followed by four smaller secondary claws. The second antenna does not contain basal spines. Natatory bristles of the second antenna also lack spines. The ultimate joint of the secondary branch of the second an· tenna is covered with fine hairs. The seventh limb bears 16 setae. The terminal part of the seventh limb resembles the "aristotle lantern" of an echinoid. Lateral eyes are not readily visible through the shell. The male is unknown. Comparisons: A. monambon is closely related to Asteropteron agassizi (Fr. Mueller) and Asteropteron fusca (G. W. Mueller). It is readily distinguished from these two species by the character of the carapace sculpture; viz., the border of A. monambon is continuous, and a horizontal ridge approximately bisects the carapace, whereas, the border of A. agassizi is discontinuous, and the horizontal ridge runs below the center. The border of A. fusca is also discontinuous. Shell measurements (in mm.) : Specimen number Length Width Height 113G (holotype) 1.25 0.63 0.90 243-1 1.14 0.58 0.83 CP38-l 0.93 0.67 CP52-4 1.52 1.04 91 0.66 0.5 119C 0.61 0.48 Material: Thirty-four specimens were collected from bottom samples. Three speci­mens (? females) were dissected. The holotype, specimen number 113G, is illustrated on Fig. 46, lla-b. Occurrence: Asteropteron monambon was fairly uniformly distributed in the Bimini area (Fig. 44). Specimens were collected in waters ranging in salinity from 31.5 to 42 parts per thousand, and in depth from three to about 20 meters. Temperature was about 29°C. Ecology and Taxonomy of Recent Marine Ostracodes 247 0 { 0 I c c 0 0 I { 0 ,' I I { 0 c ::; I 3 0 c ' ' 0 0 0 c '2 0 0 00 o. 0 0 0 0 c 00 0 0 :::: 0 0 00 0 c 3 00 ~ • 0 0 c 0 0: I I 0 co I C rf!?~ o e 0 0 n I: c " v ,:j~ ,. 0 s I• I• I• 0 0 0 00 0 0 0 Fu;. -1-1. Distribution of Asteropteron monambon Kornicker. new species. Species was absent from samples shown as empty circles. Small circles represent trawl samples: large circles represent spot samples. Famih-Sarsiellidae Genus SarsieUa l\orman Sarsiella carina.U1 A. Scott (Figs. 47, SA-B; 73, A-G; 74, A-F; 75, A-D; 88, J, l\, R; 89, E) Sarsiella carinala A. Scott, 1905. Rep. Ceylon Pearl Fisheries, Suppl. Rep, 22, P. 368, Fig. 1, 1-2; Fig. 2, 40-41. Diagnosis: The male shell has a prominent rostrum and a shallow sinus. Two longi­tudal ribs fringed with short hairs terminate in protuberances posteriorly. A series of hair-co\·ered nodes parallels the ventral margin. The male furca bears five claws, which increase in length distally. The seventh limb of the male has two lateral and four terminal bristles. and the distal end is closed. The first antenna bears a brush-like sensory organ. The female shell has ornamentation quite similar to that of the male which differs in having no antenna) sinus. Shells of mature females contain marginal radial furrows. The female shell varies in shape and would be difficult to differentiate from other sarsiellids were it not for short stiff hairs which always cover the shells of this species and which are easily visible under high magnification. These hairs are abundant along the hori­ zontal riblets which ornament the shell and around the numerous protuberances along the ventral and dorsal portions of the shell. The horizontal riblets occur, especially on immature individuals, as a series of disconnected protuberances. The female £urea bears five claws followed by two short "points." The seventh limb of the female contains six terminal and two lateral spines; the distal end of the seventh limb is open and bears "teeth." An immature female had only two lateral and two termi­nal setae on the seventh limb and possessed a closed distal end similar to that of the male. Comparisons: The female of S. carinata resembles Sarsiella globulus Brady (Brady and Norman, 1896, Fig. 60, 5-7) from which it differs in having numerous surface protuberances which are covered by short stiff hairs and in possessing fewer spines on the seventh limb. S. carinata does not resemble the type species of S . globulus (Brady, 1881-1886, Fig. 15, 8-9) . Shell measurements (in mm.) : Specimen number Length Width Height Sex 68fr-X2 1.42 0.67 1.12 Female CP38-5 1.28 0.86 Female CP38-13 1.48 1.14 Female CP52-23 1.18 0.68 Male Remarks: Females of this species have not previously been reported. Material: Seventy-eight specimens were collected from bottom samples. Eight males were among the collected specimens. Three individuals were dissected. Occurrence: Sarsiella carinata was the most abundant species of the genus Sarsiella in the Bimini area. It was especially abundant in North Bimini harbor in front of the Lerner Marine Laboratory. Specimens were collected in water ranging in depth from one to 20 meters, and in salinity from 31.5 to about 38 parts per thousand. Temperature was about 29°C. Areal distribution of the family Sarsiellidae is shown in Fig. 45. Distribution: This species has been reported from near Ceylon (A. Scott, 1905). Sarsiella capillaris Kornicker, new species (Figs. 47, 7A-B; 76, A-F and H; 89, A, B, D, F, 0) Diagnosis: The shell is subcircular in side view, with a posterior projection. Numer­ous protuberances are present on the surface of the shell. A long hair, which is annulated and becomes wider at about two-thirds of the distance from its proximal end, projects from the apex of each protuberance. Surrounding the base of the long hair are many shorter hairs which may have bulbous tips. The posterior projection of the shell bears many long and short hairs similar to those on the protuberances. The £urea bears five claws followed by several "points." The seventh limb contains four terminal and two lateral setae. The balance of the appendages are typical of the genus. The male is not known. Comparisons: Sarsiella capillaris is difficult to differentiate from some individuals of Ecology and Taxonomy of Recent Marine Ostracodes 0 0 0 0 0 c 0 c 0 "­ 0 c 0 ~ 0 c c CJj •, c '.) c 00 0 c 0 ::; 0 0 c c::: 0 I ~::: 2 ::; a:; c ~ c c r 0 ::; -J - l.. =~ c c • c 0 0 c Fu;. -t5. Distribution of the family Sar>iellidae. Genus was absent from samples shown as empty circles. Small circles represent trawl samples: large circles represent spot samples. Sarsie/la. carinata Scott, and especially from immature forms of that species. The pres­ence of short stiff surface hairs on S. carinata permit;: differentiation from S. capillaris whose surface hairs are slightly longer and appear to be more pliable. Examination of surface hairs is best accomplished under high magnification. S . capillaris also differs from S. carinala in ha,·ing fewer terminal ;:pines on the ;:ewnth limb. Shell mea;:urements t.in mm. ·1: Specimen number Length Width Height Sex 92-1 tholotype) 1.15 0.56 1.15 Female CP38-10 1.1 0.81 Female CP52-15 0.82 0.66 Female 1117(-3 0.65 0.49 Female Material: Fifteen specimens were collected from bottom samples. One ostracode was Ecology and Taxonomy of Recent Marine Ostracodes completely dissected and several others partially dissected. The holotype is specimen number 92-1, illustrated on Fig. 47, 7a-b. Occurrence: Sarsiella capillaris was found in waters ranging in depth from two to 20 meters, and in salinity from 31.5 to about 37.5 parts per thousand. Sarsiella gigacantha Kornicker, new species (Figs. 47, 8A-B; 76; 77, A-E; 88, A, F, H, I) Diagnosis: The carapace is sub-oval in outline, with the posterior truncate. The male has a rostrum and surface riblets. A prominent spine occurs in the postero-dorsal part of the shell of both sexes. The caudal lamina bears five strong claws followed by several small "points." The seventh limb of the female contains four terminal and 11 lateral setae. Comparisons: The postero-dorsal spine on the carapace and the numerous spines on the seventh limb distinguishes 5. gigacantha from previously described ostracodes of the genus Sarsiella. Shell measurements (in mm.): Specimen number Length Width Height Sex 190-lA (holotype) 1.37 0.75 1.14 Female 122D-2 1.3 0.85 Male Material: Forty-seven specimens were collected in bottom samples. Three specimens were males. Two ostracodes were dissected. The holotype is specimen number 190-lA, illustrated in Fig. 47, 8a-b. Occurrence: Sarsiella gigacantha was found in waters ranging in depth from one to 20 meters, and in salinity from 31.5 to about 38 parts per thousand. Temperature was about 29°C. Sarsiella truncana Kornicker, new species (Figs. 78, A-E; 88, Q; 89, C) Diagnosis: The shell is elliptical in lateral view, except for the truncate posterior. A shallow sulcus separates the slightly inflated postero-dorsal part of the carapace from the anterior part. The surface of the shell is covered with short hairs and minute punctae. The furca bears five to seven claws. The seventh limb contains six terminal and four lateral spines. The male is not known. Comparisons: The shell of S. truncana does not resemble any previously described species. Shell measurements (in mm.) : Specimen number Length Width Height Sex CP52-21 (holotype) 1.3 0.69 1.13 Female* CP38-5 0.8 0.67 Female CP38-6 0.41 0.18 0.35 Female 246-3 1.36 1.3 Female* 121D-5 1.01 0.92 Female* * Gravid. Ecology and Taxonomy of Recent Marine Ostracodes 251 Material: Fifty-four specimens were collected from botton samples. Two specimens ~re dissected. The holotype is specimen number CP52-21, illustrated on Fig. 88, Q; 89, C. Appendages from this specimen are figured on Fig. 78, A, B. D. Occurrence: Sarsiella trunca.na was abundant east of South Bimini and in North Bimini harbor~ Specimens were found in waters ranging in depth from one to 20 meters, and in salinity from 31.5 to about 38 parts per thousand. Temperature was about 29°C. Sarsiella punctata Kornicker, new species (Figs. 47, 3A-B; 79, A-I; 88, L, P, M) Diagnosis: The shell is sub-oval, with posterior truncated. A shallow sulcus extends from the center of the dorsal edge almost to the posterior margin. The surface of the shell is distinctly punctate and without hairs. The furca bears five claws. The seventh limb of the female contains six terminal and four lateral setae. The male has a slight rostrum and two faint horizontal riblets in the posterior part of the shell. Comparisons: Sarsiella punctata differs from Sarsiella truncana Kornicker in the ab­ sence of surface hairs and in having large punctae. 5. punctata differs from Sarsiella gracilis Scott in not having setae or spines combined with claws on the furca. Sarsiella crispata Scott differs from S. punctata in having only four furcal claws. S. punctata differs from Sarsiella murrayana T. Scott in having terminal "teeth" and more setae on the seventh limb. Shell measurements (in mm.): Specimen number Length Width Height Sex 82-1 (holotype) 246-6 119A-l CP38-4 CP38-9 1.1 0.97 0.98 1.26 1.04 0.53 0.57 0.81 0.69 0.77 1.02 LO Female Female Male Female Female Material: Nineteen specimens were collected from bottom samples. Two males were among the specimens collected. One male and one female were dissected. The holotype is specimen number 82-1, illustrated on Fig. 47, 3a-b. Occurrence: Sarsiella punctata was common in Cavelle Pond and in North Bimini harbor. Specimens were found in waters which ranged in depth from one to five meters, and in salinity from 31.5 to about 38 parts per thousand. The water temperature was about 29°C. Sarsiella costa.ta Kornicker, new species (Figs. 47, 4A-B; 80, A-E; 81, A-E; 88, B, C, G) Diagnosis: The shell is sub-oval in lateral view, with a truncated posterior. The sur­face of the shell is punctate. Surface ribs diverge from the postero-dorsal corner of the carapace and meet ventrally. A horizontal, anteriorly directed riblet branches from the lower rib in the postero-ventral quarter of the carapace. The male has a prominent . rostrum and surface rib distribution similar to that of the female. The £urea bears five claws followed by two "points." The seventh leg of the mature female contains six terminal and four lateral setae. Ecology and Taxonomy of Recent Marine Ostracodes Comparisons: The surface rib pattern differentiates Sarsiella costata from previously described species of the genus Sarsiclla. Shell measurements (in mm.): Specimen number Length Width Height Sex 156-2 (holotype) 1.02 0.58 0.79 Female CP52-7 0.97 0.61 Male CP38-8 0.73 0.26 0.47 Female CP38-16 1.14 0.84 Female Material: Forty spec: mens were collected from bottom samples. Only one male was among those collected. Three specimens were dissected. The holotype is specimen num­ber 156-2 and is illustrated in Fig. 88. C, G and Fig. 47, 4a-b. Occurrence: Sarsiella costata was especially abundant east of South Bimini. Many specimens were also found in Cavelle Pond. The depth of water from which this species was collected ranged from one to five meters, with a salinity ranging from 31.6 to 12 parts per thousand. Temperature was about 29°C. Sarsiella sculpta Brady (Figs. 47, 6A-B; 82, D, E; 88, D, K, E, 0, S) Sarsiella sculpta Brady, 1890, Trans. Roy. Soc. Edinb., Vol. 35, P. 516, Pl. 1, Figs. 17-20; 1897, Brady, Trans. Zoo!. Soc. Lond., Vol. 13, P. 93. Diagnosis: The shell is oval in side view, with a small retral process. The postero­dorsal quarter is globose. Two surface riblets diverge posteriorly'; one riblet bisects the globose postero-dorsal quarter. The shell surface is coarsely punctate. Mature individuals have radial riblets along the shell margin. The furca bears five claws followed by several minute "points." The seventh limb contains six terminal and seven lateral setae. The remaining appendages are typical of the genus. Comparisons: Brady ( 1890) in the original description of Sarsiella sculpta illustrates two individuals differing considerably in appearance. It is questionable as to whether both individuals belong to the same species. The Bahamian individuals are similar to the first individual illustrated by Brady (Pl. 1, Figs. 17, 18). Brady later (1898) illus­trated a caudal furca and the end of a seventh limb obtained from a third specimen of S. sculpta; the shell of this specimen was not illustrated. The appendages of the Ba­hamian forms are similar to those presented by Brady, with the exception of two small spines on the fourth claw of the caudal furca which are not present on the Bahamian specimen examined. Shell measurements (in mm.): Specimen number 127-1 177 Length 1.65 1.54 Width Height 1.45 1.31 Sex Female Female* *Gravid. Material: Thirteen specimens (females) were collected in bottom samples. One speci­men was dissected. Ecology and Taxonomy oj Recent ;llarine Ostracodes Occurrence: Sarsiella sculpta was collected in waters which ranged in depth from two to 20 meters. Salinity was about 37.5 parts per thousand and temperature was about 29°C. Distribution: This species has been collected from :\umea (dredged in two to four fathoms), Levuka tfrom between tidemarh). Vuna Point tfrom between tidemarh L and Flinders Passage t taken in seven fathoms) . Chelicopia Kornicker, new genus Type species: Chelicopia arostrata Kornicker, new species Diagnosis: The shell of the male and female are sub-ornl (the male is slightly more elongate than the female) . The male and female are without rostrum or sinus. The cara­pace of the type species is covered by hair. The furca contains many short and long clam;. The first antenna is fiw-jointed. The secondary appendage of the female second antenna is three-jointed, with spines on each joint. The secondary appendage of the male second antenna is two-jointed, with spines on each joint. The ultimate joint is annulated. Other appendages are similar to those of the genus Sarsiella. Comparisons: Chelicopia differs from the genu;; Sarsiella in having small claw;; be­tween the large claws on the furca and in the absence of a rostrum on the male shell. Chelicopia arostrata Kornicker, new genu;;. new specie;; (Figs. 47, 2A-B; 82, A-C; 83, A-D; 84, A-E: 89, G, K-.:\ '1 Diagnosis: The shell is sub-oval in lateral view and covered with short hairs. The male is without sinus or rostrum and is slightly longer than the female. Distinct lineated pores parallel the shell margin on the inside. The first, second, and fourth furcal claws are larger and do not have a line of demar­cation at the base. The third and remaining four smaller claws have lines of demarcation at their bases. The secondary appendage of the second antenna of the female is three­jointed (joining indistinct) and has two or three bristles on each joint. The ;;econdary appendage of the male is two-jointed. with the second _joint annulated: two bristles are on each joint. The se,·enth limb contains four terminal and two lateral setae. The re­ma"ning appendages are similar to those of the genus Sarsiella. Shell measurements (in mm.) : Specimen number Length Width Height Sex 91-1 tholotype) 1.06 0.66 0.96 ?Female CP38---16 0.78 0.76 ?Female 118A-l 0.98 0.51 0.82 ~lale Material: Five specimen;; were collected from bottom samples. Three specimen;; 1two females and one male) were dissected. The holotype is specimen number 91-1. illus­trated on Fig. 4 7, 2a-b. Occurrence: Chelicopia arostrata was found in Cawlle Pond and :\orth Bimini harbor. The waters in which specimens were collected had a salinity range of about 31.5 to 39 parts per thousand. The depth was from one to three meters. and the temperature about 29°C. Ecology and Taxonomy of Recent Marine Ostracodes Summary and Conclusions This paper describes a study of the relationship of living and dead ostracodes to the substrate, salinity and other factors in the diverse environments around Bimini. 1. Evidence was found indicating that substrate, salinity, and current velocity af­fected distribution of ostracodes. Dissolved oxygen, pH, water color, organic detritus, organic content of the sediment, water depth, and daily temperature variation had little effect in the Bimini area. Several species of Myodocopa were found to be positively re· sponsive to light. 2. Living Myodocopa were abundant, but empty carapaces were not found in the sediment possibly because of the low calcium carbonate content of the carpaces of individuals in this suborder. Common species belonging to the suborders Podocopa and Platycopa found as empty carapaces in the sediment were also collected alive in the Bimini vicinity except in the North Sound and Cavelle Pond areas where certain podocopid species occurred commonly as empty carapaces but were not found alive. 3. Brackish water ostracode remains were abundant in the sediment of Cavelle Pond, but only normal marine forms were collected alive in the pond. This is interpreted as indicating that the pond had previously contained brackish water. This interpretation is supported by the fact that a pass connecting Cavelle Pond to the sea is of recent origin. The occurrence of the relict faunal remains in the sediment of Cavelle Pond suggests that the numerous investigations being carried out today in which animal remains are being correlated with existing environments should be supplemented with a study of the living animal distribution. 4. The sediments containing the largest number of dead ostracodes did not occur in areas containing the largest number of living ostracodes. Empty ostracode carapaces were most abundant in areas where the salinity of the water was either high, low, or extremely variable. Lack of dilution by calcareous remains of other organisms is ad· vanced as the reason for high concentration of empty carapaces in these areas. S. The evidence in the Bimini area indicates that, in general, remains of the sub· orders Podocopa and Platycopa found in sediment reflect the kinds but not the absolute numbers of ostracodes of these suborders which were living in the area, and that the suborder Myodocopa was probably more widespread in the past than is indicated by its poor representation in the fossil record. 6. Ostracodes were differentiated taxonomically on the basis of appendage morphol­ogy as well as shell structure. Appendage morphology is especially important for dif­ferentiation of the suborder Myodocopa which was represented in Bimini waters by 10 genera, two of which are new (Actinoseta and Chelicopia), and 19 species, including 14 new species and one new subspecies. LITERATURE CITED Allison, L. E. 1935. Organic soil carbon by reduction of chromic acid. Soil Sci., 40: 311-320. Benson, Richard Hall. 1955. The ecology of the Recent ostracodes of the Todos Santos Bay Region, Baja California. Ph.D. thesis, Univ. of Ill., (unpublished). Brady, G. S. 1881-1886. in LES FONDS DE LA MER, vol. 4. Brady, G. S. 1890. On Ostracoda collected by H. G. Brady, Esq., L.L.D., F.R.S., in the South Sea Islands. Trans. roy. Soc. Edinb., 35: 489-525, pis. 1-4. Ecology and Taxonomy of Recent Marine Ostracodes ----. 1898. On new or imperfectly known species of Ostracoda, chiefly from New Zealand.Trans. zool. Soc. Lond., 14: 429-452, pis. 43--47. Brady, G. S., H. W. Crosskey and D. Robertson. 1874. A monograph of the post-Tertiary Entomostra· ca of Scotland including species from England and Ireland. Palaeonlogr. Soc., London, pp. i-v, 1-274, pis. 1-16. Brady, G. S., and A. M. Norman. 1896. A monograph of the marine and fresh-water Ostracoda of the North Atlantic and of northwestern Europe. Sec. 2-4, Myodocopa, Cladocopa, and Platycopa. Trans. roy. Dublin Soc., Sci., ser. 2, 5: 621-9460. Cannon, H. G. 1934. On the feeding mechanism of certain marine ostracods. Trans. roy. Soc. Edinb., 57. Dahl, Erik. 1956. Ecological salinity boundaries in poikohaline waters. Oikos, 7: 1-21. Dexter, Ralph W. 1944. The bottom community of Ipswich Bay, Massachusetts. Ecology, 25: 352-359. Harvey, H. W. 1955. The chemistry and fertility of sea waters, Cambridge University Press. 224 pp. Hedgpeth, Joel W. 1951. The classification of estuarine and brackish waters and the hydrographicclimate. Rep. Comm. mar. Ecol., 11: 49-56. ----. 1953. An introduction to the zoogeography of the northwestern Gulf of Mexico withreference to the invertebrate fauna. Puhl. Inst. of Mar. Sci., Univ. Tex., 3 ( 1) : 106-224. Hoff, C. Clayton. 1942. The ostracods of Illinois. Their biology and taxonomy. Illinois biol. Monogr., 19: 5-196, pis. 1-9. Juday, Chauncey. 1907. Ostraqida of the San Diego region. IL Littoral forms. Univ. Calif. Puhl.Zool., 3 (9 I : 135-150, pis. 18-20. Kesling, R. V. 1951. The morphology of ostracod molt stages. Illinois biol. Monogr., 21: v-viii, 1-317, figs. 1-36, pls. 1-96. Klugh, A. B. 1927. The ecology, food relations and culture of fresh water Entomostraca. Trans. R. Canad. Inst., 16: 15-99. Kornicker, L. S. 1957a. Ecology and Taxonomy of recent marine ostracodes in the Bimini Area, Great Bahama Bank. Ph.D. Thesis, Columbia University, 221 pp. Kornicker, Louis S. 1957b. Concentration of ostracodes by alcohol flotation. J. Paleont., 31. Kornicker, Louis S., and E. Purdy. 1957. A Bahamian faecal-pellet sediment. J. sediment. Petrol., 27: 126-128, fig. 1. Krumbein, W. C. and F. J. Pettijohn. 1938. Manual of Sedimentary Petrography. Appleton-Century­ Crofts, Inc., New York: 549 pp. Mueller, G. W. 1894. Die Ostracoden des Golfes von Neapel und der Angrenzenden Meeres­ Abschnitte. Naples. Sta. Zoo!., Fauna u. Flora Neapel, Monogr., 21: 1--404, pls. 1-40. Schollenberger, C. J. 1927. A rapid approximate method for determining soil organic matter. Soil Sci., 24: 65-68. Shepard, Francis P. 1948. Submarine Geology. New York, 348 pp. Skogsberg, Tage. 1920. Studies on Marine Ostracods. Zool. Bidr. Uppsala. Suppl. d. pt. 1, 784 pp. Sohn, I. C. 1951. Checklist of salinity tolerance of post-Paleozoic fossil Ostracoda. J. Wash. Acad. Sci., 41 (2) : 64-66. Swain, Frederick M. 1955. Ostracoda of San Antonio Bay, Texas. J. Paleont., 29 (4): 561-646. Tressler, Willis L. 1949. Marine Ostracoda from Tortugas, Florida. J. Wash. Acad. Sci., 39 (9) : 335-343,figs. 1-25. Tressler, W. L., and Essie M. Smith. 1948. An ecological study of seasonal distribution of Ostracoda, Solomon Islands, Maryland, Region. Puhl. Chesapeake biol. Lab., Puhl. No. 71, 57 pp., 4 pis. Turekian, K. K. 1957. Salinity variation in sea water in the vicinity of Bimini, Bahamas, British West Indies. Amer. Mus. Novit., 1822: 1-12. United States Navy Department, Hydrographic Office. 1940. Navy Air Pilot, H. 0 . No. 194. Walton, William R. 1955. Ecology of living benthonic Foraminifera, Todos Santos Bay, Baja Cali­fornia. J. Paleont. 29: 952-1018. Yonge, C. M. 1953. Marine bottom substrata and their fauna. 14 Intern. Zool. Cong. Copenhagen. lo 3o 2b 60 4 b I mm FIG. 46. la-b. Pseudophilomedes ferulana Kornicker, new species, immature instar. la. Dorsal view ; lb. Side view. No. 93-1. 2a-2. Pseudophilomedes ferulana Kornicker, new species. 2a. Dorsal view; 2b. Side view. No. 165-1. 3a-b. Philomedes multichelata Kornicker, new species. 3a. Dorsal view; 3b. Side view. No. 686Z-6. 4a-b. Philomedes paucichelata Kornicker, new species. 4a. Dorsal view; 4b. Side view. No. 287-1. Sa-b. Asteropina setisparsa Kornicker, new species. Sa. Dorsal view; Sb. Side view. No. 141K. 6a-b. Rutiderma polychelata Kornicker, new species. 6a. Dorsal view; 6b. Side view. No. 144-1. 7a-b. Philomedes lomae (Juday). 7a. Dorsal view; 7b. Side view. No. ll5-Cl. 8a-b. Rutiderma dinochelata Kornicker, new species. 8a. Dorsal view; 8b. Side view. No. S7-l. 9a-h. Asteropina setisparsa Kornicker, new species. 9a. Dorsal view; 9b. Side view. No. 48-6. lOa-b. Actinoseta chelisparsa Kornicker, new species. lOa. Side view; lOb. Dorsal view. No. 60-1. lla-b. Asteropteron monambon Kornicker, new species. Ila. Dorsal view; llb. Side view. No. ll3G. 258 Ecology and Taxonomy of Recent Marine Ostracodes 5o 3b Imm Fie. 47. la-b. Cypridina squamosa lerneri Kornicker, new subspecies. la. Side view; lb. Dorsal Yiew. No. 212-1.2a-b. Chelicopia arostrata Kornicker, new species. 2a. Dorsal view; 2b. Side view. No. 91-1. .3a-b. Sarsiella punctata Kornicker, new species, female. 3a. Dorsal view; 3b. Side view. No. S2-l.4a-b. Sarsiella costata Kornicker, new species, female. 4a. Dorsal view; 4b. Side view. No. 156--2.5a-b. Sarsiella carinata A. Scott, female. Sa. Dorsal view; Sb. Side view. No. 6S6X-2.6a-b. Sarsiella sculpta Brady, female. 6a. Dorsal view; 6b. Side view. No. 127-1.la-b. Sarsiella capillaris Kornicker, new species, female. 7a. Dorsal view ; 7b. Side view. No. 92-1.Sa-b. Sarsiella gigacantha Kornicker, new species, female. Sa. Dorsal view; Sb. Side view. No. 190-lA. 8 D Cypridina squamosa lerneri Kornicker, new subspecies. A. Furca. Specimen No. P-1 I juvenile I. B. Furca. Specimen No. 246--7 (ju\"enilel. C. Furca. Specimen No. 247-3 ( gra\"id female I. D. Seventh limb. Specimen No. 119-1 f_mature male I. Ecology and Taxonomy of Recent Marine Ostracodes ( A B E Fie. 49. Cypridina squamosa lerneri Kornicker, new subspecies. :\ .. Portion of maxilla or fifth limb of young instar. Specimen No. P-1 (juvenile). B. '.\fandihle of young instar. Specimen No. P-1 (juvenile I. C. First antenna young instar. Specimen No. P-1 Ijuvenile). D. Secondary branch second antenna. Specimen No. P-1 I juvenile) . E. Second antenna. Specimen No. P-1 Ijuvenile). Ecology and Taxonomy of Recent Marine Ostracodes D E Fie. 50. Philomedes multichelata Kornicker, new species (Male) . A. First antenna. Specimen No. 686Z-2. B. Portion of sensory setae of first antenna enlarged. Specimen No. 686Z-2. C. Distal end first antenna; sensory setae not shown. Specimen No. 686Z-2. D. Seventh limb. Specimen No. 686Z-3. E. Furca. Specimen No. 686Z-4. 262 Ecology and Taxonomy of Recent Marine Ostracodes B c A FIG. 51. Philomedes multichelata Kornicker, new species (Male). A. Second antenna. Specimen No. 686Z-2. B. Portion of sensory bristle enlarged. Specimen No. 686Z-2. C. Secondary branch of second antenna. Specimen No. 686Z-4. D. Mandible. Specimen No. 686Z-4. Ecology and Taxonomy of Recent Marine Ostracodes Frc. 52. Philomedes lomae Juday (l\!ale). A. Furca. Specimen No. CP52-12. B. First antenna. Specimen No. CPS2-12. C. Secondary branch of second antenna. Specimen No. CP52-12. D. Second antenna. Specimen No. CPS2-12. E. Sensory bristle of first antenna. Specimen No. CP52-12. 264 Ecology and Taxonomy of Recent Marine Ostracodes B FIG. 53. Philomedes lomae Juday (Male) . A. Mandible. Specimen No. CP52-12. B. Portion of sixth limb. Specimen No. CP52-12. C. Sensory appendage on mandible. Specimen No. CP52-12. D. Portion of fifth limb. Specimen No. CP52-12. Ecology and Taxonomy of Recent Marine Ostracodes FIG. 5.t. Philomedes paucichelata Kornicker, new species (Female). A. Furca. Specimen No. 156--5. B. :\landible. Specimen No. 156-5. C. Second antenna. Specimen Ko. 156--5. D. First antenna. Specimen No. 156--5. E. Part of sixth leg. Specimen No. 156--5. 266 Ecology and Taxonomy of Recent Marine Ostracodes II I I I B \ Frc. 55. Philomedes paucichelata Kornicker, new species (Female). A. Portion of fifth leg. Specimen No. 156-5. B. Seventh limb. Specimen No. 156-5. C. Portion of maxilla. Specimen No. 156-5. Ecology and Taxonomy of Recent Marine Ostracodes FIG. 56. Pseudophilomedes jerulana Kornicker. new species (Female). A. Furca. Specimen No. 118B-2. B. Distal portion of fifth limb. Specimen 1\o. 1188-3. C. Distal portion of maxilla. Specimen No. 118B-3. D. Mandibular palp. Specimen No. 1188-3. 268 Ecology and Taxonomy of Recent Marine Ostracodes F1G. 57. Rutiderma (Rutiderma) dinochelata Kornicker, new species (Female). A. Furca. Specimen No. 247-10. B. First antenna. Specimen No. 247-10. C. Second antenna. Specimen No. 247-10. 0. Frontal organ. Specimen No. 247-10. E. Seventh limb. Specimen No. 247-10. F. Portion of maxilla. Specimen No. 247-10. Ecology and Taxonomy oj Recent J/arine Ostracodes '\ -.i' I \ Fie. 58. Rutiderma ( Rutiderma) dinochelata Kornicker. new species tfemale.1. A. Fifth limb. Specimen '.\o. 2-17-10. B. Mandible. Specimen No. 2-17-10. C. Setae of fifth limb. Specimen .\"o. 2-17-10. D . .\Iandible. Specimen No. 2-17-10. Ecology and Taxonomy of Recent Marine Ostracodes A E FIG. 59. Rutiderma (A. lt ernochelata) polychelata Kornicker, new subgenus, new species (Sex unknown). A. Secondary branch second antenna. Specimen :\"o. 11(}-2. B. Seventh limb. Specimen :'\o. 110-2. C. .'llandible. Specimen ;'\o. 11(}-2. D. Proximal joint of secondary branch of second antenna. Specimen No. 11(}-2. E. Furca. Specimen :'\o. II0-2. Ecology and Taxonomy of Recent Marine Ostracodes A FIG. 60. Asteropina mulleri (Skogsberg) (Sex unknown). A. Distal end of second antenna. Specimen No. CPS-3. B. Second antenna. Specimen No. CPS-3. C. Second antenna. Specimen No. CPAC-1. D. Secondary branch of second antenna. Specimen No. CPAC-1. E. Secondary branch of second antenna. Specimen No. CP13-l. F. Furca. Specimen No. CPAC-1. Ecology and Taxonomy of Recent Marine Ostracodes FIG. 61. Asteropina mulleri (Skogsberg) (Female). A. First antenna. Specimen No. CPS-1. B. Mandible. Specimen No. CPS-1. C. Sixth limb. Specimen No. CPS-1. D. Mandibular process. Specimen No. CPS-1. E. Maxilla. Specimen No. CPS-2. F. Distal end of seventh limb. Specimen. No. CPS-2. Ecology and Taxonomy of Recent Marine Ostracodes 273 c Fu;. 62. ..J.steropina setisparsa Kornicker, new >pecies I Female) . . .\. ~landible. Specimen '.' ment of clump of hairs on shell. Specimen No. CP38-10. E. Sarsiella carinata A. Scott, female of immature instar. E, Side view. Specimen No. CP38-5. C. Sarsiella truncana Kornicker, new species, female.C, Side view. Specimen No. CP52-21. G, K-N. Chelicopia arostrata Kornicker, new genus, new species. G, Dorsal view; K, Side view. Speci­men No. 91-1. L, Inside view of detail of anterior edge (female) . Specimen No. CP38-16. M, Side view (eye is bright orange) . Specimen No. 118A-l. N, Side view. Specimen No. CP38-16. H-J, P, Q. Actinoseta chelisparsa Kornicker, new species. I, Dorsal view; Q, Side view. Specimen No. CP38-2. J, Surface pits ; H, P. Detail of surface pits. Specimen No. C-1. Figures with similar magnification, A, C, F, G, K, M, N; B; D; E; H, L, 0, P ; I, Q; J. Trematode Parasites of Donax variabilis at Mustang Island, Texas SEWELL H~ HOPKINS The A. and M. CoUege of Texas, College Station Introduction In 1951 and 1952. Harold C. Loesch1 carried out an ecological study of the popula· lions of coquina clams. Donax variabilis Say and Donax tumida Philippi, on the Gulf beach of Mustang Island. Loesch t 19571 included in his ecological report some data on parasites of Donax, with illustrations of certain species. but did not name any of them. The purpose of the present paper is to name the trematodes found in Donax wriabilis at Mustang Island, to describe them so they can be identified by other workers. and to discuss their biological significance. The author studied the parasites of Donax with Loesch in 1951 and 1952, and made a more intensive study of some of them in June. 1957, while employed on the summer staff of the Institute of Marine Science. Most of the parasites previously known from Donax have been described by European workers, mainly from the coast of France. The European species include a bucephalid cercaria identified by Giard ( 1897a) as Bucephalus haimeanus Lacaze-Duthiers, 1854, metacercariae of Gymnophallus sp., and two trichocercous cercariae, Cercaria pectinata Huet, 1891 and C. lutea rnn Bt-neden. 1870, of Giard, 1897 a. There is considerable literature on each of these species. The author has the opinion that Giard's bucephalid was not B. ha.imeanus hut a still unnamed species distinct from the one found by Lacaze-Duthiers in Ostrea edulis. Rees t 1939) believed the gymnophallids found in Donax to be metacercariae of the tailless Cercaria strigata Lebour, 1908. from sporo­cysts in Cardium edule, and thought that the adult form was some Gymnophallus of birds. most likely G. deliciosus I Olsson) of gulls. Palombi ( 1934) reported that Cercaria pectiTUJta. tor a similar trichocercous cercaria) encysted in amphipods and developed into adults of the fellodistomatine species Bacciger bacciger (Rudophi) in fishes which ate the amphipods. Dollfus ( 1946) considered the Cercaria lutea van Beneden of Giard to be a synonym of C. pectinata Huet. This species I if it is only one) has been reported from Donax rittatus Da Costa ID. anaiinus Lamarck) as well as from D. trunculus L. The only American trematode reported from Donax prior to the publication of Loesch t 1957) is the monorchiid Postmonorchis donacis Young, 1953, whose sporo­cysts, cercariae. and metacercariae are found in D. gouldii Dall on the coast of southern California. The life cycle of this species was described by Young (1953). 1 Mr. Loesch was a graduate student in biological oceanography under Dr. J. G. Mackin at the A. and l\l. College of Texas. His study was guided by Mr. l\l. D. Burkenroad and Dr. Gordon Gunter of the Institute of Marine Science at Port Aransas on Mustang Island. Descriptions of Species Parvatrema donacis n. sp. Cable (1953) described the adult, metacercaria, and cercaria of a new gymnophalline, Parvatrema borinquenae, from Puerto Rico. The new genus Parvatrema was said to differ from Gymnophallus Odhner, 1900, by having a large pit-like genital pore anterior to the ventral sucker, a short genital atrium, a well-developed pharynx, only one group of vitelline follicles, and the excretory formula 2 [ (2 + 2) + (2) ]. Metacercariae of Parvatrema borinquenae were found in the snail Cerithidea costata, and sporocysts and cercariae in the small clam Gemma purpurea. Cable obtained adults five days after feeding metacercariae to baby chicks, and guessed that the natural final host might he a wild duck. Metacercariae of the gymnophalline type, found in at least 85 per cent of the Donax variabilis on Mustang Island beach, have the features listed by Cable as distinctive for the genus Parvatrema but differ from P. borinquenae in some morphological details as well as in host and locality. They ;lre therefore considered to represent a new species, P. donacis. All metacercariae from Donax variabilis have lateral diverticula of the excre­tory bladder, posterior to the testes, which are not mentioned or illustrated by Cable. These and other structures are shown in Figure 1. Lateral papillae on the oral sucker, FIG. 1. Parvatrema donac;s metacercaria, living, dorsal view, drawn with camera ]ucida. Length 0.29 and width 0.15 mm. Note lateral diverticula of excretory bladder posterior to testes, ovary ante,rior to right testis, and Yitellarium, anterior to left testis. FIG. 2. Parvatrema donacis, older metacercaria with expanded excretory bladder, living, ventral view, drawn with camera lucida. Length 0.30 and width 0.21 mm. Note common genital pore anterior to ,-entral sucker. FIG. 3. Cercaria of "dichotoma" type, believed to be Parwtrema donacis cercaria, from sporocystin Donax variabilis, alive, dorsal view, drawn with camera lucida. FIG. 4. Cercaria believed to be Parvatrema donacis cercaria, alive, ventral view, drawn with camera lucida. FIG. 5. Cercaria choanura from sporocyst in Donax variabilis, alive, slightly flattened, stained with neutral red ; dorsal view, drawn with camera lucida. Length 0.28 and width 0.09 mm. FIG. 6. Cercaria choanura, killed in formalin and stained with alum cochineal; ventral view, drawnwith camera lucida. Length 0.275, width 0.063 mm. FIG. 7. Metacercaria dissected from cyst in Donax variabilis, believed to be the metacercarial stageof Cercaria choanura. alive, stained with neutral red; ventral view, drawn with camera lucida (com­posite of several specimens). 304. Trematode Parasites of Donax variabilis at Mustang Island like those of P. borinquenae, were seen on some specimens. Cephalic glands could not be counted accurately, even when stained with neutral red. Twelve flame cells, placed exactly as in Cable's Figure 1, were observed and drawn before Cable's drawings were seen. Connections of all tubules were not traced, but there seems lo be no reason to doubt that P. donacis has the same excretory formula as P. borinquenae, 2 [ (2 + 2) + (2) ]. Young. active metacercariae (Fig. 1) average 0.33 mm. in length and 0.17 mm. in width, with the oral sucker averaging 0.090 by 0.086 mm., the ventral sucker 0.037 by 0.039 mm., and the pharynx 0.036 by 0.030 mm. Older metacercariae ffig. 2), up to 0.80 mm. long and 0.50 mm. wide, are nearly motionless and so full of excretory concre­tions and other minute spherical bodies (parasites?) that they are quite opaque. Many of these older metacercariae eventually become calcified and sealed into the inner lining of the shell not far below the hinge, as has been reported for the metacercariae of Gymnophallus in Europe. Like other gymnophallines, the metacercariae of P. donacis remain unencysted but are surrounded by a zone of liquid or semi-liquid secretions. A furcocercous cercaria of the "dichotoma" type was found in eight of the 1017 Donax variabilis examined by Loesch in 1951 and 1952, but was not found in any of the 100 clams dissected by the author in 1957. All of the known cercariae of this type, named for Cercaria dichotoma Mueller, 1855, develop from sporocysts in marine bivalves. Markowski (1936) , partly on the basis of a clue furnished by Pelseneer (1906). suggested that "dichotoma" cercariae are larvae of gymnophallines. Loesch (1957) called the metacercariae of P. donacis "Gymnophallus metacercariae" because of their resemblance to the European species assigned to that genus, and called the "dicho­toma" cercariae "Gymnophallus cercariae" because of Markowski's theory, the close association of cercaria and metacercaria in Mustang Island Donax, and the absence of any other cercaria which might be a gymnophalline. The cercaria of P. borinqueiiae Cable, described in 1953, is so similar to the one found in Donax variabilis that it now !'eems virtually certain that the latter is the cercaria of P. donacis. The Donax cercaria (Figs. 3 and 4 \ is somewhat larger than the Puerto Rican cercaria from Gemma, live specimens averaging 0.132 mm. in length and 0.058 mm. in width, with a range of 0.120 to 0.145 mm. in length and 0.050 to 0.065 mm. in width. The oral sucker averages 0.036 by 0.032 mm., the ventral sucker 0.033 by 0.032 mm., and the pharynx 0.017 by 0.016 mm. The tail is 0.044 mm. from body to fork, and 0.050 mm. from fork to tips of furci. No papillae were noted, and no setae, either on the body or on .the tail. Cephalic glands are so inconspicuous that they could not be counted accurately. 'Flame cells and excretory tubules are exactly as in P. borinquenae, but the connection of the tubule from the second pair of flame cells co.uld not be seen, and excretory open­ings in tail furci, if present, were not noticed. These cercariae develop in short sporo­cysts, 0.40 to 0.60 mm. long and 0.14 mm. wide, in the gonad of Donax variabilis. Each sporocyst bears approximately six cercariae. The gonads of the eight infected clams contained no gametes. Parvatrema and Gymnophallus are fellodistomatids of the subfamily Gymnophallinae which have their adult forms in the intestines of birds. Cable (1953) obtained adults of Parvatrema borinquenae experimentally in baby chicks, and thought wild ducks might be the natural final hosts. Wild ducks do not feed on the Mustang Island beach. The final host of P. donacis is probably some wading bird or shore bird. Loesch (1957) ob­served black-bellied plovers, sanderlings, and Eastern willets feeding on Donax, but did not find trematodes in the few individuals he examined. The author fed metacercariae of P. donacis to two one-week-old chicks in July, 1957, but failed to find any trematodes when the intestines were examined under a binocular dissecting microscope five days later. It is possible that younger chicks would have given positive results. Cercaria choanura n. sp. This very distinctive cercaria was found in 26 of the 1017 Donax variabilis examined by Loesch in 1951-1952 and in one of the 100 examined by the author in 1957. It comes from orange-pigmented sporocysts 1.5 to 2.0 mm. long and 0.4 to 0.5 mm. wide which develop in the gonad and cause parasitic castration of the host. The stubby, transversely striated or "corrugated" tail of the cercaria is held by a muscular collar­like base, into which the tail can be withdrawn when the cercaria is crawling. This unique structure suggested the name "choanura." No other cercaria known to the author has a tail of this type. In dilute neutral red solution, the tail proper stains red but the collar remains unstained. Cercaria choanura is shown in Figures 5 and 6. Rows of minute spines or scales circling the body give it a transversely striated appearance. There are no specialized penetration spines. Cephalic glands, apparently three pairs, lie just anterior to the ven­tral sucker. They stain red in diluted neutral red. Ducts from the glands open dorsally near the anterior tip of the body. There is an anteroventral mouth, a nearly round oral sucker, a short prepharynx, a small but distinct muscular pharynx, and a narrow esophagus which runs to a point halfway between the two suckers and there divides into two sport, indistinct ceca. In stained specimens genital analagen can be seen, appar­ently anlagen of gonads just anterior to the excretory bladder and of ducts running anteriad over the dorsal side of the ventral sucker. The conspicuously granular cells of the thick bladder wall are reddish-brown in live cercariae stained with neutral red, and brown in formalin-fixed specimens stained with alum cochineal. No excretory tubes or flame cells were seen in the short time available for study of live cercariae. Nerve cords and ganglia (unstained) were seen in some cochineal-stained specimens (Fig. 6). Living, unflattened cercariae and specimens killed in cold formalin had approximately the same average measurements: Length of body 0.280 mm., width 0.080 mm., oral sucker 0.044 by 0.038 mm., ventral sucker 0.036 by 0.037 mm., pharynx 0.020 by 0.015 mm., tail collar 0.042 by 0.037 mm., and tail proper 0.024 mm. long beyond the collar. Cercaria choanura is the "Cercaria A" of Loesch (1957). Cable in private corre­spondence expressed the opinion that this cercaria belongs to the family Monorchiidae, the adult forms of which live in the intestine of fishes. The author agrees, although C. choanura is very different from the cercariae assigned to Monorcheides cumingiae by Martin (1938, 1940) and to Postmonorchis donacis by Young (1953). Both of those cercariae have tails nearly as long as the body, covered with circular rows of prominent scales, and do not have a collar-like tail base. The cercaria of M. cumingiae develops in sporocysts in the visceral mass of the marine clam Cumingia tellinoides in Massachu­setts. Metacercariae, in the same clam or in Tellina tenera, developed into adults within four weeks after they were fed to eels and flounders. The cercaria of Postmonorchis donacis develops from sporocysts in Donax gouldii on California beaches and encysts in the same clam. Encysted metacercariae fed to Embiotocidae (surf perches) developed into adults. Both Martin and Young compared their cercariae with the European Cer­caria myocerca Villot, 1878, from the marine clam Scrobicularia tenuis. Loesch and the author found encysted metacercariae apparently belonging to Cercari.a choanura in both Donax rnriabilis and D. tumida at )lustang Island, and in Dona:x tumida at Grand Isle, Louisiana, where D. rnriabilis does not occur. These metacer­cariae were most abundant in the walls of the incurrent siphon and in nearby parts of the mantle. They were seen in 72 of the 100 Donax 1;ariabilis dissected at Mustang Island in June. 1957, and in all of the relatiwly small number of D. tumida examined at Grand Isle. One Donax from Grand Isle, examined by Loesch, contained sporocysts and Cercaria choanura. .\o other trematode larrne were found on the Grand Isle beach. The metacercaria 1Fig. 7 1 beliewd to be an adrnnced stage of C. choanura is coiled "ithin a cyst usually 0.125 to 0.150 mm. in diameter, rarely as much as 0.250 mm. Pigmented eyespots, suckers, pharynx, and excretory bladder are dearly ,·isible through the thin cyst "all. Dilute neutral red stains the esophagus and intestine a conspicuous red. Only a few metacercariae were dissected intact from cysts, after many attempts. Two intact liYe specimens measured 0.25 and 0.28 mm. long, 0.12 and 0.13 mm. wide, "ith oral suckers 0.0-1-7-0.050 by 0.056-0JJ60 mm .. wntral suckers 0.0-t3 by 0.04-5 mm., and pharynx 0.030 by 0.025 mm . .\o cephalic glands were seen, ewn in neutral red solution. The pharynx was larger and the inte;;tinal ceca were longer than in the cercaria. Flame cells and excretory tubules were not seen during the short time fo·ing metacer­cariae were studied. Figure I shows one of these metacercariae. Bucephalus loeschi n. sp. Three of the 1017 Donax rnriabilis examined by Loesch in 1951-1952 and two of the 100 examined by the author in 1957 contained sporocysts and cercariae of a bu­cephalid. This specie;; was mentioned by Hopkins 119511 and Loesch 119571 but was not described. It seems best to name and describe it now even though some details of its morphology remain uncertain. The principal characteristics of the Donax bucephalid are shown in Figures 8 and 9. The species is named for its disco,·erer, Harold C. Loesch, and placed proYisionally in the genus Bucephalus rnn Baer, 1827. This is the type genus of the family Bucephalidae, and the name was first used for B. polJmorphus, the first­known bucephalid cercaria. Bucephalus loeschi closely resembles B. cuculus ~kCrady as redescribed by Hopkins I 195-l 1. It tends to be more elongate than the latter in life, being sometime;; ten times longer than wide. The pharynx is farther back, in the last fifth of the body length in elongated indiYiduals and seven-tenths of the way back in contracted cercariae. The excretory bladder is shorter than in B. cuculus, extending little more than halfway from the posterior end to the lewl of the pharynx. Fiw lfring cercariae had the following average measurements : Body length 0.382 mm., width 0.049 mm.; pharynx four-fifths of the distance from anterior to posterior end; anterior "sucker" 0.059 by 0.030 mm., pharynx 0.027 by 0.028 mm., tail stem 0.063 mm. long and 0.085 mm. wide, tail furci 0.016 mm. wide and up to 2.25 mm. long. Measurements of fiw formalin-killed speci­mens a\·eraged as follows: Body length 0.25-l mm., width 0.060 mm.; pharynx seven­tenths of distance from anterior to posterior end; anterior "sucker" 0.037 by 0.030 mm., pharynx 0.015 by 0.021 mm., tail stem 0.0-1-8 by 0.093 mm., tail furci 0.022 mm. wide and 0.60 to 0.65 mm. long. The longe;;t fo·ing cercaria was 0.440 mm. long and 0.0.t3 mm. wide. and the shortest formalin-killed specimen was 0.217 by 0.063 mm. Figures 8 and 9 giYe only an approximation or impression of the gland cells around the anterior "sucker" and pharynx, for these were newr seen very distinctly. Genital 9 Fie. 8. Bucephalus Loesch ;_ cercaria from ;oporocy;ot in Dona.\· rnriabilis. ali,·e and ;olightly Aat tened, stained with neutral red: dor~al ,-:ew. drawn with camera lucida. Length of bodY 0.23 and w:dd1 0.06 mm. . -. Fie. 9. Bucephalus loeschi, bal;oam mount, ;otained with alum cochineal. wntral Yiew. chma "·i1h c~mern lucida. anlagen. though conspicuous in all specimens. were not consistent in form and arrange· ment. It is suspected that all cercariae seen were somewhat immature: naturally emerged cercariae might show structural details more distinctly. Sporocysts of B. loeschi are long and branched like those of other bucephalids. All of those seen were in the gonadal region, where they replaced germinal tissues and caused parasitic castration of the host. Spots of orange pigment in sporocysts giw the infected gonad a yellow color, as seen with the naked eye. Bucephalid cercariae identified as Bucephalopsis haim eanus haYe been reported from Donax trunculus in Europe by Giard ( 1891 a I and others. The author doubts that the cercaria from Donax is the same as the one from the European oyster, Ostrea edulis. Cercaria caribbea :XLII. a new bucephalid cercaria found in Donax denticulata in Puerto Rico by Cable (19561, differs from the l\Iustang Island species by its smaller size, more slender tail. longer excretory bladder, and some differences in minor features. Adult bucephalids known from Mustang Island and YicinitY include Rhipidoco:yle transversale, R. lintoni, and Bucephaloides strongylurae, all from the needle gar, Strongylura marina. The cercariae of these species are unknown. Bucephalus cuculus cercariae, known to occur in Aransas Bay oysters, Crassostrea virginica, probably de­velop into Rhipidocotyle lepisostei in alligator gars, Lepisosteus spatula, but this has not been proved experimentally. An unidentified bucephalid cercaria was found once in the Gulf oyster, Ostrea equestris, at Port Aransas. Further search would probably reveal the presence of many other bucephalids, larval and adult, in this little-studied region. Lobatostoma sp. An immature aspidogastrid belonging to the genus Lobatostoma Eckmann was found in a Donax variabilis by Loesch in the summer of 1951. No additional specimens were found in the 1017 clams dissected by Loesch, nor in those examined by the author. Adults of this genus live in intestines of marine fishes, including pompanos and croakers. Discussion The trematodes now known from Donax in America belong to the three families Monorchiidae ( Postmonorchis donacis and presumably Cercaria choanura), Bucephal­idae (Cercaria caribbea XLII and Bucephalus loeschi), and Fellodistomatidae (Parva­trema donacis). The known European species belong to the last two of these families. The subfamilies Gymnophallinae, represented by Gymnophallus and Parvatrema, and Fellodistomatinae, represented by Bacciger, were placed together in the family Fellodis­tomatidae by Cable \ 1953 I. All of the known American parasites of Donax are different from any known European species. Pelseneer ( 1896, 1928) blamed heavy infections of Cercaria pectinata for periodic "decimations" of Donax vittatus on the northern coast of France. Jobert ( 1894) at­tributed an abnormal mortality of Tapes decussata and T. pullastra at Arcachon to a high incidence of a trichocercous cercaria similar to C. pectinata. Coe (1946) mentioned "trematode parasites," probably sporocysts and cercariae of Postmonorchis donacis, as one of the agents controlling California populations of Donax gouldii. Later, Coe (1955) attributed "epidemic disease" of Donax gouldii to a unicellular parasite 2-6 p. in diameter, which he compared to the oyster parasite Dermocystidium marinum. In Europe the coccidian Hyaloklossia pelseneeri Leger, which has spherical oocysts 75 to 80 p. in diameter, has been reported from the kidney of Donax. Most if not all, of the known trematode parasites of pelecypods have been reported to cause parasitic castration of the hosts. Parasitic castration of Donax was reported for "Bucephalus haimeanus" and "Cercaria lutea" by Giard (1897a), for Cercaria pectinata by Pelseneer (1896, 1928) and others, and for Postmonorchis donacis by Young (1953). Loesch and the author found that Donax variabilis was made sterile by the sporocysts of Cercaria choanura, Bucephalus loeschi, and the "dichotoma" cercaria which we believe belongs to Parvatrema donacis. Metacercariae probably do not kill or sterilize Donax. However, it is probable that heavy infections affect the growth of this small clam, in view of the adverse effects on the much larger Crassostrea gigas that Hoshina and Ogino (1952) found to be caused by metacercariae of Gymnophalloides tokiensis Fujita, 1925. Hyperparasitism of sporocysts of Cercaria pectinata in Donax vittatus by the "hap­losporidian" Anurosporidium pelseneeri Caullery and Chapellier, 1906, was first re­ported from the northern French coast by Pelseneer ( 1896) , although he did not recog­nize the nature of the protistan parasites. Cepede (1911) made a careful study of the hyperparasite. Dollfus (1925, 1946) restudied it and found the original description in error, so he placed A. pelseneeri in the genus Urosporidium Caullery and Mesnil, 1905. U rosporidium pelseneeri and a similar spec:es (described by Guyenot, 1943) infecting sporocysts in Barnea candida cause "parasitic castration" of the trematode; that is, infected sporocysts fail to produce cercariae and become filled with hyperparasites instead. Nosema legeri Dollfus, 1912, a microsporidian hyperparasite of Gymnophallus meta­cercariae in species of Donax from France was reported by Giard ( 1897b) and Leger (1897) , and was later studied by Jameson (1902), Guyenot, Naville and Ponse (1925), Dollfus (1912, 1946) , and others. According to Dollfus, the Nosema infection gradually kills the metacercariae; as these become moribund, the mantle of the clam seals them over with calcareous deposits. Similarly, Parvatrema donacis metacercariae are sealed into the shell lining after they become large, inactive, opaque, and packed with minute round bodies of unknown nature. Dr. J. G. Mackin once examined these minute bodies and was not convinced that they were parasites, but he did not have time to make a thorough study. An interesting sidelight is the relationship of the Donax parasites to the parasites which supposedly cause pearl formation in larger bivalves. Giard (1897b) found pearls formed around Gymnophallus metacercariae in several species of Tellina, and possibly in Donax. Dubois (1901) claimed that metacercariae of Gymnophallus margaritarum. but only those killed by sporozoan infections, became the nuclei of fine pearls in Mytilus edulis. Dubois (1907, 1909) found sporozoan spores in the nuclei of pearls in Margariti­fera vulgaris, Modiola barbata, and Pinna nobilis on the Tunisian coast and suspected that the pearls were formed around the cysts of trematode and cestode larvae killed by hyperparasites. Dollfus (1946) said the parasitic worms were probably killed by a Nosema, similar to the species he found killing Gymnophallus metacercariae in Donax vittatus. Dubois ( 1903) claimed to have increased enormously the pearl production of Tunisian pearl oysters, Margaritifera vulgaris, by transplanting them to waters where mussels, Mytilus galloprovincialis, were heavily infected by Gymnophallus duboisi metacercariae containing hyperparasitic sporozoans. Summary Sporocysts and cercariae of three species and metacercariae of two species occur in coquina clams, Donax variabilis Say, on the Gulf beach of Mustang Island at Port Aran­sas, Texas. In addition, an immature aspidogastrid, lobatostoma sp., was found once. The cercariae are Bucephalus loeschi, Cercaria choanura, and a "dichotoma" cercaria. The bucephalid presumably penetrates and encysts in some fish and becomes adult in a second fish that eats the first one. Cercaria choanura is presumed to develop into a meta­cercaria which is found encysted in most of the Donax of Mustang Island, and to become an adult monorchiid in some fish that eats coquina clams. Cercaria choanura. and the metacercaria presumed to develop from it also were found in Donax tumida at Grand Isle, Louisiana and at Mustang Island. A gymnophalline metacercaria described as Parvatrema donacis, presumed to develop from the "dichotoma" cercaria, is found in most of the Mustang Island Donax; it probably becomes adult in some clam-eating shore bird. All of these digenetic trematodes are considered to be new species. Previously known parasites of Donax are discussed. Parasitic castration is caused by all known sporocysts in Donax. LITERATURE CITED Cable, R. M. 1953. The life cycle of Parvatrema borinquenae gen. et sp. nov. (Trematoda Digenea)and the systematic position of the subfamily Gymnophallinae. J. Parasit., Sec. 1, 39(4): 408-421. ----. 1956. Marine cercariae of Puerto Rico. Sci. Surv. P. R., 16(4): 491-578. Caullery, M. and A. Chappellier. 1906. Anurosporidium pelseneeri, n. g., n. sp., haplosporidie infectant Jes sporocystes d'un trematode parasite de Donax trunculus L. C. R. Soc. Biol. Sci., Paris, 60 (7) : 325-328. Cepede, C. 1911. Le cycle evolutif et Jes affinites systematiques de L'Haplosporidie des Donax. C. R. Acad. Sci., Paris, 153(9) : 507-509. Coe, W. R. 1946. A resurgent population of the California bay-mussel (Mytilus edulis diegensis). J. Morph., 78: 85-103. ----. 1955. Ecology of the bean clam Donax gouldi on the coast of Southern California. Ecology,36 (3 ) : 512-515. Dollfus, R. P. 1912. Une metacercaire margaritene parasite de Donax vittatus Da Costa. Mem. Soc. zoo!. Fr., 25: 85-144. ----. 1925. Liste critique des cercaires marines a queue setigere signalees jusqu'a present. Trav. Stat. zoo!. Wimereux, 9: 43-65. ----. 1946. Parasites (animaux et vegetaux) des helminthes. Paris, Encyclopedie Biologique(ed. Paul Lechevalier), 482 pp. Dubois, Raphael. 1901. Sur le mecanisme de la formation des perles fines dans le Mytilus edulis. C. R. Acad. Sci., Paris, 133 (16) : 603-605. ----. 1903. Sur L'acclimatation et la culture des pintadines, ou huitres perlieres vraies, sur !es cotes de France, et sur la production forcee des perles fines. Ibid., 137: 611-613. ----. 1907. Sur un sporozoaire parasite de l'huitre perliere (Margaritifera vulgaris). Son role dans la formation des perles fines. C. R. Soc. Biol., Paris, 62: 310-312. ----. 1909. Contribution a l'etude des perles fines, de la nacre et des animaux qui !es produis· sent. Ann. Univ. Lyon, N. S., I. Sci., Med., 29: 1-127. Giard, Alfred. 1897a. Sur un cercaire setigere (Cercaria Lutea) parasite des Pelecypodes. C. R. Soc. Biol., Paris, Ser. 10, 49 (4) : 954-956. ----. 1897b. Sur un distome fBrachycoelium sp.l parasite des Pelecypodes. Ibid., Ser. 10, 49(4): 956-957. ----. 1907. Sur !es trematodes margaritigenes du Pas-de-calais. Ibid., 53: 416-420. Guyenot, E. 1943. Sur un Haplosporidie, parasite dans un sporocyste de la Pholade, Bam ea candida L. Rev. Suisse Zoo!., 50!15 l: 283-286. Guyenot, E., A. NaYille, and K. Ponse. 1925. Deux microsporidies parasites de trematodes. Rev. suisse Zoo!., 31(11) : 399-421. Hopkins, S. H. 1951. Studies on larrnl marine bucephalids. (Abstr.) J. Parasit., Sec. 2, 37(5): 16-17. ----. 1954. The American spec:es of trematode confused with Bucephalus (Bucephalopsis) haimeanus. Parasitology, 44 (3 & 4) : 353-370. Hoshina, T. and C. Ogino. 1952. Studien ueber Gymnophalloides tokiensis Fujita, 1925. I. Ueber die Ein"·irkung der larrnlen Trematoda auf die chemische Komponente und das Wachstum rnn Ostrea gigas Thunberg. J. Tokyo Univ. Fish., 38(3): 335-350. Jameson, H. L. 1902. On the origin of pearls. Proc. zoo!. Soc. Lond., 1902(1): 140-166. Jobert, YI. 1894. Recherches pour servir a I' histoire du parasitisme. C. R. Soc. Biol. Paris, Ser. 10, 46(1): 519-520. Leger, L. 1897. Sur la presence de glugeidees chez le; distomes parasites des Pelecypodes. C. R. Soc. Biol. Paris, Ser. 10, 49(41: 957-958. Loesch, H. C. 1957. Studies of the ecology of two species of Donax on Mustang Island, Texas. Pub!. Inst. l\1ar. Sci. Univ. Tex., 4 (21: 201-227. Markowski, S. 1936. Ueber die Trematoden-fauna der balt'scher Mollusken aus der Umgebung der Halbinsel He!. Bull. Acad. Polonaise Sci. Lett., Cl. Sci. Math. Nat., Ser. B: Sci. Nat., 2: 285-317. Martin, W. E. 1938. Studies on trematodes of Woods Hole: The life cycle of Lepocreadium seti­Jeroides (Miller and Northrup), Allocreadiidae, and the description of Cercaria cumingiae n. sp.Biol. Bull., 75(3): 463-474. ----. 194-0. Studies on the trematodes of Woods Hole. III. The life cycle of Monorcheides cumingiae (Martin) with special reference to its effect on the invertebrate host. I bid., 79 (1 I : 131-144. Palombi, A. 1934. Bacciger bacciger (Rud.) Trematode digenetico: Fam. Steringophoridae Odhner. Anatomia, sistematica e biologia. Pubbl. Staz. zoo!. Napoli, 13: 438-478. Pelseneer, Paul. 1896. Un trematode produisant la castration parasitaire chez Donax tmnculus. Bull. Sci. Fr. Belg., Ser. 4, 27 (6) : 357-363. -----. 1906. Trematodes parasites des mollusques marins. lid., 40: 161-186. ----. 1928. Les parasites des mollusques et des mollusques parasites. Bull. Soc. zool. Fr. 53: 158-189. Rees, F. G. 1939. Cercaria strigata Lebour from Cardium edule and Tellina tenuis. Parasitology, 31: 458-463. Young, R. T. 1953. Postmonorchis donacis, a new species of monorchid trematode from the Pacific Coast, and its life history. J. Wash. Acad. Sci., 43(3): 88-93. A Partially Annotated Checklist of the Marine Fishes of Texas1 HINTON D. HoEsE2 Department of Biology, A. & M. College of Texas, College Station, Texas and Marine Laboratory, Texas Game and Fish Commission, Rockport, Texas Contents INTRODUCTION -----------------------------------------------------------------------------------.. ------.. ----------....... 312 Acknowledgments PREVIOUS WORKS ON TEXAS MARINE FISHES ····-·············-·······························--·········· 313 ZooGEOGRAPHic CONSIDERATIONS --------------------------------------------------------------------------------313 ARRANGEMENT AND INCLUSION OF SPECIES ···································-···········-····--···-······-· 314 ANNOTATED LIST ------------------------------------------------------------------------------------------------------------315 LITERATURE CITED ------------------------------------------------------------------·---------···-·······--······--------349 Introduction This paper is basically the result of a literature survey with the attempt to list all the species of fishes known to occur in Texas marine waters. The annotations are not intended to be complete, except for rare or uncommon forms, but are included for the use of the interested student for further study and the establishment of a species presence in Texas waters. Doubtful records are not included in the systematic listing but are mentioned in a discussion under the proper category. As many species as possible were examined from the collections of the Institute of Marine Science of the University of Texas at Port Aransas, the Department of Wildlife Management of the A. and M. College of Texas at College Station, and the Marine Laboratory of the Texas Game and Fish Commission at Rockport. References to speci­mens in these collections are abbreviated in the annotations as follows: IMS (Institute of Marine Science), ML (Marine Laboratory), AM (A. and M. College). Other abbrevi­ations include CNHM (Chicago Natural History Museum}, FWS (Fort Crockett Labora­tory, Fish and Wildlife Service, Galveston}, USNM (U.S. National Museum). ACKNOWLEDGMENTS Without the aid of a great many people, too numerous to mention, a work of this kind would be far from possible. The writer would like to particularly thank the staff of the Marine Laboratory, namely Mr. Howard T. Lee, Director; Mr. Terrance R. Leary 1 Contribution No. 42 of the Marine Laboratory, Texas Game and Fish Commission, Rockport, Texas. 2 Present address. Field Laboratory, Texas Game and Fish Commission, Seabrook, Texas. and Mrs. Patricia Simpson. Mrs. Virginia Dickey supplied Figure I. Drs. Howard T. Odum and Richard J. Baldauf kindly permitted the writer to examine-collections of the Institute of Marine Science and the Department of Wildlife Management of A. and M. College. Drs. Henry H. Hildebrand of the l -niversity of Corpus Christi, Gordon Gunter of the Gulf Coast Research Laboratory. John C. Briggs of the Cniversity of Florida, Clark Hubbs of the Cniwrsity of Texas, and Mr. Loren Woods of the Chicago :\atural History ~foseum contributed information. suggestions, and corrections of particular value. Dr. Victor G. Springer, while at the Institute of Marine Science. constantly con­ferred with the writer and provided much of the incentive for the work. Dr. Clark Hubbs and ~fr. Han-ey Bullis of the CS. Fish and Wildlife Sen-ice kindly criticized the manu­script. The writer would finally like to acknowledge the kind assistance of Miss Carol Currie for typing the manuscript. Errors and omissions remain the full responsibility of the writer. Previous Works on Texas Marine Fishes Ewn though the ichthyofauna of Texas has receiwd more attention than any other Gulf state, excepting possibly Florida. present workers have little idea of the number of species and their distribution along the Texas coast. Baughman I 1950a, 1950b) has compiled the only extensi,-e list, including both marine and freshwater fishes. This list. ewn though incomplete and perpetuating many errors, compiles a great deal of scattered information. Since that time many taxonomic revisions have helped clarify the status of many of our marine fishes. The most important of these are Ginsburg ( 1950, 195la 195lb, 1952b, 1952c, 1954). Rims (1950, 1951). and Briggs I 1955). The literature cited includes the main systematic and ecological studies that haw bearing on Texas marine fishes. Most of the older references are not included. Baughman I 1950b I lists a fairly complete number of these. Zoogeographic Considerations The status of Texas marine fishes from the standpoint of Gulf zoogeography is not completely clear, but some progress is being made in this field. Baughman (1950a) believed that the western Gulf was a separate entity from the eastern Gulf, divided by the silt laden flood of the Mississippi. Ginsburg ( 1952a) generally supports this theory with a different line of division, based on the distribution of four species. He speculates on an east-west fauna! break from two viewpoints: First. the division of populations at the species le,-el exemplified by Archosargus probaJocephalus and Hippocampus ::.osterae, two eastern Gulf species, and their counterparts A. oi:iceps and H. regulus in the western Gulf; ;:econd, the presence of two new species on the western Florida coast not known from the western Gulf. Menticirrhus focaliger and Centropristes me/anus. Hildebrand t 1955 I subsequently questioned the value of the status of A. oviceps showing that the diagnostic characters of western Gulf Archosargus were noteworthy but not of species magnitude. Kirk Strawn (personal communication I likewise questions the status of H. ::.osterae and H. regulus and is presently revising the dwarf seahorses. Hildebrand ( 1954) also took .ll. focaliger from the Texas coast. Centropristes melanus has not been reported from Texas, although C. striatus, the Atlantic counterpart. was reported by Reed ( 1941). Even though Reed's lop. cit.) Centropristes were undoubtedly the common philadelphi­cus, me/anus may occur in the western Gulf. That there has been disagreement between authors as to a distinct qualitative fauna) difference is clearly indi cative as to the lack of information at this point. Several factors must be analyzed before any conclusions could be drawn . The number of active marine ichthyologists in the northeastern Gulf outnumbers those in the northwestern. Recent additions to the known Texas fish fauna have most always been centered near the two marine laboratories on the Central coast. The more tropical region near Port Isabel has not been sampled intensively since the work of Evermann and Kendall I 1894). As a result, the Texas fish fauna of this area is less well known. For example Baughman I 1950a, 1950b) lists near 200 species of marine fishes, where­as this list contains 424. Two studies in this area should produce significant contri­butions-those of the offshore reefs and "mud lumps," and of the jetties. Briggs (1958) has listed 108 species from the northeastern Gulf, but unreported from the northwestern Gulf. Sixteen of these are reported in this list. He explains this apparent fauna! difference by the coral sponge community present in that area but absent in the northwestern Gulf. While this difference is impressive, it still remains to be conclusively shown. Even the reef fishes, common in the southern Gulf, heretofore recorded from the Texas Coast, occur sporadically near oil rigs, jetties and banks, so even this difference may prove to be largely quantitative. The recently reported species in this area haw largely been the result of studies on the offshore and to a lesser extent pelagic fauna. Hildebrand's I 1954) study, collections by the Oregon (Springer and Bullis, 1956) and collections by Mr. E. D. McRae, formerly of the Texas Game and Fish Commission, have added greatly to the knowledge of these fishes. It seems likely that there is a much greater difference between the northern and southern Gulf of Mexico, the changes being gradual as noted by Reid I 1954) for the west coast of Florida and Hildebrand 11954, 1955) for the east coast of Mexico. Many tropical species common in the southern Gulf are either absent in Texas waters or are apparently strays during warmer months. Arrangement and Inclusion of Species Fresh-water species are included on the basis of their known existence in bay waters over ten parts per thousand total salts. This is arbitrary, but should serve to eliminate many forms which lack the osmoregulatory mechanisms to maintain a marine popula­tion. The fresh-water fishes of the saltier Texas Rivers are not included, because it is felt that these are not a true part of the marine fauna. For higher categories of bony fishes the writer has essentially followed Berg 0940). For the elasmobranchs, the classification of Bigelow and Schroeder (1948, 1953) has been used, except for the Carcharhinidae as modified by Springer 0950). For genera and species, the latest reviser of the group has been followed, where possible. In some cases the nomenclature is the writer's opinion, and here synonyms pertinent to Texas studies are included. Common names followed by an asterisk ( *) are those accepted by the American Fisheries Society (see Chute, et. al., 1948 and supplements) . Other com­mon names used were chosen on the basis of widespread usage in this area. Remarks on the abundance and distribution of the common forms are taken from the main ecological works on Texas marine fishes and from observations by the writer, and are not intended to summarize all information on a species. For the purpose of this paper Texas waters are considered to be those which lie north HEALD U.NIC 1··.... • -,/··_,,,...,·-!;....../ .. ' ..-----~::-. : ,· ' ' ;u·­ ····· ···-. .···.... ~····· · ~ ···-.. ; ·.... ' Q C _J so 0 > :c 40 u ct ~ 0 30 I­ "' 20 10 .... 0 30 40 50 60 70 BODY LENGTH (mm.l Fu;. I. Ontogenetic food progression of Anchoa mitchilli. I Body length refers to standard length.) sumed less frequently, and in the largest anchovies they made up less than two percent of the food material. Undoubtedly, micro-zooplankton is also one of the chief sources of nutrition of anchovies smaller than 30 mm. Remains of small shrimp and fishes made up a prominent portion of the food of anchovies of all sizes examined. These forms, somewhat loosely referred to as macro­zooplankton, constituted 28 percent of the material consumed by the smallest anchovies and increased to over 60 percent in th!:' largest size class. Included in this food category are adult schizopods, larval and post-larval p!:'naeid shrimp, larval fishes (including clupeids and probably some larval anchovi!:'s, again indicating cannibalism·) and a single small naked goby (Gobiosoma bosci). Small bottom-dwelling mollusks and crustaceans were scarcely represented in the diet of the smallest anchovies. This material was conspicuous in th!:' food of the two larger size classes, however, making up over ten percent of the food volume in both groups. In­cluded among these micro-bottom animals w!:'re minute snails, clams (Rangia cuneata), isopods, amphipods, ostracods, and harpacticoid copepods. The presence of such forms in the stomachs of the anchovi!:'s indicates that the speci!:'s must feed to som!:' !:'Xtent upon the bottom since the mollusks and benthic microcrustaceans apparently are unavailable elsewhere. The micro-bottom animals, as well as small quantities of sand, wer!:' most TABLJ; 1 w °' Occurrence of food items in digestive tracts of 92 Anchoa mitchilli °' 30.0-44.0 mm. 45.0-49.0 mm. 50.0-54.0 111111. 55.0-59.0 111111. 60.0-74.0 mm. IS uamint'd 25 t'xamincd 22 examined 19 examined 11 examined B wilh food 21 wilh food 19 with food 18 with food 10 wilh food ">'l 0 0 P('rcrnlaµ:t1 Pt•rct•nlnµc Pcrct'nlngt~ Pe rcenlagc Perccnlagc P c rtc<'nlagc Pcrceula~c Pt•recnlagt~ Pncenlagc Perccnla~c ~ of lrncts* ol Iola) of lrn<'ts• or lotal ul lrnct.s* of lolal of lrn<'ls* or lolnl o( lnu:Js llt of lolal t•onluininfi: slom:u·h t~onlaining slomuch t~onluininf-1 slo1111u·h 1~onlaining slonrnt·h eonlaininf!; .slonuu·h ::i::: FOOll ITEMS ilt•UJ \·olumt~ it Pm \ "Ollllllll ilt•m \•olunw ilcru \·olumt• ikm ,·olumc !:) ------·-----· ~ Rotifera 6.7 .).9 4.0 0.3 "' Ostracoda 4.0 0.8 5..) T 0 Copepoda (undet.) 26.7 11.0 24.0 9.4 4.5 21.1 1.5 -">'l Calanoid 26.7 9.0 16.0 1.4 4.5 T 21.1 3.5 27.3 0.5 i;;• Cyclopoid 4.5 0.5 ;;:,-. Harpacticoid 6.7 0.3 12.0 0.4 13.6 3.3 9.1 1.3 ~ "' !:) Mysid shrimp 20.0 28.l 48.0 52.0 50.0 43.8 42.l 34.5 45.5 40.3 ;:s lsopoda 8.0 5.4 4.5 8.5 10.5 7.3 ~ Amphipoda 16.0 6.0 9.9 1.2 10.5 2.6 ...... ;:s lnsecta o.7 1.0 4.0 0.8 ~ Mollusca ~ ..., Rangia cuneata 4.0 0.2 4.5 0.6 5.:~ 2.9 ~ Gastropoda 4.0 2.3 15.8 3.4 9.1 2.7 0-..., !:) Vertebrata Gobiosoma bosci 5.3 6.8 "' ~ Fish larvae 5.3 7.3 9.1 13.4 <::> Fish remains 27.3 19.3 -!:""< Planktonic diatoms 4.0 0.4 !:) .,.,.. Seeds 6.7 1.0 5.3 1.2 ~ Eggs and cysts 13.3 3.2 4.0 T 4.5 0.2 5.3 0.5 "1:l Organic mat. ( undet.) 7.J.3 .)4.7 48.0 6.9 68.2 3.'l.7 68.4 22.8 72.7 19.2 0 Detritus 33.3 7.2 48.0 13.4 72.7 7.3 63.2 5.1 90.9 3.3 ~ <") Sand 12.0 0.4 13.6 0.8 15.8 0.4 27 .3 ;;:,-. !:) SUMMARY ;:i. ..., Copepoda 20..'l 11.2 :rn 5.0 1.8 Q ;;· Mysid shrimp 28.I 52.0 43.8 34.5 40.3 Isopoda and amphipoda 0.0 11.4 9.7 9.9 0.0 Fishes 0.0 0.0 0.0 14.l 32.7 Miscellaneous 9.1 4.8 0.8 8.0 2.7 Incidental and undet. 41.9 20.7 41.8 28..~ 22.5 • Slomach and intestine included. frequently encountered in the stomachs of anchovies taken during the winter and spring months suggesting that more intensive bottom-feeding may take place at these seasons of the year. Undetermined organic material and detritus was most prominent in the food of the smallest size class ( 42 percent) and declined to one-fourth of the food volume of the largest fish. This substance was generally of fine particulate nature and appeared to be derived primarily by straining of suspended matter from the water rather than by scooping of surface material directly from the bottom. The decline in consumption of this detritus by the larger anchovies paralleled a decline in consumption of micro­zooplankters, as might be expected if straining were involved, and the great volu­metric preponderance of this suspended detritus in the food of the young raises the triv­ial question of which should be considered "incidental", detritus or micro-zooplankton. The fact that the detritus was decreasing as the micro-bottom animals increased points to a greater degree of selectivity by the larger anchovies, and this, in turn, was related to the increased consumption of shrimp and fishes. In summary, the Lake Pontchartrain anchovy population appears to pass through two ontogenetic feeding stages. Young individuals are plankton strainers and consume large quantities of micro-zooplankton and suspended detritus. The relative absence of phytoplankton suggests that such straining occurs near the bottom. With increasing size the species exercises greater selectivity, preying chiefly upon small shrimp and fishes and supplementing this diet with occasional bottom-dwelling mollusks and crusta­ceans. Detritus apparently forms an important nutritional supplement in both the young and adult. In connection with his studies in East Bay, Texas, Reid (1955a, b) indicated that the abundance of the anchovy population depends "upon the mass of zooplankton". Whereas this may b1) the case for the Chesapeake Bay specimens of Hildebrand and Schroeder (loc. cit.), it is only partially true for the Louisiana population which, as indicated, consumes copious quantities of detritic matter. The daily pattern of feeding activity of the anchovy is not entirely clear. Examination of Figure 2 indicates that both the young and adults feed during the middle portion of the day, although the adults seem to commence feeding activities earlier in the morning than do the young. BELONIDAE Strongylura marina (Walbaum). Atlantic Needlefish Hildebrand and Schroeder ( 1928) recorded the food of 18 specimens of needlefish. Seventeen had taken small fishes (Mugil curema, Fundulus diaphana, and silversides) while a single specimen had eaten shrimp. McLane (1948) reported that needlefish con­sume large numbers of shad (Signalosa petenensis vanhyningi) in the St. Johns River of Florida. Seven specimens of the needlefish (357-457 mm.) from Lake Pontchartrain were examined, all of which contained food. Small fishes (Anchoa mitchilli diaphana and others) made up almost two-thirds of this material. Insects (adult Diptera and Coleop­tera) and vegetation ( Cladophora sp. and some vascular plant material) made up less than five percent. Almost one-third of the stomach content consisted of undetermined organic matter which appeared to be largely fishes and invertebrates in advanced stages of digestion. Food Habits of Fishes and Invertebrates of Lake Pontchartrain eo ANCHOA Ml!CHILL! 60 ... _.. :i:: '.J _.......... AA -­ 0..J v.-----­ 40 ~"" 0 • I-;;;;. ui­ 20 80 y 60 A w~ z ..J -...J I-::> ::l"" 40 I-~ ~e... 20 0 8 9 10 II 12 13 14 I~ 1$ 17 18 TIME OF DAY FIG. 2. Hourly pattern of percent-fullness of stomach and intestine of Anchoa mitchilli (young and adult). The food habits of the needlefish are not well known. Within the size range studied, the species is a predator, feeding upon small fishes, insects, and crustaceans. This fish was frequently observed to be active near the surface in deeper water throughout the lake where it must feed on anchovies and young clupeids, although stomach analyses show that it does forage in the weedy shallows of Lake Pontchartrain. ARIIDAE Bagre marina (Mitchell). Gafftopsail Catfish The food of adult gafftopsail catfishes has been reported by Gudger (1916), Gunter (1945), Knapp (1949), Reid (1955b), and Reid, Inglis, and Hoese (1956). These authors have shown that this fish feeds primarily upon blue crabs and penaeid shrimp, although fishes (including menhaden, worm eels, and other species), and various in· vertebrates (squids, small crabs, and insects) may also be taken. A single adult speci· men from Lake Pontchartrain contained only unrecognizable material. Nothing is known of the food habits of small individuals. Galeichthys felis (Linnaeus). Sea Catfish, Hardhead Linton ( 1904) examined the food of 17 hardhead catfish from Chesapeake Bay and found clams (soft parts and valves), snails, annelids, amphipods, shrimp, sea urchin n mains, and fish bones and eye lenses, as well as some sand. Smith ( 1907) indicated that in North Carolina this fish is a bottom-loving species which feeds chiefly on worms and small crustaceans, but readily eats fish flesh or fowl, dead or alive. Perhaps, he had in mind the baits with which this catfish may be taken. Gunter (1945) examined the stomach contents of 85 individuals from Texas of which 59 contained recognizable food. These fish ranged in size from 240 to 360 mm. except for one which was only 100 mm. in length. Blue crabs (Callinect'es sapidus) and mud shrimps (CaUianassa jamaicense lousianensis) made up about 90 percent of the food of these large catfishes, and Gunter (op. cit.) suggested that this fish can root soft-shelled crabs from hide-aways in the mud. Other identified material included penaeid shrimp (P. aztecus), oyster crabs (Panopeus sp.), hermit crabs, a nudibranch, a small mullet, and a small sea catfish (indicating cannibalism) . Fish bones and a clam shell were present, and although no specific mention was made of the presence of bottom detritus, indistinguishable material was noted in five stomachs. Knapp (1949) reported that of 468 hardhead catfish examined from the Texas coast, 87 percent had consumed shrimp, whereas fishes (including menhaden) appeared in about 14 percent, crabs in 6 percent, and squids and other invertebrates were present in less than one percent. Reid ( 1954) analyzed the food of five specimens from Florida. Four small individuals (91-116 mm.) contained many copepods as well as amphipods, mysids, polychaetes, and shrimp. One adult contained crabs, shrimp, and fishes. Reid ( l 955a) later reported on the exam­ination of 14 specimens from northern Texas, of which 11 contained food. Six small in­dividuals (82-91 mm.) had eaten "myriads" of copepods, some shrimp, and unidenti­fied fish. Five larger catfishes (144-191 mm.) had consumed one xanthid crab, one fish, and quantities of "pelecypod debris". Of the specimens examined from Lake Pontchartrain, smaller individuals had con­sumed quantities of small bottom-dwelling invertebrates (Table 2). These included amphipods (10 percent), mud crabs (16 percent), and chironomid larvae and pupae (13 percent), as well as smaller percentages of hydroids, clams, snails, harpacticoid cope· pods, schizopods, isopods, and water beetles. Bottom invertebrates were also prominent in the food of the larger sea catfishes, although these invertebrates tended to be some­what larger in size and included higher percentages of both mud crabs and blue crabs. Adults consumed fewer amphipods, but the percentages of most other invertebrate groups remained relatively unchanged. Fish remains consisting mainly of bones, scales, and lenses of eyes were present in the stomachs of catfishes of both size groups. Since no soft remains of fishes were encoun­tered, however, it is assumed that no live fishes had been eaten and that the hard parts were strained from the surface of the bottom mud. These remains were least prominent in the stomachs of the adults indicating a greater apparent degree of selectivity within this group. Undetermined organic material and detritus constituted over one-third of the stomach contents in both sizes of sea catfish. This material which appeared to have been derived from the surface layer of the bottom deposits was more prominent in the diets of the young than of the adults. The Lake Pontchartrain specimens indicate that within the size range investigated both young and adult sea catfish are bottom feeders which take in quantities of benthic invertebrates together with much bottom surface debris and detritus. Although little change was evident in the types of food consumed by the two size groups, the adults ap· peared to take in larger items, and especially more crabs. Increase in size appeared to be accompanied by somewhat greater selectivity involving a decrease in the consumption of both miscellaneous bottom materials and detritus. Food Habits of Fishes and Invertebrates of Lake Pontchartrain TABLE 2 Occurrence of food items in digestive tracts of 40 Galeichthys felis 90.0-169.0 mm. 170 .()-229.0 mm. 19 examined 21 examined 19 wilh food 17 with food Percenlage Percenlage Percenlage Percentage of tracts• of Lolal of tracts• of total con~aining stomach containing stomach FOOD ITEMS item Yolume-item ,·olume Copepoda-Harpacticoid 5.2 Mysid shrimp 10.5 1.9 4.8 1.3 Isopoda 42.l 1.8 23.8 2.1 Amphipoda 68.4 9.5 38.1 4.0 Palaemonidae 4.8 0.5 Crabs (undet.) 4.8 3.8 Rithropanopeus harrisii 47.4 15.7 61.9 27.8 Callinectes sapidus 9.5 23 Insecta I un det.l 5.2 0.1 Coleoptera 10.5 0.7 28.6 3.7 Diptera-larrne 89.5 10.6 61.9 10.3 Pupae, adults 26.3 2.0 14.3 0.6 Arachnida 4.8 0.4 Mollusca Rangia cuneata 15.8 1.3 4.8 Gastropoda 5.2 4.8 Hydroids 10.5 0.1 Vertebrata Fish remains 63.2 11.l 38.l 5.2 Vascular plants 10.5 Organic mat. (undet.I 78.9 31.6 76.2 27.3 Detritus, Sand 68.4 13.6 81.0 10.8 SUMMARY Mysid shrimp 1.9 1.3 Isopoda, Amphipoda 11.3 6.1 lnsecta 13.4 14.6 Crabs 15.7 33.9 Miscellaneous 12.5 6.1 Detritus, undet. 45.2 38.1 * Stomach and inlesline included. The total evidence available indicates that during its life history the sea catfish may actually pass through three feeding stages. Zooplankton, especially copepods, appears to be an important food for individuals less than 100 mm. in length. Above this size micro-bottom invertebrates assume importance, and these grade into larger crabs and fishes in "sea cats" above 200 mm. Much bottom detritus was noted in the Louisiana specimens, although this material has not been recorded as food for the species by workers in other areas. lcTALL"RIDAE lctalurus furcatus (Le Sueur). Blue Catfish The food of the blue catfish appears to have been largely neglected in the literature. Forbes (1888) found that a single specimen had eaten only fishes. Forbes and Richard­son (1920) mentioned that fragments of bark, insect remains, and miscellaneous or­ganic debris were also consumed and that blue catfish are caught on trot-lines baited with fishes and crayfish. Hildebrand and Towers (1927) reported a single small speci­men (195 mm.) had fed on crustaceans, insects, fishes, and vegetation. Gunter (1945) found that four large blue catfish (280-305 mm .. t from brackish waters of southern Texas had consumed grass shrimp (Palaemontes sp.), algae, and some indistinguishable material. In the Lake Pontchartrain specimens of the smallest size group (60--129 mm.), zoo· plankton, including schizopods and calanoid copepods, constituted the most important single food, making up almost one-half of the stomach Yolume (Table 3 and Fig. 3) . This material, however, was barely represented in the stomachs of the larger fish. l"ndoubtedly TABLE 3 Occurrence of food items in digesti,·e tracts of 78 lctalurus jurcatus 60.0-129.0 mm. 130.()-199.0 mm. 200.0-229.0 mm . 230.0-411.0 mm. 18 rxamined 18 t"xaminf'd 17 exam'. ned 25 e:xamintd 15 with food 16 "'·ith food 15 wilh food 23 with food Perct>nlage Percrnlage Pt>rcenlage PPrcentagt> Perctnlage Pt'rcenlage Percenlagt' Perct'nla~e ot lracH• of tola1 of tracts* of tolal of tracts* of total of tracts* o( total conlainin~ stomach containing stomach containing stomach containin{!. stomach FOOD ITEMS item ,·olume itt>m volume item volume item Yolumt' Ostracoda I l.l T I7.6 0.2 Copepoda Calanoid 5.6 I2.6 Harpacticoid I l.l 0.5 Mysid shrimp 27.8 36.0 I6.i' l.5 Il.8 0.4 24.0 0.2 Isopoda 22.2 4.2 II.I 23.5 0.4 I6.0 4.8 Amphipoda 50.0 4.4 55.6 I9.I 29.4 23.0 52.0 l.8 Palaemonetes sp. 5.9 8.0 2.6 Macrobrachium ohione 4.0 LO Penaeus setiferus 4.0 2.6 Callianassa sp. 5.9 5.I 4.0 l.3 Crabs Rithropanopeus harrisii 44.4 10.9 47.I 2.4 20.0 6.4 Callinectes sapidus 5.6 5.6 LO 5.9 0.8 24.0 3.0 Insecta Coleoptera I6.7 l.7 33.3 0.3 29.4 0.5 8.0 Diptera 27.8 0.5 27.8 l.3 52.9 0.8 40.0 0.6 Hemiptera 5.6 5.9 Homoptera 4.0 Hymenoptera 5.6 Il.8 4.0 Orthoptera 5.6 5.9 4.0 Arachnida 5.9 Annelida 5.6 5.6 l.3 I2.0 0.7 Mollusca Rangia cuneata 5.6 72 .2 25.3 76.5 36.0 52.0 9.9 Mytilopsis leucopheata 5.6 23.5 6.i' I6.0 2.I Gastropoda 5.6 35.3 0.5 4.0 Hydroids 5.9 O.I 4.0 Vertebrata Anchoa mitchilli I l.8 9.9 4.0 0.5 Citharichthys spilopterus 8.0 7.3 Menidia beryllina 4.0 0.4 Micropogon undulatus 5.9 O.i' I2.0 8.3 Svngnathus sp. 4.0 0.4 Fish remains I I.I 3.5 ll.8 2.3 24.0 I3.5 Algae--filamentous 5.6 3.3 32.0 25.7 Vascular plants 5.6 0.2 5.6 5.9 24.0 0.5 Eggs and cysts 5.6 Organic mat. (undet.) 88.9 26.3 66.7 25.5 76.5 IO.I 52.0 3.3 Detritus 33.3 10.6 27.8 6.2 23.5 0.3 32.0 I.I Sand and silt II.I 3.8 I6.7 0.4 I 7.6 O.I 40.0 2.3 SU'.\L\IARY Copepoda, Ostracoda. :\Iysids 48.6 2.0 0.6 0.2 Isopoda, Amphipoda 8.6 I9.I 23.4 6.6 '.\lacrocrustacea 0.0 I l.9 8.3 I6.9 Mollusks 0.0 25.3 43.2 I2.0 Fishes 0.0 3.5 I2.9 30.4 Vegetation 0.2 3.3 0.0 26.2 Misc. Im·ertebrates 2.2 2.9 1.4 l.3 Incidental and Undet. 40.7 32.I 10.5 6.7 • Stomach and inl11'~tine include-d. 70 60 ~ 0 50 LI.I ::;: :::. ...J 40 0 > J: (.) 30 c ... 40 x -.! ..u !.. ::i; ICTALURUS FURCATUS 0 20 .... "' O...t::=======:::=:::====:::::::===:ic===::::::==:::::::==11:=:::::==::::::==================::::llll_, 7 8 9 10 II 12 13 14 I~ 16 17 18 I> HOUR OF DAY Frc. 4. Hourly pattern of percent-fullness of stomach and intestine of lctalurus furcatus. feeding must occur throughout the day with, perhaps, a diminution in feeding activity in the late afternoon and early evening. The condition of the intestines, however, points to the fact that the main feeding must take place either at night or in the early morning hours. Although the food habits of the blue catfish are quite complex, three general feeding stages are recognizable. In smaller individuals a zooplankton-feeding stage is dependent chiefly upon calanoid copepods and schizopods. This stage appears to be largely com­pleted by a size of 100 mm. A bottom-feeding stage dependent upon small bottom­inhabiting invertebrates is prominent in the 100-240 mm. catfish. Finally, a stage de­pendent primarily upon macro-mobile animals such as fishes, crabs, and shrimp achieves significance in catfishes over 200 mm. in length. This general pattern of food stages is displayed in common with a number of other species inhabiting Lake Pontchar­train, and it appears that from the standpoint of nutrition the blue cat is highly success­ful in the estuarine environment. lctalurus punctatus (Rafinesque) . Channel Catfish Food of the channel catfish in fresh water has been the subject of investigation by many workers. In a recent study, Bailey and Harrison (1948) reviewed a number of the more pertinent references including the following: Forbes (1888), Smith (1907), McAtee and Weed (1915), Shira (1917), Mobley (1931), Ewers and Boesel (1936), Aitken (1936), Boesel (1938), McCormick (1940), Dill (1944), Menzel (1945), and Dendy (1946). As indicated by these writers up to 98 percent of the diet of the very young channel catfish (less than 100 mm.) consists of small aquatic insects. With in­crease in size the catfish have been found to consume a much lower percentage of in­sects, and within wide ranges of tolerance, larger individuals appear to subsist upon whatever food happens to be locally available in quantity. Evidence indicates that many types of food materials are probably consumed in direct proportion to their availability. Groups of channel catfish have been found to specialize at least temporarily upon algae, mollusks (especially gastropods) , microcrustaceans, erayfish, insects (aquatic and ter­ restrial), and fishes. Even young of the same species have been found to enter the diet of the adults with a frequency proportional to their relative abundance in the fish pop­ ulation (Bailey and Harrison, 1948). In the Chickahominy River of Virginia Menzel ( 1945) found that algae and blue crabs formed a large proportion of the diet of this species when available. Because of this wide range of food tolerance, the channel cat­ fish is generally considered to be omnivorous. Fourteen specimens of channel catfish were examined from Lake Pontchartrain of which 13 contained food. Stomach contents of the 11 smaller individuals (76-119 mm.) consisted primarily of small bottom-living arthropods (isopods, amphipods, xanthid crabs, and chironomid larvae and pupae, as well as occasional microcrustaceans). Unde­ termined organic material and detritus made up about one-fourth of the stomach con­ tents. Traces of foraminifera, filamentous algae, and vascular plants were present, and sand made up about eight percent of the material encountered. Two larger catfishes (207-312 mm.) had consumed quantities of crustaceans (isopods, xanthid crabs, and others) , as well as smaller amounts of fish, hydroids, and undetermined organic matter. In the brackish water environment of Lake Pontchartrain young channel catfish main­ tain themselves upon a diet composed primarily of small bottom crustaceans and insects together with bottom detritus. Food of the larger fish probably includes the same ma­terials with the addition of fishes and large crustaceans as in the case of the related blue catfish. A wide variety of miscellaneous items appears to be consumed by both the young and adults. M uGILIDAE Mugil cephalus Linnaeus. Striped Mullet Species of the genus Mugil are rather widely distributed throughout the tropical and subtropical shores of the world, and in some areas they assume a position of con­siderable importance in the economy of local human societies. Hence, it is not surprising that the food habits of the group in natural as well as artificial environments have re­c~ived the attention of many workers. Much of this literature has recently been sum­marized by Pillay (1953), Thomson (1954), and Ebeling (1957) . Most species of this genus are considered to feed upon plankton, filamentous algae, diatoms and other mi­nute vegetable matter, or upon organic detritus and such nutritive material as may be procured by filtration of the bottom mud, although some divergence has been noted. Morphological and behavioral adaptations of these mugilids for obtaining and process­ ing the nutriments are well known (see Ebeling, 1957, and the bibliography listed therein) . The adults of most species browse upon the surface of shallow water sediments, rind by means of a remarkable pharyngeal filtering device (supplemented by pharyngeal l:r•;te buds ) they sort out the coarser materials which are expelled through the mouth. Finer filtered material receives mucus in the esophagus, and the mass is pulverized by the grinding action of the gizzard-like pyloric portion of the stomach. In the very long intestine the nutritive matter is digested from the mineral matter, apparently in the absence of either proteolytic or lipolytic enzymes (Ishida, 1935). The food habits of Mugil cephalus have been discussed by several writers including Gunther (1880) , Linton (1904), Jordan (1905) , Smith (1907), Jacot (1920), Hilde­brand and Schroeder (1928), Ghazzawi (1933) (not seen), Hiatt (1947a-b) , and Reid Food Habits of Fishes and lnrertebrates of lake Pontchartrain (1955b). An excellent review of most of this literature was presented by Hiatt (1947a-b), therefore, only a summary will be undertaken here. It is generally agreed that the striped mullet feeds largely upon epiphytic algae, littoral diatoms, and finely divided organic detritus scraped from the surface layer of shallow mud flats or from the surface of roots and other objects present in such habitats. Remains of larger invertebrates and vascular plants sometimes appear among the ingested detritus, and the presence of planktonic crustacea and surface algae indicates that some plankton straining must also take place, especially among the younger individuals. In the Lake Pontchartrain study 57 specimens of striped mullet (97-327 mm.) were examined, only 54 of which contained food. A few preliminary analyses of the contents of the muscular pyloric region of the stomachs yielded virtually no identifiable material, and it became necessary to include the contents of the cardiac region as well. Even examination of the material previous to grinding, however, provided little definitely identifiable matter, and since no clear distinctions could be made between the food of the clifferent size groups, all specimens are treated below as a unit. Almost one-half of the material encountered was classified as detritus and unde­ termined organic matter which seemed to have some cellular and tissue structure, al­ though it was generally in an advanced state of decomposition. More than one-third of the substance appeared to be mud. silt and sand, and only about 14 percent could be definitely recognized. Eleven percent of the material was of vascular plant origin, but could not be categorized further. About two percent was algal in nature including both flat and filamentous algae and littoral diatoms (BUldulphia sp., Terpsinoe sp., and others). Miscellaneous fish scales, foraminifera, sponge spicules, and minute gastropods each constituted less than one percent of the contents. These results indicate that in Lake Pontchartrain, as elsewhere, the striped mullets are iliophagous and that their food consists chiefly of bottom surface material. Decom­posing organic matter. detritus, and mud were much too abundant to be considered incidental in the food of the species, and it was not possible to detect any significant in­dictations of selectivity for the minor items such as algae, foraminifera, minute gastro· pods, etc. "Cndoubtedly some selection in feeding site does take place based primarily upon taste and size and consistency of the particles. This study suggests that the vege­table matter itself, or the decomposing bacteria, or a combination of the two factors may have influenced the feeding site of the mullets examined. Field observations indicated that striped mullet do feed at the surface, and schools were occasionally observed to be actively feeding upon surface scums of Anabaena spp. which were seasonally abundant. This material was not encountered, however, during the course of stomach analyses. Judging from the degree of fullness of the alimentary tracts, striped mullets appear to feed throughout the day, and on the basis of what is known of other iliophagous species (Gneri and Angelescu, 1951 I, it is probable that they feed throughout most of the night as well. ATHERJ'.\"JD.-\E .Menidia berylli'.na (Cope). Silverside Hildebrand and Schroeder (19281 examined the stomachs of 20 silversides from Chesapeake Bay and encountered the following food items I listed in order of impor­tance) : small crustaceans, small mollusks, insects, worms, and a few strands of algae. Reid ( 1954) found that in the Cedar Key area of Florida the food of this species con­ sisted primarily of plankton organisms, especially copepods, as well as some algae and amphipods. Six specimens taken under night lights contained freshly-ingested insects. In the smallest size class of silversides examined from Lake Pontchartrain, calanoid copepods made up six percent of the food, although they were practically absent from individuals above 55 mm. (Table 4 and Fig. 5) . These copepods undoubtedly represent TABLE 4 Occurrence of food items in digestive tracts of 60 Menidia beryllina 40.0-54.0 mm. 55.0-64.0 mm. 65.0-79.0 mm. 21 examined 20 examined 19 examined 19 with food 20 with food 16 with food Percentage Percentage Percenlage Percentage Percentage Percentage of tract~* of total of tracts* of total of tracts* of total containing stomach containing stomach containing stomach FOOD ITEMS ilt>m Yolume il('m \·olumP. item volume Ostracoda 5.0 T 5.3 Copepoda Calanoid 9.5 5.5 5.0 0.4 ... . Mysid shrimp 4.5 3.0 15.0 10.5 10.5 2.5 Isopoda 38.1 42.3 30.0 12.4 10.5 0.3 Amphipoda 42.9 18.9 70.0 58.7 73.7 61.0 Insecta ( undet.) 9.5 0.5 20.0 2.8 21.1 0.5 Coleoptera 5.3 Diptera-larvae 4.8 1.5 10.0 0.4 Pupae, adult~ 9.5 3.0 15.0 4.4 21.1 15.4 Hymenoptera 14.3 5.3 10.0 4.5 5.3 9.9 Arachnida 5.3 0.1 Annelida 4.8 Hydroids 4.8 Vertebrata Fish remains 9.5 5.0 T 10.5 Algae-fil~mentous 4.8 15.0 0.1 10.5 0.2 Vascular plants 4.8 0.6 5.0 Eggs and cysts 5.0 Organic mat. (undet.) 47.6 17.0 70.0 5.0 73.7 9.5 Detritus 33.3 2.5 35.0 0.3 15.8 0.5 Sand 9.5 5.0 0.7 SUMMARY Copepoda 5.5 0.4 0.0 Mysid shrimp 3.0 10.5 2.5 Isopoda 42.3 12.4 0.3 Amphipoda 18.9 58.7 61.0 Insect pupae, adults 8.8 11.7 25.8 Misc. Invertebrates 2.1 0.5 0.3 Incidental, Undet. 19.5 6.0 10.0 * Stomach and intestine included. the remnants of an earlier important zooplankton-feeding stage in individuals less than 40mm. lsopods and amphipods together made up the bulk of the food, constituting from 61 percent to 71 percent of the stomach contents in all size classes. Stomach analysis rec­ords indicate isopods were abundant in the 40-54 mm. size class and declined there­after to less than one percent in the largest fi sh. Although leptochelia sp. and a number of other tanaids and anthurids were definitely recognized, it appears now that at least some of these crustaceans identified as isopods may in reality have been dorso-ventrally flattened amphipods. Whatever the taxonomic position of these forms may be, they should eventually provide important evidence regarding the feeding site of the young silversides. Other species of amphipods displayed a progressive increase from 19 percent -.. . .... ~ 100 0 - w 80 ~ ::::> ...J 0 60 > I u 40 ....... I&. ...... ....... 60 f-­ ....... ~ • ....... !.. .... ve­ ....... ­' ....... 0 J: 40 ._ '--.. ­ ................. ct ~ .... ­ 0 I-20 >-­ (/) .... - - ...I 80 - - ...I :::> .... - I&. 0 t w z 60 ­-40 ­ AA_ ----.............................. --...... ...... __ • --- I­ ....... (/) w I­~ 20 ­ - .... - I I . I I I 9 10 II 12 13 14 15 16 HOUR OF DAY Fie. 6. Hourly pattern of percent-fullness of stomach and intestine of Menidia beryllina (young and adult J. SERRANIDAE M orone interrupta Gill. Yellow Bass Studies on the food of the yellow bass from fresh water by Forbes (1878, 1880a, b), Hildebrand and Towers (1927), and McCormick (1940) indicate that the species con­sumes · small aquatic animals including entomostraca, aquatic insect larvae, and small fishes, as well as occasional bits of vegetation and some terrestrial insects. The food of the species in brackish-water environments apparently has been studied only by Lam­bou (1952) who examined the stomach contents of 273 large specimens (ca. 127-292 mm.) taken from the marshes bordering the northeastern shore of Lake Pontchartrain. Only 167 of these fish contained food. Fish, insects, vegetable matter, and unrecogniz­able organic material were each encountered in less than five percent of the stomachs. Crabs and penaeid shrimp, each appeared in about 30 percent of the stomachs, and palaemonid shrimp in about 10 percent, indicating a high percentage of crustaceans consumed in the brackish-water marshes. In the present study crustaceans made up over half of the food material and included 24 percent shrimp (schizopods, juvenal penaeids, Pal,aemonetes sp., and Macrobrachium ohione) and 28 percent crabs (Callinectes sapidus and Rithropanopeus harrisii) (Table 5). Fishes constituted 35 percent of the stomach contents and included Cynos- TABLE 5 Occurrence of food items in digestive tracts of 27 Marone interrupta 130.Cf-195.0 mm. 27 examined 18 with food Percentage Percentage of tracts* of total FOOD ITEMS containing item stomach ,-olume Copepoda (Arguloid) 3.7 0.1 Mysid shrimp 18.5 18.2 lsopoda 7.4 0.3 Amphipoda 22.2 2.1 Palaemonid shrimp (undet.) 7.4 0.1 Palaemonetes sp. 3.7 4.8 Macrobrachium ohione 3.7 1.1 Crabs Rithropanopeus harrisii 22.2 18.0 Callinectes sapidus 18.5 9.7 lnsecta ( undet.) 3.7 Diptera 7.4 Odonata 3.7 0.1 Annelida 3.7 0.3 Hydroids 3.7 T Sponge 7.4 Vertebrata Cynoscion sp. Cyprinodon variegatus Gobiosoma bosci 3.7 3.7 3.7 7.7 0.5 4.8 Micropogon undulatus 3.7 1.1 M ollienesia latipinna 3.7 4.3 Fish remains 29.6 16.5 Algae-filamentous Organic mat. (undet.) 3.7 63.0 C.l 6.8 Detritus 18.5 3.5 SUMMARY Microcrustacea 20.7 Macrocrustacea 33.7 Fishes 34.9 Miscellaneous, Undet. 10.8 *Stomach and intestine included. cion sp., Cyprinodon variegatus, Gobiosoma bosci, Micropogon undulatus, Mollienesia latipinna, and other small forms. Undetermined organic material and detritus consti­tuted 11 percent of the food volume, and incidental items included filamentous algae, sponge, hydroid, annelids, arguloid copepods, isopods, amphipods, and odonate and dipteran insects. The above studies point to the fact that the yellow bass is a predatory species which in the brackish-water environment takes in a variety of shrimps, crabs, and fishes. The presence of bits of filamentous algae, sponge, and hydroid in the Lake Pontchartrain specimens indicates feeding near the bottom, either around the shallow margins of the lake or near the mouths of freshwater passes. CENTRARCHIDAE Micropterus s. salmoides (Lacepede). Northern Largemouth Bass The food of the largemouth bass in freshwater environments has been extensively investigated, and since the literature has recently been reviewed in some detail by Mc­ Lane (1948) it need not be reiterated here. In general, these works have shown that with increasing size there occurs a progressive change in the food of this species from microcrustaceans to insects to fishes, although crayfish and other arthropods may also form important portions of the adult diet. McLane, himself (op. cit.). carried out a study of the seasonal food habits of the largemouth bass in St. Johns River of Florida where a number of estuarine species were available for consumption. He found that very young bass consumed large amounts of cladocera and other entomostraca. In slightly larger fish these items were replaced by schizopods. Bass belonging to larger size groups fed chiefly upon macrocrustacea (Palaemonetes paludosa, Procambarus fallax, and Rithropanopeus harrisii'i and at least twenty-five species of fishes, many of which are also available in Lake Pontchar· train. Lambou (1952) examined the stomachs of 93 largemouth bass (ca. 203-432 mm.) taken from the marshes bordering the northeastern shore of Lake Pontchartrain. Fifty· three of these bass contained food. Crabs were present in 56 percent of the stomachs, "shrimp" (presumably Penae us spp.) appeared in 25 percent, "freshwater shrimp" (pre­sumably Macrobrachium ohione and Palaemonetes sp.) and insects appeared in 7 per· cent each. and vegetable matter and undetermined material in 2 percent each. Thus, crabs and shrimp made up the bulk of the diet of the largemouth bass in the marshes, whereas fish and other materials were of much less importance. Three specimens of the largemouth bass (175-209 mm.) from Lake Pontchartrain were examined in the present study, only two of which contained food. This material in­cluded a single grass shrimp (Palaemonetes sp.) and five small blue crabs (Callinectes sapidus) which together made up 97 percent of the stomach contents. Additional ma· terial included vegetation (V allisneria spiralis and the filamentous green alga, Clado· phora sp.), as well as small amounts of undetermined organic material and detritus. These studies together with those of McLane (op. cit.) and Lambou (op. cit.) demon· strate that when available a number of species of estuarine invertebrates and fishes are readily utilized by the largemouth bass. CARANGIDAE Caranx hippos (Linnaeus). Common Jackfish Food habits of the common jackfish have been reported by Linton (1904), Hildebrand and Schroeder (1928), Knapp (1949) , and Reid (1954). These authors have found this fish to be highly predatory consuming large quantities of fishes and crabs, as well as smaller percentages of squids, shrimp, and smaller invertebrates. Only a single young specimen of the common jack (79 mm.) from Lake Pontchar· train was examined, and this contained one anchovy (Anchoa mitchilli diaphana) 30 mm. in length which constituted about 98 percent of the stomach contents. A small amount of undetermined matter also was present. Although only a few jacks were cap­tured and only a single specimen was examined for food, many jacks were observed in the open waters of Lake Pontchartrain. These were generally presumed to be actively feeding at the surface upon small fishes, probably clupeids and engraulids. Sc1AENIDAE Aplodinotus grunniens Rafinesque. Freshwater Drum Studies on the food of the freshwater drum by Forbes (1878, 1880a, b) , Forbes and Richardson (1920), Hildebrand and Towers (1927), Rimsky-Korsakoff (1930) , Ewers (1934), and Dendy (1946) have indicated that in fresh waters this species normally passes through a series of ontogenetic food stages. Thus, the smallest individuals feed upon entomostracans. These are followed by aquatic insects, and the large drums feed chiefly upon clams and snails, supplemented by crayfish and other material. Dendy (1946) has pointed out that if insects and mollusks are essentially unavailable the drum may get along by abbreviating the insect-feeding stage and by replacing the mollusk­ feeding stage with one dependent upon small fishes. In the present study five specimens of the freshwater drum (211-347 mm.) from Lake Pontchartrain were examined, of which only four contained food. This material included 73 percent clams (Congeria ieucopheata and Rangia cuneata), 11 percent mud crabs (Rithrnpanopeus harrisii), 10 percent undetermined organic material, and 6 percent amphipods. Additional items present in trace amounts included remains of blue crabs (Callinectes sapidus) , gastropods, hydroids, and leaves and twigs of vascular plants. The food of this species appears to be quite similar to that of the young black drum. Bairdiella chrysura l Lacepede). Silver Perch Food habits of the silver perch have been studied by Linton (1904), Welsh and Breder ( 1923) , Hildebrand and Schroeder (1928) , Hildebrand and Cable ( 1930) , Reid (1954), and Reid, Inglis, and Hoese (1956). These workers have shown that the chief food of the smallest silver perch is copepods supplemented by smaller amounts of ostra­cods, cladocera, schizopods, amphipods, and chaetopods. With increase in size there is greater emphasis upon annelids and larger crustaceans (schizopods, amphipods, iso­pods, small shrimp, and crabs) with occasional mollusks. Largest individuals have been reported to consume a few anchovies and other fishes, as well. These considerations led Hilde}:,rand and Cable (op. cit.) to conclude that the silver perch feeds largely on the bottom and is strictly carnivorous. Food Habits of Fishes and Invertebrates of Lake Pontchartrain The food of the silver perch from Lake Pontchartrain was made up of four primary types, each constituting roughly a quarter of the total stomach volume. These included schizopods, larger shrimp, fishes, and miscellaneous material (including smaller in· vertebrates and detritus) (Table 6). Stomachs of the silver perch frequently were packed TABLE 6 Occurrence of food items in digestive tracts of 41 Bairdiella chrysura 70.0-143.0 mm , 41 examined 20 with food Percenlage Percentage o( tract s• of total FOOD ITEMS containing item stomach ,-olume Copepoda 4.8 Mysid shrimp 14.6 24.3 Isopoda 7.3 8.3 Amphipoda 2.4 0.8 Palaemonid shrimp 7.3 19.8 Penaeid shrimp 12.2 6.1 Crabs Rithropanopeus harrisii 7.3 1.0 Callinectes sapidus 2.4 2.4 Vertebrata Anchoa mitchilli 7.3 12.1 Fish remains 12.2 12.3 Vascular plants 2.4 0.2 Organic mat. (undet.) 53.7 12.5 Sand 9.8 SUMMARY Mysid shrimp 24.3 Palaemonid, Penaeid shrimp 25.9 lsopoda, Amphipoda 9.1 Crabs 3.4 Fishes 24.4 Incidental, Undet. 12.7 * Stomach and intestine included. with schizopods, and one individual contained more than 100 of these small crustaceans. Larger shrimp included grass shrimp (chiefly Palaemonetes pugio) and considerable numbers of young penaeids. Fish food included mainly Anchoa mitchilli diaphana as well as unidentified remains, and one large silver perch had swallowed a 40 mm. an· chovy. The miscellaneous material encountered in the stomachs included copepods, iso· pods (Aegathoa sp.), and crabs ( Callinectes sapidus and Rithropanopeus harrisii), as well as a small amount of vascular plant material and sand. Undetermined organic ma· terial made up 13 percent of the contents. On the basis of present knowledge of the food of the silver perch, two general feeding stages may be recognized, an early copepod-feeding stage which is completed before a length of 50 mm. and a later schizopod-palaemonid-penaeid shrimp stage. In this second stage a great variety of other invertebrates as well as occasional fishes are consumed. Further work will undoubtedly result in the recognition of subdivisions of the shrimp­feeding stage of this voracious fish. In Lake Pontchartrain the food of the silver perch overlaps that of many other species and appears especially similar to that of young speckled trout (Cynoscion nebulosus) of comparable size. Cynoscion arenarius Ginsburg. Sand Trout, Sand Squeteague The food of the sand trout in Florida and Texas has been investigated by Reid (1954, 1955b, 1956) and by Reid, Inglis, and Hoese (1956). These workers have indicated that trou't less than 60 mm. subsist largely upon small crustaceans (copepods, larval and metamorphosing shrimp and crabs, and others) . Shrimp, although consumed to some extent by trout of all sizes, appeared to be an important element in the diet of the inter­mediate-sized trout. Fishes were found to be consumed by sand trout as small as 34 mm., and they were present in the stomachs of at least four-fifths of all trout over 80 mm. in length. Among the fishes consumed, the menhaden (Brevoortia patronus) was the most conspicuous, although other clupeids, anchovies, and young sciaenids were also noted. The presence of young sand trout indicated cannibalism. Miscellaneous items included crustacean and molluscan debris, as well as undetermined organic matter. In the smallest sand trout (40-99 mm.) from Lake Pontchartrain schizopods con­ stituted almost one-third of the food volume, and traces of other small invertebrates in­cluded amphipods, palaemonid shrimp, and bits of crabs (Table 7 and Fig. 7). These small invertebrates were almost entirely absent from the food of the larger size groups. Fish remains constituted over one-half of the food of the smallest sand trout examined, and this material increased to over 90 percent of the stomach contents in the larger trout (150-225 mm.). Fishes consumed included Anchoa mitchilli diaphana and the remains of other unidentified species, probably young clupeids. Penaeid shrimp made up 8 per- TABLE 7 Occurrence of food items in digestive tracts of 64 Cynoscion arenarius 40.0-99.0 mm. 100.0-149.0 mm. 150 .0-225.0 mm. 22 examined 29 examined 13 examined 18 with food 21 with food 8 with food Percenlai;::e Percentage PercenlaJ?;e Percentage Percentage Percenlage of tracts* of lotal of tracts* of total of lracls* of lolal containing slomach containing stomach containing stomach FOOD ITEMS item ,·olume item ,·olume item ,·olume Mysid shrimp 45.5 31.9 3.4 1.0 Amphipoda 9.1 0.2 Palaemonetes sp. 4.5 2.7 Penaeus sp. 7.7 7.7 Crabs ( undet.) 4.5 2.7 ··-· Annelida 3.4 T Mollusca Rangia cuneata 4.5 Gastropoda 3.4 0.4 Hydroids 3.4 Vertebrata Anchoa mitchilli 4.5 10.2 37.9 54.1 30.8 56.7 Fish remains 45.5 44.1 37.9 28.9 53.8 34.1 Algae-filamentous 3.4 0.1 Organic mat. (undet.) 68.2 7.7 72.4 15.3 23.1 1.5 Detritus 40.9 0.4 6.9 15.4 Sand 3.4 0.1 SUMMARY Microcrustacea 32.1 1.0 0.0 Macrocrustacea 5.4 0.0 7.7 Fishes 54.3 83.0 90.8 Misc. and Undet. 8.1 15.9 1.5 *Stomach and intesline included. 100 , ..~·­CYNOSCION ARENARIUS 90 _.. -··-··-··-··-·· _.. ~ _.. _....cr··-·· 80 l1J ::!: 70 :J ...J 0 60 0­ > ::<: ~l () 50 :i:: :.> ct :;: 40 ::i n 30 20 BODY LENGTH (mm.) F1G. 8. Ontogenetic food progression of Cynoscion nebulosus. 90 percent of the food volume in the adults. Both palaemonid and penaeid shrimp were taken, and these shrimp together, made up one-sixth of the food of the 100-200 mm. individuals. The fish food of the adult trout included Anchoa mitchilli diaphana (30-60 percent), small Micropogon undulatus (7-14 percent), and large quantities of unidenti­fied fish remains (definitely clupeiform and probably Brevoortia patronus) . Small amounts of vascular plant remains were uniformly present in the stomachs of the adults. Detritus and undetermined organic material made up only a small percentage of the stomach contents, and miscellaneous items included a]gae and seeds, as well as hydroids, crabs, and insects. Although very small speckled trout (less than 40 mm.) were not included in the pres­ent study, the investigations of Moody (op. cit.) demonstrated that very young indi­viduals subsist almost entirely upon copepods. In the Lake Pontchartrain specimens a schizopod-amphipod feeding stage was indicated in the 40-100 mm. size class and, per­haps, traces of a palaemonid-penaeid shrimp stage in the 100-200 mm. group. The pre­dominant food of the subadult and adult speckled trout, however, was fishes. These entered the diet early, and they comprised one-half of the food the smaller trout (40­99 mm.) and three-quarters of the food of the adults. Judging from the degree of fullness of the stomachs (Fig. 9) , the speckled trout feeds heavily in the early to mid-morning hours and takes in little, if anything, during the afternoon. In this respect it apparently differs from the sand trout which feeds earlier in the morning and which appears to resume feeding in the mid-afternoon. A number of authors including Pearson (1929), Moody (1950), Pew (1954), and Reid (1956) have discussed the habitat relations of the speckled trout in Florida and 100 80 ::; ..J ::> IL 60 ~ !.. :c 40 0 <( :E 0 .... ---.j----.............. 20 ........__ "' ---.. 80 ..J ..J ::> IL 't 60 40 "'~ .... "' .... "' ~ 20 8 9 10 II 12 13 14 15 16 17 18 HOUR OF DAY FIG. 9. Comparison of hourly patterns of percent-fullness of stomach and intestine of Cynoscion arenarius and C. nebulosus. Food Habits of Fishes and Invertebrates of lake Pontchartrain Texas, and they have generally agreed that the speckled trout is an inhabitant of clear waters with heavy grass bottom. These authors have pointed out that young trout seem to be critically associated with quiet shallow lagoons and coves possessing grassy flats where they consume great numbers of caridean shrimp. Adults, on the other hand, have been noted to remain in deeper water although frequently coming into the shallow grassy areas to feed upon shrimp and small fishes, especially the pinfish. On a priori grounds, therefore, it might be assumed that the speckled trout would not regularly inhabit Lake Pontchartrain which is characterized by high turbidity and sparse grassy flats. In view of their presence in the lake, some deviation from the ac­cepted patterns of food and habitat utilization by resident individuals is not at all sur­prising. In the virtual absence of weed beds caridean shrimp were essentially unavailable and were insignificant in the diet of the young trout. Likewise, as mentioned above, very few penaeid shrimp were taken. This may have been due in part to the relative absence from Lake Pontchartrain of the pink shrimp ( P. duorarum) (see Darnell and Williams, 1956), which is the staple penaeid for the Florida population, or to greater difficulty in locating and apprehending the mud-loving brown and white shrimp (P. aztecus and P. setiferus) which are abundant in the lake. Finally, in the virtual absence of weed beds, the adult trout of Lake Pontchartrain were not dependent upon the pinfish as in the case of the Florida trout. Instead, it appeared that the young Louisiana speckled trout partially substituted schizopods and bottom-living amphipods for the caridean shrimp and began to exploit at an early age the enormous populations of anchovies and larval fishes which inhabit the lake. In addition, the Louisiana population essentially bypassed the penaeid shrimp stage, achieving the adult diet by a length of 100 mm. rather than 150 mm. as in the Florida population. Actually, the food of the young and adult trout indicated that in Lake Pontchartrain, neither is critically associated with shallow grassy areas. The presence of such forms as Palaemonetes sp. and the tanaid (leptochelia rapax) did indicate some association of the young with grassy shallows, but the presence of quantities of schizopods and bottom­living amphipods is taken as evidence of feeding outside of the shallow flats. Stomachs of adults also contained some Palaemonetes sp. as well as bits of grass. The anchovies, how· ever, which constituted the bulk of the adult food, are ubiquitous in the lake, and neither these nor the young croakers and penaeid shrimp give much indication of feeding habi­tat. Positive evidence that adult speckled trout feed in the deeper areas stems from the centers of successful sport-fishing activity, aimed primarily at obtaining this species. Using shrimp or small croakers as bait, fishermen regularly obtain this trout by trolling in deeper waters and by still-fishing around the wooden pilings of the railroad bridge which crosses the eastern neck of the lake. Gunter ( 1945) pointed out a relationship between the availability of shrimp and the utilization of these crustaceans as food by the speckled trout. He found shrimp to be the predominant food in Texas during the summer months and noted a shift from shrimp to fishes (Mugil sp.) when shrimp suddenly became scarce following a January freeze. In line with Gunter's observations, Moody ( 1950) demonstrated that in Florida during the spring and summer months fishes and crustaceans contributed about equally to the diet of the adult speckled trout, whereas during the winter months fishes were about three times as important as crustaceans. This did not occur, however, in the Lake Ponchartrain series. Although all speckled trout over 150 mm. which were analyzed for food were taken during the warm months (March-September), fishes greatly predominated in the diets. This again seems to point to the great availability of small fishes as opposed to penaeid shrimp in the muddy Louisiana waters.* leiostomus xanthurus Lacepede. Spot The food of the spot has been reported by Linton (1904), Smith (1907), Welsh and Breder (1923), Hildebrand and Schroeder (1928), Hildebrand and Cable (1930) , Gunter (1945), Roelofs (1954), and Reid (1954, 1955b). These investigators have analyzed spots of different sizes from various parts of their geographical range and have accumulated a long list of food items. In general, the smallest spots have been found to feed upon planktonic or benthonic microcrustaceans including chiefly copepods, ostra­cods, and amphipods. Adults have been shown to be bottom feeders, specializing upon small mollusks, polychaetous annelids, and small crustaceans. Occasional items en­countered in the stomachs of the spots have included foraminifera, nematodes, chirono­mid larvae, mysids, shrimp, mites, small crabs, sea urchin spines, bryozoa, small fishes and fish scales, diatoms and other algae, vegetable debris, "grass", seaweed, and sand. Quantities of detritus and undetermined organic matter have consistently been encoun­tered in the stomachs of all sizes, but have been especially prominent in the food of the adults. Among the spots examined from Lake Pontchartrain, the smallest individuals con­ tained three percent rotifers and schizopods, and this material may represent the vestiges of an earlier plankton-straining stage. These forms were virtually absent in the larger­ sized fish (Table 9 and Fig. 10). 50 LEIOSTOMUS XANTHURUS >I! 40 !.. II.I ::E ::> ..J 0 30 > :I: 0 ~ ::E 0 20 t- II) 10 ZOOPLANkT•Rs T~TIOM ..-··-··-0-·-··­ 0 .-.. -·-·-.. L...r.~~~~~~~-===~~~·::::V:.~:2:J-=·~·-~.-· ··:-=··-:;;;:;;;;;;;;;oA;;;;;;;__·--·~·-:··~-:·-~::;;;,1;;L.~....1..J 50 100 125 150 200 BODY LENGTH (mm.) Frc. 10. Ontogenetic food progression of Leiostomus xanthurus. *It appears to be a habit of the speckled trout to regurgitate the stomach contents at intervals following a meal. On such occasions a putrid mass of partially-digested fish appears on the surface surrounded by a small round oil slick. Commercial fishermen familiar with this habit frequently locate their quarry by the appearance and characteristic "fishy" odor of these oil spots. The low percentages of stomachs with food given in the literature undoubtedly reflect, in part, this phenomenon. TABLE 9 Occurrence of food items in digestiYe tracts of 88 Leiostomus xanthurus 40.0-99.0 mm. 100.0-149.0 mm. 150.0-203.0 mm. 22 examined 38 exam:ned 28 examined 18 with food 28 wilh food 20 with food Percentage Percenlae:e Percenlage Percentage Percenlage Percentage oi tracts* of totaf of tracts* of total of tracts* of lotal containing stomach containin@ stomach conta~ning stomachFOOD ITEMS item ,·olume item ,-olume item volume Rotifera 9.1 2.0 Ostracoda 40.9 13.2 15.8 0.6 3.6 0.3 Copepoda (undet.) 13.6 6.2 2.6 3.6 T Harpacticoid 63.6 11.3 13.2 0.7 Mysid shrimp 4.5 1.0 7.9 0.3 lsopoda 31.8 7.6 44.7 9.0 35.7 11.8 Amphipoda 31.8 7.8 42.1 10.3 32.1 7.0 Cirripedia 3.6 lnsecta Coleoptera 13.2 0.5 Diptera 22.7 1.3 44.7 1.3 50.0 9.0 Arachnida 4.5 1.0 2.6 Annelida 4.5 2.0 Mollusca Rangia cuneata 68.2 13.5 60.5 23.7 46.4 29.7 Mytilopsis leucopheata 2.6 Gastropoda 50.0 4.0 31.6 5.5 30.0 1.9 Hydroids 4.5 2.6 0.1 Foraminifera 27.3 0.5 7.9 0.1 Vertebrata Fish remains 15.8 8.4 3.6 Algae-filamentous 4.5 5.3 14.3 0.3 Vascular Plants 4.5 7.9 4.5 14.3 0.6 Organic mat. ( undet.) 63.6 26.5 52.6 20.6 67.8 19.0 Detritus 68.2 0.5 47.4 6.7 35.7 19.2 Sand 31.8 1.5 39.5 8.4 10.7 0.6 SUMMARY Rotifera, Copepoda, Ostracoda, Mysid shrimp 33.i 1.6 0.3 Gastropoda, Foraminifera 4.5 5.6 1.9 Isopoda, Amphipoda 15.4 19.3 18.8 Rangia cuneata 13.5 23.7 '29.7 Misc. Invertebrates, Vertebrates 4.3 10.3 9.0 Vegetation 0.0 4.5 0.9 Incidental, Undet. 28.5 35.7 38.8 * Slomach and inlesline included . A number of small invertebrates grouped together as micro-bottom surface animals, made up over one-half of the food volume of the smallest spots. These species declined somewhat in importance in the larger fish, and constituted only one-fifth of the food of the largest spots. Included among the micro-bottom surface animals were ostracods (in heavy concentrations), copepods (primarily harpacticoids) , isopods, amphipods, mi­nute gastropods, and foraminifera. All these species presumably could have been either scooped from the bottom surface or strained from the immediately overlying layer of water. Another group of small invertebrates consumed in quantity by the spot included the bottom-burrowers, animals which could be obtained only by digging into the bottom surface material. These forms made up only 17 percent of the food of the smallest size group, but they increased in importance in the food of the larger spots and constituted 39 percent of the material consumed by the largest individuals. Included among the bottom-burrowers were large numbers of the clam (Rangia cuneata) and smaller quanti· ties of chironomid larvae and annelids. Undetermined organic material and detritus made up a considerable proportion of the food of spots of all sizes, and this material increased from 27 percent of the food volume in the smallest fish to 38 percent in the largest. The high percentage of this material in the diet of the spot is associated with the bottom habitat of the fish, and the increase in percent consumption of this matter by larger individuals parallels the increase in con­sumption of bottom-burrowers. The stomach contents indicated that individuals occasionally specialize temporarily upon one food or another. A single stomach contained 250 harpacticoid copepods and five chironomids, another contained 200 amphipods, still another adult spot had 76 unbroken Rangia cuneata in its stomach. Most fishes, however, and especially those in the small and intermediate size categories, contained a variety of foods indicating more of a straining or sifting process than selective capture of individual animals. Filamentous green algae including Cladophora sp., Rhizoclonium sp. and others, were encountered in one-tenth of the stomachs. Some vascular plant matter (probably Val­lisneria spiralis) was found in the stomachs of eight individuals, one of which was packed with this material. The presence of this vegetation in the diet of the spot suggests some shallow water feeding. Most of the amphipods encountered were gammarids, but species of Corophium which were so abundant in the grassy shallows appeared to be entirely absent from the stomachs. Many of the invertebrate species which were abundant in the food of the spot appear to be inhabitants of the deeper waters indicating that this fish, in its feeding, utilizes such areas to a large extent. The presence of large quantities of detritus and the virtual absence of sand suggests that the species feeds largely on mud rather than on sandy bottoms. At least 6 species of minute gastropods were encountered, and the distributional patterns of these may eventually be employed to pinpoint the feed­ing ground0 of the spot. The spot appears to consume a limited amount of material throughout the morning and early afternoon. An increase in feeding intensity apparent in the late afternoon points to the fact that this fish may feed primarily at twilight or during the night hours (Fig. 11). Hildebrand and Cable (op. cit.) pointed out that the very young spots tend to swim in schools and that a number of changes take place when the fish reach a length of ap­ .... z LEIOSTOMUS XANTHURUS .,.... ...................... ...._ ~ 80 ..--­ ................. ..-..-..-........ ~........ "' ... ......-...... "' 1~-............ ~::; 60 ...J Q ::> z II. "' ~ 40 :c • " u~ ::E ~ 20 "' o._1::::::=================-c:::::=======*==================:::111• 8 9 10 II 12 13 14 15 16 17 18 HOUR OF DAY Fie. 11. Hourly pattern of percent-fullness of stomach and intestine of Leiostomus xanthums. pr.1ximately 25 mm. At about this size the schools break up, the oblique terminal mouth becomes inferior, and the food emphasis shifts from microcrustaceans to bottom ma. terial. Roelofs (1954 J observed that in laboratory aquaria spots obtained food by mak· ing shallow dives to the bottom and scooping up mouthfuls of bottom material. Ex­amining the gill structure of the spot he found that the alternating gill rakers and numer­ous fine setae comprise an excellent straining device for filtering small food particles from the water. The accumulated evidence suggests that, as a rule, young spots feed just above the bottom. With increasing size they begin to feed more upon bottom surface animals, and as they approach maturity they dig more deeply into the bottom, taking a greater quan­tity of burrowing forms. Bottom detritus has been shown to be consumed in quantity by spots in North Carolina and Texas. In Lake Pontchartrain this material constitutes a substantial portion of the diet of spots of all stages, and it is particularly pronounced in the food of the larger, bottom-feeding stages. Temporary specialization upon one or an­other type of food has been reported, but most stomachs contain a wide variety of food items. Micropogon undulatus (Linnaeus). Atlantic Croaker The food of the Atlantic croaker on the south Atlantic and Gulf coasts has been in­vestigated by Linton (1904) , Smith (1907), Welsh and Breder (1923) , Hildebrand and Schroeder ( 1928) , Pearson ( 1929), Hildebrand and Cable ( 1930) , Gunter (1945) , Mc­Lane 11948) , Roelofs (1954), Reid !l955b J, and Reid, Inglis, and Hoese !1956). These workers succeeded in compiling an extensive list of food items recognized in the stom­achs of croakers, which may be summarized as follows: mollusks (including pelecypods and gastropods), annelids, small crustaceans (copepods, ostracods, barnacle larvae, and amphipods), schizopods, mud shrimp, penaeid shrimp, crabs, bryozoans, sea urchin remains, hemichordates, tunicates, fishes (menhaden, threadfin shad, anchovies, small croakers, and gobies ., , unidentified fish vertebrae, diatoms, and incidental sand. It wa~ observed that small croakers feed largely upon planktonic crustaceans and small bottom invertebrates, whereas the adults consume larger invertebrates such as annelids, clams, shrimp, and crabs as well as small fishes. Several of the workers reported the presence of large quantities of indeterminable matter and organic debris, especially among the young and medium-sized croakers. However, due to the general complexity of the food habits of this species and to geographic and seasonal differences in availability of the various acceptable items, the food relations of the Atlantic croaker have not been clearly understood to date. In view of the obvious ecological success of the croaker in Lake Pontchartrain as evidenced by its large populations, widespread distribution, and year-around habita· tion of the lake (see Suttkus, 1954), and because of its importance as a food and game fish in the area, it was given special attention. Zooplankton, including calanoid copepods and a few schizopods, made up almost 70 percent of the food of the smallest size class (Table 10 and Fig. 12 l. This material decreased by about one-half in each succeeding size group, and in croakers of about 100 mm. zooplankton formed an insignificant por­tion of the diet. Clearly the food of the young croaker is the calanoid copepod (espe­cially Arcartia tonsa) , and whereas most stomachs contained less than 40 copepods, three small individuals contained more than 125 copepods apiece. Small bottom-inhabiting animals comprised the chief food of the 25-50 mm. group. /1~\ MICROPOGON UNDULATUS 80 ...~/ \ ti~/ \ . JI ''A__ 50 ~ '~ .·n ,.../ --......., r ~/ 4 :' ~ 40 o· \ :'P ,J \ :i ..J .rl . , I/ 0 ~/· n \..1, > . .,, ':-(I I / \ // x· J: 30 u • i / \ 1 /., c /l' ! \1 ! . ~ !!.--·-·-·-·-L. i '\. ; 1 ..' \ i;; /. -·-" ,~.z / \ :' \ ...,o? '\. '.d ' I 0 \ 20 ·-6 o"' '\. !.· \ I \ 0'~0 J>.---. ..; 'cs . 10 ~.,~~+.. "-~~/' ----~ ,. . I h ~ ,..o~ "'~;,-' '-''-. I . a_ ri' ~:.;" .~ ,..J ~--......._ '\o~,," 1Aoe11.t ./ '-u ~ ...._...... -­ eo~"' IA~c"~.;~1.s _.. -.0 T ........_ .. -:·:·-~ 25 50 75 100 125 150 200 300 o1~0~~~....Ql~~:::_~~--~~_i;;;,~..~~~~ .. :_~~......~~..___:~~~~~:;;,;;;;:a~.._~_J BODY LENGTH Imm.) F1G. 12. Ontogenetic food progression of Micropogon undulatus. These forms included both bottom-surface animals (harpacticoid copepods, amphipods, isopods, ostracods, minute snails, foraminiferans, and occasional coleopterans), as well as certain forms which have a greater tendency to burrow ( chironomid larvae and small clams with attached mussels). Some of these bottom animals were present to a very limited extent in the smallest croakers, but they reached a peak of abundance (about 43 percent) in the 25-50 mm. size class and were present in reduced volume throughout all of the larger size groups. Occasionally the young, bottom-feeding croakers would specialize upon one form of food or another and contain up to 358 chironomids, or 10 mysids, or 46 amphipods. Most individuals in the intermediate size groups, however,. exhibited a variety of food items, each present in small quantity as though general strain­ing rather than selectivity were involved. Detritus and undetermined organic material made up an important food category in croakers of all sizes, and this material was especially significant in croakers 50 to 200 mm. in length where it constituted over 40 percent of the stomach contents of each size group. In the 75-100 mm. group this material comprised over two-thirds of the food volume, and a decline in the utilization of this substance by the larger croakers was evident. No marked difference was noted between the detritus encountered in the stomachs of the croakers and that encountered in the stomachs of the spots. In both cases the material appeared to contain bits of shredded tissue, was sometimes Aocculent in consistency, and was often greenish to dark brown in color. Furthermore, although much of the identifiable detritic material appeared to be of vegetable nature, it was generally impossible to distinguish between decomposed matter of animal and plant origin. In the larger size groups the croaker was found to be a somewhat selective feeder, depending more upon the sorting or capture of larger discrete animals than upon filtra­ TABLE 10 Occurrence of food items in digestive tracts of 176 Micropogon undulatus 75.0-99 .0 mm. 100 .0-124 .0 mm . 125.0-149.0 mm. 150.0-199.0 mm. 200.0-:125.0 mm. 10.0-24.0 mm . 25.0-49.0 mm. 50.0-74.0 mm . 17 uamined 26 exarnint'd 20 examined 14 examined 30 exnmin"d 25 examim•d 21 exnmint>d 2:J examined 15 wilh food 26 wilh food 20 with food IO wilh food 28 with food 24 with rood 18 wilh food 19 wilh food Percentage Ptrcenlage Pt>rctnla~e Percentage Percentage Per1:enla~e Perc<'nlage Percentage Percentage Percentage Percentage Percentage Percenla(Ce Percenla(Ce Percentage Percenlage of tract~• of total of tracts* of lolal of tracts• of total of tracts• of total or trac1s• of total of tracts* of Iota) of tracts* of total of tracts* of total containing stomach containing stomach containing stomach conlaininA slomach containing stomach conlaining stomach containing stomach conlaining stomach FOOi> ITEMS item volume ilem \'olume item volume ilem volume ilem volume ilcm volume ilem volume ilem volume Ostracoda 15.4 0.1 10.0 0.1 6.7 12.0 T 4.8 Copepoda (undet.) 3.8 16.7 0.2 Calanoid 82.3 52.3 69.2 24.4 35.0 4.5 7.1 0.1 6.7 0.1 4.0 0.1 Harpacticoid 29.4 0.2 50.0 2.1 20.0 0.2 10.0 0.2 Mysid shrimp 23.5 16.9 19.2 12.2 30.0 16.l 21.4 10.5 40.0 3.3 24.0 1.4 14.3 1.8 8.7 T Isopoda 5.9 5.1 11.5 1.9 30.0 0.6 21.4 2.1 30.0 2.4 24.0 0.1 28.6 2.2 Amphipoda 17.6 1.8 57.7 23.2 50.0 9.9 7.1 1.3 43.3 2.3 40.0 3.7 23.8 4.3 Palaemonid shrimp 3.3 2.4 4.0 4.8 Penaeus sp. 4.8 0.4 Crabs (undet.) 21.4 1.8 4.0 0.2 14.3 1.7 8.7 4.2 Rithropanopeus harrisii 20.0 7.3 24.0 8.2 28.6 11.4 26.l 11.8 Callinectes sapidus 3.3 0.6 16.0 9.5 21.7 8.2 Insect a 3.8 7.1 0.3 20.0 0.4 4.0 4.3 Coleoptera 5.0 0.2 23.3 2.5 32.0 1.0 Diptera 5.9 50.0 15.l 65.0 9.8 50.0 1.5 56.7 5.8 56.0 11.6 19.0 2.5 13.0 0.2 Annelida 40.0 15.0 8.0 0.5 4.8 0.2 4.3 0.8 Mollusca Rangia cuneata 7.7 10.0 0.6 42.9 4.2 30.0 0.1 52.0 0.3 42.9 5.5 52.2 29.4 Mytilopsis leucopheata Gastropoda Hydroids Sponges Foraminifera Vertebrata Anchoa mitchilli Cyprinodon variegatus Gambusia a/finis Gobiosoma bosci Micropogon undulatus Myrophis sp. Syngnathus sp. Fish remains Algae-Filamentous Vascular plants Eggs and Cysts Organic mat. ( undet.) Detritus Sand SUMMARY Copepoda Mysid shrimp Isopoda, Amphipoda Insect a .Mollusca Fishes Crabs, Shrimp Miscellaneous Incidental, undet. * ~lom~ch and intestine incll!dcd. 5.9 0.1 5.9 0.3 58.8 23.2 11.8 52.5 16.9 6.9 0.0 0.0 0.0 0.0 0.1 23.5 7.7 3.8 61.5 65.4 26.9 0.3 16.3 4.3 0.2 26.5 12.2 25.1 15.l 0.0 0.0 0.0 0.4 20.8 5.0 5.0 80.0 90.0 15.0 0.2 3.3 43.3 8.7 2.2 4.7 16.1 10.5 10.0 0.8 3.3 0.0 0.1 54.2 7.1 7.1 14.2 85.7 71.4 35.7 9.1 T 40.5 26.5 1.8 0.1 10.5 3.4 1.8 4.2 9.1 1.8 0.0 68.8 3.3 3.3 3.3 3.3 13.3 3.3 16.7 83.3 43.3 0.2 1.6 1.1 3.5 0.2 1.9 35.l 14.l 0.5 3.3 4.7 8.7 0.1 6.2 10.3 17.3 49.2 12.0 8.0 4.0 4.0 4.0 12.0 8.0 20.0 88.0 60.0 12.0 0.3 2.7 4.5 1.8 6.5 0.4 41.9 4.7 0.2 0.1 1.4 3.8 12.6 0.6 15.5 17.9 0.9 46.8 19.0 4.8 4.8 4.8 14.3 9.5 28.6 71.4 47.6 14.3 20.3 39.l 9.7 8.7 0.8 4.3 1.3 4.3 2.0 4.3 9.2 4.3 9.4 1.3 1.5 4.3 1.4 0.4 13.0 0.2 2.3 8.7 T 36.0 52.2 8.9 6.6 65.2 2.8 8.7 0.0 0.0 1.8 0.0 6.5 0.0 2.5 0.2 25.8 39.9 4.1 22.0 13.5 24.2 2.9 1.0 42.6 12.9 Food Habits of Fishes and Invertebrates of Lake Pontchartrain lion of microscopic creatures or consumption of detritus. This trend of selectivity be­comes progressively more noticeable above 100 mm., and in croakers above 200 mm. larger animals constituted over 85 percent of the total food volume. Among the animals consumed by adult croakers the following made up significant percentages in some size classes: annelids, up to 15 percent in one class, palaemonid and penaeid shrimp made up less than 3 percent in any of the adult classes, crabs ( Callinectes sapidus and Rithro­panopeus harrisii) up to 20 percent, mollusks (chiefly the clam, Rangia cuneata) up to 40 percent, and small fishes (including Anchoa mitchilli, Cyprinodon variegatus, Gam­busia af finis, Gobiosoma bosci, young Micropogon undulatus, Myrophis sp., Syngnathus sp., and remains of other species) up to 22 percent. Undoubtedly, size is an important factor determining the utilization of a particular food item by the adult croaker, and even young croakers as large as 67 mm. are not free from predation by adults over 300 mm. in length. Miscellaneous items encountered in the stomachs of some croakers in­cluded sponge, hydroids, eggs and cysts, worm burrows, adult insects, filamentous algae (Cladophora sp.), and sand. On the basis of the above information it is clear that, whereas the croakers of Lake Pontchartrain consume a great variety of different food items, from young to adult they pass through a succession of four over!apping, but distinctly recognizable food stages. They specialize successively upon (1) zooplankton, (2) micro-bottom animals, (3) detritus, and (4) larger animals, the latter group including burrowers, .crawlers, and swimmers. As a rule, small croakers displayed only one or two types of food within a given stom­ach with little variation from one croaker to the next. Intermediate-sized individuals, however, feeding upon small bottom invertebrates exhibited a wide variety of foods with­in any given stomach, and differences between one individual and another were common. Adults, as a group, continued to display a varied diet, but since the food items consumed were individually much larger, the variety of items within a given stomach tended to decrease. Study of the percent fullness of the stomachs and intestines reveals a clear picture of the daily pattern of feeding of the young and adult croakers (Fig. 13). In the young plankton.feeding and bottom-browsing individuals (11.5-74 mm.) feeding intensity is low during the early morning and increases to a peak by early afternoon which gradu­ally tapers off toward evening. This may be related to a presumed vertical migratory pattern of the calanoid copepods which constitute a major element of the diet of these small croakers. Presumably, during the hottest and lightest portion of the day these copepods would be most concentrated near the bottom, and, therefore, most available as food for the young fishes. lntermediate·sized croakers (75-150 mm.) which feed upon detritus together with a general mixture of animals from the bottom water and bot­tom surface appeared to feed moderately throughout the day, increasing their feeding ac· tivity somewhat toward evening (the dip in the curve during the late afternoon appears to be an artifact resulting from small numbers in that particular time class). Large adult croakers ( 150-325 mm.) exhibit two distinct feeding periods during the day. An early peak in mid-morning is followed by a period of low feeding intensity throughout the afternoon with some indications of increased feeding activity toward the evening. This bimodality may be associated with periodic availability of its molluscan, crus­tacean, and vertebrate food, with its own necessity' of escaping mid-day predation, or with general physiological inactivity during the warmest and lightest portions of the day. Food Habits of Fishes and lni-ertebrates of lake Pontchartrain 397 100 80 MICROPOGON UNDULA !US 60 :c ..J 0-' ::i;"" 40 o~ y ... !.. Cl) 20 80 60 Z..J_, ILi­ _ I-::> (/)IL 40 ~~ zL 20 0 7 8 9 10 II 12 13 14 16 17 18 19 20 21 HOUR OF DAY F1:;. 13. Hourly pallern of percent-fullness of stomach and intrstine of Micropogon 11nd11laws (young, intermediate. and adult 1. Studies to date suggest that the food of the croaker is roughly simi!ar throughout its geographic range, but local differences in detail may be pointed out. Along the Atlantic coast polychaetous annelids are of considerable importance as food, and quantities of echinoderms and some ascidians and hemichordates are consumed in that area. Along the Gulf coast these items appear to be of less importance, whereas quantities of shrimp, fishes, and crabs are substituted. On the Atlantic coast the staple food mollusk of the adult croaker appears to be the razor clam_ which is replaced in Lake Pontchartrain by Rang:·a cuneata and Congeria leucopheata, and in east Texas by Macoma mitchelli. Several investigators on the Atlantic coast have encountered quantities of debris or undeterminable organic matter in the stomachs of young and medium-sized croakers and have generally been inclined to consider such material as "incidentar' or simply annelid or other large food digested beyond recognition. Reid I 1955b l, working in Texas, appears to have been the first to point out that organic debris may constitute an important energy source for the croaker as well as for other species of sciaenid fishes. This point is reemphasized by the food contents from Louisiana specimens. Regarding adaptations of the croaker for obtaining its food, Welsh and Breder 11923 l noted that this fish possesses a flattened profile and sensitive, pendant mandibular bar­bels typical of bottom feeders. Roelofs 0954) reported that when feeding in laboratory aquaria, croakers dive into the bottom with some force, digging as they feed, and thus they are able to obtain subsurface animals for food. He also studied the gill rakers of this fish, pointing out that they tend to be short and stout with relatively few setae. Hence. the gill structure of the adult croaker forms a coarse basket rather than a fine straining mechanism and tends to retain only the larger particulate matter. This knowledge aids in understanding the food habits of the species and in explaining the observed differences between the food habits of the adult croaker and those of the adult spot. The croaker shovels more deeply while the spot scoops the surface; also, the latter retains what the former discards. As a matter of fact, although the adults of the two species appear not to be in keen competition for food, there does exist a great deal of overlap in the food of the intermediate-sized croaker and that of the adult spot. Pogonias cromis (LinnaeusI. Black Drum Smith 11907) mentioned that the black drum is a bottom feeder taking mainly crus­taceans and various mollusks including oysters. Welsh and Breder (1923) agreed, emphasizing depredations of the black drum upon oyster beds. Reporting on the food of 117 black drums (80-990 mm.) from Texas, Pearson I 19291 found that young individ­uals (80-200 mm. I feed a great deal upon what he termed "soft foods" (including fish and annelids) as well as upon crustaceans and mollusks. With increasing size the "soft foods" were dropped from the diet, and older drums were found to feed largely upon mollusks (74 percent I and crabs (16 percent). The mollusks included mainly Mulima transversa corbuloides with some Mytilus sp. and Ostrea sp. Gunter (1945) examined the stomach contents of 124 black drums (205-460 mm.) from southern Texas, only 96 of which contained recognizable material. He found that crustaceans, molluskc, and fishes were the most abundant foods. The crustaceans included large numbers of amphipods, as well as blue crabs, penaeid shrimp, palaemonid shrimp ( Palaemonetes sp. I, pistol shrimp (Cargo sp. 'l, and oyster crabs (Panopeus sp). The mollusks included small gastropods, and pelecypods (Donax sp. and razor clams), and the only identifiable fish was the goby (Gobiosoma bosci l which appeared in the stomachs of at least 12 individuals and which numbered as high as 23 in a single stomach. Reid (1955b l examined seven young black drums from East Bay, Texas, and in the six which contained food he found masses of pelecypod shell debris. Among the small drums examined from Lake Pontchartrain, mollusks made up over 65 percent of the food material, and mud crabs (Rithropanopeus harrisii) made up about 12 percent (Table 11). A few other small invertebrates were noted. While the quantity of detritus present in the stomachs was very small, more than one-fifth of the food consisted of organic material which was not positively identifiable. This material frequently took on the appearance of a milky sludge and was probably composed of the partially-digested bodies and mucus of the many pelecypods consumed. Among the three large black drums examined, the stomach of one was empty while the remaining two were about half full of the clam (Rangia cuneata I. In addition to the above, a num­ber of large black drums were examined qualitatively in the field. These generally con· tained only the remains of Rangia cuneata, supporting the conclusion that all sizes of black drum above 100 mm. in length are primarily dependent upon this clam as the staple food while in Lake Pontchartrain. Welsh and Breder (1923) pointed out that the black drum exhibits the flattened profile and mandibular barbels typical of bottom-feed­ing fishes. Pearson 0929) correlated the food with the environment in which the fish feeds. He noted that the pelecypod food of the black drum I in this case, the clam, Mu· linia I lives in shallow muddy lagoons and bays, and that the black drum is most abun­dant where the water is always turbid, highly saline to brackish, shallow (4ft.1 , and characterized by temperature extremes. Reid 11955a) mentioned that this species, like the redfish, Sciaenops ocellata, has a characteristic habit of grazing in the salt marshes when these marshes are under water. Food Habits of Fishes and Invertebrates of Lake Pontchartrain Commercial fishermen have pointed out to the writer that black drums frequently dig for mollusks in shallow water, and in so doing they literally stand on their heads waving the caudal fins aloft. Where the water is very shallow the caudal fins may clear the sur­face and be visible for some distance. This phenomenon of "flagging frequently en­ables the commercial fisherman to locate the black drum among the marshes and tidal Hats of the Louisiana coast. Apparently the phenomenon of head-standing is not limited to shallow waters. Mr. Percy Viosca, Jr. (in personal conversation and correspondence) has indicated that while flying over Lake Pontchartrain on a still summer day he noted on the surface what appeared to be the result of large black drums feeding singly or in groups through the western sector of the lake. Digging and flagging, each drum created an upwelling of turbid water resembling smoke pouring from a chimney. These fish were undoubtedly digging for the clam, Rangia cuneata, the only abundant large pelecypod in that area of the lake. Indirect evidence indicates that the adult black drum can easily crush and discard the hard shells of adult rangia clams retaining only the soft bodies as food. In the field it was repeatedly observed that freshly-caught specimens with shell fragments in the buccal cavity contained only a milky mush in the stomach. Although a number of au­thors have noted shell fragments occurring in the stomach of this fish, such material ap­pears to have been derived from small or more fragile mollusks. It is generally accepted that the black drum is a bottom feeder, but details of the food habits of the species are still poorly known. Individuals less than 100 mm. have not been well investigated, and larger individuals have generally been lumped into a single group. On the basis of the existing knowledge of the food habits of the species two feeding stages are distinguishable, and from what is known of related sciaenids, a third may be postulated. The very young undoubtedly feed upon planktonic or bottom-dwelling mi- TABLE 11 Occurrence of food items in digestive tracks of 24 Pogonias cromis 116.0-218.0 mm. 24 examined 20 with food Percenla,r:e Pt>rcenla~e o( tracl s* of 101al containin11: !'lomar.h FOOD ITEMS ilem ,·olutnf' Isopoda 8.3 0.4 Amphipoda 4.2 Crabs Rithropanopeus harrisii 20.8 12.2 Insecta Dipter­ larvae 16.7 O.l Mollusca Rangia cuneata Mytilopsis leucopheata 75.0 12.5 55.5 9.9 Gastropoda Fish scales 20.8 4.2 O.l Algae-filamentous Organic mat. (undet.) Detritus 4.2 41.7 41.7 21.7 T Sand 12.5 SUMMARY Crabs 12.2 Mollusks 65.5 Misc. Invertebrates 0.5 Organic mat. (undet.) 21.7 • Slomach and intestine induded. Food Habits of Fishes and Invertebrates of lake Pontchartrain crocrustaceans. Following this stage, larger bottom invertebrates must enter the diet, in· dividuals within the 100-200 mm. class being dependent primarily upon small mol­lusks, crustaceans, and "soft foods" (fish and annelids) . Larger individuals (over 500 mm.) subsist chiefly upon mollusks. In Lake Pontchartrain the primary food of the middle-sized as well as the adult black drums, appears to be the clam, Rangia cuneata. Sciaenops ocellata (Linnaeus). Red Drum, Redfish The food of the redfish on the Atlantic coast has been reported by Linton ( 1904) and Hildebrand and Schroeder (1928), and food habits of this species on the Gulf coast of Texas have received the attention of Pearson (1929) , Gunter (1945), Knapp (1949), and Reid (l 955b). The smallest redfishes examined (30 mm. and over) have been found to consume small crustaceans such as schizopods and amphipods. Food of the larger size groups has been found to include primarily penaeid shrimp (P. aztecus and P. seti fer us) , blue crabs (Callinectes sapidus) and fishes (Brevoortia sp., Galeichthys felis, Anchoa mitchilli, Cyprinodon variegatus, Fundulus sp., lucania parva, Mugil cephalus, Menidia sp., Syngnathus sp., leiostomus xanthurus, lagodon rhomboides, Symphurus plagiusa, Gobiosoma bosci, and Myrophis punctatus, as well as a number of unidentified species). In addition, a variety of other organisms has been encountered in the stomachs of large redfish with less frequency. These include the following: bi­valves, squids, annelids, crustaceans including grass shrimp (P. vulgaris), mud crabs (Neopanope t. texana, and Panopeus herbstii), snapping shrimp (Crago sp.), mud shrimp (Callianassa jamaicense louisiananum), a large marsh rat, algae, and vascular plant matter. Stomach contents of 17 redfish (184-625 mm.) from Lake Pontchartrain were ex· amined. Among the 12 which contained recognizable food, crabs predominated and included both the blue crab (Callinectes sapidus) and the mud crab (Rithropanopeus harrisii) which together made up 62 percent of the food. Other important categories included fish remains (17 percent) and unidentified organic material (15 percent). Vascular plants made up five percent of the stomach contents, and amphipods, palae· monid shrimp, and hydroids constituted le!'s than one percent each. Within the above size range, large living invertebrates and fishes made up four-fifths of the material con· sumed by the redfish, and especially heavy predation upon blue crabs was indicated. Available information indicates that young redfish subsist mainly upon a diet of small crustaceans including amphipods and schizopods, with larger individuals shifting to larger crustaceans and fishes. In inside waters, as pointed out by Gunter (op. cit.) and borne out in the Pontchartrain series, the principal food of the adult redfish is the blue crab, whereas in marine waters penaeid shrimp achieve greatest significance. In both habitats a wide variety of different invertebrates is utilized as supplementary food. As pointed out by Pearson (o p. cit.) and Gunter (op. cit.) this predatory fish consumes both bottom-living and free-swimming forms, and the former author noted that in its feeding habits the redfish is intermediate between the bottom-feeding black drum and the pelagic predator, the speckled trout. It has been reported by Reid (l955a) that the redfish, like the black drum, frequently grazes in flooded salt marshes, a habit which appears to bP-of considerable significance in the ecology of the species in the marshy areas of southeastern Louisiana. SPARIDAE Archosargus oviceps Ginsburg. Gulf Sheepshead In 1894 Brooks discussed the food of the sheepshead, and on the basis of personal observations concluded that although this fish browses among algae, its food is "ex· elusively animal", including barnacles and young oysters. Linton (1904) examined the stomach contents of two specimens and found crustaceans (including hermit crabs), sea urchin fragments, shells, and gorgonian spicules. Smith (1907) mentioned that the sheepshead feeds chiefly upon mollusks and crabs, a statement which was repeated by Hildebrand and Schroeder (1928) apparently without further verification. Gunter (1945), however, examined the food of 18 specimens (190-365 mm.) from southern Texas, and he found that in most individuals the long guts were distended with great quantities of plant material ("grass" and algae). Crabs (including Callinectes sapidus), gastropods, and unidentified shells were encountered in only a few individuals. Gunter concluded, therefore, that regardless of previous statements, the adult sheepshead is largely herbivorous. Gunter (1954) later reported observing the clumsy but successful attack by an adult sheepshead on a half grown blue crab swimming at the surface. Viosca (1954) drawing upon years of personal experience indicated that in Louisiana the sheepshead feeds both upon "grass" in the beds of aquatic vegetation and upon small invertebrates inhabiting oyster reefs and other hard surfaces. As invertebrate food items, he listed sessile forms such as barnacles, mussels, small oysters, and hydroids, as well as small mobile species including crabs, shrimp, and snails. Reid, Inglis, and Hoese (1956) stated that four sheepshead (230-400 mm.) from East Bay, Texas had consumed crabs, pelecypods, and vegetable matter. In the present study 11 specimens of the gulf sheepshead (218--410 mm.) from Lake Pontchartrain were analyzed, and all contained identifiable food. Vegetation was the most abundant material encountered and included filamentous algae (Cladophora sp.) together with vascular plants (V allisneria spiralis and Ruppia maritima) which constituted over 54 percent of the stomach volume. Invertebrates included mussels (Congeria leucopheata and Mytilus recurvus) 19 percent, clams (Rangia cuneata) 8 percent, sponge (Spongilla l, lacustris) 10 percent, and mud crabs (Rithropanopeus harrisii) 1.5 percent. Young croakers and remains of other small fishes constituted three percent of the food, and blue crabs (Callinectes sapidus), amphipods, isopods, barnacles (Balanus spp.), small gastropods, and hydroids made up less than one per­ cent each. A small quantity of undetermined organic material also was present. The food of young sheepshead apparently has not been investigated. Larger individ­ uals employ the well-formed, incisor-like teeth for cropping vegetation, for picking up small active invertebrates, and for scraping sessile animals from hard surfaces. The sheepshead's habit of nibbling the invertebrate epifauna from vertical posts and pilings appears to be widely recognized among sport fishermen of the Gulf area. Individuals collected from inshore shallows such as Lake Pontchartrain and the bays of southern Texas were found to have fed chiefly upon vegetation. This species is further remark­able in its occas:onal utilization of sponge as food. Lagodon rhomboides (Linnaeus). Pinfish Linton (1904) and Smith (1907) mentioned that the food of the pinfish is quite varied, including primarily mollusks, worms, crustaceans (amphipods and small shrimps and crabs), small fishes, and vegetation ("sea weed", "sea lettuce", and "vegetable detritus"). A variety of other animal remains also was present. Hildebrand and Schroeder (1928) found that the food of 13 specimens included (in decreasing order of abundance) vegetable debris, crustaceans, mollusks, and annelids. Gunter (1945) reported the food of eight large pinfish (150-285 mm.) from Texas. One fish had consumed razor clams, and the remaining individuals all contained plant material including algae and "grass". Because of the stomach contents and the presence of large, inci:1or-like teeth, the pinfish was regarded by Gunter as being a "grazer". Reid l1954) prnvided information on the food of 65 young pinfish ( 15-128 mm.} from Florida. His data indicate that the smallest fish ( 15-50 mm.) fed most frequently upon copepods, amphipods, and shrimp. The next larger size group (51-100 mm.} con­tained (in decreasing order of occurrence) copepods, shrimp, organic detritus and mud, amphipods, mollusks, crabs, fishes, and plant detritus. The largest pinfish group (101-128 mm.) contained only copepods, amphipods, shrimps, and mollusks, each occurring in 18-25 percent of the stomachs. Although dealing primarily with smaller individuals, Reid's study seems to suggest that if invertebrates are sufficiently abun­dant, less vegetation is consumed by the pinfish. In the Lake Pontchartrain study micro-bottom animals constituted the principal food of the two smallest size classes (40-64 mm. and 65-74 mm.), making up from 60 to 80 percent of the ingested material in each class I Table 12 and Fig. 14). 8o 70 - ~ !.. 60 LI.I :::E :::> ...J 50 0 > :I: 40 0 ...J IL 60 ~ !... :I: u 40 ct :::;: 0 I-20 (/) LAGO DON RHOMBOIDES A~---------------. 80 ...J ...J ::> IL 60 -! v• • !.. L&J z 40 ~ (/) L&J I­ 20 ~ 10 II 12 13 14 15 16 17 18 HOUR OF DAY Fie. 15. Hourly pattern of percent-fullness of stomach and intestine of Lagodon rhomboid es I young and adult I. Like its relative, the sheepshead, the pinfish exhibits a row of incisor-like tce:h which are employed for grazing upon vegetation, picking up small invertebrates, and, perhaps, scraping invertebrate epifauna from solid surfaces. Size differences seem to preclude much overlap in food requirements of the adults of the two species. Com­petition of adult pinfish with young sheepshead appears to be avoided largely by dif­ferences in habitat utilization. BoTHIDAE Paralichthys lethostigma Jordan and Gilbert. Southern Flounder Gunter ( 1945) reported the examination of 16 large flounders t 240---i90 mm.) from southern Texas, eight of which contained food. Sewn had consumed fishes in­cluding mullet, anchovies (Anchoa mitchilli diaphana L pinfish (lagodon rhom­boUies), mojarra, and other unidentified species. In addition, three flounders con­tained penaeid shrimp (P. aztecus and P. setiferusl and a stone crab. McLane (19481 mentioned that in the St. Johns River of Florida, the southern flounder consumes large numbers of the locally abundant shad, Signalosa petenensis va.nhyningi. Knapp (1949) examined 24 specimens from Texas and found that fishes. including the men­haden, were present in oYer 40 percent of the stomachs, shrimp were present in 50 Food Habits of Fishes and Invertebrates of lake Pontchartrain percent and miscellaneous invertebrates occurred in less than five percent. In north­ern Texas, Reid (1955b) examined the stomachs of seven specimens (159-265 mm.) and found fish in three of the stomachs while shrimp occurred only once. A total of 19 specimens (113-380 mm.) from Lake Pontchartrain were examined, of which only 14 contained recognizable food. Fishes constituted 89 percent of the total food volume and included anchovies !Anchoa mitchilli diaphana) (41 percent), small croakers, (M icropogon undulatus) (2 percent), and a large quantity of un­\dentifiable fish remains ( 46 percent). Crahs made up about seven percent of the food and included Callinectes sapidus as well as Rithropanopeus harrisii. Schizopods, clams (Rangia cuneata) , small gastropods, an .f undetermined organic material made up less than two percent each. The above studies indicate that adult southern flounders are highly predaceous con­ suming large quantities of small fishes and smaller amounts of crabs and shrimp, which they apparently attack either upon or near the bottom. The relative percent­ages of fishes and shrimp utilized, however, appears to vary from one environment to another and may be related to availability. Young flounders probably prey upon small bottom invertebrates. SoLEIDAE Trinectes maculatus fasciatus (Lacepede). Hogchoker Hildebrand and Schroeder (1928) examined the stomachs of 17 hogchokers from Chesapeake Bay and concluded that the species feeds primarily upon annelids sup­plemented by occasional small crustaceans and strands of algae. Reid (1954) men­tioned that all stomachs which he inspected were devoid of food, although they usu­ally contained quantities of sand. Ten hogchokers (61-74 mm.) from Lake Pontchartrain were examined, but the stomachs of only three conta:ned recognizable food. Amphipods of the genus Coro­phium were present in all three stomachs and made up about 50 percent of the con­tents. Undetermined organic matter and detritus comprised the remainder. Examina­tion of the intestinal contents revealed that the hogchokers had also consumed other microcrustaceans, dipteran (chironomid) larvae, foraminifera, and seeds. Although quite meager, the above data suggest that the hogchoker subsists largely upon a diet of small bottom-living invertebrates supplemwted by quantities of organic detritus and other materials which it gleans from the surface of the bottom. INVERTEBRATES MACTRIDAE Rangia cuneata Gray. Common Rangia, Clam Three specimens of the common rangia (35-38 mm.) from Lake Pontchartrain were exam'.ned, and all of the stomachs contained food. Unidentifiable detritus made up about 70 percent of this material, and sand constituted 10 percent. Enormous num­bers of small, regular, round bodies were observed in two of the clams, and these made up an estimated 17 percent of the total food volume. These bodies, which were barely distinguishable at lOOOx magnification, were assumed to be minute protozoans or individual cells derived from disintegrated colonies of blue green algae (Anabaena Food Habits of Fishes and Invertebrates of Lake Pontchartrain spp. or Microcystis spp.). Additional material present in trace amounts included eggs (or cysts), a variety of bottom diatoms, foraminifera, and bits of vascular plant ma­ terial. In Lake Pontchartrain the common rangia is most abundant on the muddiest bot­toms in waters of maximum turbidity. As indicated by the above analyses and by observations in laboratory aquaria, the species is a typical "filter feeder'', and in view of the high percentage of detritus ingested a simple straining action with very little selectivity is suggested. Penaeus setiferus (Linnaeus). White Shrimp Williams (1955) reported on stomach analyses of 184 young and adult penaeid shrimp from the estuaries of North Carolina. He found that the stomachs were gener­ally filled in the summer and fall, but were nearly always empty in winter. Williams listed the most abundant food materials as follows (in order of decreasing frequency of occurrence), unrecognizable debris (probably a mixture of digesting tissue and organic deposit from the bottom), chitin fragments (from crustaceans), setae (prob­ably from annelids), worm jaws (annelid), plant fragments, and sand. Other ma­terials occurring in smaller quantities included foraminifera, minute gastropods and lamellibranch shells, squid suckers, complete small fishes, fish parts (scales, muscle fibers, remains of gut, mandible, ribs, eye lens), eggs, seed pods, and other materials. Williams emphasized that most of the remains which he encountered were either large or very hard, in either case, materials which would not readily pass through the strain­ing apparatus of the pyloric stomach. Since such materials might accumulate over a period of time, he suggested that unrecognized, softer, more digestable matter could easily form the bulk of the shrimp diet. Flint ( 1956) studied the food of penaeid shrimp from Grande Isle, Louisiana and Biloxi, Mississippi (localities verified by personal correspondence). Small shrimp (ca. 10 mm.) contained chiefly cropped filaments of blue green algae and such dia­ toms as ordinarily would have been found adherent to them in the natural habitat. Adult shrimp, however, were found to have consumed large quantities of bryozoans, some algae, and a few small pieces of sponge and coral. The algae included primarily lithophytic diatoms which may have been taken in incidentally, adherent to other materials. A few fragments of roots and stems of vascular plants and bits of sawdust also were encountered. Flint noted that partial or complete digestion of the algal proto­ plasts had taken place. He concluded that in the turbid coastal waters (of Mississippi and Louisiana), blue green algae which thrive in low light intensities probably serve as a major food for both young and adult shrimp, although the latter consume a diversity of materials among which bryozoans, lithophytic algae, and diatoms appear to be outstanding. In the present study the stomach contents of ten white shrimp (91-142 mm.) were examined, all of which contained food. Detritus and ground organic matter made up 58 percent of this material. Mollusks, comprising 12 percent included many minutp clames ( Rangia cuneata) and a few small gastropods. Microcrustacea made up less than four percent of the stomach contents and included ostracods and harpacticoid copepods. Insect remains were evidenced by traces of dipteran and coleopteran larvae. Numbers of foraminiferans were present in six of the stomachs, and traces of vascular plants were noted twice. In addition to the above materials, all shrimp contained quantities of sand. Pearson (1939) successfully reared post-larval white shrimp (5-35 mm.) in la­boratory aquaria. He found that green algae were consumed in quantity, but alone this material seemed insufficient to satisfy the hunger of the young shrimp. When placed in aquaria containing a shallow layer of estuarine bottom deposits of sand and organic debris, however, the young shrimp fed well and thrived on this material. At a length of about 15 mm. other food was introduced such as raw fish, angleworms, and shrimp meal which also were taken readily as food. More recently, Johnson and Field­ing (1956) reported success in rearing adult white shrimp on a diet of fish meal and also upon fresh fish. In special growth experiments shrimp were found to consume as much as 10 percent of their body weight of fresh fish per day. The accumulated evidence suggests that, whereas both young and adult white shrimp are omnivorous, they feed largely upon organic detritus of the estuarine bot­ tom. This substance is supplemented with more concentrated food material of animal and vegetable origins, both fresh and decaying, from a great variety of immediate sources. In Lake Pontchartrain large concentrations of white shrimp were sometimes observed feeding in shallow mud flats on hot summer days, especially in the rich deltaic deposit areas of tidal streams which drain the north shore marshes. Much feed­ing must also take place in deeper waters since shrimp captured a mile or more from shore generally exhibited full stomachs. PALAEMONIDAE Macrobrachium ohione (Smith). River Shrimp Ten specimens of the river shrimp (48-81 mm.) from Lake Pontchartrain were examined, and all contained food material. Finely-ground detritus and bits of or­ganic material made up over one-half of the diet, and sand constituted almost 20 percent of the stomach contents. Of the remaining identifiable material, clams ( Rangia cuneata} made up the highest percentage. These were frequently broken, although the stomach of one river shrimp contained a total of 33, mostly unbroken clams, the largest of which measured 1.5 mm. Miscellaneous items included ostracods, bits of arthropod integument, minute gastropods, hydroid "stems", sponge spicules, tintinnids, foramini­ferans, filamentous algae (Cladophora sp.), diatoms (species of Cymbella, Navicula, and Pinnularia), eggs and cysts, and remains of vascular plants. Three of the river shrimp contained fish remains including eye lenses, vertebrae, ribs, otoliths, scales, and macerated fibers which appeared to be muscle tissue. Several individuals contained quantities of minute round bodies similar to those encountered in stomachs of Rangia cuneata (mentioned earlier). These analyses indicate that the river shrimp is a general bottom scavenger feed­ing upon bits of animal and vegetable material which it sifts from the bottom surface material. The bottom detritus itself forms a significant portion, if not the bulk, of the diet of this species. The nature of the stomach contents further suggests that the river shrimp and the white shrimp were for the most part feeding on different types of bottoms. PoRTUNIDAE Callinectes sapidus Rathbun. Blue Crab In his discussion of the life history of the blue crab, Hay (1904) considered the food and the feeding habits of the species. He noted that in shallow back-waters the Food Habi~s of Fishes and Invertebrates of lake Pontchartrain 409 blue crab often stalks and consumes small fishes and that occasionally it nibbles at tender shoots of eel grass and other aquatic vegetation. He considered the "favorite" :food of the blue crab, however, to be decaying flesh and other putrid matter and noted that large numbers of crabs were attracted to the offal from stables and water closets overhanging the water. Hay also indicated that injured crabs, small crabs, and soft crabs are sometimes eaten by stronger individuals, and he concluded that while the blue crab conswnea a variety of foods, it is preeminantly a scavenger and a cannibal. In the Lake Pontchartrain study a great variety of materials was encountered in the stomachs of all sizes of crabs, although food differences between young and adults were not pronounced (Table 13 and Fig. 16). Crustacean remains constituted the most abundant material in the stomachs of small individuals, making up over one­third of the food in the 30-74 mm. group. This material decreased to 10 percent of the food volume in the largest individuals. Included in this material were the re­mains of both the blue crab ( Callinectes sapidus) and the mud crab (Rithropanopeus harrisii) , as well as unidentified pieces of crustacean exoskeletons. Greatest utiliza­tion of the mud crabs occurred in the 75-124 mm. class where they made up 16 per­cent of the food volume, and small blue crabs achieved a maximum of 13 percent in TABLE 13 Occurrence of food items in stomachs of 124 Callinectes sapidus 30.0-74.0 mm. 75.0-124.0 mm. 125.0-147.0 mm. 148.0-197.0 mm. 29 examined 31 examined 24 examined 40 examined 24 wilh food 27 with food 23 wilh food 29 wilh food Percentage Percentage Percenlage Percentage Percentage Percentage Percentage Percentage ol lracls* of total of tracts* of total of tracts* of total of tracts* of total containing stomach containing stomach containing stomach containing stomach FOOD ITEMS item volume item volume item volume item volume Crabi (undet.) 13.8 2.i 12.9 4.9 16.7 5.7 12.5 4.3 Rithropanopeus harrisii 16.l 15.6 4.2 0.2 5.0 0.1 Callinectes sapidus 10.3 1.4 3.2 2.5 8.3 13.0 7.5 5.0 Cirripedia 6.5 0.1 Crustacea (undet.) 31.0 31.7 9.7 2.0 20.1 3.5 10.0 1.0 Odonata 3.2 0.2 4.2 0.2 Annelida 6.5 0.1 4.2 T Mollusca Rangia cuneata 41.4 32.4 45.2 20.2 70.8 30.0 57.5 46.5 Mytilopsis leucopheata 6.5 0.3 25.0 19.4 20.0 11.9 Gastropoda 13.8 1.9 12.9 9.0 29.2 5.5 25.0 5.0 Hydroids 3.4 0.3 6.5 0.2 8.3 0.5 2.5 T Vertebrata Fish remains 3.4 0.5 6.5 0.4 16.7 1.6 17.5 5.4 Bottom Diatoms 3.2 0.1 Algae-filamentous 12.9 3.0 4.2 T 2.5 0.3 Vascular plants 6.9 0.4 25.8 8.1 20.8 0.8 10.0 2.0 Organic mat. (undet.) 17.2 7.7 35.5 13.9 25.0 5.9 17.5 8.8 Detritus 37.9 12.1 32.3 12.8 33.3 12.7 15.0 9.7 Sand 37.9 9.1 41.9 6.4 29.2 1.7 2.5 T SUMMARY Crabs, Undet. Crusts 35.8 25.0 22.4 10.4 Mollusks 34.3 29.5 54.9 63.4 Fish remains 0.5 0.4 1.6 5.4 Vegetation, misc. 0.7 11.8 1.5 2.3 Detritus, undet. 19.8 26.7 18.6 18.5 Sand 9.1 6.4 1.7 T * Stomach only. the d!et of the 125-UI mm. size group. Following a molt the blue crab may con­sume its own shed exoskeleton, and it is likely that much of the blue crab shell en­countered in the stomachs was simply obtained in this manner. The evidence here indicates that some cannibalism does occur, however. since fresh muscle fibers were frequently found still attached to the consumed shell fragments. Callinectes sapidus is known to be cannibalistic when confined in "holding" pens, especially if hard and soft individuals are confined together (Benedict, 1940; Cargo and Cronin, 1951), and the present work bears out the suggestion of Hay (op. cit ..1 that cannibalism is also of widespread occurrence in the natural population. Mollusks also appeared as an important elment in the diet of the blue crabs, con­stituting one-third of the food of the smallest indi,-iduals and increasing to almost two-thirds of the food rnlume of the largest size class. These mollusks included small clams (Rangia cuneata) , mussels (Conljeria leucopheata), and gastropods (}felam­pus co_ffeus and Neritins virginico, both of which are inhabitants of the marginal pus cofjeus and J\'eritina virginica, both of which are inhabitants of the marginal Fish remains were not significant in the diet of the smaller size groups, and even in the largest blue crabs they made up only 5 percent of the food. Vegetation was 100 - ~ - 0 80 IJJ ::E ::::::> ....J 60 0 > :c 40 u ..... 60 ~ 0 y -­ _...__--­-­ J: 0 40 ~­ --.. . pods and small benthic invertebrates of this turbid estuary. Such detritus has a very complex origin (cf. Gneri and Angelescu, 19511, and it has been shown by other workers to be rich in bacteria (Baier, 1935; ZoBel\ and Feltham, 1938, 1942; Burk­holder and Bornside, 1957). An ontogenetic progression of food stages was clearly demonstrated for several of the well studied species, and for one species (Atlantic croaker) as many as four distinct nutritional stages were recognized. Within a given stage considerable sub­stitution of food items was often obserYed. In some cases it was possible to correlate ontogenetic changes in food utilization with changes in morphology, habitat, feed­ing time, and behavior. As a result of their highly varied diets including much de­tritus, most, if not all, of the species examined are considered to be omnivorous. Dis­tinct trophic lewis (in the sense of Lindeman) were not recognizable in this estuarine community. Summary The physical enYironment of Lake Pontchartrain was found to be characterized by moderate temperature, generally low salinity, and very high turbidity, although considerable variations in these factors were noted. Food studies, including 1,399 quantitative and about 100 qualitative stomach analyses, were carried out on thirty­five of the most important species comprising the estuarine community. These in­cluded the following: bull shark, longnose gar, spotted gar, alligator gar, bigeye herring_ gulf menhaden, gizzard shad, threadfin shad, southern bay anchovy, gaff­topsail catfish, sea catfish, blue catfish, channel catfish, Atlantic needlefish, striped mullet, silverside, yellow bass, largemouth bass, common jack, freshwater drum, silver perch, sand squeteague, spotted squeteague, spot, Atlantic croaker, black drum, red drum, gulf sheepshead, pinfish, southern flounder, hogchoker, common rangia (clam), white shrimp, river shrimp, and blue crab. A definite ontogenetic progression of food stages was recognizable in many species, and most were considered to be omnivorous. Organic detritus was prominent in the diet of most species. Two partial food chains (planktonic and benthonic) were found to support the top predators, but distinct "trophic levels" among the consumers were not recognizable. LITERATURE CITED Aitken, W.W. 1936. Some common Iowa fishes. Iowa State Coll., Extension Cir. 224. Baier, C. R. 1935. Studien zur Hydrobakteriologie stehender Binnengewasser. Arch. Hydrobiol., 29: 183-264. Bailey, R. M., and H. M. Harrison. 1948. Food habits of the southern channel catfish (Ictalurus lacustris punctatus) in the Des Moines River, Iowa. Trans. Amer. Fish. Soc., 75 (1945) : 110-138. Baughman, J. L., and Stewart Springer. 1950. Biological and economic notes on the sharks of the Gulf of Mexico, with especial reference to those of Texas, and with a key for their identification. Amer. Midi. Nat., 44 (1) : 96-152. Bell, J.C., and J. T. Nichols. 1921. Notes on the food of Carolina sharks. Copeia., 1921 (92): 17-20. Benedict, Steve. 1940. Soft crabs-and hard. Louisiana Conser. Rev. Spring, (1940) : 11-14, 48. Bigelow, H. B., and W. C. Schroeder. 1953. Fishes of the Gulf of Maine. Bull., U.S. Bur. Fish. 74: 1-577. Boesel, M. W. 1938. The food of nine species of fish from the western end of Lake Erie. Trans. Amer. Fish. Soc., 67: 215-223. Bonham, Kelshaw. 1941. Food of gars in Texas. Trans. Amer. Fish. Soc., 70 (1940) : 356--362. Brooks, W. K. 1894. The origin of the food of marine animals. Bull. U.S. Fish. Comm., 1893, 13: 87-92. Burkholder, P. R., and G. H. Bornside. 1957. Decomposition of marsh grass by aerobic marine bacteria. Bull. Torrey. Bot. Cl., 84 ( 5) : 366--383. Cargo, David G., and L. E. Cronin. 1951. The Maryland crab industry in 1950. Contr. Chesapeake biol. Lab., 92: 1-23. Darnell, R. M. 1954. An outline for the study of estuarine ecology. Proc. La. Acad. Sci., 27: 52-59. ----., and A. B. Williams. 1956. A note on the occurrence of the pink shrimp Penaeus duo· rarum in Louisiana waters. Ecology, 37 (41: 844-846. Dendy, J. S. 1946. Food of several species of fish, Norris Reservoir, Tennessee. J. Tenn. Acad. Sci., 21 (1): 105-127. Dill, W. A. 1944. The fishery of the lower Colorado River. Calif. Fish & Game, 30 (3): 109-211. ----., and C. Woodhull. 1942. A game fish for the Salton Sea, the ten-pounder, Elops a/finis. Ibid., 28 (4): 171-174. Ebeling, Alfred W. 1957. The dentition of eastern Pacific mullets, with special reference to adapta­ tion and taxonomy. Copeia, (3): 173-185. Eddy, Samuel. 1934. A study of fresh-water plankton communities. Ill., Biol. Monogr., 12 (4): 1-93. Eigenmann, Carl H. 1902. Investigations into the history of the young squeteague. Bull. U.S. Fish. Comm., 1901, 21: 45-51. Evermann, B. W., and H. W. Clark. 1920. Lake Maxinkuckee. A physical and biological survey. Ind. Cons. Dept., 1: 1--660. Ewers, Lela A. 1934. Summary report of crustacea used as food by the fishes on the western end of Lake Erie. Trans. Amer. Fish. Soc., 63: 379-390. ----.,and M. W. Boesel. 1936. The food of some Buckeye Lake fishes. Ibid., 65: 57-69. Fisk, H. N. 1944. Geological investigation of the alluvial valley of the lower Mississippi River. Vicksburg: Corps of Engineers, Miss. River Comm., i-vi: 1-78, 78 figs., 11tables,33 plates. ----. 1947a. Geological investigation of the Veterans Hospital site, New Orleans, Louisiana. (mimeographed report) : 1-40, 7 figs., 3 plates. ----. 1947b. Fine-grained alluvial deposits and their effects on Mississippi River activity. Vicksburg: Corps of Engineers, Miss. River Comm., Vol. 2: 74 plates. Flint, L. H. 1956. Notes on the algal food of shrimp and oysters. Proc. La. Acad. Sci., 19: 11-14. Forbes, S. A. 1878. The food of Illinois fishes. Bull. Ill. Lab. nat. Hist., 1 (2) : 71-89. ----. 1880a. The food of fishes. Ibid., 1 (3l: 18-65. ----.1880b. On the food of young fishes. Ibid., 1 (3): 66-79. ----. 1888. Studies of the food of fresh-water fishes. I bid., 2: 433-473. ----., and R. E. Richardson. 1920. The fishes of Illinois. 2nd Ed., Springfield Bull., Ill. nat. Hist. Surv., cxxxvi+357 pp. Frisby, Robert. 1942. The food habit of "gars". Ohio Cons. Bull., 6(2l: 12-13. Ghazzawi, F. M. 1933. The pharynx and intestinal tract of the Egyptian mullets M. cephalus and M. capita-Part I. On the food of mullet from Egyptian waters. Notes Fish. Res. Dir., Cairo, 5. Gneri, F. S., and V. Angelescu. 1951. La nutricion de los peces iliofagos en relacion con el meta­ holismo general de! amhiente acuatico. Rev. Inst. Invest. Mus. argent. Cienc. nat., 2 (1): 1-44, 6 figs., 2 pis. Goode, George Brown. 1879. The natural and economical history of the American menhaden. Rep. U.S. Comm. Fish., 5: 1-531. Gowanloch, J. N. 1933. Fishes and fishing in Louisiana. Bull. La. Conserv. Dep., 23: 638 pp. ----. 1940. Guard well our fish and game. La. Cons. Rev., 9 (1): 17-21. Gudger, E. W. 1916. The gaff-topsail, F elichthys felis, a sea catfish that carries its eggs in its mouth. Zoologica, 2 (5) : 123-158. Gunter, Gordon. 1945. Studies on marine fishes of Texas. Puhl. Inst. Mar. Sci., Univ. Texas, 1 (1):1-190. ----. 1952. Historical changes in the Mississippi River and the adjacent marine environment. Ibid. , 2 (2J: 119-139, 1 fig. ----.1954. Sagacity of a crab. Science, 120 (3109): 188-189. Giinther, A. 1880. An introduction to the study of fishes. Edinburgh. Hay, W. P. 1904. The life history of the blue crab (Callinectes sapidus). Rep. U.S. Comm. Fish., Appendix (1905 J : 395-413. Hiatt, R. W. 1947a. Food chains and the feeding cycle in Hawaiian fish ponds. Part I. The food and feeding habits of mullet (Mugil cephalus) , milkfish (Chanos chanos) , and the ten-pounder(Elops machnata). Trans. Amer. Fish. Soc., (19441 74: 250-261. ----. 1947b. Food-chains and the food cycle in Hawaiian fish ponds. Part II. Biotic inter­action. I bid., (1944) 74: 262-280. Hildebrand, Samuel F., and Louella E. Cable. 1930. Development and life history of fourteen teleostean fishes at Beaufort, North Carolina. Bull. U.S. Bur. Fish., 46 : 383-499. Hildebrand, Samuel F., and William C. Schroeder. 1928. The fishes of Chesapeake Bay. Bull. U.S. Bur. Fish., 43 (Part IJ : 1-366. Hildebrand, Samuel F., and Irving L. Towers. 1927. Annotated list of fishes collected in the vicinityof Greenwood, Mississippi, with descriptions of three new species. Bull. U.S. Bur. Fish., 43: 105-136. Hubbs, C. L. 1921. An ecological study of the fresh-water atherine fish Labidesthes sicculus. Ecology,2 ( 4) : 262-276. Hunt, Burton P. 1953. Food relationships between Florida spotted gar and other organisms in the Tamiami Canal, Dade County, Florida. Trans. Amer. Fish. Soc., 82 ( 1952 l : 13-33. Hussakof, L. 1914. Fishes swallowed by gar pike. Copeia, 1914 (2J : 11-12. Ishida, J. 1935. The stomach of Mugil cephalus and its digestive enzymes. Annot. zoo!. jap., 15: 182-189. Ivlev, V. S. 1945. The biological productivity of waters. (Translation by W. E. Ricker I. Adv. mod. Biol., Moscow, 19: 98-120. Jacot, Arthur Paul. 1920. Age, growth and scale characters of mullets, Mugil cephalus and Mugil curema. Trans. Amer. micr. Soc., 39: 199-299. Johnson, M. C., and J. R. Fielding. 1956. Propagation of the white shrimp Penaeus setiferus (Linn. I in captivity. Tulane Stud. Zoo!., 4 16): 175-190. Jordan, David S. 1905. A guide to the study of fishes. New York, Henry Holt & Co., xxvi+624 pp. Knapp, Frank T. 1949. Menhaden utilization in relation to the conservation of food and game fishes of the Texas Gulf coast. Trans. Amer. Fish. Soc., 79 : 137-144. Kuhne, E. R. 1939. A guide to the fishes of Tennessee and the mid-south. Tenn. Dept. Conserv. Bull., 124 pp. Lagler, Karl F., and Frances V. Hubbs. 1940. Food of the long-nosed gar ( Lepisosteus osseus oxyurus)and the bowfin (Amia calva) in southern Michigan. Copeia, 1940 (4 l: 239-241. Lagler, K. P., C. B. Obrecht, and G. V. Harry. 1942. The food and habits of gars I. 1894. On the food of the menhaden. Bull. U.S. Fish. Comm., 13 (1893): 113-126, 8 pis. Peterson, C. G. John, and P. Boysen Jensen. 1911. Valuation of the Sea. I. Animal Life of the Sea Bottom, its food and quantity. Beretn. dansk. biol. Sta., 1910, 20: 1-76, 6 pis., 3 charts. Pew, Patricia. 1954. Food and game fishes of the Texas coast. Bull. Tex. Game Fish., 33: 1-68. Pillay, T. V. R. 1953. Studies on the food. feeding habits, and alimentary tract of the grey mullet, Mugil tade Forskal. Proc. nat. Inst. Sci. India., 19: 777-827. Raney, Edward C. 1942. Alligator gar feeds on birds in Texas. Copeia, 1942 (l) : 50. Reid, George K., Jr. 1954. An ecological study of the Gulf of Mexico fishes in the vicinity of Cedar Key, Florida. Bull. Mar. Sci. Gulf & Caribb., 4 (l): 1-94. ----. 1955a. A summer study of the biology and ecology of East Bay, Texas. Part I. Intro­duction. description of area, methods, some aspects of the fish community, the invertebrate fauna. Tex. J . Sci., 7 (3): 316-343. ----. 1955h. A summer study of the biology and ecology of East Bay, Texas. Part II. The fish fauna of East Bay, the Gulf beach, and summary. /bid., 7 (4) : 430-453. ----. 1956. Ecological investigations in a disturbed Texas coastal estuary. Ibid.. 8 (3): 296-327. ----,A. Inglis, and H. D. Hoese. 1956. Summer foods of some fish species in East Bay, Texas. Southwestern Nat., l (3): 100-104. Rice, Lucille A. 1942. The food of seventeen Reelfoot Lake fishes in 1941. J. Tenn. Acad. Sci., 17 (l) : 4-13. Rimsky-Korsakoff, V. M. 1930. The food of certain fishes of the Lake Champlain watershed. (in) A biological survey of the Champlain watershed. Suppl. 19th Ann. Rept., N.Y. Cons. Dept., 139-145 (1929). Roelofs, Eugene W. 1954. Food studies of young sciaenid fishes, Micropogon and Leiostomus, from North Carolina. Copeia, 1954 (2) : 151-153. Russell. R. J. 1936. Physiography of the lower Mississippi River Delta., Geo!. Bull., New Orleans, (8) : 3-199. ----. 1938. Coast of Louisiana. Bull. Soc. belg. Geo!. Pal. Hydr., 57 (2). ----. 1940. Quaternary history of Louisiana. Bull. Geo!. Soc. Amer., 51: 1199-1234. Shira, A. F. 1917. Notes on the rearing, growth, and food of the channel catfish lctalurus punctatus. Trans. Amer. Fish. Soc., 46: 77-88. Smith. Hugh 2\I. 1907. The fishes of North Carolina. Raleigh, N.C., Geo!. and Econ., Sun. II :xi -7­453 pp. Steinmeyer. R. A. 1939. Bottom sediments of Lake Pontchartrain, Louisiana. Bull. Amer. Assn. Petrol. Geo!., 23 ( 1 I : 1-23. Suttkus. Royal D. 1954. Seasonal moYements and gro,rth of the Atlantic croaker (Micropogon undu­latus) along the Louisiana coast. Proc. Gulf & Caribb. Fish. Inst., 7th Ann. Sess. November, 1954: 1-7. ----. 1956. Early life history of the Gulf menhaden Brevoortia patronus in Louisiana. Trans. N . . \mer. Wild!. ConL 21st: 390-407, 20 figs. ----. R. l\L Darnell, and J. H. Darnell. 1953-55. Biological Study of Lake Pontchartrain. Research Progress Reports. a-j. (submitted to the Commercial Seafoods Division, Louisiana Wild­ life and Fisheries Commission, State of Louisiana. I (multilithed I. Teal, J. '.\I. 1957. Community metabolism in a temperate cold spring. Ecol. l\fonogr., 27 (3): 283-302. Thomson. J. 1\1. 1954. The 2\lugilidae of Australia and adjacent seas. Austr. J. Mar. Freshw. Res., 5: 70-130. Tiffany. L. H. 1920. The gizzard shad in relation to plants and game fishes. Trans. Amer. Fish. Soc., 50: 381-386. ----. 1921. Algal food of the young gizzard shad. Ohio J. Sci. 21 (41: 113-122. Verrill. ..\.. E. 1871. On the food and habits of some of our marine fishes. Amer. Nat., 5: 397-400. Viosca. Percy. 1954. Them bait stealin' sheepshead. Louisiana Consen-., 6 (71: 5-8, (April, 1954). Weed . . \lfred C. 1923. The alligator gar. Leafl. Field 1\lus. nat. Hist., 5: 1-16. Welsh. Wm. W .. and C. 1\1. Breder, Jr. 1923. Contributions to the life histories of the Sciaenidae of the eastern United States coast. Bull. U.S. Bur. Fish., 39: 141-201. Williams. A. B. 1955. A contribution to the life histories of commercial shrimps (Penaeidae) in North Carolina. Bull. 1\lar. Sci. Gulf & Caribb., 5 ( 2 I : 116-146. Wortman, Jacob L. 1882. Ichthyological papers by George Powers Dunbar, with a sketch of his life. Amer. Nat., 16: 381-388. ZoBell. C. E., and C. B. Feltham. 1938. Bacteria as food for certain marine invertebrates. J. Mar. Res.. 4: 312-327. ----. 1942. The bacterial flora of a marine mud flat as an ecological factor. Ecology, 23 (1): 69-78. Systematics and Zoogeography of the Clinid Fishes of the Subtribe Labrisomini Hubbs1 VICTOR G. SPRINGER2 Department of Zoology and Institute of Marine Science The University of Texas Port Aransas Contents INTRODUCTION ··········· · ..... .................................................. .. ·· ········· ············.. ··········...... 418 METHODS .................................................................................................................... . 419 Abbreviations ........................................................................................................ 421 SYSTEMATICS ······································-········-·························-· ········-······························ 421 Labrisomus Swainson .......................................................................................... 422 Key to the species of Labrisomus ................................................................ 424 Labrisomus (labrisomus) nuchipinnis (Quoy and Gaimard) .................... 425 Labrisomus (gobioclinus) haitiensis Beebe and Tee Van .... . ............. 429 Labrisomus (gobioclinus) bucciferus (Poey) ................... 430 Labrisomus (gobioclinus) guppyi (Norman) ............. ......................... 433 Labrisomus (gobioclinus) kalisherae (Jordan) ........................................ 434 Labrisomus (gobioclinus) gobio (Valenciennes) ............... . ............. 436 Labrisomus (gobioclinus) albuquerquensis (Fowler) ... 437 Labrisomus (brockius) Albigenys (Beebe and Tee Van) ....................... 437 Labrisomus (brockius) nigricinctus (Rivero) ........................................... 438 Malacoctenus Gill .................................. .............................................................. 439 Key to the Pacific Species of Malacoctenus ........................... ............... .... . 440 Key to the Atlantic species of Malacoctenus ...................... . 442 Malacoctenus boehlkei .............................................................................. . 443 Malacoctenus erdmani Smith ............... 444 Malacoctenus margaritae (Fowler).... . ............... . ............ 448 Malacoctenus margaritae margaritae (Fowler) ............. 449 Malacoctenus margaritae mexicanus ........................................... . 449 Malacoctenus gilli (Steindachner) ................ . 450 Malacoctenus aurolineatus Smith .................. . 454 Malacoctenus versicolor (Poey) .......................................................... . 455 Malacoctenus africanus Cadenat ..................................................... . 456 Malacoctenus gigas ...................................................................................... . 457 Malacoctenus delalandei (Valenciennes) ............................... .................. . 460 1 This paper is extracted from a dissertation submitted to the Graduate School of The University of Texas, in partial fulfillment of the requirements for the degree of Doctor of Philosophy. 2 Present address: State Board of Conservation Marine Laboratory, Maritime Base, Bayboro Harbor, St. Petersburg, Florida. Systematics and Zoogeography of the Clinid Fishes Malacoctenus afuerae (Hildebrand) --------------········ __ _____--------------461 111a!acoctenus afuerae afuerae (Hildebrand) 462 Ma!acoctenus afuerae multipunctatus ________ ___ ___ __________ 462 Malacoctenus hubbsi ____-· ···-_·······----_______ __ __ --·· ----------463 Malacoctenus hubbsi hubbsi . -------------······ ·············· ___ _ 465 Malacoctenus hubbsi polyporosus 468 Ma!acoctenus costaricanus __ ---··············--------·· 468 Malacoctenus zacae -------·--·-····-. 470 Malacoctenus zonogaster Heller and Snodgrass _ __ __ _ ___ _____ ___ 472 Malacoctenus zonifer (Jordan and Gilbert) 473 Malacoctenus zonifer zonifer (Jordan and Gilbert) .. . .. .. 474 Malacoctenus zonifer sudensis ------·-···············---------------·--·-·--474 Malacoctenus triangulatus -·····-·-·······--·---------------···------477 Malacoctenus ebisui -----·········-····-·-----------·------··-···-· 478 ZooGEOGRAPHY -------·-············-----479 Pacific 482 Atlantic 485 SUMMARY . 486 LITERATURE CITED ---······ ···········------------------------------········ 487 Introduction The subtribe Labrisomini Hubbs, 1952, is comprised of the genera labrisomus and Malacoctenus. These two genera contain the most recognized species and are the most widely distributed of the fishes of the family Clinidae. They are, in addition to this, two of the most common genera found inhabiting the shallow waters of the rocky coastlines and coral reefs of the American tropical regions, although not restricted to these areas. Despite and because of this commonness, many of the forms have remained undescribed; many have been described under numerous synonyms; and many have been misidenti­fied in the literature and museum collections. Until recently (Hubbs, 1953), neither of these genera had been subjected to a critical revision, although Longley (in Longley and Hildebrand, 1941) apparently was considering a revision of the Atlantic forms at the time of his death. This gap in ichthyological knowledge is all the more magnified when one considers that because of the sedentary nature of these forms, a knowledge of them would contribute significantly to the delimiting and understanding of the zoo­geography of the regions they occupy. Other members of the Clinidae, especially Para­clinus (Hubbs, 1952; Springer, 1955a) and Gibbonsia (Hubbs, 1952) , have already been demonstrated as signposts of zoogeographic consequence. The present study has been undertaken for the purpose of clarifying the somewhat involved systematics of the group, and the exposition of their zoogeographic importance in the regions they occupy. ACK:\'OWLEDGMENTS I wish to express my appreciation to Dr. Clark Hubbs, my supervisor during the course of this study, who gave freely of his own limited time. His efforts on my behalf, Systematics and Zoogeography of the Clinid Fishes encouraging, culling, and directing, were of the greatest importance In bringing this work to completion. To those individuals listed below I wish to express my gratitude for their advice. cooperation, and hospitality. Dr. J. Bohlke and Mr. J. L. Hamell, Academy of Natural Sciences of Philadelphia; Drs. L. P. Schultz, E. A. Lachner, and Mr. R. H. Kanazawa, United States National Museum; Drs. V. Walters and W. N. Tavolga, American Museum of Natural History; Dr. J. Tee-Van, New York Zoological Society; Dr. R. M. Bailey, University of Michigan; Mrs. M. Dick, Harvard University; Prof. L. R. Rivas, University of Miami; Dr. G. S. Myers and Miss M. H. Storey, Stanford University; Dr. B. W. Walker and Mr. R. Rosen­blatt, l'niversity of California at Los Angeles; Dr. J. R. Pfaff, Universitetets Zoologiske Museum, Copenhagen; Dr. P. Kiihsbauer, Naturhistorisches Museum, Vienna; Mr. L. P. Woods, Chicago Natural History Museum; Mr. D. W. Tucker, British Museum (Natural History); Drs. W. F. Blair, J. A. Wilson, C. P. Oliver, W. S. Stone, J. Myers, and R. K. Selander, Mr. R. W. Axtell, and Mr. F. E. Potter, The University of Texas; Mr. T. Beim­ler, Brownsville, Texas; Dr. D. Pettus, Colorado Agricultural and Mechanical College; Dr. R. K. Strawn, Lamar Technological Institute; Dr. R. W. Harrington, Jr., Florida State Board of Health, Dr. Gordon Gunter, Gulf Coast Research Laboratory; Com­mander and Mrs. R. M. Hirsch, Hyattsville, Maryland; Mr. and Mrs. S. Weinberg, Philadelphia, Pennsylvania; and Mr. and Mrs. M. Gruschka, New York, New York. Completion of this study was made possible by a grant from the Southern Fellowships Fund. Methods Measurements were made with a pair of finely sharpened dividers and estimated to the nearest tenth millimeter on a ruler graduated in half-millimeters, or divided into other body parts. Measurements of body parts are recorded as per cent of the standard length. STANDARD LENGTH was measured from the median anterior tip of the upper lip, with the mouth in normal closed position (i.e. the upper jaw not protracted) to a point on the midlateral posterior margin of the hypural vertebra. LENGTH is always standard length. HEAD LENGTH, SNOUT LENGTH, and MAXILLARY LENGTH were meas­ ured from the same anterior point as the standard length. Head length was measured to the most posterior membranous margin of the opercle, but not including the attached branchiostegal membrane; snout length was measured to the nearest point on the anterior bony orbital margin; and maxillary length to the most posterior point of extension of the maxillary bone. BOI\Y OHBJTAL DIAMETEB is the greatest horizontal bony orbital diameter. DOHSAL SPINES were measured by pressing one point of the dividers in the posterior angle formed by the spine and the dorsal body contour, and extending the other point to the tip of the spine; any curvature of the spines was ignored. SHORT­ EST PELVIC HAY LENGTH was measured from thi> base of the ray to its tip with the ray in a straight position. This mPasurement was divided into the longest ray using tht· level of the tip of the shortest ray as the first point of reference on the longest ray. The INTERSPACE (Fig. 1) between the nuchal cirri bases was measured as the distance between the two most median points of attachment of the cirri bases, excluding the medianly overlapping edges of the laminae supporting the cirri. This measurement was then dividt>d into a single cirri base. o\4-~~~MENTAL~~~-r~· SYMPHYSIAL NUCHAL CIRRI BASE FIG. 2. Disposition and Nomenclature of FrG. l. Position of lnterspace between :\lucous Pores on the Ventral Side of the Head Nurhal Cirri Bases. in Labrisominids. The last two DORSAL and A:\AL FI.'.\ RAYS were counted separately and not recorded as one as was done by Hubbs ( 1952, 1953). Dorsal fin elements are reported as formulae actually found; hence, if dorsal counts of XIX.19; XIX,20; XIX,21; and XX,19 were made, they are reported as XIX,19 to 21 or XX,19 (instead of XIX or XX,19 to 21, which can be misinterpreted to indicate counts of XX,20 or XX,21). All of the PECTORAL rays and all segmented CACDAL rays were counted. For the latter, an almost undeviating count of 13 is obtained; this technique obviates subjective dis­cussions as to how to count reduced rays at the dorsal and ventral caudal margins. LATERAL Lll'IE SCALES were counted as all scales in the series whether bearing mucous tubes or not. The last scale of the series on fishes which have scales with tubes on the caudal fin, is the last scale with a tube. When the lateral line was incomplete, the most posterior scale originating lateral to the hypural vertebra was the last scale counted. Scales above the anterior lateral line were counted as the number of scales touched by a vertical line passing from the lateral line to the base of the fifth dorsal spine. GILL RAKERS were counted as total elements, including rudiments, on the first gill arch. SYMPHYSIAL PORES (Fig. 2) were counted while directing a stream of air in the symphysial region. All pores posterior to the two mental pores (just posterior to the ventral lip folds), medial and anterior to the series of interopercular pores, and anterior to the fold of the fused branchiostegal membranes, were counted. The BREAST was considered to be that portion of the ventral area of a specimen posterior to a line across the posterior margin of the bases of the pelvic fins and anterior to a line extending across the body between the distal margins of the fleshy bases of the pectoral fins. The VEl\TER extends from the posterior margin of the breast pos· teriorly to the anus. Counts and measurements appearing in parentheses are modes. As is often true when very large series of specimens are examined, unique specimens come to light. In the present study the very rare exceptions where one or three anal spines were present; one pelvic fin was absent; soft dorsal rays occurred between the spines; or a ctenoid scale appeared in the lateral line, are not reported in the descriptions. The descriptions are arranged so as to make possible a comparison of the species. These include the characters considered most useful for identification. The synonymies are an attempt to include all publications containing reference to the particular taxon concerned. It was not always possible to examine the material discussed in these publications, or to tell with certainty from the discussions to what taxa the names referred. All such citations are marked with an asterisk, and are included in the synonymy of that species most probably correct. Unless stated otherwise, coloration is descrihed from pre;:ervt>d specimens. ABBR HIATIO:\S Below are given the abbre\·iations used in the lists of material examined. All l "CLA numbers are colJection numbers and not specimen or lot numhers. AMNH-American Museum of '.\atural Histon-. :\ew York ; A:\SP-Academv of Natural Sciences of Philadelphia; BM'.\H-British \foseum 1:\atural Histon·). London: CAS---California Academy of Sciences, San Franeisco: C:\HM-Chic~go :\aturai History Museum, Chicago; L\fSC-lnstitute of Marine Science Collections, Port Aransas. Texas (specimens do not bear numbers) ; MCZ-:'.\fo,.eum of Comparati\-e Zoology, Harvard University, Cambridge. Massachusetts; :\YZS-:\ew York Zoological Society. New York; SU-Standford Cniversity :\atural History :'.\lm:eum. Stanford. California; TNHC-Texas Natural History Collection, l -niversity of Texas. Austin, Texas; l "CLA­ University of California at Los Angeles; CM-l-niwrsity of Miami lchthyological Collection, Coral Gables. Florida; CMMZ-l"niwrsity of :\lichigan, Must>um of Zoology, Ann Arbor, Michigan; CS:\'.\1-l-nited States :i\ational :'.\luseum, Washington. D.C.; _ UZMK--Vniversitetets Zoologiske Museum, Copenhagen. Systematics As a result of considerable parallel and/ or convergent ernlution within the genera Labrisomus and Malacoctenus, it is difficult to establish mutually exclusive generic limits for any character. The most constant differences are found in the size and dentition of the jaws. Examination of stained preparations of several species of both genera revealed no striking differences. Some difficulty may be encountered in the generic allocation of specimens of l. nigricinctus, but the following key will separate all !'pecimens of labrisomus and Malacoctenus that I have examined. Maxillary length 10.2 to 19.0 percent of standard length I usually more than 13 and rarely less than 12 percent l ; preorbital barely or not sheathing posterodorsal portion of maxillary when mouth is closed; more than two rows of small teeth always present behind large outer row in at least upper jaw (those in lower jaw frequently obscured by fleshy folds) ; teeth present or absent on palatines (if absent, scales are always present on upper margin of opercle and pectoral base) ; one or more scales usually present on upper margin of opercle (they may be deepl y imbedded, but if absent, teeth are always present on palatines) ; interspace between bases of nuchal cirri contained less than three times in a single base; membranes of fins with or without scales _ ----------__ ___ Labrisomus Maxillary length 7.9 to 11.2 percent of standard length (usually 8.5 to 9.5 and rarely more than 10.l percent); preorbital almost completely sheathing entire maxillary when mouth is completely closed; teeth uniserial in each jaw (four or five small teeth behind outer row in each jaw: M. erdmani ; teeth biserial in each jaw: M. macropus ; several rows of small teeth behind large outer row Systematics and Zoogeography of the Clinid Fishes in each jaw: M. boehlkei) ; teeth never present on palatines; scales never present on upper margin of opercle; interspace between bases of nuchal cirri frequently contained more than three times in a single base; membranes of fins never scaled _________ ____________ -----------------------------------___ _____ __ ___ _____ Malacoctenus LABRISOMUS Swainson Labrisomus Swainson, 1839: 277 (type Clinus pectinifer Valenciennes in Cuvier and Valenciennes, 1836, by original designation) . Lepisoma De Kay, 1842: 41 {type Lepisoma cirrhosum De Kay by original designa­tion). Gobioclinus Gill, l 960b: 103 (type Clinus gobio Valenciennes in Cuvier and Valencien­nes, 1836, a mixture; subsequent designation of lectotype by Longley, in Longley and Hildebrand, 1941). Labrosomus Gill, 1860b: 105 (substitute spelling for Labrisomus Swainson). Ericteis Jordan, 1904: 543 (type Ericteis kalisherae Jordan by original designation). Odontoclinus Reid, 1935: 164 (type Odontoclinus dendriticus Reid by original designa­ tion) . Ctenichthys Rivero, 1936: 69 (type Ctenichthys interrupta by original designation). Brockius Hubbs, 1953: 120, a subgenus (type Labrisomus (Brockius) striatus by orig­ inal designation) . Labrisomus, Gobioclinus, and Brockius are retained in this study as subgenera. Spe· cialization is considered to have progressed in the same direction as the sequence of these names. Within each of these subgenera the relationships are not always clear. Labrisomus philippii is considered the most primitive member of the genus based upon its large size, high lateral line and pectoral counts, and increased dentition. Specializa· tion within the genus Labrisomus is considered to have progressed in the direction of decreased size, dentition, and scalation. I feel certain that some systematists would relegate each of the above subgenera to the rank of genus, as the differences separating them are trenchant; and if some other systematist revised this group and utilized these names on the generic level, I would find no quarrel with him. The decision for the present usage of these names is based upon similarities and practicality. I believe that the inclusion of all the forms in the genus Labrisomus serves to show the close relationships of these forms, while at the same time decreasing the number of names that other workers must utilize. The various phyletic lines are brought out by the use of the subgenus and so the phylogeny of the group is not clouded. Ecologists, physiologists, and others who are usually concerned only with an identifying name, and as few names as possible, will be somewhat relieved. Past usage also makes the comprehensive name Labrisomus desirable. All major revisions and geographic studies have incorporated all the forms included here in that genus. Hubbs (1953 I has recently revised the Pacific species of Labrisomus. Although these taxa are included in the key below, they are not described in the body of the paper. Description. Fin rays: dorsal XVII to XXII, 10 to 13; anal IL 16 to 22; caudal 12 or 13 I 13) ; pectorals 13 to 16; pelvics I, 3. Lateral line 40 to 74; all scales of posterior arc of lateral line with anterior pore ex· posed (Brockius I ; not more than posterior half of scales of posterior arc of lateral line with anterior pore exposed (Labrisomus and Gobioclinus) . Cheek scaled (striatus, Systematics and Zoogeography of the Clinid Fishes socorroensis, and philippii) or naked. Dorsal margin of opercle with or without scales. Fin membranes naked to scaled, except those of pelvics. Prepectoral area and breast completely scaled, except in young. Jaws with an outer row of large conical teeth, behind which are found two or more irregular rows of smaller teeth. Smaller teeth frequently obscured by fleshy folds (La­brisomus and Gobioclinus) or clearly represented (Brockius). Most posterior of smaller teeth sometimes larger than those anterior {Gobioclinus, especially dendriticus and guppyi) or uniform in size (Labrisomus and Brockius). Vomerine teeth present. Pala· tine teeth present (Gobioclinus and Labrisomus, except xanti and socorroensis) ; same size as vomerine teeth or smaller (Labrisomus I; some noticeably larger than vomerine teeth (Gobioclinus); or palatine teeth absent (Brockius). Gill-rakers 8 to 15 on first arch. Origin of dorsal fin in region above opercle. Length of spinous dorsal base longer than that of soft dorsal. Penultimate dorsal spine about same size as ultimate (Labriso­mus) ; smaller than ultimate (Brockius and Gobioclinus). Soft dorsal higher than spi· nous portion, except L. dendriticus in which the anterior two spines may approach the soft dorsal height. Anal fin base shorter than dorsal; beginning about midventrally. First anal spine shorter than second, which is shorter than all but last few soft rays. Second to fourth segmented caudal ray from the bottom the longest; shorter than longest pectoral ray. Pectoral rays increasing in thickness ventrally; fourth to sixth from the bottom longest. Pelvic spine visible only by dissection or x-ray. Inner pelvic ray the shortest, con­tained slightly more than 1 to 3% times in middle ray, which is the longest. Cirri present on tube of anterior nostriL above eye, and on nape, increasing in num­ber with age. Interspace between nuchal cirri bases contained less than three times in a nuchal cirri base. Maxillary length 10.2 to 19.0 percent of standard length (usually more than 13 and rarely less than 12 percent). Posterodorsal portion of maxillary rarely sheathed by preorbital. Anterior margin of cleithrum never with a hook dorsally. Sexual dimorphism. Males have a papilla just posterior to the anus containing a com­ mon opening for the urinogenital system. Females have an enlarged gential aperture posterior to the anus and a fleshy rugose knob posterior to the genital aperture con­ taining several small papillae, one of which contains an opening from the kidneys. This latter opening is extremely difficult to find, except in large specimens. External genitalia in young of both sexes are similar and malelike. Secondary sexual dimorphism is exhibited in most species by a larger average size for the upper jaw in males and by increased pigmentation of the females. This pigmenta­ tion usually takes the form of heavy spotting and barring on the body and fins, and in L. nuchipinnis, a highly reticulate pattern is sometimes formed, especially on the head. In contrast, the males are usually quite uniform in color. Young specimens of both sexes are typically like the female in color, with the pattern changing gradually in males as they approach maturity. KEY TO THE SPECIES OF LABRISOMUS" A. Palatine teeth present, several of which are considerably larger than those on vomer ------------------------------------------------------------------------··-----------------··---Gobioclinus B AA. Palatine teeth, when present, same size as, or smaller than those on vomer _________J B. Lateral line scales more than 57 ----------------(Pacific) L. (Gobioclinus) dendriticus BB. Lateral line scales fewer than 55 ----------------------------______ ----------------------(Atlantic) C C. Length of shortest pelvic ray contained more than 2 times in longest; pectoral rays usually 14 .................. _..... ___ _____ .... L. (Gobioclinus) haitiensis (p. 429) CC. Length of shortest pelvic ray contained less than 2 times in longest; pectoral rays usually 13 ----------------------------------------- ------------------------------------------------------------------D D. Peritoneum uniforml y dusky gray to black; dorsal spines usually 20; anal rays usually 20; lateral line scales 45 to 48; first dorsal spine longer than fifth -------­-------..... .. ___________ _ l. (Gobioclinus) bucciferus (p. 430) DD. Peritoneum white with scattered large chromatophores; dorsal spines usually 19; anal rays usually 19; lateral line scales 48 to 53; first dorsal spine shorter than fifth --------------------------------------------------------------------------E (specimens over 40 mm.) --------------------------------------------------------------------------------G (specimens 28 to 40 mm.) E. Symphysial pores always 2 .. .. _ _ _________ L. (Gobioclinus ) kalisherae (p. 434) . EE. Symphysial pores more than 2 --------------------------------------------------------------------------------F F. A well-developed opercular ocellus present in all specimens; largest specimen, of 100 examined, 88 mm. ___________ ___ .. L. IGobioclinus) guppyi (p. 433) FF. An opercular ocellus never present; largest specimen, of 216 examined, 49.2 mm. ....... .. -----------------------________ _________ L. (Gobioclinus) gobio (p. 436) G. An opercular ocellus not present ----------------------------------------------------------------------------H GG. An opercular ocellus present ---------------------------------------------------------------------------------I H. Penultimate dorsal spine rarely more than 8 percent of standard length; anal and pectoral fins never heavily spotted ................ l. (Gobioclinus) gobio (p. 436) HH. Penultimate dorsal spine rarely less than 8 percent of standard length; anal and pectoral fins often heavily spotted _________ __ . l. (Gobioclinus) kalisherae (p. 434) I. Gill-rakers 12 to 15, rarely less than 13 ____ ________L. (Gobioclinus ) guppyi (p. 433) II. Gill-rakers 10 to 13, rarely more than 12 __ l. (Gobioclinus) kalisherae (p. 434) J. Scales fewer than 45; all scales on posterior segment of lateral line with anterior pore of canal exposed ____-------------------------------------------------------------------Brockius K JJ. Scales more than 55; at least half of scales on posterior segment of lateral line with anterior pore of canal covered by scale to the anterior ____________ Labrisomus M K. Scales present on cheek anterior to those at dorsal margin of opercle; length of shortest pelvic ray contained less than 1 4/ 5 times in longest ---------------------------­--------------------------------------------------------------------------(Pacific) l. (Brockius) striatus KK. No scales present on cheek anterior to those at dorsal margin of opercle; length of shortest pelvic ray contained 1 4/ 5 to 2% times in longest ________ (Atlantic) L L. A well-developed, black ocellus on opercle of all specimens 12 mm. or longer in standard length; body coloration un:formly white with unmarked fins (males) or body with about nine distinct, dark bands extending onto fins (females and young); gill-rakers 9 to 12 ____________ _L. IBrockius) nigricinctus (p. 438) " Pinto (1957. Bo!. Mus. nae. Rio de J. 163: 1-15) has described a new species of Labrisomus (L.trindadensis) from Trindade Is. off Brazil. LL. No black ocellus on opercle; body coloration uniformly dark with indistinct indi­cations of bands on body; fins dark and or spotted; gill-rakers 13 or H -------­------------------------_ ___ _______ _L. (Brockius) albigenys l p. 437) M. A distinct black spot or ocellus present on opercle; palatine teeth present ____________ -------------_________ ____ (Atlantic l L. l labrisomus l nuchipinnis (p. 425 l MM. No black spot or ocellus present on opercle; palatine teeth present or absent -------­--------------------------------------------------------------------------------------------------------------l Pacific) N N. Palatine teeth absent ____________________ --------------------------------______________ -------------------------0 NN. Palatine teeth present ________ ------------------------------------__ ___________________ ------------------------P 0. Many scales on cheek and along entire dorsal part of opercle ---------------------------------­------------------------------------------------------------------------------L. \Labrisomus l socorroensis4 --------------------------------------------------------_______ _____ ____________ __ ___ _____ l. (labrisomus) xanti P. First dorsal spine longer than opercle depth·' ; most conspicuous color marking a dark humeral spot ----------------------------------------------------L. (labrisomus) wigginsi'' PP. First dorsal spine shorter than opercle depth; no humeral spot present ____________ Q Q. Fewer than 63 scales in lateral Ene ; caudal peduncle depth more than 9.1 percent of standard length ________ -----------------------------------------------L. l labrisomus) jenkinsi QQ. More than 63 scales in lateral line; caudal peduncle depth less than 9.1 percent of standard length (except in L. philippii l _____--------------------------------------------------R R. Pectoral rays usually 15 or 16; dorsal spines usually 19 or 20: lateral line scales 70 to 74 (rarely 68 or 69) ; branch or preopercle canal extends less than one­half distance across opercle __------------------------------------L. \Labrisomus) philippii RR. Pectoral rays usually 14; dorsal spines usually 17 or 18; lateral line scales 65 to 69; branch of preopercle canal extends at least one-half distance across opercle ------------------------------------------------------------------------------L. l Labrisomus) multiporosus LABRISOMUS (LABRISOMUSI NlXHIPINN/S \Quoy and Gaimard) (Plate I Fig. 1) Clinus nuchipinnis Quoy and Gaimard, 1824: 255; Gunther, 186lb: 262; Steindachner, 1867: 47; Gunther, 1868: 389 l in partl ; Kner, 1868: 336; Peters. 1877: 248*; Rochebrune, 1882: 116 (misspelled nuctipinnis); Jordan and Gilbert. 1882a: 762; Vinciguerra, 1883: 616*; Guimaraes, 1884: 21*; Moquard. 1889: 40 !in part); CockerelL 1892"; Vinciguerra, 1893: 322"; Osorio, 1896: 63*: Perugia, 1896: 19*; Osorio, 1898: 198*; Boulenger, 1905: 188*; Metzelaar. 1919: 154; Roule and Angel, 1930: 104. 4 Hubbs (1953 l described this species as L. soccorroensis, naming it for Socorro Island in the Revilla Gigedo Islands. As such the original spelling must be considered in error and emended to the form in the above key. 00. Cheeks naked and fewer than 30 scales in patch on upper margin of opercle __________ 5 fo his 195.) key, Hubbs used the term ··opercle length··. a horizontal measurement. By using his original data I have ascertained that he meant ··opercle depth"', a vertical measurement. This change is necessary if his key is used. 6 Hubbs placed this species in the subgenus Odontoclinus_ which I consider the same as Gobioclinus. Labrisomus wigginsi has the vomerine and palatine teeth of equal size; 14 or more pectoral rays: the villiform teeth behind the large outer row in each jaw of about uniform size: and the penultimate dorsal spine not noticeably shorter than the ultimate. All of these are typical of Labrisomus but not of Gobioclinus. Systematics and Zoogeography of the Clinid Fishes Blennius herminier Le Sueur, 1825: 361. Clinus pectinifer Valenciennes (in Cuvier and Valenciennes), 1836: 374; Guerin-Mene­ville, 1829-1844 (date uncertain): Plate 38, Figure 2; Schomburgk, 1848: 16*; Castelnau, 1855: 26; Dumeril, 1858: 263"; Miiller, 1864: 631; Poey, 1866: 333. Clinus capillatus Valenciennes (in Cuvier and Valenciennes), 1836: 377; Schomburgk, 1848: 16*; Poey, 1866: 333. Clinus herminier, Valenciennes (in Cuvier and Valenciennes), 1836: 379; Giinther, 186lb: 264 (spelled herminieri). Clinus canariensis Valenciennes, 1839: 60. Labrisomus pectinifer, Swainson, 1839: 277; Cope, 1871 : 472. Labrisomus capillatus, Swainson, 1839: 277. Lepisoma cirrhosum De Kay, 1842: 41. Clinus fasciatus Castelnau, 1855: 26. Labrosomus pectinifer, Gill, 1860a: 21; Gill, 1860b: 105; Poey, 1861: 381. Labrosomus fasciatus, Gill, 1860b: 106. Labrosomus capillatus, Gill, 1860b: 107; Poey, 1861: 381. Labrosomus herminieri, Gill, 1860b: 108. Labrosomus nuchipinnis, Poey, 1868: 398; Goode, 1876: 28; Goode, 1877a: 3 (mis­ spelled Labrosomuc) ; Goode, 1877b: 291 ; Goode and Bean, 1883: 236; Jordan, 1887b: 908. Clinus pedatipennis Rochebrune, 1880: 165; Rochebrune, 1882: 116. Labrisomus microlepidotus Poey, 1880: 246; Jordan, 1887a: 599; Jordan and Ever­ mann, 1896: 468; Jordan and Evermann, 1898: 2363; Jordan, Evermann, and Clark, 1930: 459. Labrisomus nuchipinnis, Poey, 1880: 246; Smith, 1885: 553; Jordan, 1887a: 599; Lee, 1889: 669; Jordan and Evermann, 1896: 468; Jordan and Rutter, 1897: 133; Jordan and Gunn, 1898: 346*; Jordan and Evermann, 1898: 2362; Fowler, 1899: 119; Evermann and Marsh, 1900b: 311; Gilbert, 1900: 179 (misspelled nuchipin· nus); Evermann and Kendall, 1900: 93; Barbour, 1905: 130; Bean, 1905: 319; Bean 1906b: 84; Mclndoo, 1906: 488*; Rosen, 1911: 66 (in part); Starks, 1913: 74; Nichols and Murphy, 1914: 266*; Fowler, 1915: 49; Fowler, 1916: 251; Fowler, 1920: 145 (misspelled nuchipinuis); Nichols, 192la: 24; Metzelaar, 1922: 135; Breder, 1925: 157; Breder, 1927: 86*; Meek and Hildebrand, 1928: 936; Fowler, 1928: 467; Beebe and Tee Van, 1928: 231; Breder, 1929: 270; Fowler, 1930: 275; Jordan, Evermann, and Clark, 1930: 459; Nichols, 1930: 377; Fowler, 1931: 403¥'; Longley, 1932: 300; Beebe and Tee Van, 1933: 226; Beebe and Hollister, 1935 : 220*; Fowler, 1936: 1037 (in part); Butsch, 1939; 30*: Fowler, 1940: 794; Schmitt and Schultz, 1940: 9; Longley and Hildebrand, 1941: 249; Herre, 1942: 16; Fowler, 1942a: 178; Fowler, 1942b: 77*; Fowler, 1942c: 12; Fowler, 1944: 141 & 472; Fowler, 1945: 327*; Irvine, 1947: 193; Manter, 1947: 300*; Baughman, 1947: 280; Schultz, 1949: 178; Baughman, 1950: 253; Steinitz, 1950: 342; Fowler, 1951: 31; Fowler, 1952: 105; Fowler, 1953: 70; Hubbs, 1953: 115 & 120; Springer and Bullis, 1956: 97; Erdman, 1957: 321. Labrisomus herminier, Jordan, 1887a: 599; Jordan and Evermann, 1896: 468; Jordan and Evermann, 1898: 2361; Jordan, Evermann, and Clark, 1930: 459. Lepisoma nuchipinnis, Jordan and Thompson, 1905: 254 (misspelled nuchipinne); Ribeiro, 1915: 9. labrisomus lentiginosus Bean, 1906a: 30; Bean, 1906b: 83; Fowler, 1930: 275; Jordan, Evermann, and Clark, 1930: 459; Beebe and Tee Van, 1933: 277. labrisomus canariensis, Starks, 1913: 74; Hubbs, 1953: 115 & 120. Clinus (labrisomus) nuchipirmis,Arambourg, 1921: 1244; Cadanet, 1950: 271. labrisomus fasciatus, Fowler, 1942a: 178. labrisomus bahamensis Fowler, 1947: 8. Description. Fin rays: dorsal XVII, 12 or 13; XVIII, 10 to 13; XIX, 11 or 12; or XX, 12 (XVIII, 12 in over 90 percent of specimens) ; anal II, 17 to 19 (18 or 19) ; pectorals 13 to 15 (14). Lateral line 64 to 69 (65 to 67). Gill-rakers 10 to 13 (11or12). Head usually 28 to 31 percent of standard length in specimens over 35 mm. Bony orbital diameter less than snout length in specimens over 40 mm. First dorsal spine usually less than 9 percent of standard length; rarely longer (usually shorter) than any other dorsal spine in specimens over 70 mm.; never longer than sixth through last dorsal spines. Length of shortest pelvic ray slightly more than 1 to 11h times in longest. lnterspace between nuchal cirri bases contained 415 to 22/.'3 times in a single base. Maxil­ lary 12. 7 to 17.4 percent of standard length. Frequently reaching lengths over 100 mm.; largest specimen examined: 176 mm. Teeth present on palatines. Scales on upper margin of opercle, prepectoral area, and venter, varying with geographic distribution and size of specimen. In specimens from Natal, Brazil, scales are present on these portions of the body at less than 35 mm., and are obvious at all sizes. In specimens from Panama, scales are developed in these regions sometime between 29 and 42 mm., and are still apparent in specimens of at least 100 mm. In specimens from Marineland, Florida, scales in these body areas do not develop until sometime between 39 and 56 mm., and in large specimens the opercular patch of scales has usu­ally disappeared. Specimens from Vero Beach, Florida, may or may not have scales in these rt;gions at 38 mm. The disappearance of the opercular patch of scales is also true of large specimens from Bermuda. The largest specimens are known from Florida and Texas. Adult type color pattern is not recognizable until a length of about 20 mm. is attained. Specimens smaller than 18 mm. may be pelagic (planktonic) and are uniformly light with very few melanophores. Specimens of 19 mm., or less, taken in a plankton net off Bimini, Bahamas, are slender and light, as was a 19.3 mm. specimen from off the coast of Colombia. Of two specimens of this same size from Puerto Rico taken in a rocky area, both are greater in depth than those mentioned above; one had acquired adult type coloration, and the other showed indications of color pattern. An opercle blotch or ocellus is always present in specimens over 25 mm., and most specimens have a blotch on the anterior two or three dorsal spines ; this latter blotch decreases in distinctness with size above 120 mm. Females have a more spotted or reticulate pattern than males, and bars on the body are better exhibited. In any large series of specimens there is considerable variability, and color patterns of the opposite sex may occur in particular specimens. In life, Texas specimens may vary from a dark violet brown to a uniformly light cream with dusky stripes. The following description is based on preserved specimens from Natal, Brazil. Body with five or six irregular bands, only three or four of which are readily distin· guishable. Bars usually mottled (females) or solid (males), extending well onto the dorsal and anal (young) or only faintly so (adults) . Area between bars lighter, spotted, Systematics and Zoogeography of the Clinid Fishes or with a reticulate pattern (females), or uniform (males). Fins, especially the caudal, heavily spotted (females) or usually uniform (males). Opercle with a distinct black ocellus or a cliffuse splotch. Anterior dorsal spines with a similar mark. Usually two or more dark lines radiating posteriorly from each eye, with light bars between them. Of ten African specimens seen, the color pattern in all had deteriorated, but appeared to agree with the above description. Discussion. Although specimens from the African coast only as far south as the Gold Coast were examined, literature reference (Osorio, 1898) places the southern limit for this species in the eastern Atlantic at Annobon Island off the coast of French Equa­torial Africa. Significant variation among the populations was found only in counts of the anal rays (Table 1). I am unable to explain these differences, but it is interesting to note TABLE 1 Geographic Distribution of Total Anal Elements in Labrisomus nuchipinnis Number of Anal Elemenls 19 20 21 Number of Anal Elemenls 19 20 21 Florida Barbados 30 13 St. Augustine & Marineland 3 43 Grenada 5 4 Daytona Beach 1 Tobago 12 4 New Smyrna 1 2 Trinidad 2 Brevard County 2 Venezuela 3 1 Vero Beach 2 30 2 Colombia 1 1 Ft. Pierce 1 Panama 2 91 39 West Palm Beach 3 Albuquerque Cay 1 Ft. Lauderdale 1 Old Providence Island 1 Tavernier (Keys) 1 Swan Island Spanish Harbor (Keys) 1 Mexico Tortugas (Keys) 2 6 Campeche Banks 4 6 Vera Cruz 2 4 Bahama Islands Texas New Providence (and emirons) 13 5 Port Isabel 13 1 Bimini 2 2 Port Aransas 13 Andros 2 Abaco 1 1 Brazil Turks 3 Isla Fernando Noronha 1 1 Cay Sal Bank 5 1 Natal 2 35 65 Cuba 34 21 Porto Seguro Rio de Janeiro 2 14 Jamaica 1 2 Haiti 4 4 Bermuda Islands 42 113 Dominica 1 3 Puerto Rico 16 11 Eastern Atlantic Virgin Islands 13 3 Madeira Islands 3 St. Martins St. Lucia St. Vincent 4 4 3 3 Canary Islands Cape Verde Islands Gold Coast 2 I 4 the grouping of the regions on the basis of anal ray counts. Florida, Texas, and the eastern Atlantic fall into one group; the Bahamas, Antilles, and Panama in a second group; and Natal, Brazil, and the Bermudas into a third. Labrisomus bahamensis Fowler is included in the synonymy of this species for the first time. Fowler's species is based on a very young specimen which has the typical dor­sal, anal, pectoral, and lateral line counts of L. nuchipinnis. Relationship. This species is most closely related to L. multiporosus (Pacific) from which it differs in the possession of an opercular ocellus and a larger number of gill· rakers on the first arch. Systematics and Zoogeography of the Clinid Fishes Material. Eight hundred and eighty-eight specimens were examined. One hundred and three from Florida: ANSP 30972, 33007-08, 55908, CNHM 50n7, Slf 8188. TNHC 3877-84, UM 881, llMMZ 136494, 154825, 158658, l 'SNM 62653, 62628, 88128-29. 91417, 116814, R. W. Harrington (personal collection) H56-24, H55-9C; 63 from tht> Bahamas: AMNH 7227, 7346, ANSP 74648-53, 74732-33, 72181, 71 n4 (holotypt> of L. bahamensis) , BMNH 1936.6.5.24, UM 57, 58, 537, 882, 1296, University of Miami Marine Lab (no number), USNM 38379; 58 from Cuba: MCZ 12491-98, 12566, 12607, MCZ ! no number), UM 56, llSNM 478:1, Bm9. 24780--81, 24948, 24950, :13099, ;1:H04, :13109, :B ll8, :Bl27, 3:1129, ;1742;1, :17426, :n574, 1:13013, 154868, 37572 (holotype of /,. microlepidotus); 3 from Jamaica: ANSP 18637, BMNH 97.7.1.3, Sl l 4967; 8 from Haiti: MCZ 12609-10, lfSNM 133748; 4 from Dominican Hepublic: ANSP 165B, MCZ 34695; 29 from Puerto Rico: AMNH 2074, ANSP 12577, 236ll, 28852, SU 8247, UMMZ 160623, 17ll60, 17ll64, 171208, 171213, 171215, 171220, 171242, 17ll49, USNM 50197, 63046, nS:14, ll7438, 126187; 17 from the Virgin Islands: ANSP 74661-62, 10589, BMNH 63.8.7.168-9, USNM 120488, UZMK 12, 45-51, 65, 2; 7 from St. Martins: ANSP 10591-7; 4 from St. Lucia: Sl' 4257, USNM 42190; :1 from St. Vincent: BMNH (no number), MCZ 26101; 49 from Barbados: ANSP 74464, 74466, 74654-57, SU 32034, l 'SNM 5924; 9 from Grenada: ANSP 52494-99, 52500-02; 16 from Tobago: BMNH 1920.12.174-83, UZMK P76933; 2 from Trinidad: ANSP 74688, BMNH 1902.1.25.10; 4 from Venezuela: USNM 78251-2, 123173-74; 3 from Colombia: USNM 49077, 134905, 163276; 136 from Panama: CNHM 18520-46, 18548-62, SU 17628, 17870, 18605, l 'CLA W53-268, USNM 81908-16, 1487ll-12; 1 from Albu­ querque Cay: ANSP 74685; 2 from Old Providence Island: ANSP 74684, USNM 107114; 1 from Swan Island: BMNH 1908.7.6.25; 17 from Mexico: CNHM 46654. IMSC (no numbers), t:SNM 37793; 23 from Texas: TNHC 3885, 4313, 4364, UMMZ 166112; 2 from Isla Fernando Noronha: BMNH 88.1.19.91; 151 from Brazil: AMNH 3809, BMNH 1923.7.30.3.3-4, CNHM 59060, MCZ 4638, 12490, and one skeleton MCZ (no number), SU 22150, l'SNM 83242, UZMK 4, 6; 159 from Bermuda: AMNH 9ml, ANSP 39842-43, 74658-60, BMNH 1872.8.28.54-56, 72.8.28.56, 1880.9.14.23, CNHM 48206, 48271, 48176, MCZ 12489, 32850, 32850A, 340ll, :16434, USNM 21240, 21380, 23798, 154753, 164915; 3 from the Madeira Islands: BMNH 99.1.16.17-19; 2 from the Canary Islands: SU 4290; 1 from the Cape Verde Islands: BMNH 64.6.6.52; 4 from the Gold Coast: BMNH 1930.8.26.75, 1932.2.27.21-2, 1939.7.12.34. LABRISOMl"S 1GOBIOCLINlTS) HAITIE!\SIS Bet>bt> and Tet> Van (Plate I Fig. 4) labrisomus haitiensis Beebe and Tee Van, 1928: 2:~2; Longley and Hildebrand. 1941: 253; Manter, 1947: 300*; Fowler, 1952: 105. Descript,wn. Fin rays: dorsal XX, 10 to 12; XXI or XXII, 10 or ll (XX or XXL 10 or ll) ; anal II, 18 to 22 ( 20 or 21); pt>ctorals 13 lo ] 5 (14). Lalt>ral line 4;~ to 46 (many scales lack tubt>s t. Gill-rakers 10 to B. Head 29.5 to 31.7 pt>rcent of standard length in specimens over 20 mm. Hony orbital diameter considerably greater than snout length in all specimens (orbital diameter al­ ways more than 10.5 percent of standard length; snout length always less than 8.6 percent of standard length). First dorsal spine always t>xceeding 10 percent of standard length; longer than third. fourth, or penultimate dorsal spines. Length of shortest pelvic ray contained 21/z to 31/z times in longest. lnterspace between nuchal cirri bases con­tained 14 to 1/z times in a single base. Maxillary 12.8 to 15.9 percent of standard length (usually more than 13.5 percent). Largest specimen 58.2 mm. No differences in color pattern were noted between males and females. All fins of adults are specked or barred. The most distinct barring occurs on the anal fin. There are five or six irregular bands on the body which extend onto the dorsal in diminished intensity (well defined in young up to 30 mm.). The ground color of the body between the bands is light with scattered darker markings. The venter is irregularly dusky, as is the underside of the head. The pectoral fin has a dark crescent mark at the bases of the rays. The color of the head is similar to that of the body. The opercle may or may not have a diffuse blotch on its dorsal portion. There are two dark bars, with a lighter area be­tween, extending posterior to the eye. Discussion. There is slight variation in total dorsal and anal elements in specimens from different localities. Of eight specimens from Tortugas, Florida, six had thirty-one dorsal elements and one each had thirty and thirty-two. Of nineteen specimens from the northern Bahamas, fifteen had thirty-two dorsal elements, three had thirty-one, and one thirty. Of the specimens from Tortugas, six had twenty-two anal elements and one each had twenty and twenty-one. Of the Bahamas specimens, sixteen had twenty-three anal elements and one each had twenty-one, twenty-two, and twenty-four. Of two specimens from the northwest coast of Florida, one each had thirty-one and thirty-two dorsal ele­ments and both had twenty-three anal elements. Relationship. This species differs from all other Gobioclinus in having typically four­teen pectoral rays and the shortest pelvic ray contained in the longest more than two times. Materia 1• Thirty-one specimens were examined. The holotype, NYZS 7170, from Port­au-Prince Bay, Haiti; 19 from the northern Bahamas: ANSP 7451L 74662, 74664-6, 74709-11, 74715-6; 8 from Tortugas, Florida: USNM 88121, 116803, and oneUSNM no number; 1 from a reef thirty miles south of Miami, Florida: USNM 167668; and 2 from off t1:e northwest coa~t of Florida: USNM 73062 and 129877. In addition to the above, eight specimens from the Bahamas, USNM 38515, 17 to 18.5 mm., are tentatively assigned to this species. LABRISOMUS (GOBIOCLINUS) BUCCIFERUS (Poey) (Plate I Fig. 5) Labrosomus bucciferus Poey, 1868: 399. Labrisomus buccif erus, Jordan, 1887 a: 599; Jordan and Evermann, 1896: 468; Jordan and Evermann, 1898: 2363; Nichols, 192la: 24*; Metzelaar, 1922: 135; Jordan, PLATE I Arranged Top to Bottom FIG . i. Labrisomus nuchipinnis, UM 882, an adult male, 76.3 mm. in standard length, from Nichols­town, Andros Island, Bahamas. FIG. 2. Labrisomus guppyi, CNHM 59874, an adult male, 67 mm. in standard length, from Cam· peche Banks, Mexico. FIG. 3. Labrisomus guppyi, CNHM 61894, an adult female, 77 mm. in standard length, from Cam· peche Banks, Mexico. FIG. 4. Labrisomus haitiensis, ANSP 74715, an adult male, 46.5 mm. in standard length, from be· tween Periwinkle Rocks and Rose Island, Bahamas. FIG. 5. Labrisomus bucciferus, ANSP 74562, an adult female, 45.3 mm. in standard length, from Rose 1'l.~nd, Bahamas. Evermann, and Clark, 1930: 459; Longley, 1932: 300 (in part); Longley, 1933: 294. (in part); Longley and Hildebrand, 1941: 254; Fowler, 1944: 472; Fowler, 1950: 85 and 92; Fowler, 1953: 69; Manter, 1954: 338* ; Hubbs and Springer, 1954; 348; Springer and Bullis, 1956: 97*. Labrisomus nuchipinnis, Rosen, 1911: 66 (part not of Quoy and Gaimard); Fowler, 1936: 1037 (part not of Quoy and Gaimard). Clinus bucciferus, Metzelaar, 1919: 154. Labrisomus heilneri Nichols, 192lb: 2. Ctenichthys interrupta Rivero, 1936: 69; Hubbs and Springer, 1954: 349. Labrisomus herminier, Schmitt and Schultz, 1940: 9 (not of Le Sueur) . Ericteis kalisherae, Herre, 1942 : 15 (not of Jordan). Labrisomus gobio, Fowler, 1947: 8 (not of Valenciennes). Description. Fin rays: dorsal XIX, 11 or 12 ; XX, 10 to 12; or XXI, 10 (XX, 11 in 80 percent of specimens); anal II, 19 to 21 (20 ) ; pectorals 12 to 14 (13). Lateral line 45 to 48. Gill-rakers 11 to 14. Head 30 to 35 percent of standard length (usually 31 to 32 percent) in specimens over 30 mm. Bony orbital diameter always greater than snout length. First dorsal spine always exceeding 10.5 percent of standard length; always longer than third, fourth, fifth, pe­nultimate, and ultimate spines. Length of shortest pelvic ray contained Ilh to l 4f> times in longest. Interspace between bases of nuchal cirri contained 1h to 1 time in a single base. Maxillary 14.4 to 19.0 percent of standard length. Largest specimen 70 mm. Sexual dimorphism of color pattern is restricted primarily to the fins. In the female all the fins are profusely spotted. In males, the pectorals, caudal, soft dorsal, and posterior anal are clear. The pelvics, spinous dorsal, and anterior anal fins of males are shaded with du~ky of varying darkness. All specimens have the five bands of the body extending onto the bases of the dorsal elements, and membranes of the first three dorsal spines are more heavily marked than any of the others. The body bands are lighter below the midline. The interspaces between the bands are lighter and wider than the bands. The ventral areas of the head and body are not conspicuously marked. The head is variably dusky. There are two dark lines or bars behind the eye, between which is a lighter bar. Young of 18 to 19 mm. may be unmarked, except for a few large chromatophores atop the head and at the bases of the soft elements of the dorsal and anal. Other young of 17 mm. may already show banding of the body, and a slightly greater body depth. It is possible that the young undergo a metamorphosis, becoming shorter and stubbier when they change from a planktonic to a sedentary life. A similar type of change is indicated in L. nuchipinnis (q.v.). Relationship. Labrisomus bucciferus is most closely related to L. gobio, l. guppyi, and L. kalisherae. Its uniformly dark peritoneum and low scale count serve to separate it from all Gobioclinus. Young of this species are often confused with L. gobio, but the anal fin of the latter species, when young, is only lightly marked, and the lines behind the eye are rarely noticeable. Material. Two hundred twenty-nine specimens were examined. One hundred thirteen from the Bahamas: AMNH 7769 ( holotype of L. heilneri), ANSP 72239--40, 72228-37, 74467, 74469, 74471, 74475, 74508, 74652, BMNH 1939.4.19.38, UM 847, 883; 80 from Cuba: MCZ 34151 (holotype of Ctenichthys interrupta), UM 48, CSNM 82527-30, 107382, 107413-14; 7 from Puerto Rico: UMMZ 171188, 171200, 171205, 171212, 171229, 171237; 6 from the Virgin Islands: USNM 114754, 117441, UZMK P76256, 76249-50, 42; 14 from Barbados: ANSP 74463, 74465, SU 37268, l'SNM 86751; 1 from Old Providence Island: USNM 107113; 4 from St. Andrews Island: ANSP 72686; 1 from British Honduras: CNHM 39826; 1 from the Cayman Islands: BMNH 1939.5.12. 176-178; 1 from Swan Island: MCZ 30771; 1 from Bermuda: CNHM 48172. LABRISOMUS (GOBIOCLJNUS) GUPPY/ (Norman) Wlate I, Figs 2 and 3) labrisomus nuchipinnis, Rosen, 1911: 66 (part not of Quoy and Gaimard); Fowler, 1950: 85 (not of Quoy and Giamard); Springer, 1955b: 70 (not of Quoy and Gaimard). Clinus herminier, Metzelaar, 1919: 153 (not of Le Sueur). Clinus guppyi Norman, 1922: 533. labrisomus heilneri, Parr, 1930: 126 (not of Nichols). Mal,acoctenus bucciferus, Fowler, 1930: 215 (not of Poey). Malacoctenus bondi Fowler, 1930: 275. Labrisomus guppyi, Longley and Hildebrand (in part), 1941: 254. Starksiasanctiandrewsi Fowler, 1950: 91. Description. Fin rays: dorsal XVIII, 11 or 12; XIX, 10 to 12 (XIX, 11 in 84 per­cent of specimens); anal II, 18 to 20 (19); pectorals 12 to 14 (13) . Lateral line 48 to 53 ( 49 to 51). Gill-rakers 12 to 15 ( 13 or 14). Head 31.0 to 35.7 percent of standard length in specimens over 28 mm. Bony orbital diameter rarely less than snout length. First dorsal spine rarely exceeding 10 percent of standard length; usually exceeding only third and penultimate spines. Length of shortest pelvic ray contained l J4 to 12/.'3 times in longest. lnterspace between nuchal cirri bases contained 1 to 2 times in a single base. Maxillary 15.2 to 17.5 percent of standard length. Largest specimen 88 mm. Color pattern varies with sex. Females have all the fins heavily spotted and exhibit five or six distinct bands on the body. The bands extend onto the bases of the dorsal elements. The areas between the bands are mottled, as are the head and ventral regions. There is a dark ocellus present on the opercle and usually a clear bar extending pos­ teriorly from the eye. In contrast to the females, the males usually have the dorsal, anal, and caudal mostly clear or only faintly marked. The pelvics and pectorals are lightly spotted. The bands on the body are usually less conspicious, except in young specimens, and the body is more uniform in color than in females. The head may be mottled. There is a dark ocellus on the opercle, and indications of a light bar extending posteriorly from the eye can usually be discerned. Specimens at 18 mm. show the well-developed opercular ocellus and the body bands. Discussion. The type material of Starksia sanctiandrewsi Fowler, here included in synonymy for the first time, consists of small specimens of this species. Fowler's aberrant counts are taken from damaged fins. Relationship. This species is most closely related to L. gobio and L. kalisherae. It can be differentiated from these by those characters given in the key (also see Relation­ship under L. kalisherae) . In large specimens the large teeth behind the villiform teeth in the upper jaw may approach the large size of those teeth in L. dendriticus. Material. One hundred specimens were examined. Forty-seven from the Bahamas: AMNH 18288, ANSP 74677-79 (including one stained). UM 1297: 14 from Cuba: L'SNM 114756-58, 82532, 82533: 5 from the Virgin Islands: ANSP 74680, USNM 117426, UZMK 43, P76251-2; holotype of M. bondi from Grenada: ANSP 52504; 1 from Barbados: ANSP 74681; 4 paratypes of S. sanctiandrewsi from St. Andrews Island: ANSP 71782-5; 3 from Old Providence Island: ANSP 72685, USNM 117455; 1 from Swan Island: BMNH 1908.7.6.25; 1 from Cayman Islands: BMNH 1939.5.12. 177; and 23 from Campeche Banks, Mexico: CNHM 46655, 59874, 61894. LABRISOMUS (GOBIOCLINUS) KALISHERAE (Jordan) (Plate II, Fig. 1) Ericteis kalisherae Jordan, 1904: 543; Jordan and Thompson, 1905: 254; Jordan, Evermann, and Clark, 1930:460. Clinus bucciferus, Metzelaar, 1919: 154 (part not of Poey) . labrisomus bucciferus, Longley, 1932: 300 (part not of Poey). Labrisomus kalisherae, Longley and Hildebrand, 1941: 252; Springer, 1955b: 70. labrisomus guppyi, Longley and Hildebrand, 1941: 256 (part not of Norman). Description. Fin rays: dorsal XVIII, 11 or 12; XIX, 10 to 12; or XX, 10 or 11 (XIX, 11 in 85 percent of specimens); anal II, 18 to 20 (19); pectorals 13 or 14 (13). Lateral line 48 to 53. Gill-rakers 10 to 13 (11). Head 31.5 to 35.8 percent of standard length in specimens over 33 mm. Bony orbital diameter always greater than snout length (orbital diameter rarely less than 10 per­cent of standard length; snout length rarely more than 10 percent, usually 7 to 9 percent). First dorsal spine usually more than 10 percent of standard length; al­ways longer than third, fourth, or penultimate spines. Length of shortest pelvic ray contained 11;2 to 1% times in longest. lnterspace between nuchal cirri bases con­tained 1 to 2 times in a single base. Maxillary 15.1 to 18.9 percent of standard length. Largest specimen 68.8 mm. No differences in color pattern between males and females were noted in this species. All of the fins are densely spotted, but the spotting is not as pronounced as that in females of L. guppyi. The bands on the body are distinct and extend faintly onto the dorsal. The areas between the bands are lighter and broader than the bands. The head may be light or with chromatophores. There are indications of a light bar extending posteriorly from the eye. The opercle is dusky to clear, but the young have a distinct blotch in this area, which makes them resemble closely the young of L. guppyi. The ventral areas are uniformly dusky or light. PLATE II Arranged Top to Bottom Fu;. 1. Labrisomus kalisherae, CNHM 59877, an adult male, 66.3 mm. in standard length, from Campeche Banks, Mexico. Fie. 2. Labrisomus gobio, ANSP 74712, an adult female, 4-0.5 mm. in standard length, from HogIsland, Bahamas. Fie. 3. Labrisomus albigenys, CNHM 59875, an adult female, 51 mm. in standard length, from Campeche Banks, Mexico. Fie. 4. Labrisomus nigricinctus, ANSP 74708, an adult male, 51 mm. in standard length, from New Providence Island, Bahamas. Fie. 5. Labrisomus nigricinctus, ANSP 74708, an adult female, 53 mm. in standard length, from New Providence Island, Bahamas. .:J.36 Systematics and Zoogeography of the Clinid Fishes Discussion. There has been considerable confusion among workers as to the loca­tion of the holotype of this species. Jordan (1904) originally gave the holotype a Stanford catalogue number. Later he presented the holotype to the U.S. National Museum, where the specimen still bears the Stanford number on a metal tag in addition to the National Museum number. Longley and Hildebrand (1941) give data on the "type" of E. kalisherae in the British Museum. As this specimen was a gift from its collector, Thompson (British Museum, personal communication), it must be con­sidered the same as that specimen mentioned by Jordan in his original description as being retained by Thompson. Although labeled a holotype in the British Museum, this specimen can be considered only paratypic. Relationship. This species is most closely related to L. guppyi and L. gobio, and can be distinguished from them by those characters given in the key. Some confusion arises as to the identification of specimens of less than 40 mm. of L. guppyi and L. kalisherae. Specimens of guppyi may have one or two symphysial pores at this size, but always have more at larger sizes. Labrisomus kalisherae never has more than two symphysial pores, but it may have the opercle with a dark splotch in small specimens. In those specimens which overlap in gill-raker count, kalisherae can usually be distinguished from guppyi as the opercle blotch is less dark and sharply defined. Material. Sixty-eight specimens were examined. One from Biscayne Bay, Florida: CNHM 50247; 37 from Tortugas, Florida and vicinity: ANSP 74682, USNM 79275 (holotype of E. kalisherae), 116813, and one USNM with no number; 1 from Tobago: BMl\H 1931.12.5.365; 2 from Isla Fernando Noronha (Brazil); BMNH 1888.1.19. 100-101; 1 from British Honduras: USNM 49151; 21 from Compeche Banks, Mexico: CNHM 59877, 61892; 5 from Caho Rojo, Vera Cruz, Mexico: IMSC no numbers. LABRISOMUS (GOBIOCLINUS) COBIO (Valenciennes) (Plate II, Fig. 2) Clinus gobio Valenciennes (in Cuvier and Valenciennes), 1836: 395 (in part); Storer, 1846: 372 (in part) ; Gunther, 186lb: 276 (in part). Labrisomus gobio. Swainson, 1839 (in part) ; Longley and Hildebrand, 1941: 257; Hubbs and Springer, 1954: 348. Gobioclinus gobio, Gill, 1860b: 277 (in part) ; Jordan 1887a: 599 (in part) ; Jordan and Evermann, 1896: 468 (in part); Jordan and Evermann, 1898: 2365 (in part); Jordan. Evermann, and Clark, 1930: 455 (in part); Hubbs, 1952: 102 (in part) ; Hubbs and Springer, 1954: 346. Labrisomus heilneri Nichols, l 92lb: 2 (in part: paratype). Description. Fin rays: dorsal XVIII, 12; XIX, 10 to 12; or XX, 10 or 11 (XIX, 11 in 77 percent of specimens) ; anal II, 18 to 20 (19) ; pectorals 12 or 13 (13). Lateral line 48 to 53 ( 49 to 51). Gill-rakers 11 to 14 ( 13 or 14). Head 29 to 31 percent of standard length in specimens over 25 mm. Bony orbital diameter always more than 10 percent of standard length (usually more than 11 per­cent) ; always considerably greater than snout length, which is rarely more than 8.7 percent and always less than 10 percent of standard length. First dorsal spine about 10 percent of standard length; always exceeding penultimate spine, and always shorter than fourth and fifth spines. Length of shortest pelvic ray contained 1% to 1% times in longest. Inters pace between nuchal cirri bases contained 1h to 1 % times in Systematics and Zoogeography of thl' Clinid Fishes 4;)7 a single base. Maxillary 12.4 to 16.l percent of standard length l usually more than 14 percent I. Largest specimen 49.2 mm. This species is the least strongly marked of all labn:somus. Except for a light speck· ing of the dorsal, anal, and caudal fins of females, there is no difference in color pat· tern between the sexes. All the fins of the male are clear, as are the pelvics and pectorals of the female. There are five bands on the body which may extend just to the bases of the dorsal elements. They are about the same width as the areas between them. The band~. are darkt>r than the intnspaces, which have a light color covered by dusky markings. The ventral areas of the head and body are light and unmarked. The head is usually unmarked, hut has some diffusely marked areas on the cheek. Relationship. This species is most closely related to L. guppyi and L. kalisherae and can be distinguished from them by those characters given in the kt>y. Material. Seventy-six specimens were examined. Sixty-four from the Bahamas: AMNH 7769 (paralype of/,. heilneri), ANSP 722:)8, 74455, 74457-8, 74468, 74474, 74476. 74660, 74712 (including one stained I, UM 844; lO from Cuba: USNM 8253L 114753; 1 from the Virgin Islands: ANSP 74473; 1 from Belize, British Honduras: USNM 91817. LABRISOMUS (GOBIOCUNUS) ALBUQUERQUENSIS (Fowler) nomen dubium Malacoctenus albuquerquensis Fowler, 1944: 162 and 243; Fowler, 1953: 69. The holotype and paratype (ANSP 69987 and 69988) of this nominal species repre· sent post-larval stages of either one or two of the following: L. gobio, L. kalisherae, or L. guppyi; all of which have priority over L. albuquerquensis. Present knowledge of life histories of these species is not complete enough to enable exact identification. LABRISOMUS (BROCK/US) ALBIGENYS Beebe and Tee Van (Plate II, Fig. 3) Labrisomus albigenys Beebe and Tee Van, 1928: 233; Fowler, 1952: 105. Description. Fin rays: dorsal XVIII, 11; anal II, 18; pectorals B. Lateral line 41 to 44. Gill-rakers 13 or 14. Head 33.4 to 34.4 percent of standard length in specimens over 47 mm. Bony orbital diameter less than snout length in specimens over 47 mm. First dorsal spine less than 8 percent of standard length; usually shorter than any other dorsal spine, except penultimate, which it may exceed. Length of shortest pelvic ray contained ap· proximately 2 times in longest. Interspace between nuchal cirri bases contained -V:; lo 1 1/ 10 times in a single base. Maxillary 12.2 to 13.4 percent of standard length. Largest specimen 52.4 mm. Color pattern is slightly different in males and females. Females have the dorsal. anal, and caudal fins with several rows of dark spots. Thne are faint indications of about eight stripes on the dorsal portion of the body whieh extend onto the dorsal fin. The pelvics are light with dusky bases. The ground color of the body is a uniform dusky brown with scattered darker spots. The venter is light brown. There is a distinl'I dark stripe at the bases of the pectoral rays followed by a light stripe on thP prqwctoral area. The ground color of the head is about as dark as that of the body. The opercle has some dark spots with light areas bt'lwt>en them. Tht' underside of the head is alternately barred or mottled with dark and light. There are two bands lt>ading pos· teroventrally from the eye with a light bar between them. The male examined is more uniform in color. Faint spots are present only on the caudal. The anal is dusky. The pectorals, pelvics, and ground color of the body are the same as in the females. The head, however, is uniformly dusky below and the postero­ventral bars leading from the orbits are not so obvious. Photographs of the holotype, when fresh, indicate that this small specimen was some­what intermediate in coloration between males and females, with a more pigmented spi­nous dorsal than any of the adults. Discussion. This species is, at present, the most poorly represented of its genus in museum collections. The considerable distance between the two localities from which it is known indicates that it is probably widely distributed, and future collecting will probably establish this. The holotoype is at present in very poor condition and will probably be valueless for study in another few years. Two excellent photographs taken of the holotype were kindly given to me by Dr. John Tee Van, and these have been deposited in the FS. National Museum. Relationship. This species is most closely related to L. striatus and L. nigricinctus. It differs from the former mainly in not having scales on the cheek, and the latter in the nature of its coloration. Material. Five specimens were examined. The holotype, l\YZS 7372, from Port-au­Prince Bay, Haiti, and four specimens, C1'HM 59875, from Cayos Arcos, Campeche Banks, Mexico. LABRISOMUS (BROCK/US) NIGRICINCTUS Rivero (Plate II Figs. 4 and 5) Labrisomus nigricinctus Rivero, 1936: 68. Starksia oculata Fowler, 1947: 9; Fowler, 1950: 92. Description. Fin rays: dorsal XVII, 11; XVIII, 10 to 12; or XIX, 11 (XVIII, 11 in 90 percent of specimens) ; anal II, 17 to 20 ( 18 in 86 percent of specimens) ; pectorals 13. Lateral line 40 to 44 ( 41or42). Gill-rakers 9 to 12 (10 or 11). Head usually 29 to 31 percent of standard length in specimens over 30 mm. Bony orbital diameter usually greater than snout length. First dorsal spine always less than 8 percent of standard length; shorter than any other dorsal spine, except penultimate which it occasionally equals. Length of shortest pelvic ray contained 14,1; to 214 times in longest. Interspace between nuchal cirri bases contained % to 1% times in a single base. Maxillary 10.2 to 12.l percent of standard length. Largest specimen 53.6 mm. Color pattern is dependent upon age and sex. Females and young males have about eight or nine broad, dark bands separated by narrow light stripes. The first two bands are on the caudal. Bands and stripes on the body extend onto the dorsal and anal, and are most distinct on the dorsal. The pectorals are light with a dusky area at the base of the rays, followed by a light band and another dusky band on the prepectoral area. The pelvics are clear and other portions of the ventral region are light. The fins although barred are never spotted. There is a well-developed ocellus on the opercle, and two thin, dark lines leading posteroventrally from the eye with a broad, light band between them. A similar light band bordered by dusky areas extends anteroventrally from the eye across the maxillary. Adult males are uniformly light with some dusky markings on the body and fins. The Systematics and Zoof!,eography of the Clinid Fishes opercular ocellus, bands on the prepectoral region, and those on the head are a;; in the female. The opercular ocellus is present in specimens 12.6 mm. in standard length. Discussion. The holotype of S. oculata Fowler is a typical specimen of Rivero\; speC'ies. Relationship. This species is most closely related to L. striatus and L. albigenys. It differs from both of these in having a well-developed opercular ocellus, in the solid bar­ring, but never spotting, of the soft dorsal and caudal, in the nature of its striking st·xual dimorphism of color pattern, in having a slig:1tly shorter jaw. and in having fewer gill­rakers. In addition to these charaeters it differs from L. striatus in not having ,;cales on the cheeks. Material. Sixty-one specimens were examined. One from Sand Key, Florida: ANSP 74516; 38 from the Bahamas: ANSP 717:-35 (holotype of Starksia oculata), 74472'. 74708, 72:106-09, 74512-15, 74622 (including one stained) , 74623; 8 from Cuba: MCZ :14150 (holotype of L. nigricinctus), USNM 130790, 82538-40; 1 from the Virgin Islands: l 'ZMK 44; B from Barbados: ANSP 74517-18, 74624. MALACOCTENl'S Gill M alacoctenus Gill, 1860b: 103 (type Clinus delalandii Valenciennes, rn Cuvier and Valenciennes, 18:16, by original designation). Acteis Jordan, 1904: 543 (type Malacoctenus moorei Evermann and Marsh, 1900a, by original designation). Description. Fin rays: dorsal XVIII, 10 to 12; XIX, 9 to U; XX, 9 to 14; XXI, 8 to B; XXII, 8 to 13; XXIII, 9 to 10 or B; anal II, 17 to 24; caudal 13; pectorals 13 to 17; pelvics I, 3. Lateral line 40 to 66; scales on anterior half of posterior arc with anterior pore ex­ posed or not. Head naked. Fin membranes naked. Prepectoral area and breast naked to completely scaled. Jaws with a single row of large conical teeth (except M. erdmani with four or five small teeth behind the outer row, M. macropus with teeth biserial, and M. boehlkei with a patch of villiform teeth behind the large outer row). Vomerine teeth present. Palatine teeth absent. Gill-rakers 8 to 15 on first arch. Origin of dorsal fin in region above operde. Length of spinous dorsal longer than that of soft dorsal. Penultimate dorsal spine shorter than ultimate. Soft dorsal higher than spinous dorsal, except in occasional specimens where the anterior two spines may ap­ proach the soft dorsal height. Soft dorsal free from caudal (except M. boehlkei, M. ma­ cropus, M. erdmani, and M. gilli with the soft dorsal hound to C'audal base). Anal, caudal, pectorals, and pelviC's as in /,ahrisomus. Distribution of cirri same as in f,abrisomu.s. lnter,,.pa!'e bt'lween nuC'hal !'irri bases frequently contained more than three times in a single base. Maxillary 7.9 to 11.2 percent of standard length (u,;ually 8.5 to 9.5, and rarely more than 10.1 percent). Maxillary almost C'Ompletely sheathed by preorbital. A hook sometimes present on anterior maq!;in of deithrum in specimens of M. nd­ mani and M. au.rolineatus. Color pattern is distinet in many of the spe!'ies and is most useful in identifyin{! the young; however, specimens of one species may show color patterns similar to those of unrelated species (i.e. 111. triangulatus and M. boPhlkei). Sexual dimorphism. Crinogenital openings as in Labrisomus. There is no difference between males and females in the size of the jaws. Secondary sexual dimorphism in pigmentation may be like that found in Labrisomus, or the males may have the increased pigmentation, but without spotting. In rare instances mature females will exhibit male-type coloration; and in some populations of M. triangulatus males and females will exhibit only male-typ~ coloration. In M. macropus, M. erdmani, M. gilli, and M. aurolineatus sexually ripe males develop large fleshy pads on the innermost pelvic ray. Adult males are usually more completely scaled on the ventral areas than females. Frequently females do not develop scales in the breast region. The scales in the ventral areas of males often become fleshy and raised, increasing in degree from anterior to pos­terior. Phylogeny. Malacoctenus is considered to have branched off close to the stock which gave rise to the subgenus Brockius. The most primitive species of the genus are consid­ered to be those with high pectoral counts and increased dentition. Relationships within the genus are indicated under the appropriate heading within the species descriptions, but are to be considered only tentative. Jordan (1904) erected the genus Acteis to include Malacoctenus macropus (actually its synonym M. moorei) based upon differences in dentition and pectoral counts from other species of Malacoctenus. The fact that there is a gradation from no teeth behind the outer row in M. margaritae to a patch of teeth behind the outer row in M. boehlkei tends to negate the importance of teeth in segregating the four species with higher pec­toral counts from the other members of the genus. The higher pectoral count does, how­ever, serve to set these species apart; but even so, specimens with higher pectoral counts are occasionally found among the other species. In addition to this, the similarity of the unique sexual dimorphism of males between M. macropus and M. erdmani, and between M. gilli and M. aurolineatus, convinces me that differentiation has not proce~ded far enough to merit the erection of separate genera. In order to facilitate identification, separate keys are given for the Atlantic and Pacific species; also, it seems that the Atlantic and Pacific forms are more closely related intrageographically than intergeographically. KEY TO THE PACIFIC SPECIES OF MALACOCTENUS (N) = North of Latitude 15° N (S) = South of Latitude 13° N A. Pectoral rays 15 or more in 98 percent of specimens; scales on venter same size as those on sides . ......................... ....... M. margaritae (p. 448) (A) . Length of shortest pelvic ray usually contained more than 2% times in longest; symphysial pores more (Table 2); scale rows between anterior lateral line and dorsal contour usually 3, sometimes 4 (Table 2) . ... .......... .. ... (S) M. m. margaritae (p. 449) (AA). Length of shortest pelvic ray usually contained less than 2% times in longest; symphysial pores less; scale rows between anterior lateral line and dorsal contous usually 4, sometimes 5 ................................ . .. (N) M. m. mexicanus (p. 449) AA. Pectoral rays usually 14; scales on center reduced in size ................................... B B. Total dorsal elements 33 or more in 94 percent of specimens; interspace between nuchal cirri bases contained one time or less in a single base .............................. ................................................................................ ............ (N) M. gigas (p. 457) BB. Total dorsal elements less than 33 (33 or more in less than 1 percent of specimens of M. hubbsi, which is sympatric with M. gigas, and in less than 7 percent of specimens of M. zonogaster, which occurs only in the Galapagos) ; interspace between nuchal cirri bases contained more than one time in a single base (9/ 10 to 1 time in less than 6 percent of specimens of M. hubbsi) _______________ C C. Dorsal spines 21 or 22 in more than 98 percent of specimens; a dark round blotch covering upper half of opercle and appearing in line with a series of dark blothches which occur in the middle of each body band ............................. .. ........... ......... . ....................... .. ................... ......... (S) M. zonogaster (p. 472) CC. Dorsal spines 18 to 20 (21, never more, in less than 5 percent of specimens; never 21 in M. afuerae or M. costaricanus) ; no such blotch on upper half of opercle .... .. ....................... .. ...... ...... .................... ................... ......... D D. At least one scale on midline anterior to first dorsal spine (it may be impedded; not valid for specimens of M. zonifer of less than 25 mm.) ...... .................... E DD. No scales on midline anterior to first dorsal spine ·--·--------·---·-----------------· -·--·· G E. Preopercular canals rarely extending more than one-third (never more than two­fifths) distance across opercle; length of shortest pelvic ray contained usually less than 1% times in longest ............................................ M. zonifer (p. 474) (E). Membranes of anterior dorsal spines with a dark splotch dorsally; symphysial pores more (Table 8) ..............(S) M. z. sudensis (p. 474) (EE). Membranes of anterior dorsal spines without a dark splotch dorsally; symphysial pores less (Table 8) .................. (N) M. z. zonifer (p. 474), EE. Preopercular canals usually extending one-half distance or more across opercle; length of shortest pelvic ray contained usually more than 1% times in longest ...................... ..........................................................................................................F F. First dorsal spine less than 10.5 percent of standard length; dorsal spines usu­ally 20; anal rays usually 20; interspace between nuchal cirri bases con­tained less than 3 times in a single base; symphysial pores 1 or 2.______________ ...................... ........................................................... (N & S) M. ebisui (p. 478) FF. First dorsal spine more than 11 percent of standard length; dorsal spines 19 (only 4 specimens known); anal rays 19; interspace between nuchal cirri bases contained more than 3¥2 times in a single base; symphysial pores 4 or more ..... .. ................................. . ........................... (S) M. costaricanus (p. 468) G. Peritoneum white, sometimes with scattered large chromatophores; first dorsal spine usually less than 10 percent of standard length ........ M. a/uerae (p. 461) (G). Entire fish suffused with numerous spots of varying sizes (Plate V Fig. 5) ............................................ (N) M. a. multipunctatus (p. 426) (GG). Body more uniform in color; spots never on venter (Plate V Fig. 4) ............................................. ............ (S) M. a. afuerae (p. 462) GG. Peritoneum almost uniformly stippled with chromatophores (most dense dorsally) ; first dorsal spine usually more than 10 percent of standard length ______________ H H. A spot or blotch at the base, or just anterior to the base, of each pelvic fin (rarely absent in adults; rarely present in young under 40 mm.); prepectoral area with or without scales; interspace between nuchal cirri bases contained usually less than 3 times in a single base; anal rays 19 to 21 ( 18 in one specimen out of over 1200; less than 50 percent of specimens with 19); attaining a length of 79.5 mm. ---·----------------------------------------------------------. . . (N) M. hubbsi (p. ) (H) . Lateral line scales modally 54 to 56 (Table 5) ; symphysial pores less (Table VI) ........................................................ M. h. hubbsi (p. 465) (HH). Lateral line scales modally 51 to 53 (Table 5) ; symphysial pores more (Table VI) .................................... M. h. polyporosus (p. 468) HH. Markings at base of pelvic fins present in only 2 out of 67 specimens; prepectoral area never scaled; interspace between nuchal cirri bases contained 3 or more times in a single base; anal rays 18 or 19 (20 in 1 of 67 specimens I; largest S')ecimen 47 mm. !N) M. zaca'.! (p. 470) KEY TO THE ATLANTIC SPECIES OF MALACOCTENUS ( W) =Western Atlantic (E) = Eastern Atlantic A. Length of shortest pelvic ray contained four or more times in longest; a patch of small teeth anteriorly behind the large outer row in each jaw; pectoral rays usually 15 . ... ..... ....... ......... . ............... (W) M. boehlkei (p. 443) AA. Length of shortest pelvic ray contained less than four times in longest; small teeth, when present, in a single row behind the large outer row; pectoral rays 14 to 17 --·-··------------·---------------------·--------··--·----------··--------------·--------------------------------------B B. Pectoral rays typically 15 to 17; scales on prepectoral area, if present, about same size as those on body; small teeth present behind large outer row1 --------·---------C BB. Pectoral rays typically 14; scales on prepectoral area, if present, reduced in size; small teeth never present behind large outer row --------------------------------------------D C. Pectoral rays 15 to 17(usually16); pectoral base never scaled; small teeth behind large outer row never more than 5 in each jaw; interspace between nuchal cirri contained 1h to 1 Vs times in a single base; a distinct black blotch at bases of last dorsal spines (most prominent color marking) probably only rarely ex­ceeding 30 mm. in standard length ........................... (W) M. erdmani1 ( p. 444) CC. Pectoral rays 14 to 16 (usually 15) ; pectoral base rarely without scales; small teeth present as a series of more than 6 teeth behind each large outer row; inter­s pace between bases of nuchal cirri contained less than 1h time in a single base ; no outstanding black blotch at bases of last dorsal spines; frequently exceeding 40 mm. in standard length ..................... (W) M. macro pus (p.???) D. At least one scale on midline before first dorsal spine (it may be imbedded) ; color pattern primarily of three to five inverted triangles on sides of body; pre­pectoral area scaled . ............................ (W) M. triangulatus (p. 477) DD. No scales on midline before first dorsal spine; color pattern otherwise; prepectoral area scaled or not ----------------------------------------··------------------------------------------------------E E. Most prominent color markings a dark ocellus on anterior dorsal spines and an­other on sides of body and bases of posterior dorsal spines; scales on venter, when present, same size as those on sides ; prepectoral area without scales; soft dorsal closely bound to caudal base .......... ................. (W) M. gilli (p. 450) 1 Because of the very small size of M. erdmani and the fragility of the teeth behind the outer row, these teeth are frequently lost when a needle is inserted into the mouth of a specimen. It is only with difficulty that they can be seen. Systematics and Zoogeography of the Clinid Fishes EE. Color markings otherwise; scales on venter, when present, reduced in size; pre­pectoral area with or without scales; soft dorsal free from caudal base ____ ____ __ F F. Lateral line scales more than 56; scales usually present on prepectoral area; stripes on body extending almost to tips of dorsal spines without diminution____ G FF. Lateral line scales less than 56; scales never present on prepectoral area; stripes on body changed considerably in appearance on dorsal; not extending more than 1h distance up dorsal spines ----------------------------------------------------------------------H G. Dorsal spines 18 (28 specimens examined); at least one (usually 3) preopercle canal extending well onto opercle ----------------------------(W) M. versicolor (p. 455) GG. Dorsal spines 20 (1 specimen examined); preopercle canals scarcely encroach­ing on opercle _ __ _______ _______ ___________________ __ _ (E) M. africanus (p. 456) H. Membranes in region of first 3 dorsal spines more darkly pigmented than others; interspaces between bands on body of about uniform width; no humeral blotch present; 2 canals extending from preopercle onto opercle ----------------------------------­--__ ___________ __ _____ _ (W) M. delalandei (p. 460) HH. Membranes in region of first 3 dorsal spines containing little or no pigment; not different from remaining membranes; anterior two or three bands on body separated from posterior bands by a wider space than occurs between any of the other bands; first two bands sometimes fused dorsally into a humeral blotch; 3 canals extending from preopercle onto opercle ..(W) M. aurolineatus (p. 454) MALACOCTENUS BOEHLKE/, New Species (Plate III, Fig. 1) Description. Fin rays: dorsal XX, 13; XXI, 11 or 12; or XXII, 11; anal II, 20 to 23 (22). Lateral line 58 to 66. Gill-rakers 8 to 11 (10). Symphysial pores 1or2 (1). Head 26.4 to 32.2 percent of standard length in specimens over 25 mm. Bony orbital diameter 8.8 to 11.4 percent of standard length in specimens over 25 mm; smaller than snout in specimens over 45 mm. All dorsal spines, except rarely third and fourth, greater than 10 percent of standard length. First dorsal spine always shorter than tenth to last dorsal spines. Length of shortest pelvic ray contained 41h to l l 1h times in longest; 6 or more times in specimens over 40 mm.; frequently appearing to be absent. lnterspace between nuchal cirri bases contained 1/z to 11;'.> times in a single base. Maxillary length 8.5 to 10.7 percent of standard length; usually less than 9.3 percent. Largest specimen 50.8 mm. Scales present on midline before dorsal, on prepecloral area, and prepelvic area in adults over 40 mm. Lateral line scales are variable in size and many of those on the posterior portion have no canals. The circuli of the body scales are more prominent than in the other species of the genus. Breast scales of males are sometimes imbedded. There are two preopercular canals which just encroach upon the opercle. Coloration is different in males and females. In males all the fins are clear with some scattered dusky areas on the membranes. There is an ocellus on the membranes between the second, third, and fourth dorsal spines about midway up from their bases. There are six or seven pentagonal saddle-like markings on the body with their bases on the dorsal body contour. The saddles are darker ventrally. Below each of these saddles and between the anterior ones are darker diamond-shaped marks, each of which has a clear spot in its center. These diamonds are lighter than the saddles in some specimens. On the pre­ pectoral, prepelvic, and ventral head regions there are alternating dark and light bands, which in the latter area extend onto the cheeks. Females have the same basic color pattern as males, but are much lighter. In some cases the pattern is scarcely visible, and in all specimens the ventral areas are uniformly light. Coloration in life, as taken from color slides kindly supplied by Dr. J. Bohlke, shows that the saddles and diamonds of males are black, margined with white. The top of the head, areas between the saddles, and the dorsal and caudal fins are yellow-brown. In females the saddles are orange-brown, and the diamonds are bright orange with white centers. The belly and the areas between the diamonds are white. The fins and head are tinged with light orange. Discussion. Dr. J. Bohlke informs me that in contrast to other labrisominids, M. boehlkei is taken only in comparatively deep water: fifty feet. Relationship. This species of Malacoctenus differs from all others in its extremely long pelvic fins and in the nature of its dentition. It is most closely related to M. macropus and M. erdmani in the nature of its high pectoral count and the fact that small teeth are present behind the large outer rows. Holotype. ANSP 74719, an adult male, 42 mm. in standard length, from north of the east end of Rose Island, Bahamas (ANSP Station 292), 25° 07' 55" north latitude and 77° 05' 32" west longitude iH. 0. Chart 1611/June, 1938). Collected by C. C. G. Chaplin, E. Browing, S. Waterma!l, and J. Bohlke, May 5, 1956. Paratypes. Eighteen specimens. One from Sandy Cay. Bahamf Guaymas as compared to first dorsal spine lengths of specimens from other localities. Relationship. Malacoctenus gigas appears to be m'Jst closely related to a group of species comprising also M. afuerae in the Pacific, and M. delalandei in the Atlantic. It can be distinguished from the former and the other Pacific species by the characters giYen in the key. It differs from all the Atlantic species, except M. boehlkei, in its typi­ cally high anal and dorsal formulae. lt differs from all but M. gilli, M. erdmani, and M. boehlkei in the number of times the interspace between the nuchal cirri bases is contained in a single nuchal cirrus base. It differs from the first of these in its larger number of lateral l'ne scales, and from the latter two in the nature of its dentition and lower pectoral count. Holotype. SU 49617, an adult male, 78 mm. in standard length. from 250 yards south of the tip of Punta Cholla, Sonora, Mexico (six miles north of Punta Peiiasco). Collected by Clark Hubbs and Janet Haig, March 19, 1948. Paratyp~s. E!ght hundred forty-six specimens. Five hundred thirteen from the vicinity of Punta Peiiasco and Punta Cholla, S::>noro: SU 4 7183-84, 49618 (parachronotopo­ types) , 49624-31, 49534-42. UZMK 76956; 2 from the vicinity of San Felipe, Baja California: SU 49623, UCLA W54--193; 13 from the vicinity of Punta Diggs, Baja California: l "CLA W52-69, W52-192A, W54-194; 2 from Isla San Luis: UCLA W52­ 74; 1 from Isla Mejia: SU 49646; 10 from Isla Angel de la Guarda: UCLA W54--20; 58 from Js!a Tiburon: Cl\HM 61903. SC 17875. 49619. CCLA W56-27; 16 from just south of Bahia Los Angeles, Baja California: TNHC 2973; 1 fom Isla Partida: SU 49647; 3 from Bahia San Pedro, Sonora: SU 49649; 18 from Bahia San Francisquito, Baja California: SL' 49648; 115 from Guaymas and vicinity, Sonora: CAS 20721-22, SU 16483, 16537, 16644-45. 16655, 19180, 49643-45, 49622, UCLA W52-51; 54 from the vicinity of Santa Inez. Baja California: BMNH 1957.5.1.205-236. SU 49620-21, 49650, UCLA W53-86, USNM 174961; 30 from Isla Espiritu Santo: CAS 6041. Named gigas because of its being the largest member of its genus. MALACOCTENUS DELALANDEI (Valenciennes) (Plate V, Fig. 3) Clinus delalandii Valenciennes (in Cuvier and Valenciennes), 1836: 379; Gunther. 1861: 264 (in part). labrisomus delalandii, Swainson, 1839: 277. Clinus delalandi, Muller, 1864: 632 (in part) ; Gunther, 1868: 389 (in part). Jlalacoctenus delalandi, Gill, 1860: 103; Jordan and Evermann, 1896: 468 (in part); Gilbert. 1900: 179; Evermann and Marsh, 1900: 310; Jordan, Evermann, and Clark, 1930: 459 (in part); Nichols, 1930 : 376; Fowler, 1942A: 178. labrisomus delalandi, Jordan, 1887: 599. Ualacoctenus delalandii, Meek and Hildebrand, 1928: 938; Fowler, 1944: 472; Schultz, 1949: 179. Description. Fin rays : dorsal XIX, 9 to 11; XX, 9 to 11; or XXI, 9 (XX, 10 in 75 per­cent of specimens) ; anal II, 17 to 20 ( 19) ; pectorals 13 to 15 (14). Lateral line 48 to 54. Gill-rakers 10 to 12. Symphysial pores 1 to 6 ( 3 or 4). Head 26.5 to 29.2 percent of standard length in specimens over 25 mm. Bony orbital diameter 8.3 to 11.2 percent of standard length; always larger than snout length, which does not exceed 10 percent of standard length. First dorsal spine 10 to 13 percent of Systematic and Zoogeography of the Clinid Fishes standard length; only consistently exceeding third, fourth, fourth from last, third from last, and penultimate dorsal spines; usually exceeded by tenth dorsal spine. Length of shortest pelvic ray contained 1% to 2 times in longest. Interspace between nuchal cirri bases contained 11;2 to 8 times in a single base (usually 2 to 4 times). Maxillary length 8.2 to 9.8 percent of standard length. Largest specimen 55.8 mm. Scales not present on midline before dorsal or on prepectoral area. Breast usually scaled in adult males; mostly naked in females. The breast scales become fleshy in mature males. There are two or three preopercular canals extending onto the opercle, the more ventral is the longest, extending two-fifths to one-half the distance across the opercle. Coloration depends on sex and geographic distribution. Females have the dorsal, anal, and caudal fins densely spotted; body pattern is overlain by numerous irregular patches of dusky spots. In contrast to this, the fins of males are uniformly colored, as is the body. Both sexes have the membranes of the anterior three or four dorsal spines more heavily pigmented than any of the others, and a dusky area running subbasally the length of the spinous dorsal. There is a dark, crescent mark at the base of the pectoral rays in all specimens, otherwise the pectorals and pelvics are unmarked. The six irregular bands on the body are usually discernible in females, but rarely so in males. All specimens from Brazil, Panama, Venezuela, and Tobago have a well developed dark blotch covering most of the opercle. Only a few specimens from Guatemala and neither specimen from Puerto Rico showed indications of this blotch. The ventral portions of the body and head are uniformly dusky in males, but the ventral portion of the head is either barred with light and dark or spotted in females. The body is usually abruptly lighter below the midline. Four dark "spokes" with lighter interspaces extend from the eye: one anteroventrally, one ventrally (the most distinct) , one posteriorly, and one dorsoposteriorly. Specimens 16 mm. long show indications of adult coloration, though the fins are variably spotted at this stage. Discussion. This species was named for its collector, Delalande. It is therefore neces­sary to emend Valenciennes' name delalandii to delalandei in accordance with the rules of nomenclature. Relationship. This species appears to be most closely related to the Pacific M. afuerae and M. gigas. It differs from the former most notably in the nature of the preopercular canals on the opercle, the typical formula of the dorsal, and the color pattern; and from the latter in dorsal and anal formulae, coloration, and the number of times the interspace between the nuchal cirri bases is contained in a single base. Mate rial. One hundred nineteen specimens were examined. Thirty from Guatemala: UMMZ 144151; 2 from Puerto Rico: USNM 50179, 126125; 2 from Tobago: BMNH 1924.7.22.78-79; 11 from Natal, Brazil: CNHM 61893, SU 49584; 7 from Venezuela: USNM 123170-71; 67 from Panama: USNM 81893-97, 114751, CNHM 26001-15, 18547, Sl117870, UCLA W53-268. MALACOCTENUS AFUERAE (Hildebrand) (For synonymy see subspecies) Description. Fin rays: dorsal XVIII, 11 or 12; XIX, 10 to 12; or XX, 9 to 11 (XIX, 11 in 62 percent of specimens) ; anal II, 18 to 20 (19) ; pectorals 13 to 15 ( 14). Lateral line 53 to 61 ( 53 to 56). Gill-rakers 11 to 15 (11 to 13). Symphysial pores 3 to 13 ( 4 to 8). Systematics and Zoogeography of the Clinid Fishes Head length 27 to 31.4 percent of standard length in specimens over 15.5 mm. Bony orbital diameter larger than snout in specimens under 38 mm.; either of these measure­ments usually less than 10 percent of standard length. First dorsal spine never attaining 11.5 percent of standard length (usually less than 10 percent) ; always exceeding second through fourth and fifth from last through penultimate dorsal spines; always exceeded by other dorsal spines, except last, which it sometimes exceeds. Length of shortest pelvic ray contained 11/z to 1% times in longest. Interspace between nuchal cirri bases con­tained more than four times (sometimes bases meet) in a single base. Maxillary length 9.3 to 11.2 percent (usually less than 10.5 percent of standard length. Largest specimen 56.6 mm. Scales are not present on the midline before the dorsal fin ; sometimes present on the prepectoral area; present on the breast of males only, the scales becoming fleshy with maturity. There are two or three preopercular canals extending onto the opercle, the longest of which extends one-half to three-quarters the distance across the opercle. When two canals are present the more ventral one is the longest; when three canals are present the middle one is the longest. Coloration is discussed under the subspecies. Relationship. This species appears most closely related to the Pacific ,U. gigas and the Atlantic M. delalandei. It can be distinguished from the former by characters given in the key, and from the latter in the nature of the preopercular canals on the opercle, by the typical formula of the dorsaL and in coloration. MALACOCTENUS AFCERAE AFUERAE 1Hildebrand1 !Plate V, Fig. 4) labrisomus afuerae Hildebrand, 1946: 400. M alacotenus afuerae, Hubbs, 1952: 104. This subspecies is separated from M. a. multipunctatus under the account for that subspecies. Because of the poor, or juvenile, condition of those specimens examined, it is necessary to refer to the original description and figure for purposes of determining the color pattern. In none of the specimens examined, nor in Hildebrand's figure or account is there to be found indications of a profuse spotting of the venter or sides of the body. Despite the poor condition of the type material it is hardly considered possible that spot­tin g as found in M. m. multipunctatus ever existed, except on the wntral surface of the head. Faint indications of the body bands are still discernible in the type materiaL Material. Six specimens were examined. One from Port Parker, Costa Rica: SU 49595; 1 from Chame Point, Panama: l-S:\M 81963; 1 from Manta, Ecuador: CSNM 143030; 3 from Lobos de Afuera Bay, Peru: CS:\M 12821.3 I holotype of l. afuerae), 128214 . .MALACOCTENUS AFUERAE ML'LTIPUNCTATCS. :\ew Subspecies I Plate V, Fig. 5) .1Ialacoctenus a. multipunctatus diffffers from lit/. a. afuerae in the nature of its dense, dark spotting, which in most youn g and all adults encompasses all of the ventral regions anterior to the anus and the sides of the body and head below the midline I in females and Systematics and Zoogeography of the Clinid Fishes some males, all of the body, head, and unpaired fins). Spots on the \"enter and breast of males may be obscured by the presence of fleshy scales. Though only six specimens of M. a. a/uerae are known, none approximate the spotted condition of J/. a. m11/ti­punctat11s. Frequently males are lighter and more uniform in color than females. especially abo,-e the midline of the body. l-sually the spotting on males takes the form of a fine peppering rather than a polka-dotting as found in females. The complex color pattern is best understood by reference to the photograph rather than by a written description. The two subspecies are allopatric, separated by approximately sewn hundred miles of coastline from which no clinids are known_ -1/alacoctenus a. ajuerae is known only from those areas south of the Gulf of Fonseca and J/_ a. multipunctatus only from those areas north of the Gulf of Tehuantepec. Holotype. SC 49596, an adult male 43.8 mm_ in standard length. from the southwest end of San Lucas Bay, Baja California_ l\lexico. Collected by J. Bohlke_ D. Cohen_ and J. Lindbergh, May 20, 1952 I Field number: 44-TP-l). Paratypes. One hundred thirty-se,·en specimens from l\Iexico. One from one and one­ half miles south of Punta Peiiasco, Sonora: SC 49599; 2 from Bahia Agua Dulce. Isla Tiburon, Gulf of California: UCLA W56-26; 1 from the southwest corner of Isla Par­ tida, Gulf of California: SU 49600; 2 from Punta de las Cuerns, Sonora: CCLA W50­ 37; 90 from Bahia Santa Inez, Baja California: Cl\HM 61900, CS:\M 17--1-962; 2 from Bahia Coyote in Bahia Concepcion, Baja California: CCLA W53-91: I parachronoto­ potypes: SU 49591; 7 from Cape San Lucas, Baja California: SC 18933: 1 from Mazatlan, Sinaloa: UCLA W51-58; 19 from the Islas Venados: BM:\H 1951.5.1.11-81. UCLA W51-20, W51-33, W51-54; 2 from l\Iazanillo, Colima: l -l\f\IZ 171893; 3 from Puerto Marques, Guerrero: SU 49598. Named multipunctatus for the numerous spots which distinguish it from the other species of Malacoctenus. MALACOCTE/\'CS HL-BBSI. :\ew Species (for synonymy see subspecies I Description. Fin rays: dorsal XIX. 10 to 13: XX, 9 to 12; or XXL 10 to 12 I XX. 11 in 75 percent of specimens of fl!/. h. hubbsi and 59 percent of specimens of J/. h. poh-­porosus) ; anal IL 18 to 23 ( 19 to 21; 20 in 18 percent of M. h. hubbsi and 53 percent of M. h. polyporosus I ; pectorals 13 to 15 ( 14 I. Lateral line --1-9 to 61 I Table 5 I. Gill-rakers 10 to 14 (11or12): Symphysial pores 1to10 ITable 6 I. Head 26.0 to 30-8 percent of standard length in specimens over 22 mm. Bony orbital diameter smaller than snout in specimens over --1-5 mm.; larger than snout in specimens under 35 mm. Either of these two measurements usually less than 10 percent of standard length. First dorsal spine rarely less than 11 percent of standard length. decreasing in proportional size with increasing standard length I as much as 28 percent of standard length in specimens under 25 mm .. though usually 13 to 15 percent in specimens up to 45 mm.) ; usually exceeding second through fifth and last four dorsal spines; frequently exceeded by second and tenth dorsal spines in specimens owr --1-8 mm. Length of shortest pelvic ray contained 1 Y:? to 2 times in longest. Interspace between nuchal cirri bases contained 1 to 31h times in a single base. Maxillary length 8.6 to 10.2 percent of stand­ard length (usually less than 9.8 percent I. Largest specimen 19.5 mm. TABLE 5 Frequency Distributions of Lateral Line Scales in Malacoctenus hubbsi (Arranged North to South for Each Heading) 49 50 51 S2 S3 S4 SS S6 s; S8 59 60 61 M. h. hubbsi Pacific Coast, Baja California Puerta Mala Arrima 1 10 25 28 10 7 1 ··-· Bahia Magdalena 1 3 11 28 29 19 9 5 2 Gulf Coast, Baja California1 Isla Mejia 2 7 11 9 5 3 Isla Angel de la Guarda 6 9 7 8 2 1 Isla Partida 3 11 15 9 9 3 4 Bahia San Francisquito 6 20 36 32 12 9 1 Bahia San Carlos 2 11 22 15 6 3 Isla Santa Inez 1 1 Bahia Santa Inez 2 5 20 16 8 1 Isla Ildefonso 2 1 2 10 8 22 15 1 Isla Coronado 1 Isla Monserrate 2 3 1 2 Bahia Agua Verde 1 1 4 2 Isla Cerralvo 2 2 Bahia de los Muertos 3 2 1 Punta Frailes 4 1 San Lucas 2 Gulf and Pacific Coast, Mexico Isla Tiburon 9 10 16 10 4 Isla San Pedro Martir2 9 6 8 10 4 Isla San Pedro Nolasco 1 1 1 1 .... Bahia San Pedro 2 7 5 1 1 Guaymas and environs 1 4 12 38 58 70 39 20 9 6 M. h. polyporosus Punta Camaron 1 2 Mazatlanl 3 7 30 22 20 5 2 1 Chamela 1 Cleophas (Tres Marias) 1 Acapulco 1 l Including adjacent islands. 2 Arbitrarily included here. Scales never present on midline before dorsal; present or absent on prepectoral area and breast of adult of M. h. hubbsi; always present on prepectoral area and breast of adults of M. h. polyporosus. There are two to four preopercular canals (the most ventral of which may be present only as a pore) extending onto the opercle, the longest of which extends from two-fifths to three-fifths the distance across the opercle. Color pattern is variable with sex and locality. Specimens of M. h. hubbsi vary geographically and within a series more than do those of M. h. polyporosus. The most striking character demonstrating this is represented in Plate VI, Figure 4 showing ex· tremes in color pattern of the ventral side of the head. Some specimens of M. h. hubbsi show the spotted condition which is present in M. h. polyporosus, but in no specimen of M. h. polyporosus is the underside of the head barred. The barred pattern, or some recognizable variation of it, always constitutes a large, if not the largest, percentage of ventral head patterns present in any series of M. h. hubbsi. There is a spot almost always present at the base of each pelvic fin. TABLE 6 Frequency Distributions of Symphysial Pores in Specimens of Malacoctenus hubbsi Over 4-0 mm. in Standard Length 2 10 M. h. hubbsi Pacific Coast, Baja California Puerta Mala Arrima Bahia Magdalena 46 51 IO 25 6 28 6 Gulf Coast, Baja Californial Isla Mejia Isla Angel de la Guarda Isla Partida Bahia San Francisquito Bahia San Carlos Isla Santa Inez Bahia Santa Inez Isla Ildefonso Isla Coronado Isla Monserrate Bahia Agua Verde Isla Espiritu Santo Isla Cerralvo Bahia de los Muertos Punta Frailes San Lucas 13 13 12 59 9 2 13 24 2 2 7 13 7 36 4 15 18 3 4 2 4 2 1 13 20 25 40 7 l 2-t 20 l 3 3 1 l 2 2 2 4 3 IO 4 2 8 3 l 4 1 1 Gulf and Pacific Coast, Mexico Isla Tiburon Isla San Pedro Martir2 Isla San Pedro Nolasco Bahia San Pedro Guaymas and environs 19 11 2 8 126 18 7 1 72 18 12 3 7 113 4 1 1 23 M. h. polyporosus Punta Camaron Mazatlan1 Cleophas (Tres Marias\ 1 IO 14 1 26 2 21 20 5 3 1 1 1 Including adjacent islands. ! Arbitrarily induckd here. Males of both subspecies are usually separable from females by ha,·ing a more uniform coloration and decreased, or usually absent. spotting of the dorsal, anal. and caudal fins. Other particulars of coloration can be determined from the photographs. Relationship. Malacoctenus hubbsi appears to be most closely related to a group of species comprising also M. zacae, M. costaricanus, M. zonifer, M. zonogast.er, and M. ebisui, in the Pacific and M. triangulatus in the Atlantic. It can be distinguished from the Pacific forms by the characters given in the key. It differs from M. triangulatus in lacking scales on the midline before the dorsaL in haYing the length of the shortest peh-ic ray contained in the longest fewer times, and in coloration. Holotype. See nominal typical subspecies. MALACOCTENUS HUBBS/ HUBBS/, l\ew Subspecies t Plate VI, Figs. 1, 2, and 4) Malacoctenus delalandi, Osburn and :\"ichols, 1916: 1784 tnot of Valenciennes). Malacoctenus, new species Barnhart and Hubbs, 1946: 371 •. This subspecies is separated from M. h. polyporosus under the account for that sub­species. In Table 6 it can be seen that there is a bimodality of the symphysial pores at the 1 and 3 frequencies for this subspecies. Young specimens always have one centrally lo­cated pore. If more pores are formed at a later stage they usually appear as a pair rupturing simultaneously. The number of pores present appears to be stabilized by the time the individual attains 40 mm. Barnhart and Hubbs (1946) referred to Osburn and Nichols' (1916) specimen of M alacoctenus, from Cedros Island in Bahia San Sebastian Vizcaino, as a new species. Although this specimen has not been seen, the only specimens of Malacoctenus known from localities that far north on the Pacific coast are those of M. h. hubbsi. Holotype. SU 49651, an adult male, 57.3 mm. in standard length, from the southwest end of Isla San Pedro Martir, Gulf of California, Mexico. Collected by J. Bohlke and party, May 4, 1952 (Field number: 34-TP-2). Paratypes. Two thousand five hundred ninety-three specimens. Nine hundred sixty· four from Puerto Mala Arrima, Bahia Sebastian Vizcaino, Baja California, Pacific coast: UCLA W51-221, W51-223-4; 246 from Bahia Magdalena, Baja California, Pacific coast: SU 49659-60, UCLA W55--85, W55-93-4; 39 from Isla Mejia, Gulf of California: SU 49668; 95 from Isla Angel de la Guarda, Gulf of California: UCLA W54-258, W54-262, W56-20; 145 from Isla Partida, Gulf of California: SU 18932, 49669, UCLA W56-22; 136 from Bahia San Francisquito, Baja California, Gulf coast: SU 49670-72; 24 from Moreno Rocks, Bahia San Carlos, Baja California, Gulf coast: 49664; 3 from Isla Santa Inez, Baja California, Gulf coast: SU 49657-58; 137 from Bahia Santa Inez, Baja California, Gulf coast: BMNH 1957.5.1.150.,-204, SU 18935, 49655-56, UCLA W53-91; 1 from Bahia Coyote, Bahia Conception, Baja California, Gulf coast: UCLA W53-97; 67 from Isla Ildefonso, Gulf of California: SU 19182, 49673-75; 1 from Isla Coronado, Gulf of California: SU 18769; 8 from Isla Monser· rate, Gulf of California: SU 49665; 11 from Bahia Agua Verde, Baja California, Gulf coast: SU 18934; 2 from Isla Espiritu Santo, Gulf of California: CAS 20719, SU 49654; 5 from Isla Cerralvo, Gulf of California: SU 49663; 8 from Bahia de los Muertos, Baj a California, Gulf coast: SU 49662; 7 from Punta Frailes, Baja Cali· fornia, Gulf coast: UCLA W52-263; 6 from Cape San Lucas, Baja California: SU 49676, USNM 134904, UCLA W55-118; 111 from Isla Tiburon, Gulf of California: USNM 174965, SU 49653, 49667; 43 from Isla San Pedro Martir, Gulf of California: SU 49652 (parachronotopotypes); 6 from Isla San Pedro Nolasco, Gulf of California: SU 18968, 49666; 17 from Bahia San Pedro, Sonora: SU 18969; 521 from Guaymas PLATE VI Arranged Top to Bottom Fie. 1. Malacoctenus hubbsi hubbsi, SU 49652, an adult male, 58.5 mm. in standard length, from Bahia San Carlos, Baja California, Mexico. Fie. 2. Malacoctenus hubbsi hubbsi, SU 49664, an adult female, 57 mm. in standard length, from Bahia San Carlos,Baja California, Mexico. Fie. 3. Malacoctenus hubbsi polyporosus, UCLA W51-20, an immature male, 49 mm. in standard length, from the Islas Venados, north of Mazatltan, Sinaloa, Mexico. Fie. 4. Ventral view of Malacoctenus h. hubbsi (upper) and Malacoctenus h. polyporosus (lower) showing coloration of ventral side of head. Fie. 5. Malacoctenus zacae, SU 49685, holotype, an adult male, 43.8 mm. in standard length, from Puerto Marques, Guerrero, Mexico. Fie. 6. Malacoctenus zonogaster, SU 37583, an adult female, 58.5 mm. in standard length, from James Island, Galapagos. and environs, Sonora: CAS 20720, SU 16654, 16564, 16514, 16522, 16573, 49661, CNHM 42685, 50254, 61902. I take pleasure in naming this species for Dr. Clark Hubbs in recognition of his con­tributions to the study of clinid systematics. MALACOCTENUS HUBBS! POLYPOROSUS, New Subspecies (Plate VI, Figs. 3 and 4) This subspecies differs from M. h. hubbsi most prominently in having a higher average number of symphysial pores (Table 6) , in having a lower average number of lateral line scales (Table 5), and a lower average number of anal rays (see Description under species). The two subspecies are entirely allopatric: M. h. hubbsi is known only from the Gulf of California coast of Mexico as far south as Guaymas, the Gulf and Pacific coasts of Baja California, and the islands of the Gulf of California; M. h. polyporosus is known only from the coast of the Mexican mainland from Punta Camaron, north of Mazatlan, to Acapulco and the Islas Tres Marias. Holotype. USNM 174957, an adult male, 51 mm. in standard length, from the Islas Venados, just north of Mazatlan, Sinaloa, Mexico. Collected by G. Bartholomew and K. Norris, February 26, 1951 (Field number: W51-26) . Paratypes. Four hundred ninety-nine specimens. Three from Punta Camaron, Sinaloa: UCLA W51-44; 8 from Mazatlan, Sinaloa: UCLA W51-58; 485 from the Islas Venados: ANSP 74511, BMNH 1957.5.1.90--149, CNHM 61898, UCLA W51-20, W51-29, W51­ 54, USNM 17 4960 ( parachronotopotypes) ; 1 from Chamela, Jalisco: SU 49677; 1 from Acapulco, Guerrero: SU 49679; 1 from Isla Maria Cleophas, Tres Marias: SU 49678. Named polyporosus for its many symphysial pores. MALACOCTENUS COST ARICA NUS, New Species (Plate VII, Fig. 1) Description. Fin rays: dorsal XIX, 11; anal II, 19; pectorals 13 or 14 (14). Lateral line 50 or 51. Gill-rakers 10 or 11. Symphysial pores 4 and 7. Head 27.6 to 30.0 percent of standard length in specimens over 28 mm. Bony orbital diameter 8.4, 8.0, 10.4, 10.6, and snout length 8.8, 8.4, 9.7, 8.8 percent of standard length respectively in specimens 51.2, 49.8, 28.8, 28.3 mm. First dorsal spine more than 11 percent of standard length; exceeding second through fifth and last four dorsal spines; sometimes exceeding tenth. Length of shortest pelvic ray contained 1 % to 2 times in longest. lnterspace between bases of nuchal cirri contained more than four PLATE VII Arranged Top to Bottom FIG. l. Malacoctenus costaricanus, SU 49601, holotype, an adult female, 49.8 mm. in standard length, from Piedra Blanca Bay, Costa Rica. FIG. 2. Malacoctenus zoni/er zonifer, UCLA W51-52, an adult female, 58.5 mm. in standard length, from the Islas Venados, north of Mazatlan, Sinaloa, Mexico. FIG. 3. Malacoctenus zonifer sudensis, SU 49707, an adult female, 49 mm. in standard length, from the Gulf of Nicoya, Costa Rica. FIG. 4. Malacoctenus triangulatus, ANSP 74729, holotype, an adult male, 45 mm. in standard length, from New Providence Island, Bahamas. FIG. 5. Mal,acoctenus ebisui, SU 49614, an adult female, 31.3 mm. in standard length, from Bahia Honda, Panama. Systematic and Zoogeography of the Clinid Fishes -1-69 Systematics and Zoogeography of the Clinid Fishes times in a single base. Maxillary leng ::t'> ~ ("!) ~ ::::.. "(j ~ 8. ;· ~~~~00 _. ::::I (""'. -.._... g. ~ ~ -· -· • (/l ~ ::l ::l '""3 '"O .... .... ::l"' ;:!. to ::l"' if ("!) ::l o: ("!) a~ ~ ~ ~ ~ .,.,. '-' ct> flJ '"""" ~ ~-~ ~ ;:i ~~t:c;::i::i ~ iil CJl(Jl>~ /:;· " t:r' w .... P> ---(j ~ --§ ::;.; >:l ~ ~ u;· ;p· :=. S' o CD e;-r ~ N ..... 0 0... 0... -::i ("!) .... (") ::l"' ~ ~ ::l 0... 0 ..... ::l"' ~ ("!) ..... 0::l ("!) ("!) Vi (/l .,.,. '"O ~ N' flJ ct> >:l ..... .... >:l pO '"O "' ~ ::r '"1:l t:r' ("!) '"O ;::.­ -(") ("!) ;:i ~ _J'D ;· ~­ 0 ("!) '- -~ ~ ("!) ::l pO (Jl ~ 0... ::l -(Jl .... 0... ~ P> Q 0 ~ ::l 3 ;;· ~ 0 0 ("!) O' 15.: ~ 00 00 ~ ..... = = ("!) pO t:r' t:r' ::l = ::l ::-: (':) 00 00 ct> 0... "' :: -g -g iil ~. ;:.­ 3 ~. ~-->< <"I> "' 3~~~-d ---0 (1) ::l"' -. ..... (Jq (") Fie. 4. Distribution of the Species and Subspecies of Malacoctenus. ~ ci ::s ~ (;" ("1)3&-g.."' Symbols are as follows: M. boehlkei, a; M. macropus, b; M. erdamni. c; M. gilli, d; M. aurolineatus, e; M. triangulatus, f; M. versicolor. ::::I -ct> -· ,,.-... = ::l"' (") pO g; M. africanus, h; M. delalandei, k; M. margaritae mexicanus, I; M. margaritae margaritae, L; M. gigas, m; M. h.ubbsi hubbsi, n; M. 3 ("!) ~ :=:' g_ hubbsi polyporosus, N; M. afuerae mulaipunctatus, o; M. afuerae ajucrae. O; M. zonifer zonijer, p; M. zonifer sudensis, P; M. zacae, ~ ~ ~ 3 q; M. ebisui, r; M. costaricanus. s; M. zonogaster, t. -· (1) ct> -·::I ~ ~ 7 ; @" 0 ...... similarities: there are six members of the subgenus Labrisomus and one each of the subgenera Brockius and Gobioclinus in the Pacific, and one Labrisomus, two Brockius, and five Gobioclinus in the western Atlantic. Part of the importance of the above findings is that they serve as examples contradictory to the general findings of Ekman ( 1953) and Jordan (1905) that the tropical western Atlantic fauna is richer than the tropical eastern Pacific fauna. The clinid genera Cryptotrema, Mnierpes, Starksia, and Exerpes are found only in the tropical eastern Pacific, whereas all tropical Atlantic clinid genera are also found in the tropical eastern Pacific (pending revision of the emblem­ariids and chaenopsids I. Meek and Hildebrand 0923) indicated the existence of a tropical eastern Pacific fish fauna considerably larger than that of the tropical western Atlantic, and Briggs (1955) found a larger number of gobiesocid species in the tropical eastern Pacific than in the tropical western Atlantic. PACIFIC Hubbs (1952), in expanding on Ekman's (1935) Panamanian faunal region, pri­marily used species of Paraclinus as indicators of tropical eastern Pacific zoogeographic provinces. He considered that area to be divided as follows: I 11 a Panamanian province extending from an indefinite point in northern Peru to the vicinity of Bahia Honda, Panama, (2) a Mexican province extending from the boundary of the Panamanian province northward to a point somewhere north of Mazatlan, Sinaloa, on the mainland of Mexico, and a point near La Paz, at the southern tip of Baja California, and another point between Punta Abreojos and Port San Barto!me on the Pacific side of Baja Cali­fornia, and (3) a Gulf of California province with its southern limits near La Paz and somewhere north of Mazatlan. He proposed further subdivisions of the Mexican province as follows: (11 south of Guatemala to Bahia Honda, Panama, (21 the Mexican mainland, and (3 I the southern part of Baja California. Hubbs (1953) found what he considered only partial support for his original pro­posals from his study of eastern Pacific Labrisomus. He believed the demarcations of his Mexican from his Panamanian and Gulf of California provinces to be less sharp than he had previously proposed. He also believed his knowledge of Labrisomus distributions in the Mexican province to be incomplete on the basis of his Paraclinus studies. Briggs (1955) found what he thought were differences from the provinces as proposed by Hubbs if the boundaries were to be based solely on knowledge of gobiesocids. Briggs' data would indicate that the Mexican province should extend north of Mazatlan to Guaymas on the mainland. He interpreted Hubbs (19521 as setting a boundary "just north" of Mazatlan; whereas Hubbs' wording was "north of Mazatlan". From the data Hubbs gives on Paraclinus, this boundary could be interpreted as existing at Guaymas. It is unfortunate that he did not state this locality explicitly. Information gained from the study of Pacific Malacoctenus, and reinterpretation of Hubbs' and Briggs' findings, results in a more uniform, although by no means com­pletely satisfactory, concept of tropical eastern Pacific zoogeography. In the eastern Pacific there is an area occupying approximately seven hundred miles of coastline which has its northern border roughly at Salina Cruz at the middle of the Gulf of Tehuantepec, Mexico, and its southern border approximately at the Gulf of Fonseca, Nicaragua. From available maps and literature, the coast in this area appears Systematics and Zoogeography of the Clinid Fishes to be sand, mud, mangroves, and few or no rocks. The coast extending from this area both to the north, to Mazatlan, and to the south, to perhaps Peru, is frequently inter­rupted by rocky areas. A few collections (Beebe, 1938; Hildebrand, unpublished manuscript) are known to have been made from this intermediate block of coastline, and there are probably some others, but no specimens of the Clinidae, Gobiesocidae, or Tripterygiidae (Richard Rosenblatt, personal communication) have ever been reported from there. This Central American gap separates the subspecies of several sedentary forms: Ma!acoctenus a. afuerae from south of the gap and M. a. multipunctatus from the north; M. ;:;. sudensis from south of the gap and M. z. zonifer from the north; M. m. margaritae from south of the gap and M. m. mexicanus from the north. Hubbs ( 1953) re­ported distinctions between populations of labrisomus multiporosus from either side of the gap (l. multiporosus, which occupies the most northern locality of those species of labrisomus found south of the gap, reaches north, in that area, only to middle Pana­ma). Malacoctenus hubbsi, M. zacae, and labrisomus striatus are not found further south than the northern border of the gap, and M. costaricanus is not known to reach north to its southern boundary. Al-Uthman (1955) found that an undescribed species of the clinid genus Brannerella was separated into two morphological types which had their limits north and south of the gap respectively (a third morphological type was described from the Galapagos). Richard Rosenblatt informs me that there are several closely related tripterygiids which occur only on one side of the gap or the other. Gins­burg (1947) described Bathygobius r. curticeps from north of the gap and B. r. ramosus from south of it; he did, however, group a small sample of specimens from Ecuador and Peru with the northern form. Ginsburg (1953) also describes Dormitator l. mexicanus from north of the gap and D. l. latifrons from south of it. Only two gobiesocids (Tomi­codon petersi and Gobiesox adustus) are considered by Briggs (1955) to be undifferen­tiated on either side of the gap. Of nine other species inhabiting the mainland coast of the tropical eastern Pacific, eight are restricted to areas either to the north or to the south of the gap, and one (G. papillifer) is separated into different subspecies on either side of the gap. There is also evidence from the invertebrate fauna which suggests that the gap is real. Bayer (1953) found no species of the coelenterate family Gorgoniidae from the area of the gap. The relationships between the species from the two areas on either side were not mentioned. Ekman (1953) gives similar distributions for the crab genus Mithrax (after Rathbun, 1925). A perusal of Rathbun's paper indicates that many of the species are the same from either side of the gap; however, it must be pointed out that the sys­ tematics of infraspecific categories in marine invertebrates has received little attention. In light of the above evidence a southern boundary for the Mexican Province would be placed in the region of Tangola-Tangola Bay, as suggested by Briggs (1955). The northern limit of the Panamanian Province would be shifted to the Gulf of Fonseca. The area between these two points should perhaps be defined separately, for which I propose the name "Pacific Central American Fauna] Gap." There are two clinids which are apparently undifferentiated on either side of the Gap: Malacoctenus ebisui and Paraclinus mexicanus, although there appear to be subtle dif­ferences in color between specimens from the two localities in the first of these. Non­sedentary forms do not appear to be affected by the gap. If the "rncuum·' of the gap as defined above is an artificial one, it can at least be surmised that there is a quick change from the Panamanian to the Mexican provinces in that area. There is some paleogeological information which may have bearing upon the prob­lem. Schuchert I 1955 L in his paleogeological map of the Miocene, indicates an area between what are now the limits of the Gulf of Tehuantepec through which the Atlantic and Pacific were joined (Not all geologists agree that a Tehuantep~c oceanic bridge existed. See Durham, Arellano. and Peck. 1952). A second oceanic connection is shown occupying an area roughly betw(>en the Gulf of Fonseca, Nicaragua, and the Gulf of Chiriqui. Panama. The first channel was supposedly present during upper Miocene and lower Pliocene. and the second during lower and middle Miocene (Schuchert, 1935), although the possibility of the two having been for a short period contemporaneous is not negated. An area of coastline is situated between these two channe!s which is almost identical with that of the fauna I gap proposed here. This mass of land was most probably continuous with Jamaica in the Atlantic ISchuchert. 1955). What the marine fauna! distributions were during Miocene times cannot be told from present knowledge. but it is commonly believed that specific differences b!"tween the tropical Atlantic and Pacific fauna in this hemisphere did not occur until after the oceanic bridges had ceased to exist in either upper l\Iiocene or lower Pliocene. lnfraspecific differences (or in some cases specific differences I on either side of the Gap would have occurred after the oceanic connections had been closed for some time. This explanation still leaves unanswered the problem of how the species were able to maintain their integrity in earlier times in the areas on either side of the Gap, unless no barrier to gt-ne dispersal was present. There is no reason, on the basis of known geologv, to suspect that the coast was ecologically dif­ferent in Miocene-Pliocene time than at the present. Mountain building had already established a rocky coast north and south of the Gap by the Miocene. The mountains of the inland area comprising the Gap are of recent development and do not extend onto the coast as do the older northern and southern ranges. As stated above. Mazatlan is the approximate northern end of the rocky coast on the l\Iexican mainland south of the Gulf of California. From Mazatlan to Guaymas, on the mainland, the coast consists primarily of mangroves and river deltas. From Guaymas, which is rocky, north to Punta Pefiasco the coast is sandy in nature with isolated rocky points, the most northern of which is Punta Pefiasco. The extreme northern end of the Gulf is occupied by the delta of the Colorado River. The entire Gulf coast of Baja Cali· fornia can be considered essentially rocky. as are the numerous islands of the Gulf. This is also true for much of the Pacific coast of the peninsula. For the purposes of this study the Pacific coast is important only as far north as Bahia Sebastian Vizcaino, where con­siderable cold-,rnter upwelling occurs. This upwelling acts as a demarcation line be­t\\·een the tropical and temperate eastern Pacific faunas of the northern hemisphere. I am informed (Dr. B. \\'. Walker I, however. that most of the tropical forms drop out at a point further south. Bahia .Magdalena. My findings agree with those of both Hubbs (Mazatlan) and Briggs (Guaymas) for the northern boundaries of the :'.\Iexican proYince. :\Iazatlan is the northern limit for M. z. zonifer, M. hubbsi polyporosus, M. zacae, M. ebisui. and probably Labrisomus striatus on the Mexican mainland. Guaymas is the northern limit of :ll. margaritae mexicanus and the southern limit for M. gigas and M. h. hubbsi on the mainland. The area between Mazatlan and Guaymas is devoid of clinids (and gobiesocids), except for a single specimen of L. xanti known from Topolobampo Bay. The area between Mazatlan and Guaymas has probably acted as the isolating barrier separating M. hubbsi into two subspecies. Workers at the University of California at Los Angeles, inform me that there are many species of different ecological requirements which are separated by the Mazatlan· Guaymas faunal gap. Malacoctenus afuerae multipunctatus and Labrisomus multiporosus exceed both of the boundary lines for the northern limits of the Mexican province. On the Gulf coast of Baja California, M. z. zonifer reaches Espiritu Santo Island, but M. zacae and L. striatus scarcely reach the mouth of the Gulf. The northern limit of the Mexican province on the Pacific coast, Bahia Sebastian Vizcaino, is reached by L. xanti, L. multiporosus, and M. h. hubbsi. The Gulf of California province is represented in this study by only one endemic Malacoctenus, M. gigas; as are the Galapagos Islands with only one species of Malacoc­tenus: the endemic M. zonogaster. ATLANTIC Results of the present study agree with Ekman ( 1953) in indicating that the tropical western Atlantic is essentially a single zoogeographic unit without notable subdivisions. The coasts of Central and South America, the Gulf of Mexico, and Bermudas have a paucity of species as compared to the coasts of southern Florida, the Bahamas, and the Antillean chain, and no species are present in the former areas which are not found in the latter. Only one labrisominid occurring in this region, M. delalandei is not known from above 20° N. latitude. Of the sixteen species of labrisominids in the western At­lantic, the largest number, thirteen, are known from the northern Bahamas, and one, M. boehlkei, is known only from that area. This concentration of species compared to the number known from the eastern Antillean chain may be an artifact of collecting, al· though Erdman (1957) notes a general decrease in the number of fish species from the West Indies as one proceeds from west to east. Because of the extreme distance separating the Bermudas from other tropical coastlines, a paucity of tropical fish species in those islands is not to be unexpected. There are three areas which are not known to have clinid faunas. These are the coast· lines of the northern Gulf of Mexico, northern South America between eastern Vene· zuela and Natal, Brazil, and between eastern Honduras and middle Panama. All of these are characterized by the presence of any or all of the following: large river deltas, mud, sandy beaches, and mangroves. These areas do not appear to have an isolating effect on the populations on either side of them as the currents of the region wash all shores from south to north and are able to distribute the planktonic larvae of some of the species throughout the region. An example of this is well demonstrated by the sandy coasts of northern Mexico and Texas which are not proper habitats for labrisominids. About seventy years ago rocky jetties were built at Brownsville, Port Aransas, and Galveston, Texas, providing a suitable habitat in these areas for the first time. The jetties at the first two localities are now inhabited by Labrisomus nuchipinnis; those at Galveston do not support any clinids, possibly because of the low salinities of this locality. The most interesting element of the Atlantic labrisominid fauna is L. nuchipinnis. This species has the widest distribution of any known clinid and exceeds the ranges of any of the other clinids in the area it occupies. It is interesting to note that L. multi­ porosus, the nearest relative of L. nuchipinnis, has the widest distribution of any Pacific Labrisomus species. Within the tropical western Atlantic, variation does occur in several of the labriso­ minid species. Such variation is most notable in L. nuchipinnis, L. haitiensis, M. ma­cropus, M. erdmani, and M. gilli, but there is no discernible geographic pattern to this variation which would afford an explanation for it. Only two clinids are found in the tropical eastern Atlantic, L. nuchipinnis and the endemic M. africanus. The range of the former shows only slight agreement with the limits of the tropical eastern Atlantic as given by Ekman (1953, after W. Michaelson; reference not given l: Cape Verde Islands to Mossamedes, Angola. Labrisomus nuchi­pinnis is known from as far north as the Madeira Islands and only as far south as An­nobon Island. This latter limit is of interest as it lies in the area where the warm Guinea Current from the north meets the cold Benguela Current from the south and together leave the coast as the South Equatorial Current. The problem arises as to the locality or origin of Labrisomus and Malacoctenus. The absence of living species or fossils from the Indo-Pacific probably indicates their ab­sence from that region at any time. The presence of a Miocene fossil, Labrisomus pro­nuchipinnis Arambourg !1921, 1927), from the Mediterranean coast would tend to support Steinitz (1950) conclusion that Labrisomus on the African coast is a Tethys relic. Further support for this argument would be the endemic Malacoctenus africanus, known only from Dakar. The form?r existence of a uniform Tethys fauna would still not explain how species crossed the Atlantic. The predominent east-to-west flow of the oceans would favor an African origin, but the significantly larger American fauna would indicate a center of origin in this hemisphere. Hubbs (1953) believes that La­brisomus, with a long pelagic larval stage, crossed from Brazil to Africa via an equa­torial counter-current. If this is true, we are still in need of an explanation for the distri­bution of Malacoctenus which apparently does not have planktonic larvae (specimens of 10 or 12 mm. are collected with adults). It is interesting that Cadenat (1950) reports numerous predominantly western Atlantic fishes from the west African Coast, and Burkenroad !1939) reports that the western Atlantic shrimp, Penaeus duorarum is also found on the west African coast. Because of the presence of a Miocene fossil Labriso· mus, crossings would have to have occurred prior to, or during, the Miocene. The presence or absence of a counter equatorial current at that time is not known, neither is the mechanism allowing L. nuchipinnis to remain undifferentiated on either side of the Atlantic. Further speculation on these matters is considered useless until more his­torical and oceanographic evidence is available. Summary The subtribe Labrisomini is comprised of the genera Labrisomus and Malacoctenus. These fishes are sedentary and predominantly restricted to the rocky coasts and coral reefs of the tropical marine waters of the western hemisphere. A systematic revision of these genera reveals the existence of sixteen species, and no subspecies, of Labrisomus, and eighteen species, and eight subspecies, of Malacoctenus. The following taxa are described as new: Malacoctenus boehlkei and Malacoctenus triangulatus from the western Atlantic; M alacoctenus hubbsi hubbsi, M alacoctenus hubbsi polyporosus, Malacoctenus margaritae mexicanus, Malacoctenus gigas, Mala­ Systematics and Zoogeography of the Clinid Fishes coctenus a/uerae multipunctatus, Malacoctenus zoni/er sudensis, Malacoctenus zacae, Malacoctenus costaricanus, and Malacoctenus ebisui from the eastern Pacific. No species of either Malacoctenus or Labrisomus are common to both the Atlantic and Pacific. The present study indicates five expanses of coastline in this hemisphere which are de­void of clinid fishes, as well as other groups of sedentary fishes and invertebrates: 1) between Guaymas, Sonora, Mexico, and Mazatlan, Sinaloa, Mexico, 2) between the middle of the Gulf of Tehuantepec, Oaxaca, Mexico, and the Gulf of Fonseca, Nicaragua, 31 the northern coast of the Gulf of Mexico, 4.) the Atlantic coast between eastern Hon­duras and central Panama and 5) the coast between eastern Venezuela and Natal, Brazil. All of these sections of coastline are characterized by the presence of river deltas, sand, mud, or mangroves; conditions unsuitable as clinid habitats. The first two areas listed act as barriers to gene, dispersal since several species are represented on either side of their boundaries by subspecies; other species do not exceed either one or the other of their boundaries. On the basis of the present study the northern limit of the Panamanian marine fauna] province is located at the Gulf of Fonseca, Nicaragua, and the southern limit of the Mexican marine faunal province is placed at about the middle of the coast of the Gulf of Tehuantepec, Mexico. For the area between these two localities is proposed the name "Pacific Central American Fauna] Gap." The present study indicates that the northern boundary of the Mexican marine fauna! province on the Mexican mainland can be placed at either Mazatlan, Sinaloa, or Guaymas, Sonora, depending upon the species used as indicators. LITER..\TURE CITED Al-Uthman, H. S. 1955. The classification and anatomy of the blennioid fishes of the tribe Starksiidi. Ph.D. Dissertation, Univ. of Texas. Arambourg, C. 1921. Sur la faune ictyologique d'Oran. C. R. Acad. Sci., Paris, 172 (20): 1232-1245. ----. 1927. Les poisson fossiles d'Oran. Texte. Mater. Carte geol. Alger., Ser. 1, Paleontologie, 6: 1-298. Barbour, T. 1905. Notes on Bermudian fishes. Bull. l\lus. comp. Zool. Harv., 46 (7): 109-134. Barnhart, P. S., and C. L. Hubbs. 1946. Pontinus vaughani, a new scorpaenid fish from Baja Cali· fornia. Bull. Scripps lnstn Oceanogr., 5 I 51 : 371-390. Baughman, J. L. 1947. Fishes not previously reported from Texas, with miscellaneous notes on the species. Copeia, 4: 280. ----. 1950. Random notes on Texas fishes. Part II. Tex. J. Sci., 2 (2): 242-263. Bayer, F. :\I. 1953. Zoogeography and eYolution of the octocorallian family Gorgoniidae. Bull. Mar. Sci. Gulf & Caribb., 3 (2 I: 100-119. Bean, B. A. 1905. Fishes of the Bahamas Islands. In The Bahama Islands, ed. G. B. Shattuck. Johns Hopkins Press. pp. 293-325. Bean. T. H. 1906a. Descriptions of new Bermudian fishes. Proc. biol. Soc. Wash., 19 : 29-34. ----. 1906b. A catalogue of the fishes of Bermuda. Fieldiana, Zool., 7 (21: 21-92. Beebe, W. 1938. Eastern Pacific expeditions of the New York Zoological Society, XIV. Zoologica, N.Y., 23 !3J: 287-298. ----and G. Hollister. 1935. The fishes of Union Island, Grenadines, British West Indies,with description of a new species of star-gazer. Zoologica, N.Y., 19 (61 : 209-224. ----and J. Tee Van. 1928. The fishes of Port·au·Prince Bay, Haiti. Zoologica, N.Y., 10 (1): 1-270. ----and . 1933. Field book of the shore fishes of Bermuda. N.Y., G. P. Putnam's Sons, 337 pp. Bohlke, J. 1953. A new stathmonotid blenny from the Pacific coast of Mexico. Zoologica, N.Y., 38 (31 : 145-149. Borodin, N. A. 1928. Scientific results of the yacht "Ara" expedition during the years 1926 to 1928,while in command of William K. Vanderbilt. Bull. Vanderbilt Oceanogr. (Mar.l Mus. 1 OJ: 1-37. Boulenger, G. A. 1899. Viaggio de! Enrico Festa nell'Ecuador e regioni vicine. XIV. Poissons de l'Equateur. Boll. Mus. Zoo!. Anal. comp. Torino, No. 335, 14 : 1-8. ----. 1905. Poissons de la Guinee Espangnole. Mem. Soc. Esp. Hist. Nat., 1 (9): 188. Breder, C. M., Jr. 1925. Notes on fishes from three Panama Localities. Zoologica, N.Y., 4 (4): 137-158. ----. 1927. Scientific results of the first oceanographic expedition of the "Pawnee," 1925. Bull. Bingham, oceanogr. Coll., 1 (1): 1-90. ----. 1929. Field book of marine fishes of the Atlantic coast from Labrador to Texas. N.Y., G. P. Putnam's Sons, 332 pp. Revised, 1948. Briggs, J. C. 1955. A monograph of the clingfishes (Order Xenopterygii) . Stanf. ichthyol. Bull., 6: 1-224. Burkenroad, M. D. 1939. Further observations on Penaeidae of the northern Gulf of Mexico. Bull. Bingham oceanogr. Coll., 6 ( 6) : 1-62. Butsch, R. S. 1939. A list of Barbadian fishes. J. Barbados Mus., 7 (1): 17-31. Cadena!, J. 1950. Poissons de mer du Senegal. lniats. Africa. III. Dakar, Inst. Fran. d'Afriq. Noire, 345 pp. Castelnau, F. L. 1855. Expedition clans Jes parties centrales de l'Amerique du Sud, de Rio de Janeiro aLima, et de Lima au Para, executee ... pendant Jes annees 1843 a 1847. 3 (7 J: 112 pp. Cockerell, T. D. A. 1892. A provisional list of the fishes of Jamaica. Bull. Inst. Jamaica, 1: 1-16. Cope, E. D. 1871. Contribution to the ichthyology of the Lesser Antilles. Trans. Amer. phil. Soc., 14: 445-483. Cuvier, G. L. C. F. D. and A. Valenciennes. 1836. Histoire naturelle des poissons. Paris. 11: 506 pp.Octavo edition. De Kay, J. E. 1842. Zoology of New York, or the New York fauna. pl. IV. Fishes. Albany, W. and A. White, 415 pp. Dumeril, A. 1858. Poisson de la cote occidentale d'Afrique. Arch. Mus. Hist. nat. Paris, 10: 241-264. Durham, J. W., A. R. V. Arellano, and J. H. Peck, Jr. 1952. No cenozoic Tehuantepec seaways (ab·stract). Bull. geol. Soc. Amer., 63 (12, pt. 2) : 1245. Ekman, S. 1935. Tiergeographie des Meeres. Leipzig. Akad. Verlagsges., 512 pp. ----. 1953. Zoogeography of the sea. London, Sidwick &Jackson, Lmt. 417 pp. Erdman, D.S. 1957. Recent fish records from Puerto Rico. Bull. Mar. Sci. Gulf & Caribb., 6 (4): 315­ 340. Evermann, B. W. and W. C. Kendall. 1900. Checklist of the fishes of Florida. Rep. U. S. Comm. Fish., 25: 35-103. Evermann, B. W. and M. C. Marsh. 1900a. Descriptions of new genera and species of fishes from Porto Rico. Rep. U.S. Comm. Fish., 25: 351-362. ----and . 1900b. The fishes of Porto Rico. Ibid., 20: 51-350. Fowler, H. W. 1899. A list of fishes collected at Port Antonio, Jamaica. Proc. Acad. nat. Sci. Philad., 51: 118-119. ----. 1915. A list of Santo Domingo Fishes. Copeia, 24: 49-50. ----. 1916. Cold-blooded vertebrates from Florida, the West Indies, Costa Rica, and eastern Brazil. Proc. Acad. nat. Sci. Philad., 67: 244-269. ----. 1920. Notes on tropical American fishes. Ibid., 71: 128-155. ----. 1928. Fishes from Florida and the West Indies. Ibid., 80 : 451-473. ----. 1930. The fishes obtained by Mr. James Bond at Grenada, British West Indies, in 1929. I bid., 82: 269-277. ----. 1931. Fishes obtained by the Barber Asphalt Company in Trinidad and Venezuela in 1930. lbid., 83: 391-410. ----. 1936. The marine fishes of west Africa. Bull., Amer. Mus. nat. Hist., 70 (2) : 607-1493. ----. 1938. The fishes of the George Vanderbilt South Pacific Expedition. Monogr. Acad. nat. Sci. Philad., 2: 349 pp. ----. 1940. Fishes obtained by the Wilkes Expedition, 1838-1842. Proc. Amer. phil. Soc., 82 (5) : 733-800. - ---. 1941. Notes on Florida fishes with descriptions of seven new species. Proc. Acad. nat. Sci. Philad., 93: 81-106. ----. 1942b. Fishes observed or obtained in Cuba in 1934 (con't.). Progr. Fish Cult., 21 (10): 75-77 and 79. ----. 1942c. Notes on marine fishes from Honduras. Ibid., 22 (2): 9-12. ----. 1944. Results of the Fifth George Vanderbilt Expedition (1941) .... The fishes. Monogr. Acad. not. Sci. Philad., 6: 57-529. ----. 1945. A study of the fishes of the southern piedmont and coastal plain. Ibid., 7: 408 pp. ----. 1947. Notes on Bahama fishes obtained by Mr. Charles G. Chaplin in 1947, with descrip­tions of two new species. Notul. nat. Acad. Philad., 199: 1-14. ----. 1950. Results of the Catherwood-Chaplin West Indies Expedition, 1948, Pt. III. The fishes. Proc. Acad. nat. Sci. Philad., 102: 69-93. ----. 1951. The Brazilian and Patagonian fishes of the Wilkes Expedition 1838-1842. Bol. Inst. Paul. Oceanogr., 2 (1): 3-39. ----. 1952. The fishes of Hispaniola. Mem. Soc. cubana Hist. nat., 21 (1) : 83-115. ----.1953. The shore fishes of the Colombian Caribbean. Caldasia, 6 (27): 43-73. Garman, S. 1896. Report on the fishes collected by the Bahama Expedition of the State University of Iowa, under Professor C. C. Nutting, in 1893. Bull. Labs. nat. Hist. Univ. la., ~: 7&--93. Gilbert, C. H. 1890. Scientific results of explorations by the U. S. Fish Commission Steamer Alba­tross. XIX. A supplementary list of fishes collected at the Galapagos Islands and Panama, with descriptions of one genus and three new species. Proc. U.S. nat. Mus., 13: 449-455. ----. 1900. Results of the Branner-Agassiz Expedition to Brazil. III. The fishes. Proc. Wash. Acad. Sci. 2: 161-184. ----. and E. C. Starks. 1904. The fishes of Panama Bay. Mem. Calif. Acad. Sci., 4: 1-304. Gill, T. 1860a. Notes on the nomenclature of North American fishes. Proc. Acad. nat. Sci. Philad., 12: 19-21. ----. J860b. Monograph of the genus Labrosomus. Sw. Ibid., 102-108. Ginsburg, I. 1947. American species and subspecies of Bathygobius, with a demonstration of a sug­ gested modified system of nomenclature. J. Wash. Acad. Sci., 37 (8) : 275-284. ----. 1953. Ten new American gobioid fishes in the United States National Museum, including additions to a revision of Gobionellus. I bid., 43 (1) : 18-26. Goode, G. B. 1876. Catalogue of the fishes of the Bermudas. Bull. U.S. nat. Mus., 5: 1-82. ----. 1877a. Provisional catalogue of the fishes of Bermuda. 1-8 (Possibly this is a news­ paper item. A letter addressed to the editor of the Bermuda Royal Gazette is frontispiece. A copy is on file at the U.S. National Museum). ----. l877b. A preliminary catalogue of the reptiles, fishes, and leptocardians of the Ber­mudas. Amer. J. Sci. 14: 289-298. ----and T. H. Bean. 1883. A list of the species of fishes recorded as occurring in the Gulf of Mexico. Proc. U.S. nat. Mus., 5: 234-240. Guerin-Meneville, F. E. 1829-1844 (exact date not established). Iconographie du Regne Animal de G. Cuvier. Tome I. Planches des animaux vertebres. Paris. Guimaraes, A. R. P. 1884. Lista dos peixes de Ilha da Madeira, A~ores e das possessoes Portugezas d'Africa que existem no Museu de Lisboa. J. Sci. math. phys. nat., Lisboa, 9: 11-28. Gunther, A. 1861A. On a collection of fishes sent by Capt. Dow from the Pacific coast of Central America. Proc. zoo!. Soc. Lond., 1861: 370-376. ----. 186lb. Catalogue of the acanthopterygian fishes of the British Museum. London, Adlard & Son (Photolith). Vol. 3: 1-586. ----. 1868. An account of the fishes of the states of Central America, based on collections made by Capt. J.M. Dow, F. Godman, Esq., and 0. Salvin, Esq. Trans. zoo!. Soc. Lond., 6 (7): 377-494. Heller, E. and R. E. Snodgrass. 1903. Papers from the Hopkins-Stanford Galapagos Expedition 1898­ 1899. XV. New fishes. Proc. Wash. Acad. Sci., 5: 189-229. Herre, A. W. C. T. 1936. Fishes of the Crane Pacific Expedition. Fieldiana, Zoo!., 21: 472 pp. ----. 1942. Notes on a collection of fishes from Antigua and Barbados, British West Indies. Stan£. Univ. Puhl., Biol. Sci., 7 (2) : 1-21. Hildebrand, S. F. 1946. A descriptive catalogue of the shore fishes of Peru. Bull. U. S. nat. Mus., 189: 530 pp. Hubbs, C. 1952. A contribution to the classification of the blennioid fishes of the family Clinidae, with a partial revision of the eastern Pacific forms. Stanf. ichthyol. Bull., 4 (2) : 41-165. ----. 1953. Revision of the eastern Pacific fishes of the Clinid genus Labrisomus. Zoologica, N.Y., 38 (3): 113-136. ----and V. G. Springer. 1954. The taxonomic position of Gobioclinus gobio and Ctenichthys interrupta, two names for blennioid fishes. Bull. Mar. Sci. Gulf & Caribb., 4 (4): 346-c350. Irvine, F. P. 1947. The fishes and fisheries of the Gold Coast. London, The Crown Agents for the Colonies, 352 pp. Jordan, D. S. 1885. A list of the fishes known from the Pacific coast of tropical America from the Tropic of Cancer to Panama. Proc. U. S. nat. Mus., 8: 361-394. ----. 1887a. A preliminary list of the fishes of the West Indies. Ibid., 9: 554-608. ----. 1887b. A catalogue of the fishes known to inhabit the waters of North America, north of the Tropic of Cancer, with notes on the species discovered in 1883 and 1884. Rep. U.S. Comm. Fish., 13: 789-973. ----. 1889. List of fishes collected by Alphonse Forrer about Mazatlan, with descriptions of two new species-Heros beani and Poecilia butleri. Proc. U. S. nat. Mus., 11: 329-334. ----. 1895. The fishes of Sinaloa. Proc. Calif. Acad. Sci., ser. 2, 5: 377-514. ----. 1904. Notes on fishes collected in the Tortugas Archipelago. Bull. U. S. Fish Comm., 22: 539-544. ----. 1905. A guid 2 to the study of fishes. Vol. I, New York, Henry Holt Co., 624 pp. ----and B. W. Evermann. 1896. A check-list of the fishes and fishlike vertebrates of North and Middle America. Rept. U.S. Comm. Fish., 21: 209-584. -----and . 1898. The fishes of North and Middle America. Bull. U. S. nat. Mus., 47 (3 l : 2183a-3136. ----and -----. 1900. Ibid., (4): 3137-3313. ----and H. W. Clark. 1930. Checklist of the fishes and fishlike vertebrates of North and Middle America north of the northern boundary of Venezuela and Colombia. Rep. U.S. Comm. Fish, 1928, (2): 1-670. ----and C. H. Gilbert. 1882a. Synopsis of the fishes of North America. Bull. U. S. nat. Mus., 16 : 1018 pp. ----and . 1882b. Descriptions of 33 new species of fishes from Mazatlan, Mexico. Proc. U.S. nat. Mus., 4: 338-365. ----and . 1883. List of fishes collected at Mazatlan, Mexico by Charles H. Gilbert. Bull. U.S. Fish Comm., 2: 10~108. ----and J. A. Gunn, Jr. 1898. List of fishes collected at the Canary Islands by Mr. 0. F. Cook, with descriptions of four new species. Proc. Acad. nat. Sci. Philad., 50: 339-346. ----and C. Rutter. 1897. A collection of fish es made by Joseph Seed Roberts in Kingston, Jamaica. Ibid., 49: 91-133. ----and J. C. Thompson. 1905. The fish fauna of the Tortugas Archipelago. Bull. U. S. Fish Comm., 24: 229-256. Kanazawa, R. H. 1952. More new species and new records of fishes from Bermuda. :Fieldiana, Zoo!. 34: 71-100. Kendall, W. C. and L. Radcliffe. 1912. Report of the scientific results of the explo.ration to the eastern tropical Pacific by the "Albatross." The shore fishes. Mem. Mus. comp. Zoo!. Harv., 35 (3): 75­ 171. Kner, R. 1868. iv. Folge neuer Fische aus dem Museum der Herren Joh. Caes. Godeffroy und Sohn in Hamburg. S. B. Akad. Wiss. Wien, 58: 293-356. Lee, T. 1889. List of fish taken by the steamer Albatross among the Bahama Islands and at the Nassau fish-market during March and April 1886. Rep. U. S. Comm. Fish. 1886, (14) : 669-672. Le Sueur, C. A. 1825. Description of the two new species of the Linnaean genus Blennius (B. herminier, B. hentzi). J. Acad. nat. Sci. Philad., 4: 361-364. Longley, W. H. 1932. Preparation of a monograph on the Tortugas fishes. Year B. Carneg. lnstn., 31: 299-301. ----. 1933. Ibid., 32 : 293-295. ----and S. F. Hildebrand. 1941. Systematic catalogue of the fishes of Tortugas, Florida, with observations on color, habits and local distribution. Pub!. Carneg. Inst. 435, Pap. Tortugas Lab., 34: 1-331. Lockington, W. N. 1881. List of fishes collected by Mr. W. J. Fisher upon the coasts of Lower Califor· nia, 1876-1877, with descriptions of new species. Proc. Acad. nat. Sci. Philad., 33: 113-120. Manter, H. W. 1947. The digenetic trematodes of marine fishes of Tortugas, Fla. Amer. Midi. Nat., 38 (2): 257-416. ----. 1954. Trematoda of the Gulf of Mexico. Fish. Bull., U.S., 89: 335-350. Mdndoo, N. E. 1906. On some fishes of western Cuba. Proc. Acad. nat. Sci. Philad., 58 : 484-488. Meek, S. E. and S. F. Hildebrand. 1923. The marine fishes of Panama. Pt. 1, Fieldiana Zool., 215: 1-330. ----and . 1928. Ibid., Pt. 3, 244: 708-1045. Metzelaar, J. 1919. Over tropisch Atlantische visschen. Amsterdam, A.H. Kruyt. 314 pp. ----. 1922. On a collection of marine fishes from the Lesser Antilles. Bijdr. Dierk., Arnst., 22: 133-141. Mocquard, F. 1889. Revision des Clinus de la collection du museum. Bull. Soc. philom. Paris, ser. 8, 1: 40-46. Miiller, J. W. 1864. Reisen in den Vereinigten Staaten, Canada, und Mexico. Vol. III. Leipzig. 643 pp. Nichols, J. T. 192la. A list of Turk Islands fishes, with a description~ of a new flatfish. Bull. Amer. Mus. nat. Hist., 44 (3): 21-24. ----. 192lb. A new pomacentrid and blenny from Bahamas. Amer. Mus. Novit., 26: 2 pp. ----. 1930. The fishes of Porto Rico and the Virgin Islands: Pomacentridae to Ogcocephalidae. Sci. Surv. P.R., 10 (3) : 297-399. ----and R. C. Murphy. 1914. Fishes from South Trinidad Islet. Bull. Amer. Mus. nat. Hist., 33 (20) : 261-266. ----and----. 1944. A collection of fishes from the Panama Bight, Pacific Ocean. Ibid., 83 (4): 217-260. Norman, J. R. 1922. Fishes from Tobago. Ann. Mag. nat. Hist., ser. 9, 9: 533-535. Osburn, R. C. and J. T. Nichols. 1916. Shore fishes collected by the "Albatross" expedition in Lower California with descriptions of new species. Bull. Amer. Mus. nat. Hist., 35: 139-181. Osorio, B. 1896. Les Poissons d'eau douce des iles du Golfe de Guinee. J. Sci. math. phys. nat., Lisboa, 2, ser. 4: 59-64. ----. 1898. Da distribuic;ao geographica dos peixes e crustaceos colhidos nas possessoes Portu­guezas d' Africa. Ibid., 2, ser. 4: 185--202. Parr, A. E. 1930. Teleostean shore and shallow-water fishes from the Bahamas and Turks Islands. Bull. Bingham oceanogr. Coll., 3 (41: 1-148. Perugia, A. 1896. Sopra alcuni pesci raccolti alle Antille de! Cap. Guiseppe Capurro. Ann. Mus. Stor. nat. Genova, 16 (2): 14-19. Peters, W. 1877. -Ober die von Professor Dr. Reinhold Buchholz in Westafrica gesammelten Fische. Monatsb. Akad. Wiss., Berlin. pp. 244-252. Poey, P. 1861. Memorias sobre la historia de la Isla de Cuba. Vol. II. Havana. pp. 95-427. ----. 1866. Revista de los tipos Cuvierianos y Valenciennianos correspondientes alos peces de la Isla de Cuba. Vol. I. Havana. pp. 193-338. ----. 1868. Synopsis piscium Cubensium. Vol. II. Havana. pp. 279-484. ----. 1875. Enumeratio pisium cubensium. An. Soc. esp. Hist. nat., 5: 131-218. ----. 1880. Revisio pisium Cubensium. Ibid., 9: 243-261. Quoy, J. R. C. and P. Gaimard. 1824. Voyage autour du monde ... execute sur Jes corvettes de S. M. "L'Uranie" et "La Physicienne" pendant Jes annees 1817-20. Poissons. Paris. pp. 192-401. Rathbun, M. J. 1925. Spider crabs of America. Bull., U.S. nat. Mus., 129: 613 pp. Reid, E. D. 1935. Two new fishes of the families Dactyloscopidae and Clinidae from Ecuador and the Galapagos. Copeia, 4: 163-166. Ribeiro, A. M. 1915. Fauna Brasiliense. Peixes V. Arch. Mus. nae. Rio de J. 15: pagination not sequential. Rivero, L. H. 1936. Some new, rare and little-known fishes from Cuba. Proc. Boston Soc. nat. Hist., 41 (4): 41-76. Rochebrune, A. T. 1880. Description de quelques nouvelles especes de poissons propres a la Sene­gambie. Bull. Soc. philom. Paris, ser. 7, 3 (3): 159-169. ----. 1882. Faune de la Senegambie: Poissons. Act. Soc. Jinn. Bordeaux, ser. 4, 25 (6) : 23-190. Rosen, N. 1911. Contributions to the fauna of the Bahamas 1-3. Acta. Univ. Jund. Afd. 2, 7 (5): 1-72. Roule, L. and F. Angel. 1930. Larves et alevins de poissons provenant des croisieres du Prince Albert ler de Monaco. Result. Camp. sci. Monaco, 79: 148 pp. Schmitt, W. L. and L. P. Schultz. 1940. List of the fishes taken on the Presidential cruise of 1938 Smithson. misc. Coll., 98 (251 : 1-10. Schomburgk, R. H. 1848. A description of some new species of fishes from the sea surrounding the Island of Barbados. Ann. Mag. nat. Hist., ser. 2, (21 : 11-20. Schuchert, C. 1935. Historical geology of the Antillean-Caribbean region. New York, John Wiley and Sons, Inc., 8ll pp. -----1955. Atlas of Paleographic maps of North America. New York, John Wiley and Sons, Inc., 177 pp. Schultz, L. P. 1949. A further contribution to the ichthyology of Venezuela. Proc. U. S. nat. Mus., 99: 1-211. Seale, A. 1940. Report on fishes from Allan Hancock Expeditions in the California Academy of Sciences. Allan Hancock Paci£. Exped., 9 (1): 1-46. Smith, C. L. 1957. Two new clinid blennies (Malacoctenus) from Puerto Rico. Occ. Pap. Mus. Zoo!. Univ. Mich., 585: 1-15. Smith, R. 1885. Notes on fishes collected at San Cristobal Lower California, by Mr. Charles H. Townsend, Assistant, U.S. Fish Commission. Proc. U.S. nat. Mus., 7: 551-553. Snodgrass, R. E. and E. Heller. 1905. Papers from the Hopkins-Stanford Galapagos Expedition, 1898-1899. 17. Shore fishes of the Revillagigedo, Clipperton, Cocc.s, 1rnd Galapagos Islands. Proc. Wash. Acad. Sci., 6: 333-427. Springer, S. and H. R. Bullis, Jr. 1956. Collections by the Oregon in the Gulf of Mexico. Spec. Sci. Rep. U.S. Fish Wild!. Serv., 196: 134 pp. Springer, V. G. 1955a. Western Atlantic fishes of the genus Paraclinus. Tex. J. Sci., 4 (41: 422-441. -----. 1955b. The taxonomic status of the genus Stathmonotus, including a review of the Atlantic species. Bull. Mar. Sci. Gulf & Caribb., 5 ( 1) : 66-80. Starks, E. C. 1913. The fishes of the Stanford expedition to Brazil. Stan£. Univ. Puhl. 77 pp. Steindachner, F. 1867. lchthyologische notizen (VI! VII. Uber einige neue oder seltene Fischarten von Westindien und Surinam. S. B. Akad. Wiss. Wien., 56 (1): 347-357. ----. 1876. Ichthyologische Beitrage (V). V. Uber einige neue oder seltene Fischarten dem Atlantischen, lndischen, und Stillen Ocean. I bid., 74 (1) : 203-240. Steinitz, H. 1950. On the zoogeography of the teleostean genera Salarias, Ophioblennius, and Labrisomus. Arch. zool. (ital.), Napoli., 35: 325-348. Storer, D. H. 1846. Synopsis of the fishes of North America. Mem. Amer. Acad. Arts Sci. Boston, 2: 253--550. Swainson, W. 1839. The natural history and classification of fishes, amphibians, and reptiles, or monocardian animals. Vol. II. London. 452 pp. Tortonese, E. 1951. Materiali per lo studio sistematico e zoogeographico dei pesci delle coste occi­dentali del Sud America. Rev. chil. Hist. nat., '1947-1949 r: 33-ll8. Valenciennes, A. 1839. lchthyologie des iles Canaries. In Webb and Berthelot. Histoire naturelle des iles Canaries. Paris, 2 (2) : 109 pp. Vinciguerra, D. 1883. La crociere dell'yacht "Corsaro" de! Capitano Armatore Enrico d'Albertis. III. Pesci. Ann. Mus. Stor. nat. Genova., 18: 607--020. ----. 1893. Catalogo dei pesci delle !sole Canarie. Atti Soc. ital. Sci. nat., 34: 295-334. J\( FLORIDA STRAITS 0 10 KILOMETERS '===================~~~~~~~~~==:I SC ALE LAND ROCK BOTTOM MIXED SAND OOLITE SAND DENSE GRASS SPARSE GRASS LO NORMAN D. NEWELL JOHN IMBRIE BIMINI CAT CAY LOUIS S. KORNICKER PRELIMINARY MAP EDWARD G. PURDY DISTRIBUTION OF BOTTOM TYPES JANUARY. 1957 WILLIAM G. HEA SLIP Figure 26 in Korn1cker, L S 1958 Ecology and Taxonomy of Recent Marine Ostracodes in the B1m1n1 Area, Great Bahama Bank Puhl Inst Mar Sci Texas, Volume 5 The Publications of the Institute of Marine Science is printed annually by The University of Texas including papers of basic or regional importance by Gulf workers or on Gulf waters, in the fields of Bacteriology, Botany, Chemistry, Geology, Meteorology, Physics, Zoology and other marine sciences. Papers are read by three referees; papers by the editors are processed by the Chairman of the BudgetCouncil. Editorial Staff for Volume 5 includes: Editor, H. T. Odum; Assistant Editor, William N. McFarland;and Editorial Assistant, Mrs. Anna K. McFarland.Note to Librarians: Beginning with this publication, each annual issue will be numbered as a separatevolume. EDITORIAL ADVISORY COMMITTEE OF THE PUBLICATIONS W. Frank Blair, Department of Zoology, The University of Texas, Austin, Texas.Harold C. Bold, Department of Botany, The University of Texas, Austin, Texas.Albert Collier, U. S. Fish &Wildlife Service, Galveston, Texas. R. L. Folk, Department of Geology, The University of Texas, Austin, Texas.Marcus A. Hanna, Gulf Oil Corporation, Houston, Texas. Willis G. Hewatt, Department of Biology and Geology, Texas Christian University, Fort Worth, Texas.Donald W. Hood, Department of Oceanography, Agricultural and Mechanical College of Texas, College Station, Texas. Sewell H. Hopkins, Department of Biology, Agricultural and Mechanical College of Texas, College Station, Texas. Clark Hubbs, Department of Zoology, The University of Texas, Aust~n, Texas.Edward Jonas, Department of Geology, The University of Texas, Austin, Texas. R. J. LeBlanc, Shell Oil Company, Houston, Texas. Howard T. Lee, Marine Division, Texas Game and Fish Commission, Rockport, Texas. W. Armstrong Price, Wilson Building, Corpus Christi, Texas. Cecil Reid, Sportsmans Club of Texas, Austin, Texas. Robert 0. Reid, Department of Oceanography, Agricultural and Mechanical College of Texas, College Station, Texas. R. C. Staley, Meteorology Section, Department of Aeronautical Engineering, The University of Texas, Austin, Texas. INSTRUCTIONS TO AUTHORS Manuscripts should be prepared according to directions given in the inside cover of the journal,Limnology and Oceanography. All bibliographic abbreviations should be taken from World List ofScientific Periodicals, N. Y., Academic Press, 1952. Price of this volume $4.25