Copies of this publication may be procured from the GENETICS FOUNDATION, DEPARTMENT OF ZOOLOGY The University of Texas, Austin, Texas 78712 • studies zn GENETICS· V Dedicated to the Memory of Professor Wilson S. Stone The research articles presented here are contributions from the faculty, staff, pre-and postdoctoral students and guest investigators working on some of the research programs of the Genetics Foundation, Depart­ment of Zoology, the University of Texas, or on the program on endemic Drosophilidae of the Hawaiian Islands. We are pleased to dedicate this volume to the memory of the man who was largely responsible for the developm::nt of these projects-Wilson Stuart Stone. EDITED BY MARSHALL R. WHEELER PROFESSOR OF ZOOLOGY THE UNIVERSITY OF TEXAS AUSTIN 1 9 6 9 The benefits of education and of useful knowledge, generally diffused through a community, are essential to the preservation of a free government. SAM HOUSTON Cultivated mind is the guardian genius of Democracy, and while guided and controlled by virtue, the noblest attribute of man. It is the only dictator that freemen acknowledge, and the only security which freemen desire. MIRABEAU B. LAMAR THE UNIVERSITY OF TEXAS PUBLICATION NUMBER 6918 SEPTEMBER 15, 1969 PUBLISHED TWICE A MONTH BY THE UNIVERSITY OF TEXAS AT AUSTIN, UNIVERSITY STATION, AUSTIN, TEXAS 78712. SECOND-CLASS POSTAGE PAID AT AUSTIN, TEXAS. PROFESSOR WILSON S. STONE W. S. Sitione ua mae'a au galuega fita, A ua e mapu atu nei i le tiasa 0 mo'omo'oga sa i le nu'u ma le lalolagi atoa 'Aua ne'i tulolo Zou to'oto'o ne'i sofa le 'oa A e ua le ogatasi lava le futia ma le umele U a amia nei 'oe e le Silisili' ese 0 au galuega lelei mo Zou nu'u ma le vasaloloa E le mafai ona gala i atunu'u uma, Aemaise Samoa . . . . traditional Samoan Chief's eulogy. Contributed by friends in Apia, Upolu, Western Samoa. TRANSLATION : Page 339 ~e()ication This latest collection of articles on the genetics and evolution of Drosophila is dedicated to the memory of Wilson S. Stone. The first of these volumes appeared nearly three decades ago. At that time the University of Texas already enjoyed a reputation as a leader in the field, but this series has done much to enhance and solidify that reputation. Wilson Stone was involved in the whole series, and in several ways-as an author, as editor, and probably most important of all as one who furnished ideas and stimulus to other authors. A great many of the articles are the better for his guiding infiuence in the research itself and in the writing. W. S. Stone was born in Junction, Texas, and except for a period during World War II he spent his entire life in the State. He died in 1968 at the height of his academic and scientific career, when much work was unfinished and when the University needed him badly. As an undergraduate student at the University of Texas he had an undistinguished academic record until he took a course in genetics with H. ]. Muller. Muller realized that in Stone he had found a young man with the requisites for an outstanding career in genetics research---'-a keen analytical and creative mind and a deep curiosity. For Stone this was the beginning of a consuming interest in Drosophila genetics that was to continue through his life. Later he entered into the scientific partnership with J. T. Patterson that lasted until Patterson's retirement and death. The two men complemented each other magnificently with Patter­son supplying the driving momentum that carried things through while Stone provided the deeper ana­lytical insights. Those of us who knew Stone well recall the quiet, almost lethargic manner that belied his quick, pene­trating, and active mind. We remember how helpful he was in research problems. We remember how beautifully clear and analytical his lectures were­despite the absence of every last one of the conven­tiontil attributes of classroom showmanship. We re­member the easy, kindly relation between his graduate students and himself. We remember his strong social convictions, his belief in human dignity, and his active participation in preserving democratic institutions. We remember his total absence of pretense and his com­plete honesty even when the truth was unpleasant. We also remember that, despite his preference for a more contemplative life, he willingly accepted various administrative and Committee assignments in the Zoology Department, in the larger University, and in the Federal Government. All of these he did con­scientiously and well, and with no thought of personal aggrandizement. Ten years ago one of the volumes in this series was dedicated to J. T. Patterson. It is fitting, and no one would be more in agreement than Patterson himself, that Stone be honored in this volume. With great respect, with genuine affection, and with deep sorrow for the life and work left uncompleted I join the vari­ous contributors to this volume in dedicating it to Wilson S. Stone. JAMES F. CROW Genetics Laboratory University of Wisconsin Contents D~k~~ JAMES F. CROW v I. Studies on cytology and differentiation in Sciaridae. III. Nuclear and cytoplasmatic differentiation in the salivary glands of Bradysia sp. A. BRITO DA CuNHA, J. S. MORGANTE, C. PAVAN and M. C. GARRIDO 1 II. A study of radiation-induced dominant lethals virilis in an oxygen atmosphere at 4°C. GLENN E. OMUNDSON m Drosophila 13 III. A study of radiation-induced translocations in Drosophila virilis in an oxygen atmosphere at 4°C. GLENN E. OMUNDSON 25 IV. Descriptions of new Hawaiian Drosophila D. ELMO HARDY and KENNETH Y. KANESHIRO 39 V. A study of the relationships of Hawaiian Drosophila species based on external male genitalia. KENNETH Y. KANESHIRO 55 VI. Notes on Hawaiian "idiomyia" (Drosophila) D. ELMO HARDY 71 VII. The Drosophila crassifemur group of species in a new subgenus KENNETH Y. KANESHIRO 79 VIII. Polytene chromosome relationships in Hawaiian species of Dro­sophila. IV. The D. primaeva subgroup . HAMPTON L. CARSON and HARRISON D. STALKER 85 IX. Variations in the metaphase sophilidae . FRANCES E. CLAYTON chromosomes of Hawaiian Dro­ 95 X. Enzyme variation in natural populations of Drosophila mimica E. S. ROCKWOOD 111 XI. Studies on interspecific hybridization within the picture-winged group of endemic Hawaiian Drosophila . HEI YuNGYANG and MARSHALL R. WHEELER 133 XII. Neutron activation techniques for labeling Drosophila in natural populations R. H. RICHARDSON, R. J. WALLACE, JR., S. J. GAGE, G. D. BoucHEY and MARGARET DENELL 171 XIII. Isozyme variat10n in Drosophila island populations. II. An analysis of Drosophila ananassae populations in the Samoan, Fijian and Philippine Islands. F. M. JoHNSON, K. KoJIMA and M. R. WHEELER 187 XIV. Cytogenetic relations in the Drosophila nasuta subgroup of the immigrans group of species FLORENCE D. WILSON, M. R. WHEELER, MARGARET HARGET and MICHAEL KAMBYSELLIS 207 XV. Courtship and mating behavior of the Drosophila nasuta subgroup of species . HERMAN T. SPIETH 255 XVI. Polymorphism in esterases and hemoglobin in wild populations of the house mouse (Mus musculus) . RoBERT K. SELANDER, SuH Y. YANG and W. GRAINGER HUNT 271 I. Studies on Cytology and Differentiation in Sciaridae. III. Nuclear and Cytoplasmatic Differentiation in the Salivary Glands of Bradysia sp.1 ' 2 3 3 4 A. BRITO DA CUNHA, J. S. MORGANTE, C. PAVAN 3 AND M. C. GARRID0 INTRODUCTION Great progress in the understanding of gene action and cell differentiation in higher organisms has been due to studies of polytene chromosomes (see Kroeger and Lezzi, 1966; Pavan and da Cunha, 1969a, b, for reviews). Organisms having cells with clear-cut differentiated cytoplasm and polytene chromosomes, especially when in the same organs, are ideal to study the relations between nucleus and cytoplasm in differentiation. Several Diptera have been found which have sa­livary glands with polytene chromosomes and have segments which are different cytoplasmically. Well-studied instances are those of Camptochironomus pal­lidivitatus (Beermann, 1961; Grossbach, 1968), Acricotopus lucidus (Baudisch, 1964), Dasyneura crataegi (Henderson, 1967) and Lestodiplosis sp. (Henderson, 1967). Concomitant changes in the chromosomes and the cytoplasm were ob­served in the different segments of the salivary glands of these species. This paper will report a new instance of a particularly clear case of nuclear and cytoplasmic differentiation in cells of a sciarid, Bradysia sp. MATERIAL AND METHODS The flies, Bradysia sp., were collected at a farm in Mogi das Cruze::;, State of Sao Paulo, and brought to the laboratory by our student, Miss Marlies Schneider. The species will be described elsewhere by one of us (J. S. M.). The larvae were dissected, and the salivary glands fixed with ethanol acetic 3: 1 for ten minutes. The fixe:l material was put for one minute in a drop of 60% acetic acid to which a drop of lactic-acetic-orcein solution was added. After one minute in this solution, a coverslip was placed on the material and lightly pressed. The glands were studied and photographed with a Zeiss-photomicroscope under normal light and phase contrast. The glands to be studied with other staining methods were also fixed with ethanol acetic 3: 1 and squashed in 60% acetic acid solution. The coverslip was 1 In memoriam of Professor Wilson S. Stone, great scientist and excellent friend. 2 Research sponsored by grants from The Rockefeller Foundation, Fundacao de Amparo a Pesquisa do Estado de Sao Paulo, Conselho Nacional de Pesquisas (Brazil) and the National Institutes of Health (Public Health Service Grant GM-15769). 3 Departamento de Biologia Geral, Universidade de Sao Paulo, Caixa Postal 8105, Sao Paulo, Brasil. 4 Department of Zoology, The University of Texas, Austin, Texas. STUDIES IN GENETICS V. Univ Texas Puhl. 6918, Sept., 1969. removed after freezing with liquid nitrogen, and the material submitted to the several staining processes. The staining methods used were: Feulgen, Periodic Acid of Schiff (PAS), PAS-Feulgen-Azur A-naphthol yellow S in accordance with the method of Himes and Moriber (1956), alloxan-Schiff as devised by Yasuma and Ichikawa and alcian blue as described by Martoja and Martoja'" Pierson (1967) and Sudan black after fixation with 10% formalin. Biuret reac­tions, Millon test and RNAse digestion were also used. The digestion with RNAse was made using a solution of 1 mg of RNAse per ml of distilled water. The glands were incubated at 40C0 for 4 hours and agitated every hour. The RNAse solution was kindly prepared and given to us by Prof. F. J. S. Lara. The Millon test was made according to Baker's methods as described in Casselman ( 1959). Synthesis of DNA was studied in salivary glands of larvae injected with 2JL1 of tritiated thymidine (1.9c/mM; lmc/ml, Schwarz). The injected larvae were killed after 3 hours, and the salivary glands fixed and squashed, and the coverslips were removed as above. The preparations were covered with KODAK AR.10 autoradiographic stripping film and treated according to Prescott's recommenda­tions after 11 days of incubation in the dark. Synthesis of RNA was studied in larvae injected with tritiated uridine (1.3c/ mM; 0.25mc/ml; Schwarz) and processed as above for thymidine. The larvae were injected also with 2,u.l, killed after 65 minutes. The incubation period for the autoradiography was 11 days. RESULTS The Cytoplasm. The salivary glands of Bradysia sp. have four sharply dif­ferentiated regions. For the sake of simplicity the regions will be called A, B, C and D, proximally to distally. The regions A and C have cylindrical cells, while the cells of region B are cuboidal and those of region Dare cuboidal or pyramidal. The cells of region A have a homogeneous liquid secretion. Region B has only six cells. Due to their large sizes they give a swollen aspect to region B. The secretion of the B cells has a very peculiar disc-like appearance, as shown by the photomicrographs of Figs. 7 and 8. The secretion discs are more numerous in the vicinity of the nuclei. The cells of region C are characterized by the great abundance of cytoplasmic gran­ules. During most of the larval life, the cytoplasm of the C cells is completely filled with the granules. These granules are sometimes associated in conglom­erates. Dissociation of the C cells under pressure on the coverslip shows that the granules are attached in long ramified ribbons by a matrix (Figs. 5 and 6). The D region is the thinnest and longest segment of the salivary gland. The D cells have very small granules dispersed in their cytoplasm. The D granules are never associated in conglomerates as the C granules. Regions A and Bare PAS-positive (Fig. 1). The A cells located closest to the B region ordinarily are more strongly stained than the others. While the cyto­plasm of the A cells stains uniformly with PAS, the staining of the B cells is more intense in the vicinity of the nuclei where the secretion discs are concentrated (Fig. 1). These results show that the secretions of the A and B cells, however dif­ferent, have a polysaccharidic nature. Remarkable results were obtained where cells of Bradysia sp. were stained by the Himes and Moriber method (Figs. 2 and 3). The regions A and Bare stained by the PAS. The cytoplasmic .granules of region C are heavily stained in blue by the Azur A, while the cytoplasm and granules of the D cells are stained in yellow by the naphthol yellow S, the granules more intensely. This result indi­cates that the granules of the D cells have a proteinaceous nature. The staining of the C cells' granules with the Azur A is very surprising. Ac­cording to what is known about the Himes and Moriber method, only chromatin should be stained with the Azur A. Of all the staining methods used, Feulgen, PAS, PAS-Feulgen-Azur A-naphthol yellow S, alloxan-Schiff, alcian blue, Sudan black, only the Times and Moriber method stained the C cell granules. Millon and biuret reactions also gave negative results. The treatment with RNAse pre­vious to the staining with the Himes and Moriber method did not affect the stain­ing of the C granules by the Azur A. DNAse treatment was not used because the granules are Feulgen negative. The quantity and distribution of the C granules varied with the age of the larvae. During all larval life, the C cells have granules in the cytoplasm. The cells of the young larvae are filled with the granules (Figs. 2 and 3). The number of granules decreases, but their distribution is still uniform all over the cytoplasm in late larval life. At prepupa stage the granules are concentrated at the apical poles of the cells while the basal region of the cells is completely free of granules. It is interesting to notice that there is a very sharp separation between the dif­ferentiated regions A, B, C and D in the gland (Figs. 2 and 3). There is no transi­tion from one region to the next; the neighboring cells of two different regions are sharply distinct. The Nuclei. The nuclear differentiation in the salivary glands of Bradysia sp. is as sharp as the cytoplasmic. Figure 4 shows A nuclei at the bottom, three B nuclei at the center and four C nuclei at the top right in a Feulgen stained prepa­ration. Figures 17 to 20 show, under the same magnification, A, B, C and D nuclei from a same salivary gland. The nuclei in Figs. 17 to 20 belong to larvae of exactly the same developmental stage as the larvae from which the glands of Figs. 1 to 4 were removed. These photomicrographs show very clearly the sharp differences between the nuclei of A, B, C and D cells. At this stage, eleven days before prepupa stage, the DNA synthesis is very "strong" in cells A, C and D and slight in cell B, as shown by the four photomicrographs taken from the same gland (Figs. 9 to 12). The same occurs with RNA synthesis. The uridine incorporation shown in the photomicrographs (Figs. 13 to 16) indicates low RNA synthesis in cell B. These four photomicrogra phs were also taken from the same gland as in the case of thymidine incorporation. The C cells are by far those which produce the largest amount of secretion grains during the larval period. However, they are the cells with the smallest and thinnest chromosomes. A great number of bands of the chromosomes of region C have a very compact appearance (Fig. 18). At the stage in which the DNA and RNA synthesis was studied, 11 days before prepupa stage, very little DNA and RNA synthesis is occurring in B cells (Figs. 10 and 14). Their chromosomes have already attained their largest size. The 2 .• 100..u amount of PAS-positive material decreases toward prepupa stage. The DNA and RNA synthesis, in this stage, is still very intense in A, C and D cells (Figs. 9, 11, 12, 13, 15 and 16). These three types of cells continue to produce secretion until prepupa. At prepupa stage, the synthesis of secretion stops in C cells. The secre­tion granules of the C cells accumulate at the apical pole. A large quantity of free C granules may be seen in the lumen of the gland. The cell nuclei are spherical and have distended chromosomes until prepupa stage; then they become flattened and compact at the cell base at prepupa. It is interesting to notice that in Bradysia sp. the intensive production of the secretion in the four regions of the salivary gland during larval development occurs without the appearance of the large and conspicuous puffs which charac­terize intense metabolic activities in polytene chromosomes of other forms of Sciaridae. It is of interest also to notice that the region of the gland in which the cells show the greatest amount of secretion granules is the one in which the polytene chro­mosomes are less developed. The probable explanation for this is that these chro­mosomes, although less developed (less polytene), have a higher RNA production than the chromosomes of other regions. DISCUSSION Many attempts have been made in the past to correlate nuclear and cytoplas­matic differentiation. Barigozzi ( 1953) wrote a good review of what was known about the microscopic structure of the interphase nucleus and differentiation. Great advances in these studies came after the systematic work done on the puffs of the polytene chromosomes after 1950 (Breuer and Pavan, 1952; Beer­mann, 1952); this has been reviewed by Kroeger and Lezzi (1966) and by Pavan and da Cunha ( 1969a, b) . Beermann ( 1961) was able to show the correlation between the presence of a specific Balbiani ring and the occurrence of a specific type of granular secretion in the salivary glands of Camptochironomus palli­divitatus. There is no doubt that puffs are indications of gene activity; they are formed at specific times of larval development and many of them are also tissue specific. However, puffing is not the only way gene activities in polytene chro­mosomes may be manifested. The observations reported here show that no evi- FIG. 1. A salivary gland of Bradysia sp. stained by PAS reaction. The lower stained segment is region A. Region Bis marked by the PAS positive reaction at the proximity of the nuclei and is formed by six cells. The A cells closer to the B region are more intensely stained. FIG. 2. A salivary gland of Bradysia sp. stained by the Himes and Moriber method. The reddish yellow region below is formed by A and B cells stained by naphthol yellow S and PAS. The blue segment is the region C stained by Azur A. The yellow segment above is the D region stained by naphthol yellow S. FIG. 3. A photomicrograph of regions A, B and C stained by the Himes and Moriber method. The C cells are stained in blue by Azur A. The three cells with large nuclei just below the blue segment, are B cells. The cells below the three B cells are A cells. FIG. 4. A photomicrograph of nuclei of A, B and C cells stained by Feulgen reaction. The four nuclei below are from A cells. The three larger nuclei at the center are B cell nuclei. The four nuclei in the row at the top right belong to C cells. Fm. 5 and 6. The typical aspect of secretion granules of C cells. The granules are attached in ribbons. Lactic-acetic-orcein and phase. Frns. 7 and 8. The s2cretion of the B cells. The PAS-positive secretion is organized in discs which are seen in lateral view in Fig. 7 and in frontal view in Fig. 8. Lactic-acetic-orcein and phase. Frns. 9, 10, 11 and 12. Tritiated thymidine incorporation in the nuclei of an A cell (Fig. 9), a B cell (Fig. 10), a C cell (Fig. 11) and a D cell (Fig. 12). Autoradiography and Giemsa. Frns. 13 and 14. Trit.iated uridine incorporation in the nuclei of an A cell (Fig. 13) and a B cell (Fig 14). Autoradiography and Giemsa. Fies. 15 and 16. Tritiated uridine incorporation in nuclei of a C e;ell (Fig. 15) and a D cell (Fig. 16). Photomicrographs 13 to 16 were taken from the same gland. Autoradiography and Giemsa. FIGs. 17, 18 and 19. The chromosomes of an A cell (Fig. 17), a C cell (Fig. 18) and a D cell (Fig. 19) of the same gland and under the same magnification. Lactic-acetic-orcein and phase. FIG. 20. The chromosomes of a B cell from the same gland as used for Figs. 17, 18 and 19 under the same magnification. The granules at the right belong to a C neighbor cell. Lactic­acetic-orcein and phase. dent puff occurs in the salivary gland cells of Bradysia sp. during the larval stage, when-the synthesis of cytoplasmic granules is very intense. The very generalized incorporation of H 3 uridine in chromosomes of the cells of regions A, C and D, shown by the autoradiography, indicates a widespread, rather than a specialized, activity of a few genes. The salivary glands of Bradysia sp. are differentiated in four regions which are easily identifiable by the type of secretions present in the cytoplasm of their cells. The cytoplasmic differentiation is sharp, and there is no transition from one cell type to another. Cytoplasmic and nuclear differentiation are concomitant. There are four con­figurations of chromosomes, each one related to a specific type of cytoplasm. The four types of chromosomes differ in size, the degree of condensation of the chro­matic material, and the rates and periods of DNA and RNA synthesis. The dif­ferentiation of the four types of chromosomes is as sharp as the differentiation of the four types of cytoplasm. The nucleus and cytoplasm constitute an interacting system in which the mutual effects cause their concomitant differentiation. SUMMARY The salivary glands of Bradysia sp. are described. The glands have four regions which are sharply differentiated according to the nature of the cytoplasmic secre­tions and the aspect of the nuclei of their cells. Four types of differentiated nuclei were described, one for each type of cell cytoplasm. The chromosomes differ, in the four cell types, in their sizes, condensation, rates and period of DNA and RNA synthesis. No large puffs were observed during the periods of intense syn­thesis of secretion. ACKNOWLEDGMENTS The authors are very grateful to Miss Marlies Schneider, who collected and brought Bradysia sp. to the laboratory. Great help was given by Misses Jaire Marques, Ivonete Romeo, Ilze L. Jorge, Thelma Picard, Therezinha M. Ungaretti and Joao Benedito de Campos to whom we express our gratitude. LITERATURE CITED Barigozzi, C. 1953. La struttura microscopica del nucleo durante il riposo. Experiencia, 8: 133-136. Baudisch, W. 1964. Untersuchungen sur physiologischen Charaketerisierung der einzelnen Spechieldriisen lappen von Acricotopus lucidus. Struktur und Funktion des genetischen Materials. Erwin-Baur-Gedachtnisvorlesungen III. 1963. Akademie Verlag., Berlin 1964. Beermann, W. 1952. Chromomerenkonstanz und spezifische Modifikationen der Chromen­struktur in der Entwicklung und Organdifferenzierung von Chironomus tentans. Chromo­soma, (Berl.) 5: 139-198. ----, 1961. Ein Balbiani -Ring als locus einer Speicheldriisen-Mutation. Chromosoma (Berl.) 12: 1-25. Breuer, M. E. and C. Pavan. 1952. Gens na diferencia\.ao. Ciencia e Cultura, 4: 115. Casselman, W . G. B. 1959. Histochemical technique. Methuen and Co. London. Grossbach, V. 1968. Cell differentiation in the salivary glands of Camptochironomus tentans and C. pallidivitatus. Annales Zoologici Fennici, 5: 37-40. Henderson, S. A. 1967a. The salivary gland chromosomes of Dasyneura crataegi (Diptera: Cecidomyidae). Chromosoma (Berl.) 23: 38-78. 1967b. A second example of normal coincident endopolyploidy and polyteny in salivary gland nuclei. Caryologia, 20: 181-186. Himes, M. and L. Moriber. 1956. A triple stain for deoxyribonucleic acid, polysaccharides and proteins. Stain Technology, 31: 67-71. Kroeger, H., and M. Lezzi. 1965. Regulation of gene action in insect development. Annual Review of Entomology, 11: 1-22. Martoja, R., and R. Martoja-Pierson. 1967. Initiation aux techniques de l'histologie animale. Masson et Cie., Paris. Pavan, C., and A. B. da Cunha. 1969a. Chromosomal activities in Rhynchosciara and other Sciaridae. Annual Review of Genetics, 3 (in press) . 1969b. Chromosome activities in normal and infected cells of Sciaridae. Nucleus, 4 (in press). Prescott, D. M. 1964. Autoradiography with liquid emulsion. In: Methods in cell physiology. Ed. D. M. Prescott, p. 365-370. New York: Academic Press. II. A Study of Radiation-Induced Dominant Lethals in Drosophila virilis in an Oxygen Atmosphere at 4°C.1 2 GLENN ERWIN OMUNDSON INTRODUCTION The majority of studies thus far carried out have been designed so that muta­tions are produced by directly irradiating the test organism, although Stone, Wyss, and Haas (1947) and others have shown that irradiation of the substrate can lead to the induction of mutations. Three direct studies have shown that the physical nature of the radiation employed (Lea, 1956; Oster, 1959; Powers, Ehret, Bannon, and Prock, 1957; Powers, Webb, and Ehret, 1959), the environ­ment of the test organism before, during, and after irradiation (Giles, Beatty, and Riley, 1951, 1952; Haas, Dudgeon, and Stone, 1953; Sollunn and Stromnaes, 1964), and the biological state of the organism with regard to its state of cell cycle, hydration, metabolism, and other factors (Savhagen, 1961; Schmid, 1961; Strangio, 1962; Tallentire and Powers, 1963), are important modifying factors which can greatly alter the extent and effect of genetic damage. Of primary con­cern in this study are the modifications of radiation damage brought about by oxygen and temperature. The Oxygen Effect Although recogmt1on of the "oxygen effect" is usually credited to Thoday and Read (1947), evidence for its presence has been available from the irradia­tion of the eggs of Ascaris (Holthusen, 1921) . In Drosophila, both Demerec and Fano ( 1944) and Lea and Catcheside ( 1945) studied the effects of radiation in producing dominant lethals and chromosomal aberrations in D. melanogaster, but neither group recognized any effect of oxygen. The work of Baker ( 1949) also fails to recognize any modifications due to oxygen. However, in a series of papers immediately following (Baker and Sgourakis, 1950a, 1950b; Baker and Edington, 1952; Baker and von Halle, 1952, 1953), Baker and co-workers noted an increase in the frequencies of dominant and re­cessive lethals, and an increase in the frequency of X-ray induced translocations in relation to increases in the oxygen concentration. They also noted a significant increase in the number of mutations induced when the treatment in oxygen was at cold temperatures than when the same treatment was at room temperature. They attributed this effect to the differences in oxygen solubility and relative rates of respiration at these temperatures. At room temperatures, Baker and co­ 1 This work was supported in part by a Public Health Service research grant, GM-11609, and training grants, GM-337-06 and GM-00337-07, from The National Institutes of Health. 2 Present address: Department of Biology, University of Mississippi, University, Miss.. The author wishes to express his great indebtedness to the late Professor Wikm S. Stone for his advice, support and confidence in directing the research reported here. STUDIES IN GENETICS V. Univ. Texas Puhl. 6918, Sept., 1969. workers found that at low oxygen concentrations, small increases in oxygen ten­sion causes rapid increases in the amount of damage produced up to about 11 % oxygen. Above this level, they found little increase in damage with increasing concentrations. Bonnier and Liining (1950) and Liining (1952a, b, c) studied primarily the production of X-ray induced dominant lethals and asymmetrical interchanges in D. melanogaster and found, as had other early workers (Muller, 1930, 1940, 1954a, 1954b, 1958; Bauer, Demerec, and Kaufmann, 1938; Kaufmann, 1946, 1954), that the stage of gametogenesis al~ered the production of these aberrations, without implicating the role of oxygen. faining (1954), however, noted that there were induced more vital minute and gross rearrangements in sperm ready the first day after irradiation as compared with those from later days after irradiation in air or in nitrogen, and each sperm batch was more aberrant after treatment in air than in nitrogen. Work by Stone, Haas, Alexander, and Clayton (1954), Alexander and Stone ( 1955), and Stone ( 1955) showed similar results primarily for dominant lethals, and noted changes in susceptibility to radiation induced damage corresponding to various stages of spermatogenesis. Alexander and Stone ( 1955) correlated these variations in susceptibility to changes in the coat proteins of chromosomes, noting that sperm undergo a coat change from histone proteins to protamine proteins. Schmid (1961) did note that the X-ray induced dominant lethal frequency in the germ cells of D. virilis was greatly increased by the presence of very small quantities of oxygen when carbon monoxide was used as a respiratory inhibitor. Sobels (1960) found similar enhancement of X-ray induced damage using cya­nide, a very strong respiratory inhibitor, as a postirradiation treatment. Studies by Alexander and co-workers (Alexander, 1958a, b; Alexander, Ber·· gendahl, and Brittain, 1959) have recognized the various facets of the oxygen problem and its association with radiation damage in Drosophila. Tests of 1960 (Alexander, 1960a) and 1962 (Alexander and Bergendahl, 1962) did not indicate preferential breakage in oxygen with the failure to heal or restitute more than breakage from irradiation in nitrogen as very low values for oxygen-nitrogen and nitrogen-oxygen were obtained for mature sperm in D. virilis. Temperature Before Mull~r's historic work with X-rays (1927), Plough ( 1917) recognized the influence of temperature on the rate of crossing-over in D. melanogaster. Mul­ler (1930) indicated a slightly higher though not a significant increase in the number of sex-linked lethals induced in flies X-rayed at 8° over those at 34°C. Medvedev (1935, cited by Baker, 1949) found a significant increase in the num­ber of lethals produced· in flies maintained at 0° than in ones treated at 20°C. These results were substantiated by King (1947) who also found a two-to three­fold increase in the mutation rate of those flies irradiated at 0°C. over those at room temperature. Similar results were found by Papalashwili ( 1935, cited by Sollunn and Stromnaes, 1964) . Baker and Sgourakis (1950a) found a significantly greater number of muta­ tions induced when flies (D. melanogaster) were irradiated in oxygen at cold temperatures than when irradiated in oxygen at room temperatures, although this temperature effect was not noted when irradiation was in nitrogen. Controls in each gas and at each temperature, and their combinations showed no alteration in the rate of spontaneous mutations. Novitski (1949) also found that the mu­tation frequency could be increased by use of lower temperatures, with greater effects of cold treatment being found after irradiation than during irradiation. Sollunn and Stromnaes ( 1964) found that changes in the temperature during irradiation were responsible for large increases in the induced mutation rates. They hypothesized that in addition to the oxygen effect operating, the low tem­peratures may have inactivated the radio-protective systems of the cell, and as such, the temperature effect may be acting primarily on the initial breakage pro­cess and secondarily on restitution, or on both. As the data available for D. virilis consist primarily of two groups, one dealing with the effect of oxygen concentration independent of temperature, and the other wtih the effects of temperature independent of oxygen concentration, it is hoped that this work reported here will help bridge this gap and partially show the interaction possible between oxygen and low temperature at the time of ir­radiation. MATERIALS AND METHODS The test organism used was Drosophila virilis, which was maintained by mass culture in half-pint milk bottles with a cornmeal-agar food at 20-22°C. Males for all phases of research were taken from the Texmelucan strain (UT1801.1). These were taken within three hours of eclosion and aged 19-22 hours prior to treat­ment. Virgin females for dominant lethal studies were derived from a cross be­tween the strains of Chile (males) and Argentina (females) . In all experiments, the males were irradiated in one atmosphere of oxygen (99.5% oxygen, 0.5% argon) at 3.50°C., plus or minus 1°C. Oxygen was al­lowed to flow continuously during pretreatment ( 10 minutes), treatment (3-4 minutes), and posttreatment ( 10 minutes). Three minutes prior to the adminis­tration of X-rays, the aluminum cylinder housing the irradiation chamber was immersed in an ice-water bath and remained in this bath six to seven minutes. The intake valve to the cylinder was immediately preceded by 12 feet of one­quarter inch coiled copper tubing to cool the incoming oxygen to the required temperatures. Temperature measurements were taken of the oxygen immediately before it entered and immediately after it left the cylinder, both before and after irradiation. Radiation was administered by a Westinghouse Quadrocondex X-ray machine operated at 250 kvp at 15 ma. Filtration of relatively soft X-rays was provided by a filter of Imm aluminum and Yz mm copper. Dose rate in all experiments was 500 r/min. and was determined by a Victoreen condenser dosimeter before each irradiation. The frequencies of radiation induced dominant lethals were used to measure damage. Immediately after r~covery of "rn~rmal activity" after the treatment, individual males were mated simultaneously to four virgin females, one of which was from the Argentina-Chile cross, the ~ther three from the marker stock. To determine damage at each sucCf\.Ssive stage of spermatogenesis, each male was re­mated to a group of four females every other day for seven periods (A through G) starting the sixth day after eclosion, before being discarded. Following the proce­dure established by Stone (1955) and Alexander and Stone (1955) for D. virilis, A represents mature sperm; B, sperm bundles; C, D, and E, spermiogenesis; F, meiosis; and G, late spermatogonia. Following the work of Alexander (1959, 1962.a), dominant lethality, measured as combined chromosomal and cytoplasmic damage, was scored as the percentage of eggs which failed to pupate. The Argentina-Chile females were placed in fresh egg-count vials at the end of each two-day mating period. They were allowed to lay eggs for one day before being transferred to fresh vials. Three such transfers were made, thus allovving for four daily egg-counts per female. Females were dis­carded at the end of this time. Eggs counted daily were allowed to hatch and the larvae were kept through pupation. RESULTS The frequencies of radiation induced dominant lethals are given in Table 1. It is obvious that in the absence of irradiation, the combination of oxygen and cold temperature has little effect on the production and transmission of viable sperm, TABLE 1 Frequencies of induced dominant lethals Period A B c D E F G Series Or a b c d 5759 6278 8.5 o. 7 6214 6903 10.0 0.7 7562 7917 4.5 1. 5 7292 7709 5.4 1..6 5880 6416 8.3 0.7 6330 7240 12.6 0.8 6181 6843 9.5 0.7 500r a b c d 5111 6442 20.7 1.0 4783 6635 27.9 1.1 3526 7585 53.5 1.1 3523 9201 61. 7 1.0 1609 7735 79.2 0.9 1788 4193 57.4 1. 5 3720 6964 46.5 1.2 lOOOr a b c d 5567 9438 41.0 1.0 3292 9490 65.3 1.0 667 6700 90.0 0.7 453 10346 95.. 6 1.3 235 4909 95.2 1.9 166 2184 92.4 1.1 1558 3852 59.6 1..6 1500r a b c d 1924 7040 72. 7 1.1 682 7619 91.0 0.7 77 10264 99.2 0.2 13 11443 99.9 0.1 18 7198 99.8 0.1 1 1106 99.9 0.2 10 2129 99.5 0.3 2000r a b c d 682 5728 88.1 0.9 233 7599 96.9 0.4 14 7628 99.8 0.1 10 7609 99.9 0.1 0 2928 100.0 1 1745 99.9 0.1 8 1976 99.6 0.1 a= number of eclosions; b =number of eggs; c = %lethality; d = 2S.E. as the average fertility for periods A through G in the controls (0 r) is about 91.7%. Under optimal conditions, this value would be expected to range from about 90 to 95%. Thus the combined effect of oxygen and cold temperature in the absence of irradiation may be considered as negligible. With the addition of the X-ray treatment, considerable damage is noted in all mating periods, even at .500r. In the 500 r series, dominant lethal damage increases rapidly from a minimum of 20.7% in period A (mature sperm) to a maximum of 79.2% in period E, and then decreases to 46.5% in period G. An additional period (H, spermatogonia) was checked in this series only, and only 26.4% lethality was recorded, although in no cases were eggs checked for the presence or absence of sperm. In the 1000 r series, the percentage of dominant lethality increased from 41.0% in period A to 95.6% in period D. Period G still had 59.6% remaining. In the 1500 r series, period A showed 72.7%, while periods C through G were all above 99.0%. In the 2000 r series, results similar to the 1500 r series were obtained, with period A showing 88.1 % lethality, and periods C through Gall above 99.5%. In all series, damage as measured by dominant lethals showed mature sperm (period A) to be the most resistant period of development to radiation damage, closely followed by sperm bundles (period B). D1scuss10N The frequency of radiation induced dominant lethality was determined by the percentage of eggs laid which failed to develop into pupae, thus any factor which prevented this development, or which blocked the transfer of sperm would lead to increased frequencies. Lefevre and Jonnson ( 1962b) have demonstrated that irradiation of Drosophila at cold temperatures produces a mass of dead sperm that effectively blocked the transfer of later developing sperm. Early studies by Pontecorvo (1942) and Demerec and Fano (1944) indicated that most dominant lethality at low doses of radiation would be determined by single break sister unions, while at higher levels, deletions, translocations, and similar two break occurrences would become more operative. Stone, Haas, Alexander, and Clayton ( 1954) found that a large proportion of dominant lethality resulted from the failure of irradiated males to produce functional sperm. Similar findings were re­ported by Yanders (1959). Von Borstel (1963) also noted the failure of func­tional sperm as a cause, and the contributions of chromosome breakage to in­creased lethality. Comparisons of the data obtained in these experiments with those of other recent workers show that the combination of cold temperatures and oxygen tend to produce more dominant lethality than previously reported. Alexander (1958b), using 14-mev neutron bombardments on D. virilis in an oxygen atmosphere, found more damage with 1000 r doses only for periods A and B (68.7 to 41.0%, A; 70.0 to 65.3%, B). Periods C through Gin these experiments reported here show significantly higher frequencies than those of Alexander. With 2000 r of X-rays in air, Alexander (1959) found 99% of the meiotic (F) and spermato­gonial cells (G) contained lethals. This is less than that reported here, where dominant lethality exceeded 99.5 % for these stages as well as for those cells in spermiogenesis (C, D, and E). Similar comparisons can be made with the work of Alexander, Bergendahl, and Brittain (1959) in which D. virilis males were treated in air at 27°C with 200 kv X-rays, 22 Mv X-rays, and 1.17-1.33 Mev gamma rays. From the data of Alexander, et al., 500 r of X-rays in period B show no effects as these workers reported an average of 5.5% lethality in their controls. From this low value, damage increased to a maximum in period E of 43.0%. From the data obtained in the present experiments, dominant lethality in period B was 27.9%, and rose to a maximum, also in period E, of 79.2%, then declined to 46.5% in period Gas compared to 17.0% reported by Alexander, et al. Thus lethality was from almost two times to five times greater than that previously reported at 500 r of X-rays. At 1000 r, Alexander, et aL report 11.0% lethality for period B, a m"'ximum of 83.4% in period E, and only 31.2% in period G. This compares to 65.3% in period B, maxima of 95.6% and 95.2% for periods D and E, and 59.6% in period Gin the experiments recorded here, again indicating significantly higher damage than previously reported. Similar results are seen at 2000 r, where Alexander, et al.~ report 44.0% for period B, a maximum of 97.2% in period E, and only 55.2% in period Gas compared to 96.9%, 100.0%, and 99.6% dominant lethality for the corresponding periods reported here. The data found here support the hypothesis of Baker and co-workers who at­tributed the increase in radiation induced aberrations at low temperature in oxygen compared to room temperature in oxygen to the differences in oxygen solubility and relative rates of respiration at these temperatures. The results of Haas, Dudgeon, Clayton, and Stone ( 1954) also support this interpretation, and they also suggested that the greater activity of normal biological protective sys­tems or agents at room temperature account for much of the temperature differ­ential. As normal metabolic activity is greatly slowed or stopped at 4°C, these protective systems would similarly be impeded at this temperature, and thus the data obtained here would support this hypothesis. This is also in agreement with the data obtained by Wedvik and Stromnaes (1963) and Sollunn and Stromnaes (1964) for D. melanogaster. Tests by Clayton (1962), using D. virilis males treated with 150 kv X-rays at 4° to 5°C in air, allow for a comparison only at 1000 r for period A. In Clayton's work, 1089 r of X-rays were administered which resulted in 45.6% lethals in period A. This is in close agreement with the data obtained in these experiments at 1000 r, namely 41.0%. In similar tests under the same conditions, and for period A, Clayton ( 1965) reports for 500 r of X-rays approximately 14%; 1000 r, 29%; 1500 r, 46%; and 2000 r, 58% lethality. These data compare with 20.7%, 41.0%, 72.7%, and 88.1 % for 500 r, 1000 r, 1500 r, and 2000 r, respectively for period A as reported here. Comparisons where possible with earlier works (Baker, 1949; and Stone and co-workers) show that results obtained in these experiments tend to produce significantly more radiation induced dominant lethality for all periods or cell stages than previous! y reported. Response of Spermatogenic Cells It has been reported as early as 1934 by Moore that detection of dominant lethals varied with the different stages of spermatogenesis. Muller (1950), Oster (1959), Alexander (1959, 1960b), Chang (1962), Clayton (1962), Elequin (1966), and others have found that spermatogonia are less sensitive than those stages following, with those cells in spermiogenesis being the most sensitive to radiation damage. Trout (1964), Clayton (1965), Sobels (1963a, b; 1964, 1965) and others have reported that spermatozoa released the first day after treatment were more sensitive than those released on the second day after treatment. In the experiments reported here, no attempt was made to differentiate between these two possible classes, as mating periods were of two-day lengths. In the data obtained for dominant lethals, in all series, period E was found to be the most sensitive to radiation damage, while mature sperm (A) were the most resistant. These data are in good accord with those previously reported, ex­cept as previously indicated, damage per unit of X-rays tends to be greater. The data indicate that for dominant lethals, the combination of cold temperatures and an oxygen atmosphere significantly enhance radiation damage. A very probable factor influencing this enhancement is the blocking action by dead sperm as re­ported by Lefevre and Johnson ( 1962a) . SUMMARY Males of the Texmelucan strain of D. virilis were collected within three hours of eclosing. At the age of 19-22 hours (from eclosion), these males were irradi­ated with 500 r, 1000 r, 1500 r, or 2000 r of 250 kvp X-rays at 500 r/min. in an oxygen atmosphere at 3.5°C. Following recovery from this treatment, males were mated individually to four virgin females, one of which was from the Argentina­Chile cross, and the other three from the marker stock. Starting the sixth day after eclosion, these males were remated to a new group of females and were remated every other day for seven periods (A through G). Normal females of these mat­ing periods were placed in fresh egg-count vials daily, and dominant lethality was determined as the percentage of eggs which failed to pupate. Controls for dominant lethals came from the same treatment as described above with the ex­ception that males were not X-rayed. The data obtained indicate that: 1) The combined effect of oxygen and cold temperatures in the control series was negligible, as the average fertility for periods A through G was approximately 91.7%. 2) With the addition of only 500 r of X-rays, radiation damage rose rapidly as measured by dominant lethals, reaching a peak of 79.2 % in period E. 3) At 1000 r, dominant lethality exceeded 95% in both periods D and E, a value higher than that previously recorded. Similar though higher, values were found in the 1500 rand 2000 r series. These data were briefly compared to those of Alexander, Bergendahl, and Brittain (1959), and others. 4) In all series, damage as measured by dominant lethals indicated that cells in spermiogenesis were the most sensitive to radiation damage, while mature sperm were the most resistant. The data, taken as a whole, indicate that the combined effect of oxygen and cold temperatures at the time of irradiation produce results that are qualitatively similar to those found by other workers~ but quantitatively more damage is in­duced per unit of X-radiatio;n than previously reported, thus indicating a rela­tively greater efficiency of enhancement of radiation damage. BIBLIOGRAPHY Alexander, M . L. 1958a. Radiation damage in the developing germ cells of Drosophila virilis from fast neutron treatment. Genetics 43: 458-459. 1958b. Biological damage in developing germ cells of Drosophila virilis in oxygen and nitrogen with 14-mev neutrons. Proc. Natl. Acad. Sci. U.S. A. 44: 1217-1228. 1959. The effect of radiations of different ion densities on the germ cells of Drosophila virilis. In "Radiation Biology and Cancer." Austin, Texas: Univ. of Texas Press. 1960a. Dose rate effects and genetic recovery with cobalt-60 gamma rays in atmospheres of oxygen and nitrogen. Genetics 45: 971. 1960b. Radiosensitivity at specific autosomal loci in mature sperm and sperma­togonial cells of Drosophila melanogaster. Genetics 45: 1020-1022. 1962a. The role of recovery mechanisms and oxygen effects upon changes in radiation sensitivity in sperm treated in mature males and fertilized females of Drosophila. Genetics 47: 1505-1518. Alexander, M . L. and J. Bergendahl. 1952. Biological damage in the mature sperm of Dro­sophila virilis in oxygen and nitrogen with different dose intensities of gamma rays. Genetics 47: 71-84. Alexander, M. L., J. Bergendahl, and M . Brittain. 1959. Biological damage in mature and immature germ cells of Drosophila virilis with ionizing radiations. Genetics 44: 979-999. Alexander, M. L. and W . S. Stone. 1955. Radiation damage in the developing germ cells of Drosophila virilis. Proc. Natl. Acad. Sci. U.S. A. 41: 1046-1057. Baker, W. K. 1949. The production of chromosome interchanges in D. virilis. Genetics 34: 167-193. Baker, W. K. and C. W. Eddingt'.)n. 1952. The induction of translocations and recessive lethals in Drosophila under various oxygen concentrations. Genetics 37: 655-677. Baker, W . K. and E. Sgourakis. 1950a. Alteration of X-ray sensitivity of Drosophila by means of respiratory inhibitors. Genetics 35: 96. 1950':>. The effect of oxygen concentration on the rate of X-ray induced mutations in Dronphila. Proc. Natl. Acad. Sci. U.S. A. 36: 176-184. Baker. W. K. and E. S. von Halle. 1952. The effect of oxygen concentration on the induction of dominant lethals. Genetics 37: 565. -----1953. The basis of the oxygen effect of X-irradiated Drosophila sperm. Proc. Natl. Acad. Sci. U . S. A. 39: 152-161. Bau~r. H ., M . Demerec, and B. P. Kaufmann. 1938. X-ray induced chromosome alterations in Drosophila melanogaster. Genetics 23: 610-630. Bonnier, G. and K. G. Li1ning. 1950. X-ray induced dominant lethals in Drosophila melano­gaster. He:-editas 36: 445-456. Chang, T. H . 1962. Further observations on the relation between gas pressure and the X-ray dam::ige in Drosophila melanogaster. Univ. of Texas Puhl. 6205 : 385-393. Clayton, F. E. 1962. Effects of X-ray irradiation in D. virilis at diffe:::-ent stages of sperm3to­ genesis. Univ. of Texas Puhl. 6205: 345-373. 1965. Genetic effects from simultaneous irradiation of immature and mature Drosophila virilis males. Gent>tics 52: 1081-1092. Demerec, M . and U. Fano. 1944. Frequency of dominant lethals induced by radiation in sperms of Drosophila melanogaster. Genetics 29: 348-360. Elequin, F. T. 1966. Modification of induced genetic damage in Drosophila melanogaster by oxygen and argon treatments between two doses of X-rays. Univ. of Texas Publ. 6615: 177­ 193. Giles, N. H., Jr., A. V. Beatty, and H. P. Riley. 1951. The relation between the effects of temperature and of oxygen on the frequency of X-ray-induced chromosome aberrations in Tradescantia microspores. Genetics 36: 552. 1952. The effect of oxygen on the production by fast neutrons of chromosomal aberrations in Tradescantia microspores. Genetics 37: 641-649. Haas, F. L., E. Dudgeon, and W. S. Stone. 1953. Frequency of chromosomal rearrangements as related to rate of irradiation, temperature, and gases. Genetics 37: 589-590. Haas, F. L., E. Dudgeon, F. E. Clayton, and W . S. Stone. 1954. Measurement and control of some direct and indirect effects of X-radiation. Genetics 39: 453-471. Holthusen, H. 1921. Beitrage sur Biologie der Strahenwinkung Untersuchungen an Askari­deneinern. Pflugers Arch. ges. Physiol. 187: 1-24. Kaufmann, B. P. 1945. Organization of the chromosome. I. Break distribution and chromo­some recombination in Drosophila melanogaster. J. Exptl. Zool. 102: 293-320. 1954. Chromosome aberrations induced in animal cells by ionizing radiations. Chap. 9, Radiation Biology, A. Hollaender, ed. New York: McGraw-Hill, pp. 627-711. King, E. D. 1947. The effects of low temperature upon the frequency of X-ray induced mu­tations. Genetics 32: 1.61-164. Lea, D. E. 1956. Action of Radiation on Living Cells, 2nd ed. London: Cambridge Univ. Press. Lea, D. E. and D. G. Catcheside. 1945. The relation between recessive lehtals, dominant lethals, and chromosome aberrations in Drosophila. I. Genet. 47: 10-24. Lefevre, G., Jr. and U. B. Jonnson. 1962a. The effect of cold shock on Drosophila melanogaster sperm. Drosophila Information Service 36: 86-87. 1962b. Sperm transfer, storage, displacement, and utilization in Dros~phila me­lanogaster. Genetics 47: 1710-173(;). Liining, K. G. 1952a. X-ray induced dominant lethals in different stages of spermatogenesis in Drosophila. Hereditas 38: 91-107. 1952b. X-ray induced mutations in Drosophila melanogaster. Hereditas 38: 108­ 109. 1952c. X-ray induced chromosome breaks in Drosophila melanogaster. Hereditas 38: 321-338. Liining, K. G . 1954. Effect of oxygen on irradiated males and females of Drosophila. Heredi­tas 40: 295-312. Medvedev, N. N. 1935. The contributory effect of cold with irradiation in the production of mutations. Compt. Rend. Acad. Sci. U. R. S. S. 4: 283-285. (Cited by Baker, 1949) Moore, W. G. 1934. A comparison of the frequencies of visible mutations produced by X-ray treatment in different developmental stages of Drosophila. Genetics 19: 209-222. Muller, H.J. 1927. Artificial transmutation of the gene. Science 66: 84--87. 1930. Radiation and genetics. Am. Naturalist 64: 220-251. 194-0. An analysis of the process of structural change in chromosomes of Dro­ sophila. J. Genet. 40: 1-66. 1950. Some present problems in the genetic effects of radiation. J. Cell. Comp. Physiol. 35 (Suppl. 1): 9-70. 1954a. The nature of the genetic effects produced by radiation. Chap. 7, Radiation Biology, Vol. 1., A. Hollaender, ed. New York: McGraw-Hill, pp. 351-474. 1954b. The manner of production of mutations by radiation. Chap. 8, Radiation Biology, Vol. I., A. Hollaender, ed. New York: McGraw-Hill, pp. 475-626. 1958. Advances in radiation mutagenesis through studies of Drosophila. Proc. 2nd. U. N. Intl. Conf. on Peaceful Uses of Atomic Energy 22: 313-321. Novitski, E. 1949. Action of X-rays at low temperatures on the gametes of Drosophila. Am. Naturalist 83: 185-193. Oster, I. I. 1959. The spectrum of sensitivity of Drosophila germ cell stages to X-irradiation. Proc. 2nd Australian Conf. on Radiation Biology, J. H. Martin, ed. London: Butterworth Scientific Publications, pp. 253-265. Papalashwili, G. 1935. The effects of combined action of X-rays and low temperature on the frequency of translocations in Drosophila melanogaster. Biolo. Zh., Moscow, 4: 587-591. (Cited by Sollunn and Stromnaes, 1964) Plough, H. H. 1917. The effect of temperature on crossing-over in Drosophila. J. Exptl. Zool. 24: 147-209. Pontecorvo, G. 1942. The problem of dominant lethals. J. Genet. 43: 295-300. Powers, E. L., E. F. Ehret, A. Bannon, and A. Prock. 1957. Dependence on dose rate and wavelength of sensitivity to X-rays of dry spores of Bacillus megaterium. Rad. Res. 7: 443. Powers, E. L., R. B. Webb, and C. F. Ehret. 1959. Modification of sensitivity to radiation in single cells by physical means. Proc. 2nd Intl. Conf. on Peaceful Uses of Atomic Energy 22: 40~8. Savhagen, R. 1961. The relation between X-ray sensitivity and stages of development of treated cells in spermato-and spermiogenesis of Drosophila melanogaster. Hereditas 47: 43-68. Schmid, W. 1961. The effect of carbon monoxide as a respiratory inhibitor on the production of dominant lethal mutations by X-rays in Drosophila. Genetics 46: 663-670. Sobels, F. H. 1960. Chemical steps involved in the production of mutations and 'Chromosome aberrations by X-irradiation in Drosophila. I. The effect of post-treatment with cyanide in relation to dose-rate and oxygen tension. Intl. J. Rad. Biol. 2: 68-90. 1963a. Radiosensitivity and repair in different germ cell stages of Drosophila. Genetics Today. Proc. XI Intl. Congr. of Genetics, Vol. 2, S. J. Geerts, ed. New York: Per­gamon Press, 1965, pp. 235-255. 1963b. The contrasting effects of oxygen and nitrogen in determining initial sensitivity and post-radiation recovery in Drosophila sperm by spermatids. Intl. J. Rad. Biol. 7: 505. 1964. Post-radiation reduction of genetic damage in mature Drosophila sperm by nitrogen. Mutation Res. 1: 472-477. 1965. The role of oxygen in determining initial radiosensitivity and post-radi­ation recovery in the successive stages of Drosophila spermatogenesis. Mutation Res. 2: 168­ 191. Sollunn, F. J. and 0. Stromnaes. 1964. The effect of temperature during irradiation on in­duced mutation frequency in Drosophila melanogaster sperm. Hereditas 51: 1-12. Stone, W . S. 1955. Indirect effects of radiation on genetic material. Brookhaven Symp. in Biol. 8 (Mutation): 171-190. Stone, W. S., F. L. Haas, M. L. Alexander, and F. E. Clayton. 1954. Comments on the mech­anism of action of radiations on living systems. Univ. of Texas Publ. 5422: 244-271. Stone, W. S., 0. Wyss, and F. L. Haas. 1947. The production of mutations in Staphylococcus aureus by irradiation of the substrate. Proc. Natl. Acad. Sci. U. S. A. 33: 59-66. Strangio, V. A. 1962. Radiosensitivity during spermatogenesis in Drosophila melanogaster. Am. Naturalist 96: 145-149. Tallentire, A. and E. L. Powers. 1963. Modification of sensitivity to X-irradiation by water in Bacillus megaterium. Rad. Res. 20: 270-287. Thoday, J. M. and J. Read. 1947. Effect of oxygen on the frequency of chromosome aber­rations produced by X-rays. Nature 160: 608. Trout, W. E. 1964. Differential radiosensitivity as an explanation for so-called recovery in Drosophila sperm. Genetics 50: 173-179. von Borstel, R. C. 1963. Effects of radiation on germ cells of insects: dominant lethals, gamete inactivation, and gonial-cell killing. Radiation and Radioisotopes Applied to Insects of Agri­cultural Importance. Vienna: Intl. Atomic Energy Agency, pp. 367-383. Wedvik, H. and 0. Stromnaes. 1963. The effect of temperature during irradiation on the brood-pattern of dominant lethals induced in Drosophila melanogaster sperm. Intl. J. Rad. Biol. 7: 369-375. Yanders, A. F. 1959. The effects of X-rays on sperm activity in Drosophila. Genetics 44: 545. III. A Study of Radiation-Induced Translocations in Drosophila virilis in an Oxygen Atmosphere at4°C.1 GLENN E. OMUNDSON2 INTRODUCTION The conversion of energy from physical radiations into observable biological damage has long been recognized. Muller ( 1927) showed that deleterious muta­tions could be induced in Drosophila by exposing them to X-irradiation. Since then, other forms of ionizing radiation, and ultraviolet light, have been shown to be capable of inducing various forms of damage to biological systems ranging from viruses through man. Of primary importance to the system involved is the genetic damage which is frequently incurred, and then is multiplied by trans­mission to cellular and organismal descendants (Muller, 1954). In addition to gene or point mutations, chromosomal breaks can be produced which may cause aneuploidy, translocations, inversions, deletions, and other types of abnormal structural conditions. Though the results are fairly well understood, the operative mechanisms of the causative agents have only been partially revealed. The Oxygen Effect Cleveland (1949) noted that oxygen alone was capable of partial destruction of the chromosomes of the flagellate Trichonympha, while Conger and Fairchild ( 1952) noted similar damage to the microspores of Tradescantia paludosa. In experiments designed to show the oxygen effect in Tradescantia microspores, Giles and Riley ( 1949, 1950) , Giles and Beatty (1950), and Riley, Giles, and Beatty (1951, 1952) showed that the presence of oxygen greatly enhanced X-ray induced chromosomal and chromatidic aberrations. These experiments indi­cated that the presence of oxygen during the actual exposure time was the im­portant factor, and that pre-and posttreatments with oxygen were of secondary significance. Beatty, Beatty, and Collins (1956), postulated two methods by which oxygen could enhance radiation damage, namely by influencing the number of primary breaks formed, and by influencing the mechanisms of re­joining or restitution. Later work by Beatty and Beatty (1959, 1960, 1966) sup­ported this concept. Gaulden, Nix, and Moshman ( 1953) suggested for the neuroblast cells of Chortophage that the influence of oxygen tension on the response of cells to irra­ 1 This work was supported in part by a Public Health Service research grant, GM-11609, and training grants, GM-337-06 and GM-00337-07, from the National Institutes of Health. 2 Present address: Department of Kology, University of Mississippi, University, Miss. The author wishes to express his great indebtedness to the late Professor Wilson S. Stone for his advice, support and confidence in directing the research reported here. STUDIES I:\ GENETICS V. UniY. Texas Puhl. 6918, Sept., 1969. diation involved more than just the change in the amount of injury produced by irradiation. Whiting ( 1954) used the eggs of the wasp, H abrobracon, and found some evidence in favor of the conclusion that a reduction in the oxygen concen­tration reduced the number of primary breaks and that other factors influencing the final conditions found were unchanged in the irradiated cell. In Drosophila, Baker and co-workers (Baker and Sgourakis, 1950a, 1950b; Baker and Edington, 1952; Baker and von Halle, 1952, 1953) found an increase in induced translocation frequencies when flies were X-rayed in higher concen­trations of oxygen. Similar increases were noted at 2°C. Baker and co-workers claimed that their data supported the differential reunion hypothesis of oxygen action. In D. virilis, Haas, Dudgeon, and Stone (1953) and Haas, Dudgeon, Clay­ton, and Stone ( 1954) found that at relatively slow rates of irradiation and at low temperatures, there was only a slight decrease in the rate of X-ray induced trans­locations with a decrease in oxygen concentration from that of air down to 5%. With respect to total genetic damage, these data showed that for both slow and fast rates of irradiation, increases in the oxygen concentration at the time of ir­radiation resulted in an increase in the measurable genetic damage over all parts of the temperature range used. Differential susceptibility of the various stages of spermatogenesis was noted by Stone (1955), Stone and Alexander (1955), and Schmid (1961). Schmid proposes as a possible explanation that in mature sperm, which are relatively non-susceptible to radiation damage compared to early spermatids, chromosomes are packed side by side, while in spermatids, there is adequate space available and thus the caryolymph can be saturated with oxygen. The conclusions of Oster ( 195 7 a, 195 7b) parallel this as his data indicated that the higher sensitivity of spermatids compared to mature sperm possibly was due to more intra-and/or intercellular oxygen being normally available to these cells. Gray, Chase, Desch­ner, Hunt, and Scott (1958) estimated that there was ten times as much oxygen normally present in spermatids than there was in mature sperm. Sobels (1965) found oxygen to nitrogen ratios of 1.6 for sperm, 3.3 for spermatids, 2.4 for sper­matocytes, and 3.8 for spermatogonia. Kaufmann and Gay (1963) and Das, Kaufmann and Gay ( 1964a, 1964b) found that although the protamine type of protein does not develop from histones as spermatids mature, there is a de:rease in the lysine content of protein and an increase in the arginine content of mature sperm. Further evidence of such a transition was obtained by Abeleva ( 1966) when his data indicated that 2,4-dinitrophenol was effective during the period of spermiogenesis at the time of synthesis of arginine-rich nucleohistones. As a comparison from another system, Searle ( 1964) compared the frequencies of translocations in spermatozoa and spermatogonia of X-rayed mice and found a ratio of 6: 1, although indications from Russell and Russell (1959) and Auerbach and Slizynska ( 1964) are that this is perhaps too high an estimate. Sobels ( 1963, 1964, 1965) found that after irradiation in nitrogen, postirradi­ation recovery in spermatozoa was favored by nitrogen, while recovery was en­hanced in both spermatids and spermatocytes by oxygen. Late spermatids and spermatogonia showed no significant postirradiation recovery. Recovery of sperm by nitrogen posttreatment was also shown in sperm treated in inseminated fe­males. Yoon ( 1965) found in using a series of gases that applications of any of these gases for a duration of four minutes in posttreatment was too short to modify the induced damage. From the data he obtained, Rinehart ( 1963) felt that there was no posttreatment effect due to oxygen per se, but rather posttreat­ment damage was believed to be caused by nitric oxide which could act after irradiation. In studying stage 7 and 14 oocytes of D. melanogaster, Rinehart ( 1964) found that pre-and posttreatment with helium progressively degraded the systems responsible for protection and repair, but these mechanisms could be regenerated by the administration of oxygen. Temperature Plough ( 1917) recognized the influence of temperature on the rate of crossing­over in D. melanogaster. He found that an increase in the induction rate of trans­locations accompanied a change in temperature either above or below room temperature. Mickey (1938, 1939) found that a significantly higher frequency of translocations was induced in flies which were irradiated at colder tempera­tures than from those of room temperature, although Muller (1940) could find no increase in the number of translocations induced at 5° and 37°C. Muller attributed the detection of a temperature effect found by earlier workers to poor experimental procedures and to the inclusion of immature sperm in their results. Darlington and LaCour ( 1945) also could find no temperature effects on the frequency of induced chromosomal aberrations, and attributed the earlier results to changes in the nucleic acid metabolism or time-effected changes in nuclear development. Baker (1949) interpreted his results in D. virilis as showing that an increase in the rate of radiation induced translocations occurred with changes in tempera­ture either above or below room temperatures. Similar results in increased fre­quency of translocations were found by Haas, Dudgeon, Clayton, and Stone ( 1954). They indicated that the increase in oxygen concentrations in the germ cells at low temperatures relative to room temperatures probably account for some portions of the resultant increases in damage. They also suggested that the greater activity of normal biological protective agents at room temperature might account for much of the temperature differential. The histological and cytological examinations of Clayton ( 1962, 1965) and others have shown that the stages of spermatogenesis are more distinct in D. virilis than in D. melanogaster. Further separation of the stages is obtained in D. virilis in which the accessory gland can deplete the sperm supply (author's observations), while jn D. melanogaster the sperm supply can deplete the acces­sory gland (Lefevre and Jonnson, 1962). Thus more accurate measurements of the effects of radiation damage occurrjng in the separate stages of spermato­genesis may be obtained. D. virilis also has the advantage of a greater chromo­some number than D. melanogaster, thus allowing for a more accurate determina­tion of the rate of induction of translocations in a biological system. For these reasons, the experiments in this study are designed to study radiation induced damage as measured by translocations in D. virilis when they are treated in an oxygen atmosphere at 4°C. MATERIALS AND METHODS The test organism used was Drosophila virilis, which was maintained by mass culture in half-pint milk bottles with a cornmeal-agar food at 20-22°C. Males for all phases of research were taken from the Texmelucan strain (UT1801.1 ) . These were taken within three hours of eclosion and aged 19-22 hours prior to treatment. Homozygous recessive virgin females used in the translocation tests were genetically marked such that the second chromosome carried the gene broken (b, 188.0); the third chromosome, tiny bristle (tb, 104.0) and gap (gp, 118.5); the fourth chromosome, cardinal (cd, 32.2); and the fifth chromosome, peach (pe, 203.0). All females were aged for 7-10 days prior to mating. The minor dot autosome, chromosome 6, was not used in these experiments. Also, the X chromosome was not marked or used, although participation of the Y chromo­some in translocations was recorded by noting the distribution of classes with respect to sex. Translocations were scored by tests using the markers b, gp, cd, and pe. Treated males were mated to virgin marker females, and the resultant heterozygous F 1 males were individually back-crossed to three marker virgins. The presence of induced translocations in the F2 generation was scored by noting the absence of certain segregation groups. Control of temperature, oxygen, X-rays, and other procedural methods was as exercised earlier by Omundson (in this bulletin) . RESULTS The results are summarized in Tables 1 through 3. TABLE 1 Distribution of breaks Chromosome Y 2 3 4 5 Series Total lOOOr 47 120 112 111 92 482 1500r 36 69 85 71 76 337 2000r 21 41 ...!!.!!_ _1i 51 191 Totals 104 230 241 216 219 1010 Percent 10.3 22.8 23.8 21.4 21. 7 100.0 Trans locations The degree of radiation induced damage as measured by translocations was determined for the 1000r, 1500r, and 2000r of X-rays series. These data are briefly summarized in Table 1 according to series and chromosome involved. Table 2 presents the data obtained according to the experimental series and total TABLE 2.A The frequencies of translocations Period A B c D E F G Series lOOOr a b c 30 391 7.7 41 292 14.1 52 270 19.3 71 246 29.9 24 103 23.3 2 77 2.6 1 211 0.5 1500r a b c 60 258 22.2 80 218 36.7 10 17 58.8 1 3 33.3 2 2000r a b c 43 224 19.2 41 91 45.1 8 Total a 133 162 62 72 24 2 1 b 873 601 287 249 103 77 221 a= number of sperm with one or more translocations b = number of sperm tested c = %of sperm with one or more translocations TABLE 2.B* The frequencies of translocations Period A B c D E F G Series lOOOr a b c 4 307 1.3 54 349 15.5 57 270 21.1 2 211 0.9 2000r a b c 33 290 11.4 71 174 40.8 10 14 71.4 2 13 15.4 a = number of sperm with one or more translocations b = number of sperm tested c = %of sperm with one or more translocations *Adapted from Alexander, Bergendahl, and Brittain (1959) by permission. The University of Texas Publication TABLE 3A Distribution of translocations-1000r series Period A B c D E F G T e Total Sim,Ele Y-2 3 4 1 8,1* 16,1* Y-3 2 1 1 4 1 9 Y-4 2 2 2* 1,1* 5,3* Y-5 l* 2 2 5 2 11,1* 2-3 5 6 11 7,1* 3,2* 32,3* 2-4 6 6 8 11 2 33 2-5 5 6 4,1* 1 16,1* 3-4 3,1* 6 6 9,1* 1 1 26,2* 3-5 1 3 5 8 1 1 19 4-5 4 7 3 4 2,1* 20,1* Multi,Ele 2-3-4 3 1 4 2-3-5 3 2 1 6 2-4-5 1 1 1 2 1 6 3-4-5 3 1 1 4 9 2-3-4-5 - 1 1 2 .!. cOm,Elex Y-3-5 1 1 Number of translocations 31 41 52 74 26 2 1 227 Number of S,Eerm with one .2!. more translocations 30 41 52 71 24 2 1 221 'Number of S,Eerm tested 391 292 270 246 103 77 411 lS90 .~."sperm ·with ·one or more translocations 7.7 14.1 19.3 29.9 23.3 2.6 0.5 * Indicates involvement in complex of more than one translocation per sperm nucleus. number of translocations scored with respect to mating period. Tables 3A through 3C represent the data according to the types of translocations scored and their frequencies in each mating period. Due to severe lethality and sterility factors operating at 1500r and 2000r~ only certain periods (A, B, C, D, and Gin 1500r; A, B, and Gin 2000r) could be checked for translocations, and these can not be considered as complete due to the small numbers of sperm producing viable off­spring. Because of these reasons, standard statistical treatment of the data was omitted. It should be noted that no spontaneous translocations or translocations from the TABLE 3B Distribution of translocations-1500r series Period A B c D E-G Type Total Simple Y-2 3,1* 1 4,1* Y-3 3 6 1,1* 10,1* Y-4 2 5 7 Y-5 3 l* 3,1* 2-3 7 8 15 2-4 9 7 16 2-5 8 10 1 19 3-4 10,1* 7,1* 2 19,2* 3-5 9 12 1 22 4-5 3 7 1,1* 1 12,1* Multiple 2-3-4 1 1 2-3-5 3 3 2-4-5 1 3 4 3-4-5 2 2 2-3-4-5 1 1 2 Y-complex Y-3-4 1 1 Y-3-5 1 2 3 Y-4-5 1 1 Y-2-3-4 1 1 Y-2-3-5 1 1 Y-2-3-4-5-1 1 2 Number of translocations 61 81 11 1 154 Number of sperm with .2.!!!:. ..£!: ~translocations 60 80 10 1 151 Number of sperm ·tested 258 218 17 3 2(G) 498 ! sperm with one ·or more trartslocations 22.2 36.7 58.8 33.3 • Indicates involvement in complex of more than one translocation per sperm nucleus. female line were detected as has been the case in the work of others. Alexander ( 1962) discusses the detection and separation of the various types. Response of Spermatogenic Cells As determined by translocations at 1000r, fewer translocations were found in period E (23.3% of all sperm tested having one or more translocations per sperm The University of Texas Publication TABLE 3C Distribution of tran.:locations-2000r series Period A B C-G Type Total Simple Y-2 4 1 Y-3 2 3 Y-4 1 1 Y-5 2 2 2-3 5,1* 4,1* 2-4 1 3,1* 2-5 7 2 3-4 2 4 3-5 5 6,1* 4-5 5,1* 6,1* Multiple 2-3-4 1 2-3-5 1 3 2-4-5 2 2 3-4-5 1 Y-complex Y-3-4 1 Y-3-5 1 Y-4-5 1 Y-2-3-5 1 Y-2-3-4-5 1 Number of translocations - 44 43 Number of sperm with ~or ~translocations 43 41 Number of sperm tested 224 91 ! sperm with one .2!. ~transloc.ations 19.2 45.1 5 5 2 4 9,2* 4,1* 9 6 11,1* 11,2* 1 4 4 1 1 1 1 1 1 87 84 8(G) 321 *Indicates involvement in complex of more than one translocation per sperm nucleus. nucleus) than in period D (29.9%). In period F (meiosis), due to increasing sterility and lethality factors, only 77 viable sperm were obtained for testing. Induced damage measured at 1500r showed that no sperm that gave viable off­spring were produced in periods E and F; two in G; three in D, and only 17 in period C. Similar data were obtained at 2000r of X-rays, where no viable sperm for testing were produced in periods C through F; eight in G; and only 91 in period B. Thus the data obtained for these last two series indicated that maximum damage was induced in the early stages of spermiogenesis (E) and meiosis (F), than in later stages of spermiogenesis ( C and D) and spermatogonia ( G) . In each series, damage as measured by translocations showed mature sperm (A) to be the most resistant period of development to radiation damage, closely fol­lowed by sperm bundles (B). DISCUSSION Trans locations Bauer, Demerec, and Kaufmann (1938), and Kaufmann (1939, 1941) have reported that breaks are randomly distributed and proportional to the mitotic lengths in both hetero-and euchromatic regions of chromosomes. These re­ports have been reiterated by Kaufmann ( 1946, 1954). Baker (1949), Haas, Dudgeon, Clayton, and Stone (1954), and Clayton ( 1962, 1965) found for D. virilis that the autosomes were about equally involved in the formation of translocations, but that the Y chromosome participated only about half as fre­quently as the major autosomes. Elequin (1966) and Yoon (1965) reported even lower values in D. melanogaster. All these data agree with those previously re­ported by Alexander and co-workers for both D. virilis and D. melanogaster. In the present experiments (Table 1), the Y chromosome participated in 10.3 % of the breaks, or 0.460 as frequently as the autosomes. This compares to the data of Baker (1949), which was 9.6% and 0.422. Somewhat higher results were re­ported by Haas, Dudgeon, Clayton, and Stone (1954) of 12.7% and 0.583, and Clayton (1965) of 12.6% and 0.573, although Clayton (1962) earlier reported a low of 7.3% and 0.316. Haas, et al. (1954) reported that of the autosomes, chro­mosome 2, which is slightly longer than the other autosomes, participated more frequently than did the others (2-23.0%; 3-21.7%; 4-21.0%; 5-21.6%). Clay­ton (1962) found that chromosome 5 was the most frequent participator (2­22.4%; 3-22.0%; 4--19.5%; 5-28.8%), but in another series of experiments, chromosome 4 was (1965: 2-20.0%; 3-22.4%; 4-24.1%; 5-20.9%). In the present experiments, chromosome 3 was the most frequently involved (2-22.8%; 3-23.9%; 4-21.4%; 5-21.7%). Thus it is concluded that with respect to the distribution of breaks in the Y chromosome and in the autosomes, the combination of both oxygen and cold temperatures during the time of irradiation had no effect. Response of Spermatogenic Cells Translocation frequencies are known to vary with the stage of spermatogenesis (Auerbach, 1953; Alexander and Stone, 1955; Glass, 1955a, 1955b; Oster, 1957a, 195 7b; Alexander and Bergendahl, 1962, 1964) . These workers found greatest resistance among spermatogonia, while those cells in spermiogenesis were the most susceptible. The data for the 1000r series (Table 3A) are in agreement with the data previously reported. For this series, maximum damage as determined by translocation frequencies is found in period D (29.9%, 71 sperm with one or more translocations per sperm/246 sperm tested). Only one translocation was found in 211 tested sperm in period G (spermatogonia), which was the lowest frequency recorded (0.5%). In period F. only two translocations were found in 77 tested sperm (2.6%), but F1 lethality and sterility factors were important in reducing the number of sperm producing viable offspring for testing. At 1500r and 2000r of X-rays, the lack of sperm producing viable offspring in the F1 males was such that valid interpretations of the results and comparisons with the data of others is difficult. Tables 2A and 2B compare the results reported here and those of Alexander, Bergendahl, and Brittain (1959) for 1000r and 2000r of X-rays. In the 1OOOr series~ Alexander, et al., report 1.3% ( 4 of 307) of the sperm tested in period B carried translocations, while periods D and E were 15.5% and 21.1 % respectively. In the experiments reported here, 14.1 % of the sperm tested in period B had one or more translocations per sperm nucleus. Maxi­mum damage was found in period D, where 29.9% of the sperm tested carried translocations, while 23.3% did in period E. In period F, Alexander, et al., found that the rate of induced translocations dropped to only 0.9% (2 of 221), but in the data reported here, the rate was 2.6% (2 of 77). In the work of Alex­ander, et al., at this radiation dose, F 1 lethality and sterility factors were not im­portant in reducing the number of sperm producing viable offspring for testing, although these were important in periods E and Fat 2000r. At 1000r in the re­sults reported here, these factors were operative in both periods E and F, and at 2000r were such that no sperm produced viable offspring in periods C through F, only 8 in period G, and only 91 in period B where the translocation rate was 45.1 % ( 41 of 91) . For this same period, Alexander, et al., reported only 11.4%, and 40.8% in period D. The combined treatment procedure used in this work appears to have been much more effective in the enhancement of radiation dam­age than methods heretofore employed. It is to be noted that the peaks of damage as measured by dominant lethals reported by Omundson (this Bulletin) and by translocations do not coincide. This has previously been reported by Stone and co-workers, Alexander and co-workers, and by others who have studied the effects of both dominant lethals and trans­locations with respect to the various stages of spermatogenesis. SUMMARY Translocation tests were made by mating the treated males of the 1000r, 1500r, and 2000r series to marker virgins. The resultant heterozygous F1 males were then individually backcrossed to three marker virgins. The presence of induced translocations in the F2 generation were scored by noting the absence of certain segregation groups. Y-chromosomal participation in translocations was recorded by noting the distribution of the segregation groups with respect to sex. The sixth autosome and the X chromosome were not followed. The data obtained indicate that: 1) The degree of radiation damage as measured by translocations was recorded for 1000r, 1500r, and 2000r of X-rays, but due to severe lethality and sterility factors operative at the higher doses, only approximately one-half (8 of 14) of the mating periods could be scored, and these could not be considered as complete due to the small numbers of sperm producing viable offspring. Comparisons of these data were briefly made with those of Alexander, et al. (1959), and with others, and indicated that the procedure used here more efficiently enhanced radiation damage than was previously reported. No spontaneous translocations were recorded. 2) The distribution of the breaks within the Y chromosome and in the major autosomes is comparable to that previously reported, thus indicating that the treatment conditions employed here do not alter this distribution. 3) In each series, damage as measured by translocations indicated that cells in spermiogenesis were the most sensitive to radiation damage, while mature sperm were the most resistant. 4) Comparisons of data obtained from translocation studies with data obtained from dominant lethal studies indicate that the peaks of damage of the two differ­ent measures do not coincide. This is in accordance with previous reports. Using the frequencies of radiation induced translocations as the measure of damage, the data indicate that the combined effects of an oxygen atmosphere aml cold temperature at the time of irradiation produce results that are qualitatively similar to those found by other workers, but quantitatively more damage is in­duced per unit of X-irradiation than previously reported, thus indicating a rela­tively greater efficiency of enhancement of radiation damage. BIBLIOGRAPHY Abeleva, E. A. 1966. The influence of 2,4-dinitrophenol on the incidence of mutations in the spermia and spermatids of the Drosophila irradiated with gamma rays. Genetics U.S.S.R. 1: 159-162. Alexander, M. L. 1962. Detection and separation of spontaneous and induced translocations in mature and immature germ cells of Drosophila. Am. Naturalist 96: 309-315. Alexander, M. L. and J. Bergendahl. 1962. Biological damage in the mature sperm of Drosophila virilis in oxygen and nitrogen with different dose intensities of gamma rays. Genetics 47: 71-84. 1964. Dose rate effects in the developing germ cells of male Drosophila. Genetics 49: 1-16. Alexander, M . L., J. Bergendahl, and M. Brittain. 1959. Biological damage in mature and immature germ cells of Drosophila virilis with ionizing radiations. Genetics 44: 979-999. Alexander, M. L. and W. S. Stone. 1955. Radiation damage in the developing germ cells of Drosophila virilis. Proc. Natl. Acad. Sci. U.S.A. 41: 1046-1057. Auerbach, C. 1953. Sensitivity of Drosophila germ cells to mutagens. Heredity 6: 247-257. Auerbach, C. and H. Slizynska. 1964. The frequency of induced translocations in sperma­ togonia. Mutation Res. 1: 468. Baker, W. K. 1949. The production of chromosome interchanges in D. virilis. Genetics 34: 167-193. Baker, W. K. and C. W. Eddington. 1952. The induction of translocations and recessive lethals in Drosophila under va~ious oxygen concentrations. Genetics 3 7: 665-677. Baker, W. K. and E. Sgourakis. 1950a. Alteration of X-ray sensitivity of Drosophila by means of respiratory inhibitors. Genetics 35: 96. 1950b. The effect of oxygen concentration on the rate of ·X-ray induced muta­tions in Drosophila. Proc. Natl. Acad. Sci. U.S.A. 36: 176-184. Baker, W. K. and E. S. von Halle. 1952. The effect of oxygen concentration on the induc­tion of dominant lethals. Genetics 37: 565. 1953. The basis of the oxygen effect on X-irradiatcd Drosophila sperm. Proc. Natl. Acad. Sci. U.S.A. 39: 152-161. Bauer, H., M. Demerec, and B. P. Kaufmann. 1938. X-ray induced chromosome alterations in Drosophila melanogaster. Genetics 23: 610-630. Beatty, A. V. and J. W. Beatty. 1959. Metabolic inhibitors and chromosome rejoining. Am. J. Botany 46: 317-323. 1960. Postirradiative effects on chromosomal aberrations in Tradescantia micro­spores. Genetics 45: 331-344. 1966. Reduction of radiation damage to Tradescantia chromosomes by Adenosine Triphosphate, Proline, and Histidine. Genetics 53: 47-54. Beatty, A. V., J. W. Beatty, and C. Collins. 1956. Effect of various intensities of X-radiation on chromosomal aberrations. Am. J. Botany 43: 328-332. Clayton, F. E. 1962. Effects of X-ray irradiation in D. virilis at different stages of sperma­togenesis. Univ. of Texas Publ. 6205: 345-373. 1965. Genetics effects from simultaneous irradiation of immature and mature Drosophila virilis males. Genetics 52: 1081-1092. Cleveland, L. R. 1949. The whole life cycle of the chromosomes and their coiling systems. Trans. Am. Phil. Soc. new series 39: 1-100. Conger, A. D. and L. M. Fairchild. 1952. Breakage of chromosomes by oxygen. Pro:::. Natl. Acad. Sci. U.S.A. 38: 289-299. Darlington, C. D. and L. F. LaCour. 1945. Chromosome breakage and the nucleic acid cycle. J. Genet. 46: 180-251. Das, C. C., B. P. Kaufmann, and H. Gay. 1964a. Autoradiographic evidence of synthesis of argenine rich histone during spermiogenesis in Drosophila melanogaster. Nature 204: 1008-1009. 1964b. Histone-protein transition in Drosophila melanogaster. I. Changes during spermiogenesis. Exptl. Cell Res. 35: 507-514. Elequin, F. T. 1966. Modification of induced genetic damage in Drosophila melanogaster by oxygen and argon treatments between two doses of X-rays. Univ. of Texas Publ. 6615: 177-193. Gaulden, M. E., M. Nix, and J. Moshman. 1953. Effects of oxygen concentration on X-ray induced mitotic inhibition in living Chortophaga neuroblasts. J. Cell. Comp. Physiol. 41: 451-470. Giles, N. H ., Jr. and A. V. Beatty. 1950. The effect of X-radiation in oxygen and in hydrogen at normal and positive pressures on chromosome aberration frequency in Tradescantia microspores. Science 112: 643-6'45. Giles, N. H., Jr. and H. P. Riley. 1949. The effect of oxygen on the frequency of X-ray induced chromosomal rearrangements in Tradescantia microspores. Proc. Natl. Acad. Sci. U.S.A. 35: 640-646. 1950. Studies on the mechanisms of the oxygen effect on the radiosensitivity of Tradescantia chromosomes. Proc. Natl. Acad. Sci. U.S.A. 36: 337-344. Glass, B. 1955a. A comparative study of induced mutation in the oocytes and spermatozoa of Drosophila melanogaster. I. Translocations and inversions. Genetics 40: 252--267. 1955b. Differences in mutability during different stages of gametogenesis in Drosaphila. Brookhaven Symp. in Biology 8 (Mutation): 148-170. Gray, L. H., H. B. Chase, E. E. Deschner, J. W. Hunt, 0. C. A. Scott. 1958. The influence of oxygen and peroxide on the response of mammalian cells and tissues. Proc. 2nd. U.N. Intl. Conf. Peaceful Use of Atomic Energy 22: 413-419. Haas, F. L., E. Dudgeon, and W. S. Stone. 1953. Frequency of chromosomal rearrangements as related to rate of irradiation, temperature, and gases. Genetics 37: 589-590. Haas, F. L., E. Dudgeon, F. E. Clayton, and W . S. Stone. 1954. Measurement and control of some direct and indirect effects of X-radiation. Genetics 39: 453-471. Kaufmann, B. P. 1939. Distribution of induced breaks along the X-Chromosome of Drosophila melanogaster. Proc. Natl. Acad. Sci. U.S.A. 25: 571-577. 1941. Induced chromosomal breaks in Drosophila. Cold Spring Harbor Symp. Quant. Biol. 9: 82-92. 1946. Organization of the chromosome. I. Break distribution and chromosome recombination in Drosophila melanogaster. J. Exptl. Zool. 102: 293-320. Kaufmann, B. P. 1954. Chromosome aberrations induced in animal cells by ionizing radia­tions. Chap. 9, Radiation Biology, A. Hollaender, ed. New York: McGraw-Hill, pp. 627-711. Kaufmann, B. P. and H. Gay. 1963. Cytological evaluation of differential radiosensitivity in spermatogenous cells of Drosophila. In Repairs from Genetic Radiation Damage, F. H. Sobels, ed. New York: Pergamon Press, pp. 375-408. Lefevre, G., Jr. and U. B. Jonnson. 1962. Sperm transfer, storage, displacement, and utiliza­tion in Drosophila melanogaster. Genetics 47: 1710-1736. Mickey, G. H. 1938. Effect of temperature on frequency of translocations produced by X-rays. Genetics 23: 160. 1939. The influence of low temperature on the frequency of translocations pro­duced by X-rays in Drosophila melanogaster. Genetics 21: 386-407. Muller, H.J. 1927. Artificial transmutation of the gene. Science 66: 84-87. 1940. An analysis of the process of structural change in chromosomes of Dro­sophila. J. Genet. 40: 1-66. 1954. The nature of the genetic effects produced by radiation. Chap. 7, Radiation Biology, Vol. I., A. Hollaender, ed. New York: McGraw-Hill, pp. 351-474. Oster, I. I. 1957a. Suggested mechanisms underlying the differential radiosensitivity of cells having-condensed chromosomes. Genetics 42: 387. 1957b. Modification of X-ray mutagenesis in Drosophila. II. Relative sensitivity of spermatids and mature spermatozoa. Advances in Radiobiology, G. de Hevesey, A. Fors­berg, and J. Abbott, eds. Edinburgh: Oliver and Boyd, pp. 475. Plough, H. H. 1917. The effect of temperature on crossing-over in Drosophila. J. Exptl. Zool. 24: 147-209. Riley, H. P., N. H. Giles, Jr., and A. V. Beatty. 1951. The oxygen effect on X-ray induced chromatid aberrations in Tradescantia microspores. Rec. Genet. Soc. Am. 20: 120. 1952. The effect of oxygen on the induction of chromatid aberrations in Trades­cantia microspores by X-irradiation. Am. J. Botany 39: 592-597. Rinehart, R. R. 1963. Some effects of nitric oxide and oxygen on dominant lethal production in X-irradiated Drosophila virilis males. Genetics 48: 1673-1681. 1964. Influence of oxygen, helium and metabolic inhibition on X-ray induced dominant lethality in stage 7 and stage 14 oocytes of Drosophila melanogaster. Genetics 49: 855-863. Russell, W. L. and L. B. Russell. 1959. The genetic and phenotypic characteristics of radia­tion induced mutations in mice. Rad. Res. Suppl. 1: 296-305. Schmid, W. 1961. The differential susceptibility of sperm and spermatids to ionizing radia­tion. Am. Naturalist 95: 103-111. Searle, A. G. 1964. Genetic effects of spermatogonial X-irradiation on productivity of F 1 female mice. Mutation Res. 1: 99-108. Sobels, F. H. 1963. The contrasting effects of oxygen and nitrogen in determining initial sensitivity and postradiation recovery in Drosophila sperm and spermatids. Intl. J. Rad. Biol. 7: 505. 1964. Post-radiation of genetic damage in mature Drosophila sperm by nitrogen. Mutation Res. 1: 472-477. 1965. The role of oxygen in determining initial radiosensitivity and post-radia­tion recovery in the successive stages of Drosophila spermatogenesis. Mutation Res. 2: 168-191. Stone. W. S. 1955. Indirect effects of radiation on genetic material. Brookhaven Symp. in Biol 8 (Mutation): 171-190. Whiting. A. B. 1954. The effects of oxygen on the frequency of X-ray-induced mutations in Habrobracon eggs. Genetics 39: 851-858. Yoon, J. S. 1965. Genetic X-radiation damage and its modification by some gases and a time factor in Drosophila melanogaster. A dissertation, University of Texas. IV. Descriptions of New Hawaiian Drosophila1' 2 D. ELMO HARDY AND KENNETH Y. KANESHIR03 The following ten new species belong in a variety of species groups and are being described at this time to provide names needed for some of the genetic studies being reported on in this bulletin. The drawings have been prepared by Miss Geraldine Oda, University of Hawaii. Drosophila assita n. sp. (figs. 1 a-c) Because of the all rufous thorax in combination with the presence of a large brown spot in the middle of cell R1 and the tiny, seta-like preapical dorsal bristles of the front tibiae, this would run to couplet eight of our key to picture-winged species (Hardy and Kaneshiro, 1968: 174), and the species is closely related to vesciseta Hardy and Kaneshiro. It differs by having the labella and palpi all yel­low, rather than black; by having a clump of long curled dorsal cilia on the front tibia of male and complete rows of anterodorsal and posterodorsal cilia extending almost the full length of the segment (fig. 1a), rather than lacking the long basal cilia and having only about four long cilia on the posterodorsal surface of front tibia; by having a row of six long anterodorsal cilia on front basitarsus, and with four or five short posterodorsal cilia on basal two-thirds, rather than having over a dozen long black cilia arranged over the dorsal surface as in vesciseta. The wing markings are very similar in the two, the brown spot at middle of cell Riis larger, however, in assita, being distinctly longer than the spot at apex of cell Ri; in vesciseta the middle spot is shorter, less than the length of the apical wing spot. Compare figure 1 b with figure 26b (Hardy and Kaneshiro, 1968: 232) . The geni­talia differ as shown in figure 1c, compared with figure 26c (Zoe. cit.). MALE. Head: Proclinate bristles equal in size to anterior reclinates, the latter situated one-third to two-fifths the distance between the proclinates and upper reclinates. The upper bristle on the sides of the oral margin is strong, approxi­mately equal in size to the upper reclinate bristles, the second bristle in each row is about half the length of the first. Head and appendages entirely yellow, except for a tinge of brown over the upper median portion of the occiput, extending over the ocellar triangle. The front is opaque, golden yellow over median portion, yellow along orbits. The arista has about ten dorsal and three to four ventral rays in addition to the small apical fork and the inner surface is setose almost to its base. Palpus with a single apical bristle, subequal to second bristle of vibrissal row. Labellum fleshy. Thorax: Yellow to rufous in ground color, tinged faintly 1 Published with the approval of the Director of the Hawaii Agricultural Experiment Station as Journal Serjcs No. 1087. 2 This work was made possible by NIH Grant No. GM-10640. 3 Department of Entomology, University of Hawaii, Honolulu, Hawaii. STUDIES IN GENETICS V. Univ. Texas Pub!. 69 18, Sept., 1969. with brown on each mesopleuron, upper pteropleuron, metapleuron and over the pleurotergon; also, the mesonotum with a faint tinge of brown down median por­tion. Two strong humeral bristles present, anterior dorsocentrals situated just slightly in front of a line drawn between second anterior supraalars. Scutellum pale yellow. Legs: Yellow except for a faint preapical tinge of brown on hind tibiae, similar in many respects to punalua with front femur slender, flattened ventrally. Tibiae and tarsi bearing long cilia as noted above and as in figure 1a. Wings: Subhyaline with the subbasal brown marginal spot not continuous to r-m crossvein; with the spot in middle of cell R1 continuous across cell R3 almost to vein RH5 and distinctly longer than the spot at apex of cell R1 • Spot at apex of R1 continuing through upper two-thirds of cell R3 • Spot at apex of vein R4+s rather FIG. 1. D. assita n. sp. a, front tibia and tarsus of male; b, wing: c, male genitalia. small not contiguous with other apical spots; also with a prominent spot at the apex of vein Mi+2 and a large rather dumbbell spot over them crossvein as in figure 1 b. The r-m crossvein situated near basal one-fourth of cell 1st M2• Ab­domen: Largely subshining black in ground color covered with grey-brown pol­len. First tergum largely yellow, with just a faint streak of brown through median portion and with posterolateral margins of terga 2--4 yellow, covered with yellow­brown pollen. Venter entirely yellow; genitalia are as in figure 1c. Length: body and wings, 4.25-4.50 mm. FEMALE unknown. Holotype male and two male paratypes, Mt. House, above Naalehu, Hawaii, southern slopes of Mauna Loa, 2750', December 18-19, 1967, "L19" (H. L. Car­son and K. Y. Kaneshiro). Type in the B. P. Bishop Museum, paratypes in the collections of the U. S. National Museum and the University of Hawaii. Drosophila attigua n. sp. (fig. 2) This is obviously a very primitive species and is externally indistinguishable from D. primaeva Hardy and Kaneshiro (1968: 258). It can be separated from primaeva by lacking setae on the aedeagus and by having setae on the parameres. It has been demonstrated that these are chromosomally distinct and this species is being discussed in a paper in this journal by Dr. H. L. Carson. For description and figures of primaeva refer to the above cited reference. The primaeva complex of species is recognized by the lack of ciliation or ornamentation on the male legs, by having a black sclerotized rim on each labellum in combination with the predominantly rufous body, the evenly yellow-brown tinged wings, slightly darker over the m crossvein and the large size, compared to most rufous species which lack ornamentation on the legs. FEMALE. Not distinguishable from primaeva. Holotype male, allotype female and 32 paratypes, 19 males, 13 females, Mt. Kahili, Kauai, 2500', March 19-20, 1968, "L41C21" (J.P. Murphy). Type and allotype in the B. P. Bishop Museum, paratypes in the collections of the U.S. National Museum, British Museum (Natural History), and the Uni­versity of Hawaii. Drosophila claytonae n. sp. (figs. 3a-c) This species is almost identical with D. limitata Hardy and Kaneshiro in most respects but is differentiated by having the brown markings on the mesonotum semi-shining, not having a dense covering of yellow-grey over the mesonotum almost obscuring the brown color; also by having the sternopleura largely yellow, with the bristles situated on a yellow background, rather than sternopleura being mostly brown to black with the bristles situated on a brown background. The subbasal wing spot is pale brown in limitata, also the wings in claytonae are more slender and elongate, nearly three times longer than wide, rather than approxi­mately two and one-half times longer than wide. The wing markings are very similar in the two (fig. 3a, c.f. with fig. 8b, Hardy and Kaneshiro, 1968: 195). The ornamentation of the front legs appears to be the same in both species, both are Frn. 3. D. claytonae n. sp. a, wing; b, front leg of male; c, male genitalia. characterized by having an abundance of rather short cilia over the dorsal and anterodorsal portion of the front tibia and basitarsus; also the preapical dorsal bristle of the tibia is strong, extending almost twice its length beyond apex of tibia (fig. 3b). The genitalia are as in figure 3c. Otherwise fitting the description of limitata refer to Hardy and Kaneshiro (op. cit.: 194). Holotype male, allotype female and nine paratypes, five males, four females, Upper Olaa Forest, Hawaii, 4,000', July 8-9, 1968, F1 generation, "L89L1" (F. E. Clayton). Type and allotype in B. P. Bishop Museum. Paratypes in the collections of the U.S. National Museum, British Museum (Natural History) and the University of Hawaii. It should be noted that some specimens of claytonae show indications of spurs developed along vein Ml+2 , the obvious beginning of the extra crossvein charac­teristic of the "idiomyia" group of species. These short spur veins are also found frequently in specimens of limitata. It is a pleasure to name this species after Dr. Frances E. Clayton, University of Arkansas, whose contributions in cytogenetics and laboratory rearing techniques have been most valuable to our study of the Evolution and Genetics of Hawaiian drosophilids. Drosophila hystricosa n. sp. (fig:;. 4a-d) Because of the dark brown to black body, head and appendages, the presence of ciliation on the front basitarsi and lack of markings in the wings, this species resembles D. caccabata Hardy. The two are not related, however, and the re­semblance is superficial. It differs from caccabata by having the mouthparts 'F FIG. 4. D. hystricosa n. sp. a, mouthparts of male; b, palpus; c, apex of front tibia and front tarsus of male; d, male genitalia. highly modified, the labella spinose around the margins, rather than mouthparts not ornate; by the dorsal cilia on front tarsi more elongate and much different in arrangement as in figure 4c, compared with figure 55a (Hardy, 1965: 197) ; by having the palpi densely setose dorsally; three ventral rays on the arista rather than two, and the anterior reclinate bristles smaller than proclinates. D. hystricosa is more closely related to biseriata Hardy, from Oahu, but that species differs by having the lower half of the front, the face, genae, palpi, and two basal seg­ments of antennae yellow, rather than black; each labellum with a series of four to five black spines on upper apical portion (ref. fig. 45a, Hardy, 1965: 180) rather than with a complete border of spines (fig. 4a) ; by having the legs yellow, rather than front femora brown; by the cilia on front tarsi being less numerous, compare fig. 45b (Hardy, 1965: 180) with figure 4c. MALE. Predominantly dark brown to black species, grey-brown pollinose, subshining. Head: Higher than long, completely dark colored, mostly black with lower portion of front, median portion of face, and lower portion of occiput tinged with rufous in the ground floor. Face raised down the median portion with antennal furrows rather well developed. Sides of face converging below, narrowest in area occupied by oral vibrissae. The vibrissae small, hair-like and uniform in size. Clypeus and mouthparts dark brown to black. Palpi densely setose dorsally (fig. 4b). Labella with strong spines around margins as in figure 4a. Anterior reclinate bristles of front slightly shorter than proclinates and situated almost opposite the latter. Arista with six dorsal and three ventral rays in addition to apical fork. Thorax: Entirely dark colored, mesonotum rather densely grey­brown pollinose but with the ground color slightly shining through. Pleura faintly tinged with rufous in the ground color. Two pairs of well-developed humeral bristles. Anterior dorsocentral bristles about three-fifths as long as posterior pair and situated approximately opposite the second pair of supraalars. Halteres yel­ low. Legs: Coxae and front femora brown, middle and hind femora yellow, tinged lightly with brown. Tibiae and tarsi yellow. Front tibia lacking long cilia but with numerous erect hairs over the posterodorsal and posterior surfaces, these hairs are equal to just slightly longer than the width of the tibia. Front tarsi with long, slightly curled cilia, arranged down the dorsal surface in three irregular rows from the base of tarsus to apex of the third tarsomere (fig. 4c); fifteen or more long hairs on the basitarsus. Basitarsus just slightly over half as long as tibia and over two times longer than second tarsomere, preapical dorsal bristle small, extending approximately to apex of segment. Wings: Subhyaline, lacking dark markings. Costal fringe extending one-third to two-fifths the distance be­tween apices of veins R2+3 and R4 +5. Fourth costal section about one-sixth as long as third. Last section of vein M 1 +2 one-half longer than penultimate section. Ab­domen: Entirely dark brown to black, subshining. Genitalia as in figure 4d. Length: body and wings, 3.6 mm. FEMALE unknown. Type male and four paratypes, Keanae Valley, Maui, 1500', July 21, 1965, as­sociated with Clermontia kakeana, "C134.2" (H. L. Carson). Type in the B. P. Bishop Museum. Paratypes in the collections of the U. S. National Museum and the University of Hawaii. Drosophila kambysellisi n. sp. (figs. 5a-d) This species is very close to mimica Hardy, occupies the same habitat, and has probably been confused with mimica in the field in some situations. The two are similar in most respects, but kambysellisi is slightly smaller, and the mesonotum, scutellum and front are dark brown to black, rather than reddish brown. D. kambysellisi is readily differentiated by having the anterior reclinate bristles FIG. 5. D. kambysellisi n. sp. a, front leg of male; b, wing; c, mouthparts of male; d, male genitalia. situated opposite the proclinates rather than distinctly in front of the latter; by having the palpi broad, not long and slender; by having long curved spines along the apex of the labellum and with the apex drawn out into a slender process (fig. 5c), compare with figure 139a (Hardy, 1965: 366); also by having the anterior dorsocentrals moderately developed, not rudimentary. This species breeds in the rotting leaves of Pisonia and to date has been col­lected only at Bird Park, Hawaii National Park, Kilauea (Volcano), Hawaii, where it lives in close association with mimica but it has not been recorded to date from other areas where mimica occurs. According to Dr. M. P. Kambysellis and as discussed in his paper currently in preparation, this species has an average of only seven ovarioles per ovary in the female, whereas mimica has fourteen. MALE. Head: Front largely brown to black in ground color, rather densely silvery pubescent, yellow on the anterior and anterolateral margins. Front broad, compared to its length, measured from median ocellus to anterior median margin the front is two times wider than long. Vertex and upper median portion of oc­ciput dark brown to black, grey pollinose. Remainder of head yellow, except for a faint tinge of brown on lower portion of face. The sides of face are convergent basally, and the median portion is slightly raised, the antennal furrows are shal­low. Oral margin with four or five distinct bristles in a row just above genal bristle and approximately equal in size to the genal bristle; the upper vibrissae are repre­sented by short hairs. Frontal bristles are as noted in introduction above. Third antenna! segment yellow-brown, second yellow, brown dorsally. Arista with seven dorsal and three ventral rays in addition to the apical fork and with setae on the apical two-thirds of the inner surface. Mouthparts yellow. Labella rather slender, with strong curved yellow spines at apices. Apical bristle of palpus long, yellow, equal to or slightly longer than the segment (fig. 5c). Thorax: Dark brown to black on dorsum, pale yellow on pleura. Postscutellum and metanotum yellow-brown, covered with grey pollen. Two humeral bristles present. Anterior dorsocentrals about equal in size to upper humeral bristle and situated opposite second pair of supraalar bristles. Legs: Entirely pale yellow. Front tibiae lacking ornamentation, the preapical dorsal bristle moderately developed, extending ap­proximately to apex of segment. Front basitarsus and second tarsomere with numerous moderately long anterior cilia arranged irregularly in two rows (fig. 5a). Wings: Subhyaline, very similar to those of mimica but lacking distinct in­fuscation over r-m crossvein. With a faint brown infuscation over m crossvein, and often with faint indications of brnwn infuscation at apices of veins Rz+3 and R.i+:; (fig. 5b). Costal fringe extend:ng half the distance between R::!+ ~ and R.i+ :;· Abdomen: First tergum and narrow base of second yellow, remainder of abdomen dark brown to black covered with brown pollen. Genitalia are as in figure 5d. Length: body, 2.7 mm.; wings, 2.9 mm. FEMALE unknown. The females have not been associated with the males and are probably being confused with mimica. Holotype male and thirteen male paratypes from Bird Park, Kilauea (Vol­cano), Hawaii, type collected May 21, 1968 (M. P. Kambysellis), paratypes col­lected: same data as type, and June 2, 1968, same collector as type, also July 1963, "G36A", (W. B. Heed) and three specimens collected by L. Throckmorton, no date given. Type and some paratypes in the B. P. Bishop Museum. The remainder of the paratypes in the collections of the U. S. National Museum, British Museum (Natural History) , and the University of Hawaii. This species is named after Dr. M. P. Kambysellis, Harvard University, whose ovarian transplantation and ovarian development studies have made very im­portant contributions to our knowledge of the Hawaiian Drosophila. Drosophila murphyi n. sp. (figs. 6a-c) This species runs to couplet 16 of our key to the picture-winged species (1968: 175), and is very closely related to orphnopeza Hardy and Kaneshiro. It is differ­entiated by having the front basitarsus of the male very densely ciliated on the dorsal surface (fig. 6a), the cilia are arranged in numerous rows covering the en­tire dorsal surface of the basal two tarsomeres. Rather than, as in orphnopeza, with approximately a half dozen pairs of setae or cilia arranged on the dorsal surface of the tibia. Wings: As in figure 6b. The male genitalia are very similar in the two but can be differentiated by the shape of the parameres, c.f. figs. 6c and 13c (Hardy and Kaneshiro, 1968: 205). Otherwise fitting the description of orph­nopeza for both sexes. Fm. 6. D. murphyi n. sp. a, front leg of male; b, wing; c, male genitalia. Type male and allotype female, Pololu Stream, North Kohala, Hawaii, 3300', June 12, 1968, "L79G1 ", (H. L. Carson). Twenty paratypes, 15 males, five fe­males from the following localities on the Island of Hawaii: same as type; Honau­nau Forest Reserve, January 31, 1967 (K. Y. Kaneshiro); Olaa Forest Reserve, September 3, 1965, 3775' (J. K. Fujii and F. Kamiya); Awini Trail, 2200', August 2, 1966 (D. E. Hardy); Awini Camp, August 2, 1966 (J. K. Fujii); and East of Puu Ohu, South Kohala Mountains, 3500', October 26, 1967 (J.P. Murphy). This species has been reared from rotting bark of Cheirodendron sp. Type, allotype and some paratypes in the B. P. Bishop Museum, remainder of paratypes in the collections of the U. S. National Museum, British Museum (Natural History), and the University of Hawaii. This species is named after Mr. John P. Murphy who has been in charge of the laboratory and the field coordinator for the evolution and genetics project since 1965. He has been very active in field work and has made very important contributions to this project. Drosophila ocellata n. sp. (figs. 7a-e) Because of the ornamentation of the front tibiae of the male (presence of a clump of long black cilia at base of segment and numerous cilia extending most of the length) this would appear most closely related to punalua Bryan and re­lated species. Itis readily differentiated from other known species in this complex by the strange development of the male palpi and by the presence of a large, round, dark brown spot on vein M1 +2 beyond them crossvein (fig. 7d), also the ciliation on the front tibiae and tarsi differs as do the thoracic markings and the male genitalia. MALE. Head: Front opaque yellow-brown, more distinctly tinged with yellow in the ground color along the orbits. Remainder of head yellow, except that the upper median portion of the occiput and the vertex are tinged with brown. Face slightly raised down median portion and gently produced on lower margin. Procli- Fm. 7. D. ocellata n. sp. a, front leg of male; b, base of front tibia, dorsal; c, palpus; d, wing; e, male genitalia. nate bristles small, scarsely larger than the post-occular setae on the occiput, and situated very near middle of front, measured from lower ocellus to anteromedian margin. Anterior reclinate bristles tiny, about two-thirds as long as proclinates and situated approximately two-fifths the distance from the proclinate to upper reclinates. Ocellar bristles about equal in size to upper reclinates. One strong bristle developed at upper edge of each vibrissal row, this is .equal in size to the post-ocellar bristles. Genal bristles small, approximately equal in size to the proclinate bristles of the front. Mouthparts yellow. Palpi elongate, terminating in a sharp point which is formed by two or three closely appressed bristles at the apex (fig. 7c). The palpus is very sparsely short setose. Mentum with six or eight moderately long hairs just before apex. Thorax: Yellow in ground color, marked with brown on the pleura and on the dorsum. Mesopleura dark brown on upper two-thirds; sternopleura each with a brown marking on upper median portion and metapleura largely brown. Each humerus with a prominent brown spot in upper median portion and mesonotum with three indistinct broad brown vittae extending most of the length. The mesonotum is rather densely yellow-brown pollinose, partially obscuring the ground color. Scutellum dark brown, except for a yellow spot on each basolateral margin. Anterior dorsocentral bristles situated well behind anterior supraalars. Two strong humeral bristles present. Legs: Yel­low with a tinge of brown before apices of tibiae. Front femora distinctly con­torted, swollen dorsally, flat on the ventral surface. Tibia with a clump of long black, dorsal setae at base and a row of about fourteen anterior cilia extending from base to about apical fourth of segment (fig. 7a) and about four long curved basal bristles on tibia (fig. 7b); no preapical dorsal bristles present. Tarsi not ornate, lacking long cilia. Wings: Lightly fumose with a subbasal brown spot at apex of second costal cell extending to base of vein R4 +5 ; a less distinct brown marking near middle of cell R1 extending through cell R~ into the upper portion of cell R;,; with the apical portions of veins R2+3, R4 +;; and M1-1-2 covered by dark brown spots, these are at least partially convergent in the apical portion of cell R~ . Them crossvein covered by a large brown spot and a large round spot is situated on vein M1+2 beyond m crossvein as in figure 7d. The r-m crossvein is situated at about basal one-third of cell 1st M2 • Abdomen: Dark brown to black covered with brown pollen. Genitalia are as in figure 7 e. The shape of the aedeagus, hypan­drium and the presence of setae on the parameres and inner surface of the hy­pandrium are characteristic of the paucipuncta subgroup (Kaneshiro, same bul­letin). Length: body,4.5-4.7mm.; wings,4.7-5.0mm. FEMALE. Fitting the description of the male in most respects. The proclinate bristles, however, are well developed, equal in size to the post-ocellar bristles. The palpi are normal in development, each has a small black apical bristle. The front legs are normal in development. Holotype male, allotype female and 45 paratypes, 26 males, 19 females, Mt. Kualapa, Kauai, 1400', March 27-28, 1968, "L45" (K. Y. Kaneshiro and H. L. Carson). Three male paratypes, same data as type. This species has been reared from rotting Pisonia sp. ? bark. Type, allotype and some paratypes in the B. P. Bishop Museum. Remainder of the paratypes in the collections of the U.S. National Museum, British Museum (Natural History) and the University of Hawaii. Drosophila ornata n. sp. (figs. 80.-e) This species belongs in the adiastola complex because of the broad, flattened, densely setose front basitarsus of the male, the strong, curved, preapical posterior bristle of front femur, and by the pattern of the wing markings. It differs from other species in this complex by having the wings hyaline with numerous brown to black marks extending across cells R1, R3, R5 and 1st M2 (figs. 8c-d), rather than having the wings brown with rather large hyaline spots as in other species of this complex; by the densely setose front tibia (fig. 8a); by having dense yel­low hair on the apical margin of each labellum of the male (figs. 8b), as well as by genital characters (fig. 8e). Fw. 8. D. ornata n. sp. a, front leg of male; b, mouthparts of male; c, wing of male; d, wing of female; e, male genitalia. MALE. Head: Front opaque golden, tinged with brown, opaque yellow along the orbits. The remainder of head is yellow except for reddish brown eyes with a faint tinge of brown on the vertex and upper median portion of occiput. Face yellow-white, very slightly raised down median portion and with shallow anten­na! furrows. Oral margin lacking distinct bristles, with fine black hairs on upper portion, these are arranged in two irregular rows. Anterior reclinate bristle thin, rather hair-like, scarsely over four-fifths as long as proclinate bristle and situated about two-fifths the distance from proclinate to upper reclinate. Antennae en­tirely yellow, second segment vvith one moderately strong dorsal bristle; this is almost equal in size to the genal bristle, also with two to three shorter dorsal bristles. The arista has eight dorsal and three ventral rays in addition to the apical fork. Occiput rather densely black setose on sides, having two to four irregular rows in addition to the ocular row. Mouthparts entirely yellow. Palpi approxi­mately four times longer than vvide, and with numerous black setae on the apical portion and with a dense clump of yellow setae on the median portion of the inner edge of each palpus. One moderately small apical bristle present, this is approxi­mately half the length of the proclinate bristle of the front. Mentum with pale thin hairs scattered over the apical half. Mouthparts large and conspicuous. Men­tum and labella equal or slightly longer than the height of the head. Mentum with sparse scattered setae almost the entire length (fig. 8b). Labella with densely placed brownish yellow hairs at the apical margins. Thorax: Yellow in ground color, tinged with brown over dorsum and over upper half of each pleuron. With a pair of rather indistinct yellowish vittae extending down mesonotum from an­terior margin in line with dorsocentral bristles and extending almost to the pos­terior dorsocentrals. Scutellum opaque brownish yellow on the disc, yellow on the sides and venter. Postscutellum and metanotum yellow. Anterior dorsocentral bristles situated distinctly behind a line drawn between anterior supraalars. Legs: Yellow except for the dark brown to black apical tarsomeres. Front legs as noted above and as in figure 8a. The preapical posterior bristle of the femur is approximately equal in length to the basitarsus. The tibiae are flattened laterally and densely covered with erect hairs over the entire posterior surface. Also, the posterior surface of the basitarsus is densely haired. The preapical dorsal bristle of front tibia is approximately equal in length to the hairs on the posterior surface. Wings: Predominantly subhyaline with numerous dark brown to black marks across the cells as noted a hove (fig. 8c), also with a brown to dark brown streak ex­tending from apical portion of second costal cell over the r-m crossvein into the upper portion of base of cell 1st M 2 and with extreme basal portion of the wing tinged yellow-brown. Also the cells in the posterior portion of wing are light fumose. The r-m croswein is situated at the basal one-fifth of cell 1st Mz. Ab­domen: Opaque brown, tinged with yellow on the first tergum and on basal por­tion of second. Genitalia are as in figure 8e. The shape of the aedeagus would definitely place this specjes in the adiastola subgroup. The parameres are cylindri­cal with elongate apical sensilla. Length: body and wings, 6.25-6.5 mm. FEMALE. Similar to the male except for differences in wing markings as in figures 8c and 8d, and for secondary sexual characters. Also one prominent oral bristle is developed, equal in size to the proclinate bristle of the front. Ovipositor rather short, the blades slender. Holotype male, Kanaele Swamp, Kauai, 2500', March 19-20, 1968, "L41" (H. L. Carson). Allotype female, Pouli Stream, Hanalei Distr., Kauai, March 13, 1968, "L37G17" (H. L. Carson). Fifteen paratypes: same as type and allotype. Type and some paratypes in the B. P. Bishop Museum. The remainder of the paratypes in the collections of the U. S. National Museum, British Museum (Natural History), and the University of Hawaii. Drosophila paenehamifera n. sp. (figs. 9a-h) Fitting in the varipennis complex which is differentiated by the presence in the male of a large preapical dorsal hook on the front tibia, by the highly modified mouthparts (figs. 9b-c), by the strongly arched costal margin (fig. 9f), the pe­culiar spatulate aedeagus (fig. 9h), as well as by other characteristics. Refer to description and figures of hamifera Hardy and Kaneshiro (1968: 254, fig::;. 35a-h). D. paenehamifera is differentiated by having only four or five anterodors­al cilia on the front tibia just beyond the base of the segment and one or two ante­rodorsal cilia near the hook (fig. 9a), rather than having a continuous row of ante­rodorsal cilia from base to the hook. The costal margin is not so highly arched in paenehamifera as in hamifera (fig. 9f compare with fig. 35e, Zoe. cit.). The wing markings are distinctly different in the two. In paenehamifera the wings are pre­dominantly yellowish to brown. The r-m crossvein is situated beyond the middle of cell 1st M2 rather than slightly before the middle as in ha.mi/era. The hyaline mark in the middle of the wing from the costa through cell M2 is much narrower, not so well developed in paenehamifera not extending to base of cell R1, also the apical half of the wing lacks hyaline markings. In hamifera a prominent hyaline Fm. 9. D. paenehamifera n. sp. a, front leg of male; b, mouthparts of male, lateral; c, mouth· parts, ventral; d, male palpus; e, female palpus; f, male wing; g, female wing; h, male genitalia. band extends transversely from apex of cell R1 through most of cell 2nd M2 and the extreme apex of the wing, through cells R3 and R5, is hyaline (refer to figure 35e, Zoe. cit.). In other respects, fitting the description of hamifera, the mouthparts are as in figures 9b-d, the palpi are enlarged, clavate; the wings are as in figure 9f, and the genitalia are as in figure 9h. FEMALE. Fitting the description of hamifera except that the r-m crossvein is situated at the middle of cell 1st M2 rather than slightly before the middle, the costa is not arched and the markings are slightly different (fig. 9g), also the palpus is more elongate and slender (fig. 9e). Holotype male, Hanaula, West Maui, 4000', May 7-8, 1968, "L61" (H. L. Carson). Allotype female, same data, "L61B12" (K. Y. Kaneshiro). Seven para­types, four males, three females, same data as type and allotype (collected by H. L. Carson, K. Y. Kaneshiro, andJ. P. Murphy). Type and allotype in the B. P. Bishop Museum. Paratypes in the collections of the U. S. National Museum, British Museum (Natural History), and the Uni­versity of Hawaii. Drosophila turbata n. sp. (figs. 10a-c) This species fits in the hawaiiensis complex by having a transverse band across the wing at level with them crossvein (fig. 10c), it is closely related to recticilia Hardy and Kaneshiro from Maui and is the species keyed in couplet 44 ( 1968: 178) with recticilia as "n. sp. or pale specimens of above" from Oahu. After exam­ining further specimens from Oahu, it is obvious that these are distinct species. D. turbata specimens are smaller than recticila; the mouthparts and legs are entirely yellow except for brown rim on labella and a tinge of brown on the sub-apical and the subbasal portions of the hind tibia. In recticilia, the palpi and labella, also the front coxae and femora are brown. The ornamentation of the front legs are very similar in these species, the females have previously been con­fused with gradata Hardy and Kaneshiro. The two species are very similar in coloration and wing markings. The males are readily differentiated by having the fro.nt tibia ciliated over the entire length, and by the cilia on the front basitarsus being shorter, all approx;mately equal in length (fig. 1 Oa) ; as well as by genital characters. MALE. Head: Front opaque brown to black, tinged with rufous in the ground color of the lower half and with a narrow streak of yellow extending from median ocellus over middle part of the face ending opposite or just slightly beyond procli­nate bristles. Eye orbits yellow to rufous in ground color, densely yellow-grey pollinose. Vertex and upper median occiput black in ground color, covered with grey pollen. Remainder of the head yellow-white, except for reddish brown com­pound eyes. Second and third antennal segments largely brown to black, with yellow on apex and ventral portion of second. Arista rather short, scarsely one­third longer than combined length of antenna, with six dorsal and three ventral rays in addition to apical fork and with inner surface setose almost to base. Palpi and mouthparts yellow except for a narrow rim of ·brown on each labellum. In recticilia the rim is broadly and conspicuously blackened. Palpus with two hair­like apical bristles, the longest is approximately equal to upper bristle of vibrissal row and the shorter of the two equal to secondary bristle of vibrissal row. Mentum yellow-brown, setose only on apical portion. Anterior reclinate bristle two-thirds to three-fourths as long as the proclinate and situated distinctly above the latter. Thorax: Largely yellow in ground color, covered with yellow-grey pollen, with a pair of dark brown to black submedian vittae extending the entire length of seg­ment, with a prominent black spot on each side just in front of suture and a brown to black vitta extending from behind suture to a level with the posterior dorso­central bristle. The two brown vittae are separated on each side by a very narrow, somewhat indistinct, yellow line extending down dorsocentral row. The lateral margins of mesonotum are broadly yellow. Each humerus is yellow except for a small brown mark on upper hind margin. Scutellum pale yellow. Sternopleura largely brown to black bordered with yellow and with a dark brown to black mark in upper portion, a small spot on pteropleuron and with a faint tinge of brown on lower portion of each metapleuron. The metanotum is polished dark brown to black, densely covered with grey pollen, also the postscutellum is yellow­brown. Two well developed humeral bristles present. Anterior dorsocentral bris­tles almost in line with second pair of anterior supraalars. Halteres, pale yellow. Legs: Yellow except for preapical and prebasal tinges of brown on hind tibiae. Front tibia with two rows of moderately long anterodorsal cilia extending the whole length of segment (fig. 10a) and preapical dorsal bristle rather small, scarsely differentiated from surrounding setae. Front basitarsus with fourteen to sixteen long cilia arranged in pairs down anterodorsal surface, the entire length of the tarsomere. Second tarsomere with two anterodorsal cilia. Wings: The sub­basal spot through apex of second costal cell extends indistinctly to r-m crossvein, but does not enclose the crossvein. The transverse band across middle of wing at level with m crossvein is broad in the middle of cell R1, becoming narrower in The University of Texas Publication cells R3 and R5 (fig. 1 Oc). The brown mark in apex of wing is continuous through the cells, not interrupted. The r-m crossvein situated near basal one-fourth of cell 1st M 2 • The costal fringe extends approximately two-thirds the distance between the apices of veins R2 +3 and RH5 • In the discussion of recticilia (Zoe. cit.) it was stated in the key that in this species the costal fringe extends one-half the distance between apices of vein R.i+r. and M] +2 • This was an error, it should have read between R2+s and R-i+ ;;, and it is so stated under the description of that species. After examining further specimens we find this character to be variable and ap­parently of no value. Abdomen: Dark brown to black down median portion and on posterior and lateral margins of terga. leaving a large basal lateral yellow spot on each side of each terga. These areas are yellowish pollinose. The brown areas are brown pollinose. The genitalia are as in figure 1 Ob. The narrow and elongate aedeagus with an insignificant preapical protuberance would definitely place this species in the hawaiiensis subgroup. The parameres are elongate with minute apical sensilla. Length: body and wings, 3.5 mm. FEMALE. Fitting the description of the male except for secondary sexual char­acters. The palpus has only one apical bristle, this is approximately equal in size to the second bristle of the oral margin. Holotype male, Peacock Flat, Oahu, 1400', May 23, 1968, "L67" (H. L. Car­son). Allotype female, Opaeula, Oahu, July 1964, "HN91" (W. B. Heed). Nine paratypes, eight males, one female, from the following localities on Oahu: Trail to-Kaau Crater, 1300', May 20, 1968, "L65" (M. P. Kambysellis); Mt. Kaala, June 11, 1950 (No collector given-probably Gordon B. Mainland); Manoa Valley (no date given) (G. B. Mainland); and Haleanau Valley, Mt. Kaala, April 21, 1949, banana trap (G. B. Mainland). Type and some paratypes in the B. P. Bishop Museum. The remainder of the paratypes in the collections of the U. S. National Museum, British Museum (Natural History) and the University of Hawaii. REFERENCES CITED Hardy, D. E. 1965. Insects of Hawaii, Vol. 12, Diptera: Cyclorrhapha II, Series Schizophora, Section Acalypterae I, Family Drosophilidae, 814 pp. University of Hawaii Press. Hardy, D. E. and K. Y. Kaneshiro. 1968. New picture-winged Drosophila from Hawaii. Univ. Texas Publ. Studies in Genetics 6818: 171-262. V. A Study of the Relationships of Hawaiian Drosophila Species Based on External Male Genitalia 1 ' 2 KENNETH Y. KANESHIR03 INTRODUCTION Sturtevant (1919) first mentioned the importance of the use of the external male genitalia of Drosophilidae as a taxonomic tool in distinguishing between closely related species. Since then, investigators such as Hsu (1949), Malogolow­kin (1952, 1953), Nater (1953), Okada (1953, 1954, 1955), Spassky (1957), and Takada ( 1965, 1966, have made extensive studies of the external male genitalia of Drosophilidae. Snodgrass (1957) states: "The great diversity in structural detail of the geni­talia gives these organs a value for the identification of insect species almost equal to that of fingerprints for identification of human individuals. On the other hand, the very structural diversity of the organs makes it difficult to understand their fundamental nature and the homologies of their parts." The lack of uniformity in the concept of homologies is certainly true in the case of the male genitalia of Drosophilidae. This has led to a confusing situation where a taxonomic specialist of one group of species adopts terminology which is un­familiar to another specialist. Diagrammatic sketches (figs. 1.1 and 1.2) are presented herein to clarify any misinterpretation which may arise in reading this paper. The terminology used is based on Takada's (1966) study of the external male genitalia of Hawaiian Drosophilidae. Carson et al. (1967) presented a diagram giving the relationships of 22 species of endemic Drosophila of Hawaii. This diagram is in the form of a phylogenetic tree and is based on the relationships of the polytene chromosome banding se­quences of the 22 species with the sequences of D. grimshawi as the arbitrary standard. Carson and Stalker ( 1968a, b, c), have since expanded the phylogeny to include over 50 species of Hawaiian Drosophila. In most cases, chromosomally similar species are morphologically similar and vice versa. Carson's findings, how­ever, also show that there is sometimes remarkable chromosomal similarity be­tween species which show pronounced morphological differences; i.e., obvious differences in leg ornamentation, wing pattern, etc. This study is an attempt to form species subgroups based on relationships of external male genitalia, especially the phallic organs. In addition, an attempt will be made to establish a correlation between the relationships of species based on 1 Published with the approval of the Director of the Hawaii Agricultural Experiment Station as Journal Series No. 1101. 2 This work was supported by the National Institute of Health Grant No. GM10650-05 to Dr. D. E. Hardy, and is a portion of the thesis submitted to the University of Hawaii in partial ful­ fillment of the requirements for the degree of Master of Science. 3 Department of Entomology, University of Hawaii, Honolulu. STUDIES IN GENETICS V. Univ. Texas Puhl. 6918, Sept., 1969. . 2 FIG. 1. Diagrammatic sketches of external male genitalia; .1 lateral view, .2 ventral view of hypandrium. A. aedeagus, A.G. anterior gonapophysis (paramere), A.S. apical sensillum, B.A. basal apodeme of aedeagus, C. clasper, H. hypandrium, P.P. preapical protuberance of aedeagus, P.S. paramedian spine, 9T. ninth tergum (genital arch), 10T. tenth tergum (anal plate). male genitalia and relationships based on the banding sequences of the chromo­somes. Therefore, species with large morphological differences, but with similar chromosomal makeup, will be shown to share similar genital apparatus. Also, species which may appear to belong to the same species subgroup based on ex­ternal morphology, may in actuality belong to a separate subgroup based on genital characteristics. This is corroborated in most cases by the chromosomal relationships of the species. All of the information on the chromosomal makeup of the species studied is taken from Carson's reports; however, not all of the species whose genitalia have been studied, have been studied by Carson. Based on genitalia relationships, it is possible to make some predictions as to the approximate placement of these species in Carson's phylogeny. Most investigators have used genitalia characteristics to separate closely re­lated species; i.e., species which cannot be easily separated on the basis of ex­ternal morphological characteristics. In this paper, however, emphasis will be placed on the usefulness of the similarities of the genital apparatus for assigning the different species to particular subgroups. All of the species studied here, are considered to be distinct species based on consistent morphological characteristics; and in most cases, a study of the genital apparatus is not needed to distinguish be­tween species. In fact, it was found that in many cases, there is a slight intra­specific variability in the shapes of the phallic organs and that the aedeagus of some specimens of a particular species show a strong tendency to resemble those of another species. This strongly supports the theories which will be presented later. It will be shown that despite the slight intraspecific variability which may be present in the shapes of the phallic organs, they are nevertheless consistent characteristics of the respective species subgroups. MATERIALS AND METHODS Most of the genitalia studied were dissected from freshly killed specimens which were either collected in the field or taken from cultures reared at the Evo­lution of Hawaiian Drosophila Laboratory at the University of Hawaii. Some specimens were obtained from the general collection of Drosophilidae at the Uni­versity of Hawaii. The species studied belong to the so-called "picture-winged" species (Hardy and Kaneshiro, 1968) of Drosophilidae. They are generally large species with markings at least on the base of the wing and the m-crossvein. They usually have ciliation on the front legs, and usually also have fleshy labella (i.e. , not modified with spines or setae) except for D. neogrimshawi in the D. adiastola subgroup. The species are listed in Table 1 according to species subgroups. The numbers correspond to the drawings of the respective genitalia in figures 2 through 9. 0 .4mm FIG. 2. Phallic organs of D. adiastola subgroup; .1 D. adiastola, .2 D. cilifera, .3 D. peniculi­pedis, .4 D. spectabilis, .5 D. setosimentum, .6 D. ochrobasis, .7 D. clavisetae, 8. D. neogrimshawi. .7 .8 .5 .6 0.4rnrn Frn. 3. Phallic organs of D. paucipuncta subgroup; .1 D. paucipuncta, .2 D. uniseriata, .3 D. prolaticilia, .4 D. basiseta.e. Ventral view of hypandrium of D. paucipuncta subgroup; .5 D. pauci­puncta, .6 D. uniseriata, .7 D. prolaticilia, .8 D. basisetae. It was necessary to mount the genitalia of each specimen on a slide in order to study them in detail. The following is the procedure used in preparing the genitalic materials. It is a modification of the procedure used by Kambysellis and Wheeler (1966). ( 1) The tip of the abdomen of the fly was clipped with a pair of microscissors, then placed in a test tube with 10% KOH, and boiled for 10-to 15­minutes. (2) The KOH was then removed with a pipette and the material washed with water. (3) The water was drained and replaced with a small amount of stain [four (4) parts of Gage's Stain (acid fushin, 0.5g; 10% HCl, 25.0cc; distilled water, 300.0cc) with one ( 1) part glacial acetic acid]. ( 4) The material immersed in the stain was then heated until boiling then the stain was replaced with 95 % ethyl alcohol. ( 5) The material was transferred to a small watch glass containing a drop or two of creosote. (6) At this point, extraneous material was removed from the genitalia using a pair of fine probing needles. (7) The cleaned material was then mounted on a slide using euparol as the medium. For most of the species studied, several specimens each were examined to study the variability which may exist in the structure of the genitalia within a species. For some of the species, only a very few or even only one specimen was available. Although the phallic organs of some specimens of a particular species had a strong FIG. 4. Phallic organs of D. pilimana subgroup; .1 D. pilimana, .2 D. glabriapex, .3 D. dis­creta, .4 D. fasciculisetae, .5 D. lineosetae. tendency to resemble those of another species, it was found that for this study, the intraspecific variability of the shape of the phallic organs was of little im­portance. OBSERVATIONS After making a detailed study of the external male genitalia of several "picture­winged" species of Hawaiian Drosophila, it became evident that there are char­acteristic similarities in the phallic organs among the species which indicated group relationships. The general shape of the aedeagus (especially the shape of the preapical protuberance), and the shape of the hypandrium and the ninth tergum (at least in the D. paucipuncta subgroup) are of special interest. Based on these characteristics, it was possible to set up at least eight species subgroups. Except for the D. paucipuncta subgroup, the overall shapes of the hypandrium, the anal plate, and ninth tergum do not show striking characters which are of importance here. The shape of the anterior gonapophyses (parameres) seem to be a very useful character in distinguishing between two species which otherwise have very similar genital characters. The anterior gonapophyses are not im­portant, however, for indication of any significant tendency in the formation of the species subgroups. In several species, there are minute setae in addition to the apical sensilla present on the surface of the anterior gonapophyses but these species do not appear to be very closely related (except D. paucipuncta and D. uniseriata in the D. paucipuncta subgroup; and D. digressa and D. virgulata in the D. vesciseta subgroup) as will be pointed out later. In most of the species studied, there is a definite protuberance close to the apex of the aedeagus when viewed from the lateral aspect. Figures of the aedeagus of species such as D. pilimana (fig. 4.1), D. grimshawi (fig. 6.1 ), and D. orphnopeza (fig. 9.1) illustrate this characteristic. The shape and size of this protu·berance are .2 0.4mm FIG. 5. Phallic organs of D. vesciseta subgroup; .1 D. vesciseta, .z D. hexachaetae, .3 D. virgu­lata, .4 D. digressa. very useful characteristics in grouping species which later were shown to have similar chromosomal makeup. The eight species subgroups which include a total of 41 species will be identified by the name of the species most representative of each group as for example, the "D. adiastola" subgroup, and the "D. hawaiiensis" subgroup. The D. adiastola Subgroup This subgroup is comprised of eight species (Table 1). The characteristic fea­tures of the D. adiastola subgroup occur in the phallic organs. All species belong­ing in this subgroup have a characteristic bend at a:bout the middle of the aedeagus (fig. 2) ; also, the preapical protuberance of the aedeagus is small and insignificant as compared to species such as D. pilimana (fig. 4.1), and D. grimshawi (fig. 6.1). The basal apodeme of the aedeagus is relatively narrow and elongate. The pa­rameres of this subgroup are typically narrow and elongate and with an elongate apical sensillum. The other structures of the external genitalia are quite variable between species and may resemble those species from another subgroup. The D. paucipuncta Subgroup This subgroup is comprised of four species (Table 1). In this subgroup, the shapes of all the structures of the external male genitalia are very consistent. The aedeagus is relatively short, with a pronounced preapical proturberance and a broad but shortened basal apodeme, (fig. 3). The shape of the hypandrium is also characteristic in that the width at the widest points is about equal to or, as in D. basisetae (fig. 3.8) , slightly greater than the length. In other picture-winged species, the length of the hypandrium is generally, at least one-half times greater than the width (fig. 1.1). In two species, D. paucipuncta (fig. 3.1), and D. uni­seriata (fig. 3.2), the median surface of the hypandrium just ventrad to the para­median spine is covered with minute setae. Also, in these two species, the apical TABLE 1 Species Studied and Their Assignment to Different Subgroups Fig. No. 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 3.1 3.2 3.3 3.4 4.1 4.2 4.3 4.4 4.5 5.1 5.2 5.3 5.4 6.1 6.2 6.3 6.4 6.5 7.1 7.2 7.3 7.4 7.5 7.6 8.1 8.2 9.1 9.2 9.3 9.4 9.5 9.6 9.7 D. adiastola Subgroup D. adiastola Hardy D. cilifera Hardy-Kaneshiro D. peniculipedis Hardy D. spectabilis Hardy D. setosimentum Hardy-Kaneshiro D. ochrobasis Hardy-Kaneshiro D. clavisetae Hardy D. neogrimshawi Hardy-Kaneshiro D. paucipunta Subgroup D. paucipuncta Grimshaw D. uniseriata Hardy-Kaneshiro D. prolaticilia Hardy D. basisetae Hardy-Kaneshiro D. pilimana Subgroup D. pilimana Grimshaw D. glabriapex Hardy-Kaneshiro D. discreta Hardy-Kaneshiro D. fasciculisetae Hardy D. lineosetae Hardy-Kaneshiro D. vesciseta Subgroup D. vesciseta Hardy-Kaneshiro D. hexachaetae Hardy D. virgulata Hardy-Kaneshiro D. digressa Hardy-Kaneshiro D. grimshawi Subgroup D. grimshawi Oldenberg D. disjuncta Hardy D. bostrrcha Hardy D. crucigera Grimshaw D. balioptera Hardy D. hawaiiensis Subgroup D. hawaiiensis Grimshaw D. recticilia Hardy-Kaneshiro D. silvarentis Hardy-Kaneshiro D. gradata Hardy-Kaneshiro D. villitibia Hardy D. hirtipalpus Hardy-Kaneshiro D. ochracea Subgroup D. ochracea Grimshaw D. limitata Hardy-Kaneshiro D. orphnopeza Subgroup D. orphnopeza Hardy-Kaneshiro D. sodomae Hardy-Kaneshiro D. orthofascia Hardy-Kaneshiro D. prostopalpis Hardy-Kaneshiro D. engyochracea Hardy D. sproa.ti Hardy-Kaneshiro D. villiosipedis Hardy 0 . 4•• F1G. 6. Phallic organs of D. grimshawi subgroup; .1 D. grimshawi, .Z D. disiuncta, .3 D. bos­trycha, .4 D. crucigera, .5 D. balioptera. two-thirds of the parameres are covered with minute setae in addition to the min­ute apical sensilla (fig. 3.5 and 3.6). The other two species in this subgroup, D. prolaticilia (fig. 3.3), and D. basisetae (fig. 3.4) do not have setae on the median surface of the hypandrium (figs. 3.7 and 3.8) and the parameres, but the shape of the aedeagus and the hypandrium would definitely place them in this sub­group. The shape of the ninth tergum (or genital arch) of this subgroup also appears to be quite characteristic. It is narrow on the dorsal portion, gradually widening at the ventral margins. The D. pilimana Subgroup This subgroup is comprised of five species (Table 1). The phallic organs of this subgroup are characterized by the pronounced preapical protuberance (fig. 4), which is quite different from those of the D. paucipuncta subgroup. Also, the overall length of the aedeagus is longer than those of the D. paucipuncta subgroup relative to the basal apodeme. The parameres of this subgroup are typically nar­row and elongate but in contrast to the D. adiastola subgroup, the apex is blunt and rounded rather than pointed. Also, the apical sensilla are minute rather than elongate. The D. vesciseta Subgroup This subgroup is comprised of four species (Table 1). The characteristic feature of this subgroup is the very narrow and elongate aedeagus with a small and in­significant preapical protuberance (fig. 5). Also, the parameres are broad and .I .3 0.4111'111 Frn. 7. Phallic organs of D. hawaiiensis subgroup; .1 D. hawaiiensis, .2 D. recticilia, .3 D. sil­varentis, .4 D. gradata, .5 D. villitibia, .6 D. hirtipalpis. have minute apical sensilla except those of D. digressa (fig. 5.4), which have elongate apical sensilla. The parameres of D. virgulata (fig. 5.3), and D. digressa (fig. 5.4) are covered with minute setae in addition to the apical sensilla while those of D. vesdseta (fig. 5.1) and D. hexachaetae (fig. 5.2) are bare except for the apical sensilla. All four species belong to the same subgroup on account of the shape of the aedeagus. The D. grimshawi Subgroup This subgroup is comprised of five species (Table 1). The phallic organs of these five species are characterized by the aedeagus having a significant preapical protuberance (fig. 6) which is quite similar yet readily distinguishable from those of the D. pilimana subgroup. The preapical protuberance is narrower at the base and more rounded at the apex rather than wide at the base and angular at the apex as in the D. pilimana subgroup. The basal apodeme is short and quite similar in shape to the D. pilimana subgroup in that it is truncate at the apex. The para­meres are also quite similar to those of the D. pilimana subgroup although they appear to be somewhat more rounded at the apex. The dorsal surfaces of the para­meres of D. balioptera (fig. 6.5) are covered with minute setae in addition to the The University of Texas Publication FIG. 8. Phallic organs of D. ochracea subgroup; .1 ochracea, .2 D. limitata. apical sensilla while those of the remaining four species are bare except for the minute apical sensilla. The D. hawaiiensis Subgroup This subgroup is comprised of six species (Table 1). The species in this sub­group are characterized by the absence or very small preapical protuberance of the aedeagus (fig. 7). This character alone will readily separate this subgroup from the others. In D. gradata there appears to be a depression at the position where the protuberance normally occurs (fig. 7.4). The parameres of this sub­group are typically broad at the base and narrowing at the apex and with minute apical sensilla. The basal apodeme is typically triangular in shape and is very short in relation to the length of the aedeagus. The D. ochracea Subgroup This subgroup is comprised of only two species (Table 1). These two species appear to be closely related due to the overall shape of the aedeagus (fig. 8). Spe­cifically, there is a bend at about the middle of the aedeagus plus another bend at the junction of the basal apodeme. Also, the swelling ventrad to the preapical protuberance of the aedeagus is characteristic of the two species. The parameres of these two species, however, are different from each other. In D. limitata (fig. 8.2), they are elongate and narrow at the apex; in D. ochracea (fig. 8.1), short and quite rounded at the apex. The parameres of both species have elongate apical sensilla but those of D. ochracea are on short projections which are somewhat subapical. The D. orphnopeza Subgroup This subgroup is comprised of seven species (Table 1). In the previous sub­groups, there were several characteristics which were used to group the species Fm. 9. Phallic organs of D. orphnopeza. subgroup; .1 D. orphnopeza., .z D. sodomae, .3 D. engyochracea, .4 D. sproati, .5 D. orthofascia, .6 D. prostopalpis, .7 D. villosipedis. into their respective subgroups. This was not the case for the D. orphnopeza sub­group. There is one character which is shared by all of them; i.e., the preapical protuberance is slanted toward the base, each species to a different degree (fig. 9). A distinct similarity in the shape of the aedeagus can be seen in D. engyochracea (fig. 9.3), and D. sporati (fig. 9.4); D. orphnopeza (fig. 9.1) and D. sodomae (fig. 9.2); and D. orthofascia (fig. 9.5), and D. prostopalpis (fig. 9.6). It would appear that there are three separate subgroups involved; however, due to the intraspecific variability of the shape of the aedeagus of D. orthofascia (of which several specimens were available), a definite relationship between the species of this subgroup can be seen. Although the genitalia of D. prostopalpis and D. sodo­mae were studied from single specimens, the very fact that they resemble D. orthofascia and D. orphnopeza respectively, would be enough evidence to include them in this subgroup. Undoubtedly, if more specimens were available, a more concrete relationship between the species of this subgroup could be made. D1scussrnN In the D. adiastola subgroup, there are two (2) species D. cLavisetae and D. neogrimshawi which were until recently considered to belong to a different genus, Idiomyia. Hardy (1965) stated that the only reliable character which will sep­arate I diomyia from Drosophila is the presence of an extra crossvein in cell R5 of the wing. lnt~restingly, however, Carson et al. ( 1967) presented chromosomal evidence which indicate that I diomyia is cogeneric with Drosophila. Consequent­ly, ldiomyia has been sunk as a synonym of Drosophila by Hardy and Kaneshiro (1968). During this sh1dy, it was found that these two species definitely belong in the D. adiastola subgroup on the basis of the male genitalia. This is corroborated by Carson's chromosomal relationship. It should be restated here that the possi­bility that morphologically (external) dissimilar species actually being closely rela~e::l, i.e., if based on male genitalic characters is evidenced strongly by the D. adiasto!a rncgroup nn::l i.:; corrob::r:.·a~~:.~ by Car.::on's chromo::c!!1al r2lation­ships. In the D. paucipuncta subgroup, the three species D. paucipuncta, D. uni­seriata, and D. basisetae could very easily be considered to belong to the same subgroup on the basis of external morphology. The male genitalia shows that they are closely related species; and Carson's chromosomal study shows that D. paucipuncta and D. uniseriata are chromosomally homosequential. However, D. prolaticilia would probably have been placed in a separate subgroup from these three on the basis of the wing markings and general body shape. The male geni­talia show that D. prolaticilia is nonetheless closely related to the D. paucipuncta subgroup. Chromosomally, it has only one fixed inversion difference from D. paucipuncta and D. uniseriata. Here again is good evidence that species which can be shown to be closely related species on the basis of male external genitalia, especially the phallic organs, despite other morphological differences, can also be shown to be chromosomally closely related. Carson et al. ( 1967) showed that D. punalua is chromosomally homosequential with D. paucipuncta and D. uniseriata of the D. paucipuncta subgroup. The shape of the aedeagus of D. punalua appear to fit the characteristics of the D. pauci­puncta subgroup; however, the shape of the hypandrium and ninth tergum, which are consistent characteristics of the D. paucipuncta subgroup, are very different in D. punalua. The hypandrium is at least one-half times longer than wide in D. punalua, and the ninth tergum is of uniform length; i.e., not narrow­ing at the dorsal portion. Therefore, although chromosomally D. punalua is homo­sequential with D. paucipuncta and D. uniseriata and although some genitalic characteristics indicate a possible relationship of D. punalua to others of the D. paucipuncta subgroup, because of the difference in the hypandrium and the ninth tergum. D. punalua will not be included in this subgroup. This does not necessari­ly contradict Carson's chromosomal evidence because it has been shown that speciation and evolution of the picture-winged species of Hawaiian Drosophila may in a number of cases be based on mutational changes occurring at the sub­microscopic level rather than on changes in the sequences of the chromosomal bandings (Carson et al. 1967). Therefore, based on genitalic characteristics, two species \vhich are chromosomally homosequential may be shown to belong to two separate subgroups; i.e., each is more closely related to another species with which it may not be chromosomally homosequential but to which it is similar in geni­talic characteristics. Similar cases will be presented to emphasize this situation. In the D. pilimana subgroup, D. lineosetae has not yet been studied by Carson; but on the basis of the shape of the aedeagus, it is predicted that chromosomally it will certainly belong to this subgroup. Chromosomally, D. pilimana and D. glabriapex are homosequential (Carson and Stalker, 1968). The preapical pro­tuberance of the aedeagus of D. pilimana is somewhat variable. In some speci­mens, the aedeagus tends to resemble and in some cases is indistinguishable from that of D. glabriapex. Morphologically, these two species are also very similar but are distinct species, based on consistent morphological (external) character­istics. In the D. vesciseta subgroup, there are two species, D. vesciseta and D. hexa­chaetae, which are small; and the other two, D. virgulata and D. digressa~ which are at least twice as large. It is difficult to visualize that these four species belong to the same subgroup. Not only are there extreme size differences between the small and large species, but there are also distinct differences in the leg orna­mentation and wing markings. The phallic organs, nevertheless, show that these four species definitely belong to a single subgroup. Carson's chromosomal study shows that D. hexachaetae is indeed clo:::ely related to D. virgulata. Also, D. vesci­seta is shown to differ from D. virgulata by only one chromosomal inversion and from D. hexachaetae by only two chromosomal inversions. However, he has also found that D. vesciseta is chromosomally homosequential with D. pilimana and D. glabriapex of the D. pilimana subgroup. This does not contradict the genitalic evidence which places D. vesciseta in a separate subgroup from that of D. pili­m~.ma 1md D. glabriapex, because it has been shown that speciation of picture­wing2d species of Hawaiian Drosophila is not always accompanied by changes in the bandjng arrangements of the polytene chromosomes. In a number of cases, speciation has occurred due to mutational changes on the submicroscopic level of the chromosomes without alteration of the banding sequences. Therefore, it can be shown that chromosomally homosequential species may belong to separate subgroups on the basis of genitalic evidence. The shape of the aedeJ.gus of D. vesciseta shows that it is more closely related to species of the D. vesciseta sub­group rather than those of D. pilimana subgroup. Therefore, in this particular situation, the genitalia evidence provide supplementary information on the rela­tionships of these species. In the D. pilimana subgroup, close examination of the intraspecific variability of the shape of the preapical protuberance of the aedeagus and the overall shape of the aedeagus show that there is a slight resemblance to those of the D. grim­shawi subgroup. Interestingly, Carson's chromosomal study shows that there is only one chromosomal inversion difference between D. pilimana and D. grim­shawi. There appears to be a definite relationship between the two subgroups; however, the general shape of the preapical protuberance of the aedeagus will readily differentia~e between the two subgroups. Specifically, in the D. pilimana subgroup, the preapical protuberance is in general, wider at the base and typically more angular rather than rounded at the apex. The fact that there is a tendency for the shape of the aedeagus in some specimens of the D. pilimana subgroup to resemble those of the D. grimshawi subgroup but not vice versa, gives reason to believe that these two subgroups are indeed closely related subgroups but are probably separate subgroups. In the D. grimshawi subgroup, D. grimshawi, D. bostrycha, and D. disiuncta are chromosomally homosequential species. However, D. orphnopeza. and D. villosipedis of the D. orphnopeza subgroup are also homosequential with these three species. The external male genitalia show that the latter two species defi­nitely belong to a separate subgroup from the first three. Also, chromosomally, D. balioptera shares a common inversion with D. orthofascia and D. engyochracea which would make D. balioptera appear to belong to the D. orphnopeza subgroup. However, the male genitalia of D. balioptera would definitely place this species in the D. grimshawi subgroup; and chromosomally, D. balioptera differs from D. grimshawi by only one inversion. Both of these situations do not contradict the chromosomal evidence in the evolution of these species. Rather, the genitalia evi­dence certainly appears to supplement the chromosomal evidence and perhaps provides a better picture of the relationships between these species. It was stated earlier that in some species, there are setae on the parameres. The species with minute setae on the parameres are D. paucipuncta and D. uniseriata of the D. paucipuncta subgroup; D. virgulata and D. digressa of the D. vesciseta subgroup; and D. balioptera of the D. grimshawi subgroup. These five species in their respective subgroups do not, however, share any other similarities in their phallic organs and therefore are not considered to be related to each other on the basis of the setae on the parameres. White (1968) states that there are only two "sure" exceptions to the generali­zation, that as far as the higher animals are concerned, even the most closely re­lated species are usually different in chromosomal karyotypes. These two excep­tions occur in certain complexes of Drosophila. Wasserman ( 1962) presented evidence that D. mulleri, D. aldrichi, and D. wheeleri of the D. mulleri complex do not differ at all in the banding sequences of their polytene chromosomes. Car­son et al. ( 1967) found that this same situation occurs in several species com­plexes of Hawaiian Drosophila. All of these species which have been found to be chromosomally homosequential, have been described as distinct species on the basis of consistent morphological characteristics, and in most cases, the species concept has been verified by behavioral studies and hybridization experiments. Carson has shown that speciation in the picture-winged species of Hawaiian Drosophila has resulted in pronounced morphological diversity but with remark­able stability of the chromosomal banding arrangements. The phallic organs of these species are also very "stable" characteristics of the species subgroups. Species which may be morphologically (external) very different may be shown to belong to the same subgroup based on their genitalic characteristics. In cases where chromosomally homosequential species occur, the genitalic characteristics sup­plement the chromosomal evidence in the relationships of the species involved. This situation is illustrated in the case where D. vesciseta is shown to be chromo­somally homosequential with D. pilimana and D. glabriapex of the D. pilimana subgroup but is found to be more closely related to D. virgulata and D. hexachae­tae on the basis of genitalic characteristics. CONCLUSION In this study it was found that grouping the picture-winged species of Hawaiian Drosophila on the basis of external male genitalia closely resemble the groupings based on chromosomal evidence. Thus, in cases where the chromosomal banding arrangements of a particular species cannot be studied due to difficulty in rearing in the laboratory, a careful study of the phallic organs of a field-collected adult male, would help to determine its relationship to other species. There have been several instances where formerly only a single male specimen represented the whole of the collection of a particular picture-winged species and its relationship to a particular group of species was determined on the basis of its genital ap­paratus. Later, when females of this same species were collected and when chro­mosomal evidence became available, its relationship to the same group of species (as had been determined on the basis of genitalic characteristics) was verified. Therefore, the study of the external male genitalia can play an important role in the study of the evolution of Hawaiian Drosophilidae. In a few instances, the subgroupings may not appear to be on as sound a basis as the others. This was due to a lack of a sufficient number of specimens of several species. With more material, these subgroupings would no doubt be more firmly established. ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to the members of his thesis committee, Drs. Wallace C. Mitchell, Ryoji Namba, and Hampton L. Car­son for their valuable advice and assistance. I am also grateful to Dr. D. E. Hardy for his enthusiastic guidance throughout my research. I am especially grateful to my wife, Betty, for her moral support and understanding throughout the course of this research. LITERATURE CITED Carson. H. L., F. E. Clayton and H. D. Stalker. 1967. Karyotypic stability and speciation in Hawaiian Drosophila. Proc. Nat. Acad. Sci. 57(5): 1280-1285. Carson, H. L. and H. D. Stalker. 1968a. Polytene chromosome relationships in Hawaiian Drosophila. I. The D. grimshawi subgroup. Univ. Texas Publ. 6818: 335-354. ----. 1968b. Polytene chromosome relationships in Hawaiian Drosophila. II. The D. planitibia subgroup. Univ. Texas Puhl. 6818: 355-365. ----. 1968c. Polytene chromosome relationships in Hawaiian Drosophila. III. The D. adiastola and D. punalua subgroups. Univ. Texas Publ. 6818: 367-380. Hardy, D. E. 1965. Insects of Hawaii. Vol. 12. Drosophilidae. University of Hawaii Press, 814 pp. ----. 1966. Descriptions and notes on Hawaiian Drosophilidae (Diptera). Univ. Texas Publ. 6615: 195-244. ----and K. Y. Kaneshiro. 1968. New picture-winged Drosophila from Hawaii. Univ. Texas Publ. 6818: 171-262. Hsu, T. C. 1949. The external genital apparatus of male Drosophilidae in relation to system­atics. Univ. Texas Publ. 4920: 80-142. Malogolowkin, Chana. 1952. Sombre a genitalia dos Drosophilidae (Diptera). III. Grupo wi­llistoni do genero Drosophila. Rev. Brasil. Biol., 12(1): 79-96. 1953. Sombre a genitalia dos Drosofilideos. IV. A genitalia masculina no sub­genero Drosophila (Diptera, Drosophilidae) . Rev. Brasil. Biol. 13(3) : 245-264. Nater, Hans. 1953. Vergleichend-morphologische Untersuchung des ausseren Geschlechtsap­parates innerhalb der Gatting Drosophila. Zool. Jb. (Systematik) 81: 337. Okada, T. 1953. Comparative morphology of the Drosophilid Flies. III. The "Phallosomal Index" and its relation with systematics. Zool. Magazine (Dobutsugaku Zasshi) 62(8): 278-283. -----1954. Comparative morphology of the Drosophilid Flies. I. Phallic organs of the melanogaster group. Kontyu 22: 36-46. 1955. Comparative morphology of the Drosophilid Flies. II. Phallic organs of the subgenus Drosophila. Kontyu 23: 97-104. Snodgrass, R. E. 1957. A revised interpretation of the external reproductive organs of male insects. Smithsonian Misc. Collection 135 (b): 1-11. Spassky, B. 1957. Morphological differences between sibling species of Drosophila. Univ. Texas Puhl. 5721: 48--61. Sturtevant, A. H. 1917. A new species closely resembling Drosophila melanogaster. Psyche 26: 135-155. Takada, H. 1965. Differentiation of the external male genitalia in the Drosophilidae (In Japanese; English Summary pp. 131-132). Kushiro Women's College Puhl. 1: 30-50. 1966. Male genitalia of some Hawaiian Drosophilidae. Univ. Texas Puhl. 6615: 315-333. Wasserman, Marvin. 1962. Cytological studies of the repleta group of the genus Drosophila: V. The mulleri subgroup. Univ. Texas Puhl. 6205: 85-117. Wheeler, M. R. and M. P. Kambysellis. 1966. Notes on the Drosophilidae (Diptera) of Sa­ moa. Univ. Texas Puhl. 6615: 533-565. White, M . J. D. 1968. Models of Speciation. Science 159(3819): 1065-1070. VI. Notes on Hawaiian "idiomyia" (Drosophila)1' 2 3 D. ELMO HARDY Fifteen species of Drosophila are now recognized which have an extra crossvein in cell R5. This has been previously treated as a generic character and on the basis of the extra crossvein the genus I di omyia Grimshaw ( 1901: 50) was erected. This would appear to be a most excellent generic character, very distinctive and characteristic of only this group of Hawaiian species. These are the largest known species of Drosophilidae in the world, some have a wing expansion of 18-20 mm. ldiomyia has been synonymized with Drosophila by Carson, et al. (1967: 1284) and it has been demonstrated that, chromosomally, species of "Idiomyia" are closely related to other picture-winged Drosophila and the presence or absence of the extra crossvein has no validity as even a species group character. The extra crossvein character apparently originated on the island of Maui, where seven known species possess this character; none is present on Kauai, five occur on Oahu, two on Hawaii and one on Molokai. Spur vein abnormalities are frequently found in field collected specimens of Drosophila, suggesting that there may be genetic variability for this character in nature and it should be possible to develop an "idiomyia" in the laboratory. The drawings have been prepared by Miss Geraldine Oda, University of Hawaii. Key to species of Drosophila possessing an extra crossvein in cell R, ("idiomyia") 1. Head normal in shape, front about as long as wide ---------------------------­ 2 Head very short and broad, in the male the front is three times wider than long, and the eyes are strongly protuberant. In the fe­ male the front is about two times wider than long. The abdominal terga have bright yellow spots on the sides -----------------------­Hawaii -----------------------------------------------------------------------­heteroneura (Perkins) 2(1). Extra crossvein in cell R;; situated at, near, or beyond m crossvein. Abdomen of male lacking clavate hairs on posterior portion -----------­ 3 Extra crossvein situated far anterior to the norm, about half way between crossveins r-m and m (fig. 11c, Hardy, 1966: 220). Pos­ terior portion of abdomen of male with long capitate or clavate hairs (fig. 11e, Hardy op. cit.) ________ Maui ________ clavisetae (Hardy) . 3(2). Wings brown, with numerous small hyaline spots or with basal two-thirds and apex brown -----------------------­-----------------------------------------­ 4 Wings predominantly hyaline, with brown markings on crossveins and at apex ----------------------------------­-------------------------------------------------­------­ 5 1 Published with the approval of the Director of the Hawaii Agricultural Experiment Station as Journal Series No. 1090. 2 This work was ma:le possible by NIH Grant GM10640. 3 Department of Entomology, University of Hawaii. STUDIES IN GENETICS V. Univ. Texas Publ. 6918, Sept., 1969 4(3). Wings brown with hyaline spots ________ Oahu-----------------------------------­--------------------------------------------------neogrimshawi Hardy and Kaneshiro Wings brown on basal two-thirds and at apex ____________ Oahu -----------­ ----------------------------------------------------------------------------------nigribasis Hardy 5 ( 3). Anterior margin of wing almost completely brown, from base of cell Sc to apex of vein M1 +2 (figs. 226c and 234b, Hardy, 1965: 546, 563) --··········----------------------------------------------------------------:-------------------6 Cell R1 lacking brown markings except at extreme base and apex -----.-7 6 ( 5). Extra crossvein in cell R:; situated well before the r-m crossvein. Crossvein r-m near basal third of cell 1st M2• Front tibia of male slender, nearly two times longer than basitarsus. Tibiae and basi­ tarsi of front legs (males) bearing long erect ciliation on the dorsal surfaces ____________ Oahu ----------------------------------------hemipeza (Hardy) Extra crossvein situated well beyond the m crossvein. Crossvein r-m at middle of cell 1st M 2 • Front tibiae short, rather thick, about one-third longer than basitarsi and lacking ciliation --------------Oahu --------------------------------------------------------------------------substenoptera n. nom. for ldiomyia stenoptera Hardy, preoccupied in Drosophila (Hardy 1965:473). 7(5). Arista with 9-10 or more dorsal rays and 3-4 ventral rays in ad­dition to apical fork, and with the anterior surface densely covered with fine pile _____________ .________ .__________________________ ..______________________ .________________ . 8 Arista with about five dorsal rays and two ventral rays in addition to apical fork, anterior surface bare. Thorax predominantly black, with three yellow pollinose vittae ____ __ ______ Maui, Molokai ? ___ ____ ____ _ ----------------------------------------------------------neopicta Hardy and Kaneshiro 8 ( 7). Male with anterior margin of wing strongly arched. Vein R2 +a strongly curved --------------------------------------------------------------------------------------9 Anterior margin of male wing straight or nearly so and vein R2+3 gently curved ----------------------------------------------------------------------------------------12 9 ( 8). Head and appendages predominantly black, including face, genae, antennae, mouthparts and usually lower occiput ------------------------------10 Head and appendages mostly yellow, antennae tinged with brown, black only on upper lateral margins of front, ocellar triangle and upper median portion of occiput and with a brown spot on gena; wing with a prominent brown spot in cell R1 below the arch in the costa ------------Oahu--------------------------------oahuensis (Grimshaw) 10 (9). Legs mostly or entirely yellow. Mesonotum and pleura conspicu­ously marked with rufous and scutellum rufous, except for the sides. Crossvein m slightly sinuate ------------------------------------------------------11 Legs and thorax nearly all black, scutellum with a narrow yellow apex and tibiae and tarsi reddish brown. Crossvein m gently con­cave, not sinuate ------------Maui ----------------melanocephala (Hardy) 11 (10). Front tibia densely long haired dorsobasally (fig. 1 b). Bases of femora broadly brown. Sides of scutellum entirely black ---------------­Molokai __________ _______ _________ ___ _________ __neoperkinsi Hardy and Kaneshiro Front tibia sparsely short haired dorsobasally (fig. 1a). Femora entirely yellow except for a slight tinge of brown on the apices. Scutellum rufous, tinged faintly with brown on sides ............ Maui ······················································-------------··-······-····--· cyrtoloma n. sp. 12(8). Front velvety black except for pale yellow lower margin. Palpi yellow, long, slender, straight sided densely black bristled at apex, about two times length of mentum. Extra crossvein in cell Rs situ­ated distinctly before m crossvein. Scutellum brown to black ......... . Hawaii---·······-·-·········-····················-·······----········· silvestris (Perkins) (nigrifacies Hardy is a synonym) Not as above, if the extra crossvein is well before m, the head is sharply pointed anteriorly and the other characters differ (plani­tibia Hardy) ··········---·--····-····---····················--··········--···-······················-· 13 13(12). Head only slightly pointed, about as high as long (fig. 2c). Front tibiae not flattened or curved and the promip.ent cilia of front legs confined to dorsobasal portion of tibia. Scutellum rufous medianly, brown to black on sides ····-······················-········-----····················---········ 14 Head sharply pointed anteriorly, distinctly longer than high (ref. fig. 13a, Hardy, 1966: 226). Front tibia flattened dorsally and slightly curved and moderately long cilia extend the full length of the tibia and the basitarsus. Scutellum black on the disc, yellow on margin·-·-····-··· Maui········-·················-·····----······ planitibia (Hardy) 14(13). Femora mostly brown to black. Abdomen entirely black. Meso­notum predominantly brown to black .......... East Maui, Haleakala ·······-····-········--··········-···-··················--·········---obscuripes (Grimshaw) Femora yellow with faint tinges of brown before apices and at bases. Abdomen yellow in background on bases and posterolateral margins of segments. Mesonotum rufous with tinge of brown on sides and sometimes on posterior portion ........................ West Maui ·-········---········-····--······--·-·······-----······-·····-····················· hanaulae n. sp. Drosophila cyrtoloma n. sp. (figs. 1 a-b) The genetic studies have demonstrated that what I had previously considered to be ldiomyia perkinsi (ldiomyia is a synonym of Drosophila, and perkinsi is preoccupied under this combination and has been changed to neoperkinsi Hardy and Kaneshiro, 1968: 261), is actually a species complex. Two species on Maui and one on Molokai have been demonstrated to be genetically distinct. In the previous treatment by Hardy (1965: 556) no specimens had been seen from Molokai except for the type male in the British Museum (Natural History) which is labelled "Molokai Mts. above 4000'." We have since collected a large series of specimens in the mountains of East Molokai which is probably typical neoperkinsi. That species is characterized by having longer, more dense black hair on dorsobasal portion of front tibia (fig. 1b), the longest hairs are nearly three times the width of the tibia. Also, by having longitudinal preapical dark brown to black streaks on each of the anterior and posterior surfaces of the front femora, the posterior surfaces of the middle femora and the anterior surfaces of the hind. Also, the bases of the femora are broadly brown. The specimens of neoperkinsi The University of Texas Publication FIG. 1. Drosophila cyrtoloma n. sp. a, front tibia of male; D. neoperkinsi Hardy and Kane­shiro, b, front tibia of male. on hand have a broad brown marking on each side of the mesonotum in the area of and lateral to the dorsocentral bristles, also the sides of the scutellum are en­tirely black. D. cyrtoloma is differentiated by having the front tibia sparsely setose on the posterobasal portion as in figure 1a, the longest setae are not much longer than the width of the tibia. The femora are entirely yellow, except for a slight tinge of brown on the apices. The mesonotum is not so distinctly blackened as in the specimens of neoperkinsi on hand and the scutellum is rufous, tinged lightly with brown on the sides. I see no other characteristics for separating these species. The characters of the wing and the genitalia are apparently identical. This species is highly distinctive in that it has extra heterochromatin on all chro­mosomes, an absolutely unique feature which was discovered by Dr. Frances E. Clayton and no other species or strain has this amazing feature. Fitting the description of neoperkinsi in other respects, this is apparently the most common species on the Island of Maui, it is very abundant in the Waikamoi region on the slopes of Haleakala, but according to cytological studies, it appears probable that two or more sibling species may occur, and I am designating only specimens from collection no. L86 (Waikamoi) which have been studied cyto­logically and from collection nos. L7, L8, and L13. from Kipahulu Valley, which have been proved by laboratory crossing to be conspecific with the series from W aikamoi. In the female, the face, genae, and occiput (except for upper median portion) are yellow to rufous, with a tinge of brown in the area occupied by the oral vibrissae, also the mentum is rufous and the palpi are brown, tinged with rufous. Holotype male "L86" and allotype female "L86Q3", W aikamoi, Maui, June 28-29, 1968 ( S. K. Ochikubo), also five paratype males same data (some collected by K. Y. Kaneshiro) . The testes of the males of this series have been dissected out and the chromosomes studied by Dr. Francis E. Clayton. Also ten paratypes, nine males, one female, from Kipahulu Valley, Maui, 4000'-6000', August 16-19, 1967, "L7, L8, L13'' (H. L. Carson and R. Iwamoto). This series was proved by laboratory crossing to be conspecific with the specimens from W aikamoi. At least fifty additional specimens are at hand. These are not being designated as part of the type series since there is a possibility of two sibling species being involved. Type, allotype, some paratypes in the B. P. Bishop Museum. Other paratypes deposited in the collections of the U. S. National Museum, British Museum (Natural History) and the University of Hawaii. Drosophila hanaulae n. sp. (figs. 2a-d) A population on hand from the mountains of West Maui appears to be dis­tinctive. According to Dr. Hampton L. Carson, the salivary chromosomes differ from other species of this complex. It has been temporarily treated in our field and laboratory studies as "light neoperkinsi-like." The species resembles neo­perkinsi Hardy and Kaneshiro, hut has the mesonotum and scutellum predom­inantly rufous rather than predominantly black, and the costa of the male is only slightly curved (fig. 2a). In neoperkinsi the costa is strongly arched (refer to Hardy, 1965: 557, fig. 231b). D. hanaulae fits closest to obscuripes (Grimshaw) based upon external morphological characters. It differs by having the femora yel­low with faint tinges of brown before apices and at bases, by having the mesono­tum and scutellum predominantly rufous, tinged with brown on sides and by hav­ing abdomen predominantly yellow to rufous, brown on apices of first two terga and tinged with brown on medium portions of terga 1-3. In obscuripes the femora are mostly brown to black, the abdomen entirely black and the mesonotum pre­dominantly brown to black. Dr. Carson (in litt.) has indicated that the chromo­somal differences between hanaulae and obscuripes are quite considerable "in fact, no other species on Maui has the chromosomal conditions of hanaulae." MALE. Head: Front golden, tinged with brown. Vertex, upper occiput, ocellar triangle and an extension on each side through area occupied by fronto-orbital The University of Texas Publication bristles brown. Face and genae black. Head slightly pointed at the bases of the antennae and approximately as high as long (fig. 2c). Antennae black. Arista with eleven or twelve dorsal and three or four ventral rays in addition to the apical fork, and densely short-haired on inner surface. Palpi and mouthparts black. Thorax: Predominantly rufous, tinged with brown on the sides of mesono­tum just outside the dorsocentral rows. Sides of scutellum brown. Propleura, pro­sternum, anterior two-thirds of each mesopleuron and median portion of each sternopleuron dark brown to black. Sides of metanotum tinged with brown. Legs: Predominantly yellow with the front coxae brown, tinged with yellow, and with faint tinges of brown at bases and apices of middle and hind femora and on bases and apices of hind tibiae. Front legs as in figure 2b. The long cilia on the front tibia confined to the basal portion. Basitarsus elongate, nearly two-thirds as long as tibia and with three erect setae at the apex. Wings: As in figure 2a. Abdomen: Predominantly yellow to rufous vvith apices of second tergum and median por­tions of terga 1-3 tinged with brown. Male genitalia as in figure 2d. I see nothing distinctive in the genital characters. Length: body, 7.0 mm; wings, 7.25 mm. FEMALE. Fitting the description of the male in most respects. The ovipositor blades are rather elongate and slender, similar to those of other species of this complex. Holotype male, Hanaula, West Maui, 4000', July 9-10, 1968, "L91" (K. Y. Kaneshiro). Allotype female, same locality and elevation, May 7, 1968, "L61C18" (J. P. Murphy). Sixteen paratypes, eleven males and five females, same data as type and allotype, one collected by S. Ochikubo. The metaphase chromosomes have been studied by Dr. F. E. Clayton and the salivaries by Dr. H. L. Carson. Type and allotype in the B. P. Bishop Museum, paratypes in the collections of the U.S. National Museum, British Museum (Natural History), and the Uni­versity of Hawaii. Drosophila nigribasis NEW NAME (fig3. 3a-b) ldiomyia bruneipennis Hardy, 1965, Ins. of Hawaii 12: 541, figs. 224a-d. Pre­ occupied in Drosophila by brunneipennis Malloch, 1923, Proc. Linn. Soc. N. S. Wales 48: 617. This species has previously been known only from the males. Females have been seen but because of the striking sexual dimorphism they were not associated with brunneipennis and had been set aside as "n. sp. ? ~ rel. oahuensis." The females have now been definitely associated with the males. The female differs from the male by having the face, palpi, mentum, and front coxae yellow, not dark brown to black. The wings lack the large brown marking filling the basal two-thirds (Hardy, 1965: 542, fig. 224b) but have only a sub­basal dark brown mark from apical portion of second costal cell over r-m crossvein. A brovvn transverse band extends over extra crossvein in cell R5, and a streak of brown extends along vein M1 +2 for a short distance before m crossvein (fig. Fm. 3. D. nigribasis NEW NAME. a, wing; b, female ovipositor. 3a), in addition to the apical brown mark which is also present in the male. The ovipositor is rather long and slender, shaped as in figure 3b. This species is present in both the W aianae and Koolau Mountains, Oahu. REFERENCES CITED Carson, H. L., F. E. Clayton, and H. D. Stalker. 1967. Karyotypic stability and speciation in Hawaiian Drosophila. Proc. Nat. Acad. Sci. 57: 1280-1285. Grimshaw, P.H. 1901. Fauna Hawaiiensis 3 (1): 51-73. Hardy, D. E. 1965. Insects of Hawaii, Diptera:Drosophilidae, Univ. Hawaii Press, 12: 814 pp. 1966. Descriptions and notes on Hawaiian Drosophilidae, Univ. Texas Publ. Studies in Genetics, 6615: 195-244. ----and K. Y. Kaneshiro. 1968. New Picture-Winged Drosophila from Hawaii, Univ. Texas Publ. Studies in Genetics, 6818: 171-262. VII. The Drosophila crassif emur Group of Species in a New Subgenus1 KENNETH Y. KANESHIRd Throckmorton (1966) showed that Drosophila crassifemur and D. nasalis had some internal characteristics which are more typical of Scaptomyza rather than of Drosophila. Spieth ( 1966) observed that D. crassifemur and D. nasalis "display a typical scaptomyzoid pattern of mating behavior.'' Clayton ( 1966 and 1968) found that the metaphase configuration of the chromosomes of these two species are like those of most Hawaiian Scaptomyza species. Hardy (1966), in describ­ing a new species closely related to crassifemur, indicated that this group of species should probably be placed in the genus Scaptomyza on the basis of in­ternal morphology, egg characters, and mating behavior; but that on the basis of external morphology, the crassifemur group of species would undoubtedly fit the present concept of the genus Drosophila. The evidence presented by Throck­morton, Spieth, and Clayton plus the relative complexity of the external male genitalia as compared with most species of Hawaiian Drosophila, warrants at least the removal of this group of species from the subgenus Drosophila but left in the genus Drosophila. The name Engiscaptomyza will be used for this new sub­genus. Its prefix is derived from the Greek word "engys" which means "near" or "close to." In Hardy (1965) D. crassifemur was indicated as being found on Maui, Molo­kai, Hawaii, Oahu, and Kauai. Takada in 1964 (as cited by Hardy, 1966), dis­covered that there are distinct differences in the male genitalia of specimens of crassifemur from Kauai from those specimens from the other islands. Hardy (1966) described the Kauai species as amplilobus and indicated that a complex of species is probably present which fit the description of crassifemur. A detailed study of the male genitalia including the phallic organs show that a complex of species is indeed involved here, and that a separate species each fitting the original description of crassifemur is present on each of the major islands of the Hawaiian chain: crassifemur on the Maui complex (The Maui complex is comprised of the Islands of Maui, Molokai, and Lanai as discussed by Carson and Stalker, 1968), amplilobus on Kauai, reducta on Hawaii, and a new species on Oahu (in flatus). At present, there are only two described species of the nasalis subgroup which is obviously closely related to the crassifemur subgroup on the basis of the ex­ternal morphological characteristics. Distinct differences in the male genitalia, however, indicate a completely separate complex of species. The two species of this subgroup include nasalis from the Maui complex, and undulata from Hawaii. 1 Published with the approval of the Director of the Hawaii Agricultural Experiment Station as Journal Series No. 1102. This work was made possible by the financial support of the Na­tional Institute of Health Grant GM 10640. 2 Department of Entomology, University of Hawaii, Honolulu. STUDIES IN GENETICS V. Univ. Texas Puhl. 6918, Sept., 1969. Subgenus Engiscaptomyza New Subgenus Apices and basis of middle and hind tibiae with prominent brown bands; six to eight acrostichal rows; mesonotum typically with five dark brown to black vittae extending the full length except for the lateral vittae being interrupted at the suture; arista with five or six dorsal rays plus two or three ventral rays in ad­dition to the apical fork; body typicallybrown with abdomen dark brown to black; wings pale brown; male genitalia very complex with enlarged claspers, anal plates with a finger-like projection on the ventral margins, presence of a sclero­tized ventral fragma, aedeagus bulbous or conical with the ba~al apodeme as long as or longer than the aedeagus. Type of subgenus: Drosophila crassifemur Grimshaw The crassifemur Subgroup This subgroup is composed of four species, one of which is new and will be described here. The males of this subgroup are characterized by the swollen front femora, the presence of a pair of very conspicuous membranous lobes arising from the ventral edge of the ninth tergum, and by the venter of the front femur being densely yellow pubescent. The females are characterized by the strongly scle­rotized ovipositor blades which are pointed at the apex. Both male and female specimens have two or three pairs of small setae on the scutellum in addition to the normal two pairs of scutellar bristles. On the basis of their external morphol­ogy, the species of this complex are indistinguishable. There are no reliable char­acters which will separate these four species except for the differences in the male genitalia. There are distinct differences in the shapes of the aedeagus (fig. 1a, e, i, m), the hypandrium (fig. 1b, f, j, n), and the lobe at the base of each clasper (fig. 1c, g, k, o) . The females can be easily separated by distinct differ­ences in the shapes of the ovipositor blades (fig. 1d, h, 1, p) . The species of this subgroup are as follows: Drosophila (Engiscaptomyza) crassifemur (Grimshaw), new comb. Drosophila crassifemur Grimshaw, 1901, Fauna Hawaiiensis 3 (1): 66 Redescription by Hardy, 1965, Insects of Hawaii 12: 229-230. Figures of aedeagus, hypandrium, clasper and ovipositor as in Figure 1a-d re­spectively. Distribution restricted to the Maui complex. Drosophila (Engiscaptomyza) amplilobus (Hardy), new comb. Drosophila amplilobus Hardy, 1966, Univ. Tex. Publ. 6615: 197-200. Figures of aedeagus, hypandrium, clasper and ovipositor as in Figure 1e-h re­spectively. Distribution restricted to Kauai. Drosophila (Engiscaptomyza) reducta (Hardy), new comb. Drosophila reducta Hardy, 1965, Insects of Hawaii 12: 445-446. Figures of aedeagus, hypandrium, clasper and ovipositor as in Figure 1i-l re­spectively. Distribution restricted to the Island of Hawaii. Drosophila (Engiscaptomyza) infiatus new species Here again, specimens of this species fit so nearly the description of typical crassifemur (in Hardy, 1965, 12:229-230) and amplilobus (in Hardy, 1966, Aedeagus Hypandrium Clasper Ovipositor ompli/oba ue reducto k \_-; ~ inflotus ~ ~ 0 .53 mm .53 mm 032 mm 032 mm FIG. 1. The D. (E.) crassifemur subgroup. 6615: 197-200) that it would be repetitious to redescribe the general external features. The species can be differentiated by distinct differences in the shapes of the aedeagus, hypandrium, and clasper of the male genitalia as in Figures 1m-o respectively, and in the ovipositor blade as in Figure 1 p. Distribution restricted to the Island of Oahu. Holotype male plus one paratype male from Mt. Kaala, Oahu, 4,000', July 2-3, 1968, (K. Y. Kaneshiro). Allotype female plus five paratype females, from Wili­wilinui Ridge, Oahu, May 11, 1965, (K. Y. Kaneshiro). Three other paratypes, two males and one female from Mt. Kaala, Oahu, September, 1952 (N. Morton) and Mt. Kaala, Oahu, April 22, 1965, (M. Delfina do). Type and allotype in the B. P. Bishop Museum. Paratypes to be distributed among the following collections: U.S. National Museum, British Museum (Nat­ural History), and the University of Hawaii. The nasalis Subgroup This subgroup is comprised of two species, one from the Maui complex (Maui and Molokai) and the other from Hawaii. The species of this complex are easily distinguished from those of the crassifemur subgroup by the following charac­teristics: the front femora not swollen, absence of extra setae on the scutellum aside from the normal two pairs of scutellar bristles, absence of densely yellow pubescence on the venter of the front femora of the males, and by the absence of the fleshy membranous lobes arising from ventral margin of the ninth tergum of the male genitalia. The females of this subgroup are readily separated from those of the crassifemur subgroup by the greatly swollen, yellow seventh sternum (very conspicuous in situ) and by the broad, blunt ovipositors. The two species in this subgroup are easily separated due to the conspicuous markings in the wings of undulata (fig. 203b, Hardy, 1965, 12: 494). Also, there are distinct differences in the shapes of the head, palpi, and aedeagus. The head of undulata is nearly triangular in shape as seen from a direct lateral view with the frontal margin at least three times longer than the lower margin. In nasalis, the head is nearly quadrate with the frontal margin only about one-half times longer than the lower Aedeagus Aedeagus (lateral) (ventral) nasalis a undulata e .84 mm FIG. 2. The D. (E.) nasalis subgroup. margin. The palpus of undulata is very narrow and long rather than very broad and short as in nasalis. The differences in the shapes of the aedeagus are as in Figure 2a, b and d, e. Drosophila (Engiscaptomyza) nasalis (Grimshaw), new comb. Drosophila nasalis Grimshaw, 1901, Fauna Hawaiiensis 3 (1): 66 Redescription by Hardy, 1965, Insects of Hawaii 12: 380-381 Shapes of aedeagus and ovipositor as in Figure 2a-c. Distribution restricted to the Maui complex of Islands. Drosophila (Engiscaptomyza) undulata (Grimshaw), new comb. Drosophila undulata Grimshaw, 1901, Fauna Hawaiiensis ( 3 ( 1): 58 Redescription by Hardy, 1965, Insects of Hawaii 12: 493-495 Shapes of aedeagus and ovipositor as in Figure 2d-f. Distribution restricted to the Island of Hawaii. LITERATURE CITED Carson, H. L. and H. D. Stalker. 1968. Polytene Chromosome Relationships in Hawaiian species of Drosophila. I. The D. grimshawi subgroup. Univ. Te:xias Puhl. 6818: 335-354. Clayton, F. E. 1966. Preliminary report on the karyotypes of Hawaiian Drosophilidae. Univ. Texas Puhl. 6615: 397-404. ----. 1968. Metaphase configurations in species of the Hawaiian Drosophilidae. Univ. Texas Puhl. 6818: 263-278. Grimshaw, P.H. 1901. Diptera. Fauna Hawaiiensis. 3(1): 1-77. Hardy, D. E. 1965. Insects of Hawaii. Vol. 12, Drosophilidae. University of Hawaii Press, 814 pp. ----. 1966. Descriptions and notes on Hawaiian Drosophilidae (Diptera). Univ. Texas Puhl. 6615: 195-244. Spieth, H. T. 1966. Courtship behavior of endemic Hawaiian Drosophila. Univ. Texas Puhl. 6615: 245-313. Takada, H . 1966. Male genitalia of some Hawaiian Drosophilidae. Univ. Texas Puhl. 6615: 315-333. Throckmorton, L. H. 1966. The relationships of the endemic Hawaiian Drosophila. Univ. Texas Puhl. 6615: 335-396. VIII. Polytene Chromosome Relationships in Hawaiian Species of Drosophila. IV. The D. primaeva Subgroup.1 HAMPTON L. CARSON AND HARRISON D. STALKER2 This paper is the fourth in a series describing the giant chromosome relation­ships among members of the subgenus Drosophila endemic to Hawaii. The first three papers (Carson and Stalker 1968 a,b,c) have dealt with species characteris­tically having dark maculations on the wings (the "picture-winged'' species). These fall into four rather close subgroups: I. The D. grimshawi subgroup (29 species), II. The D. planitibia subgroup (9 species), III. The D. adiastola sub­group (9 species) and IV. The D. punalun subgroup (6 species). Carson et al., (1970) have given preliminary data on fourteen more species of these subgroups bringing the totals to 36, 14, 11 and 6, respectively. This makes a total of 67 species which have been cytologically deciphered. The present paper presents chromosome maps and cytological recognition fea­tures for two more species which can be linked with the picture-winged flies, despite the fact that they themselves lack wing maculations. These two species, D. primaeva and D. attigua, are both endemic to the wet forests of Kauai and are morphologically very close indeed. Despite the fact that these species are distinc­tive and superficially appear to be quite distant from the picture-wings, this paper documents the fact that the banding order of each of their chromosomes can in­deed be completely interpreted according to the picture-wing Standard, Dro­sophila grimshawi. With the addition of these two species as subgroup V, the D. primaeva subgroup, the number of Hawaiian Drosophila related by completely­read polytene sequences totals 69 (see also Carson et al., 1970). MATERIAL AND METHODS The methods followed have been described in detail in Carson and Stalker 1968a. Chromosomal data are given here for 26 wild strains of D. primaeva (Table 1) caught between December 1965 and August 1968. Each strain origi­nated from a single wild female. Only a single wild specimen of D. attigua has been recognized. This strain (L41C12) was derived from a female captured by J. P. Murphy at Kahili, Kauai on March 19-20, 1968. A number of specimens of D. primaeva (e.g. L41C11,20,21, Table 1) were captured at the same time and in the same small baited area. D. attigua was described by Hardy and Kaneshiro 1969. 1 This work has been supported by the Evolution and Genetics of Hawaiian Drosophilidae Project, Grant No. GM 10640 to the University of Hawaii from the National Institutes of Health and by GB-3147 to Washington University and GB 711 to the University of Texas from the Na­tional Science Foundation. Personal acknowledgements will be found in paper I of this series. 2 Present address: Department of Biology, Washington University, St. Louis, Missouri. This paper is dedicated to the memory of Wilson S. Stone, teacher, colleague and friend. STUDIES IN GENETICS V. Univ. Texas Pub!. 6918, Sept., 1969. TABLE 1 Chromosome polymorphism in three populations of Drosophila prima.eva on Kauai locality and no. of' wild no. of gene arrangements obaervod in collection no. chromosomes chromosome chromosome chromosome tested x 2 4 Waimea District x A + g2 g212 + q + .2 Mohihi G2o.;s ; 4 0 4 0 4 Kokee J8101 ; 4 2 0 2 2 0 4 Total 6 8 ; 2 2 6 0 4 per cent 50.0 ;;.; 16.7 25.0 75.0 0 100 Koloa Diatrict Kahili L41C11, 20,21; GB,11-14,16, 19,21,25,26; P4-6; M8J1,2. 48 68 0 4 44 5; 15 66 2 per cent o.o a.; 91. 7 77.9 22.1 97.1 2.9 Hanalei District Pouli Stream L;7B6; G2,;,7-9 18 24 0 15 ; 18 6 24 0 per cent o.o a;.; 16. 7 75.0 25.0 100.0 o.o Photographic chromosome maps, prepared according to the method of Stalker ( 1965), have been made of D. primaeva and the sequences of this species and of attigua described in terms of the former (Figs. 1-3). The sequences of these two species were compared both with the Standard D. grimshawi sequences and those of D. picticornis, D. ornata (new species "A", Carson and Stalker 1968 c) and other Kauai flies. This work was facilitated by the method of table-level matching of unknown sequences with photographic map cut-outs. This was done using a compound microscope fitted with a drawing tube (Wild-Heerbrugg Instruments Inc.) as described in Carson and Stalker 1968 a. RESULTS The X chromosome of D. primaeva differs in banding order from the Standard D. grimshawi (Carson and Stalker 1968 a) by a minimum of 10 fixed inversions (Figure 1, lower chromosome). Plotting of the break-points of these inversions reveals that three of them have identical break-points with three previously known inversions, namely, Xi, Xk and Xo. This is of very great interest in that it is precisely these three inversions which are common to the D. adiastola and D. planitibia subgroups. The break-points of these inversions have been depicted in Figure 1 of Carson and Stalker 1968 b and Figure 1 of Carson and Stalker 1968 c. The eight remaining inversions, however, are new. The upper chromo­ i 2 12 92 1 \ ) J 1 ~H\' \l.~ , I .mr '\~ ~\~ \ . t I \ 2 f2 e2 ltd x2 ~ x2 FIG. 1. Chromosome X (below) of D. primaeva. Above: X chromosome map of D. grimshawi with inversions Xiko. The distal ends in this and in Figs. 2 and 3 are to the left. For details, see text. ~ c ;::s ~ ;::s ~ V:) ....... ~ ~ (J ~ "( c ~ c "> c ~ c .._, ...... ~ (1:) ~ ::i. s ~ ~ "> ~ ~ "( c ~ ~ "'l The University of Texas Publication ·-·-­ some of Figure 1 shows a photographic map of the Standard D. grimshawi X chromosome which has been cut and refitted to give it the order represented as Xiko. Were the inversions marked to be made, the order would be identical with the D. grimshawi Standard (see Carson and Stalker 1968 a). The lower figure shows the D. primaeva X chromosome. It may be converted to Xiko fairly easily, because the inversions occur in two separate groups, those which are distal, i.e. in the region of Xo, and those which are proximal, i.e. in the region of Xi and k. Thus, to achieve the Xo order, k2 and j2 should be made first. As k2 is made, the proximal break of h 2 should be carried along with the proximal break of k2 • The next inversion to be made is i2 The d2 and x2 breaks should not be moved along • with i2 • Finally, when h2 is made, the Xo order is obtained. To obtain the Xik d2 order at the proximal end, starting with the primaeva X, £2, and e2 should be made in any order. Three alternative sequences occur in the X chromosome of D. primaeva, result­ing in considerable chromosomal polymorphism in natural populations. The order which is closest to Standard D. grimshawi is Xikod2e2£2h2i2pk2• The second order differs from the above by a single inversion, Xg2 and the third differs from the second by having another inversion, Xl2, in addition to Xg2 • Xl2, however, has not C\I C\J --.........-c >. >.­ - ~............... '° !()..Cu-...... ­ / / ·~ phila. II. The D. planitibia subgroup. Univ. Texas Puhl. 6818: 355-365. ----. 1968c. Polytene Chromosome Relationships in Hawaiian Species of Drosophila. III. The D. adiastola and D. punalua subgroups. Univ. Texas Puhl. 6818: 367-380. Clayton, F. E. 1968. Metaphase Configurations in Species of the Hawaiian Drosophilidae. Univ. Texas Puhl. 6818: 263-278. ----. 1969. Variations in the metaphase chromosomes of Hawaiian Drosophilidae. This bulletin. Hardy, D. E. and K. Y. Kaneshiro. 1969. Descriptions of New Hawaiian Drosophila. This bulletin. Stalker, H. D. 1965. The salivary chromosomes of Drosophila micromelanica and Drosophila melanura. Genetics, 51: 487-507. ----. 1968. The phylogenetic relationships of Drosophila species groups as determined by the analysis of photographic chromosome maps. Proc. Xllth Int. Congr. Genet. 1: 194. ERRATA The caption for Figure 4, Studies in Genetics IV, 1968, p. 375, should read: "Chromosome X (top) and chromosome 4 (below) of D. punalua. For details, see text." The caption for Figure 5, p. 376, of the same article should read: "Chromo­some 3 (top) and chromosome 5 (below) of D. punalua. '' IX. Variations in the Metaphase Chromosomes of Hawaiian Drosophilidae1 FRANCESE.CLAYTON2 The present investigation is a continuation of karyotype determinations for species of the Hawaiian Drosophilidae. In the last publication (Clayton, 1968) the karyotypes present among 117 species in the genera Drosophila, Antopocerus, Scaptomyza, and Titanochaeta were summarized. The present tabulation includes twenty karyotypes for species not previously determined, eighteen in the genus Drosophila and two in the genus Antopocerus. The species listed in this report have been described by Hardy (1965, 1966, 1969) and Hardy and Kaneshiro ( 1968, 1969). Preliminary results on a study of added heterochromatin are also described. MATERIALS AND METHODS Metaphase configurations were determined from analysis of squash prepara­tions of larval brain tissue or testes dissected from adult males. The techniques for photography, slide preparation, and maintenance of the males in the labora­tory are identical wtih those described previously (Clayton, 1968). Small cages were established in an attempt to obtain larvae from some of the larger species of Hawaiian Drosophila which had not been reared successfully in vials. Cages consisted of glass lantern globes with the tops covered by several layers of cheesecloth. The globes were placed on wet paper toweling in plasticized paper plates. The toweling was moistened daily to keep the interiors of the cages humid, and small squares of sponge were placed in the floors of the cages to absorb any excess water. Wheeler-Clayton medium (Wheeler and Clayton, 1965) was placed into small plastic cups and a small piece of absorbent tissue was pushed into the food at the side of each cup. Fresh food was placed into the cages every three or four days, but old food cups were not removed for two weeks. All cups, however, were taken from the cages daily and the food surface was rubbed with a small spatula to smooth down the growth of mold and microorganisms. It was found that some females would deposit eggs in the tissue or old food which had been in the cages for as long as two weeks; they did not lay eggs in fresh food. For those species which would not deposit eggs in the food cups, various rotting plant ma­terials were added to the cages. The plants used were among those found to be utilized by Drosophila as breeding sites (Heed, 1968). Eggs were recovered from rotting Charpentiera and Clermontia bark and from the cut end of rotting 1 This investigation was supported, in part, by Public Health Service Research Grant GM 10640 from the National Institutes of Health, grant GB-711 from the National Science Foundation, and by the University of Arkans3s Graduate School through a Biomedical Science Support Grant from the National Institutes of Health. 2 Department of Zoology, University of Arkansas, Fayetteville. STUDIES IN GENETICS V. UniY. Texas Pub!. 6918, Sept., 1969. Cheirodendron stem. When the plant tissue was removed from the cage it was placed into a vial of Wheeler-Clayton medium and the larvae moved into the food from the rotting plant. RESULTS AND DISCUSSION Karyotypes A summary of metaphases determined since the last published list (Clayton, 1968) is given in table 1. This tabulation includes 326 strains of 59 species of the Hawaiian Drosophilidae, 5 7 Drosophila species and two Antopocerus species. Among the Drosophila, eighteen karyotypes are reported here for the first time. Seventeen of these species retain the primitive haploid configuration of five rods and one dot; "melanocephala ?" has five pairs of rods and one pair of V-shaped chromosomes. These species, which retain the primitive karyotype, may now be added to the previously reported eighty species, birnging the total to 97 of 111 sp?ci'3c; (87.4%) which retain haploid sets of five rods and one dot. The configura­tion of five rods and one V-chromosome has not been reported for any other Hawai.ian species of Drosophila. During 1965 and 1966, adult males tentatively identified as Idiomyia perkinsi from W aikamoi, Maui, were dissected to determine the karyotype of the species from spermatocyte configurations. The first metaphase figures were unlike any meiotic cells observed previously; the chromosome complement seemed to consist of five sets of three rods each and one set of two rods. This was referred to as a "triploid" condition. Three larvae were obtained for dissection in August 1966; metaphases from all three larvae consisted of five pairs of rods and one pair of dots. This situation was reported by Carson, Clayton, and Stalker (1966). No additional larvae were obtained until the fall of 1967 when females, tenta­tively identified as ldiomyia perkinsi, were collected in Kipahulu Valley, Maui. A sufficient number of larvae were obtained for the cytological study and to main­tain a stock of this strain. The larval metaphase chromosomes consisted of five pairs of V's and one pair of J's with satellites. Primary spermatocytes from F1 males had the "triploid" condition. Additional larvae were obtained from Waika­moi females and metaphases from these specimens also consisted of five pairs of V's and one pair of J's. Larvae from a cross of Kipahulu Valley females and males from Waikamoi had the same configuration, with no irregularity in pairing of the chromosomes. Hardy and Kaneshiro ( 1968) placed the "idiomyia" species in the genus Dro­sophila and the name of "perkinsi" was changed to neoperkinsi. Collecting trips to Hanaula, West Maui in May and July, 1968 yielded specimens tentatively identified as neoperkinsi. Most of the males and females were placed in cages but two males were dissected; first metaphase figures were typical of those species with haploid sets of five rods, one dot. Metaphases from larvae and F1 males con­firmed the configuration of five pairs of rods and one pair of dots. In July, 1968, neoperkinsi males and females were collected at Nawaihulili Stream, East Molo­kai and these were placed in cages, with the exception of four males which were dissected. All dividing spermatocytes had five pairs of rods and one pair of dots; this configuration was confirmed by metaphases from larvae and F1 males. TABLE 1 Metaphase configurations of Hawaiian Drosophilidae (j Spec:e.; Metaphase Collection Number Locality Drosophila adiastola asketostoma *attigua balioptera basisetae bostrycha cilifera *claytonae r.onspicua crucigera 5R, 1D 6R 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R. 1D 5R, 1D L25P7; L35P23 L61C10, G14, G16, G19, G20, G23, G24, G25, G44 K42C10 L41C12 L97G5, G6, B25 L71G4, G5 L19B36, B35, G2 L97B6, G3, L97G0 , L79B0 L98G21 L89L1 L19B26, B28, G6 K5C11, L26B2 L23G7, G9, G11, G16 L27B2 L38G1 L60G13, G15, G23 L88G2; L92G13, G17, G21 L87G28 L64G11, G12, G17, G42, G43 G46, G53, G54, G59 L64GH X C53.11 ~; L64G10 c1 X C53.11 ~; L64G28c1 X C53.11 ~ L67G7, GS, G13, G14, G15, G17, G25 ·~ "'4o 0 Waikamoi, Maui ~ Hanaula, W . Maui ~ .§ ~ Haleakala Crater, Maui Mt. Kahili, Kauai ~ ~ Apee, E. Molokai ~ (j Upper Olaa Forest Reserve, Hawaii ~ ......, Mountain House, Hawaii 0 Apee, E. Molokai ~ N awaihulili, E. Molokai 0 Vi Upp~r Olaa Forest Reserve, Hawaii 0 ~ Mountain House, Hawaii ~ Mauna Kapu, Oahu .Q_ Trail to Kaau Crater, Oahu Waimano Trail, Oahu ~ ~ Kilohana Crater, Oahu ~. Kupaua Valley, Oahu E). Makaleha Valley, Oahu ;:::i Mt. Kaala, Oahu tj ..., Lulumahu Falls, Oahu 0 CF> 0 Lulum3hu Falls x Tantalus, Oahu ~ ...... .......... ~ Peacock Flats, Oahu <.O 'l (.0 TABLE 1-Continued Species Metaphase CollecLion Numbet· cyrtolomat discreta disiuncta *disticha *distinguenda engyochracea fasciculisetae gradata *hanaulae heteroneura hirtipalpus 5V, 11 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D L45B16, B18, B19, B21, B22, B23, G8 L84B2 L84G1 ~ X C53.11 9 L85B1, Gt J9L C140.20°; FC3.5°; C125.14B 0 ; C3901°; J9B 0 ; J9D 0 ; J9G0 ; 0 ; J16H0 ; J23J°; J11K2°; J98A12°; K1L1°; L47C3; L86Q3; L86Q0 ; L86B0 L13G9-11 9 9 X L48G ¢ ~ F1 K75G27, G28, P132 L61BS, B7, C6, C7, CB; L91Q3, Q4 L83L1 M3B7, C9 L35P66 L92G1, G2, G3, G4 J29H2, H5 K75G18, K89G7 A L97B0 ; L97B11, B13 L98G0 L65P1 L67G3, G4 L92Q6 L61B0 ; L91B0 ; L91Q10 M3B0 L70G1, G7 K75P82 L61B8 Locality Mt. Kualapa, Kauai Iliiliula River, Kauai Iliiliula River, Kauai x Tantalus, Oahu Wailua River, Kauai Waikamoi, Maui ~ ~ (1:) ~ ;:::s ...... ~ ~ ~ "> Kipahulu Valley, Maui x Waikamoi, Maui ...... ....... Waikamoi, Maui ~ Hanaula, W. Maui 0._ ~ Waikamoi, Maui ~ i:::i Kaulalewelewe, W. Maui "> W aikamoi, Maui ~ !::: Makaleha Valley, Oahu \)"< ........ Puu Laalaau, Hawaii (:)· ~ Waikamoi, Maui ....... ...... Apee, E. Molokai 0 ;:::s Nawaihulili Stream, E. Molokai Kaau Crater Trail, Oahu Peacock Flat, Oahu Makaleha Valley, Oahu Hanalua, W. Maui Kaulalewelewe, W. Maui Puu-oo Volcano Trail, Hawaii Waikamoi, Maui Hanaula, W. Maui TABLE 1-Continued Species Metaphase Col lection Number Locality *inedita 5R, 1D L88G1 Makaleha Valley, Oahu ()·­ infuscata 5R, 1D L77G11 W. of Puu Ohu, S. Kohala Mts., Hawaii ~ ~ L78G3 Alakahi Stream, S. Kohala Mts., Hawaii ...... 0 *lineoseta.e 5R, 1D L61C1-5, B2 Hanaula, W. Maui ~ melanocephala 5R, 1D L91B0 Hanaula, W. Maui ~ *melanocephala? 5R, 1V M12L2 Waikamoi, Maui ~ ...... mitchelli 5R, 1D K77C3, C4 Alakahi Stream, S. Kohala Mts., Hawaii -§ K79B14 E. of Puu Ohu, S. Kohala Mts., Hawaii ~ ~ K80G4 W. of Puu Ohu, Kohala Mts., Hawaii ~ *murphyi 5R, 1D L79G1, G2 Polulu Stream, N. Kohala Mts., Hawaii CJ ~ L82B1, B2, G1; L89C12, C21, Upper Olaa Forest Reserve, Hawaii ""C 0 C24, C25, C26, C27, C33, G5, 3 G7, G10, G14, G1S, G19, G23 0 Vi *musaphila 5R, 1D L42G1 Alexander Reservoir, Kauai 0 ;3 *neoperkinsi 5R, 1D L9SG0 ; L9SG6, G8, G7-9-13; Nawaihulili Stream, E. Molokai ~ L98B9-12, B17-23 0-.. neopicta 5R, 1D L91B12 Hanaula, W. Maui ::t: M12L1 Waikamoi, Maui i::i M16D Paliku, Haleakala, Muai ~ nigribasis2 5R, 1D LS7B 0 ; L87G7-11, B5-9, G3, G4 Mt. Kaala, Oahu ~ .......... ~ nigrifacies 5R, 1D L34P3; L90C7°, CS 0 , C9° Upper Olaa Forest Reserve, Hawaii ;:::i L70G5, GS Puu-oo Volcano Trail, Hawaii tj ..., L22G2, Gt 1, G12; L52G21, G22, Kipuka #9, BM 510S, Saddle Road, Hawaii 0 [J) G23, G25, G2S, G30, G32; 0 >a L74G2, G3, G4, G7, GS, G9 e-: obscuripes 5R, 1D M16D Paliku, Haleakala, Maui .......... PJ *ocellata 5R, 1D L45B1, BZ, B12, B13, B14, G7 Mt. Kualapa, Kauai ochracea 5R, 1D L19B31, G9, P91 Mountain House, Hawaii ochrobasis 5R, 1D L22P1 x G9; L52G1, G4, G7, GS, Kipuka #9, BM 510S, Saddle Road, Hawaii G12, G13, G-4. G15, G17, G1S <.o <.o ~ 0 0 TABLE 1-Continued Species Metaphase Collection Number ornata • paenehamifera pa:tcipuncta • peniculipedis picticornis pilimana planitibia primaeva prolaticilia punalua •quasianomalipes *seiuncta *setosifrons setosimentum 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D 5R, 1D L78G1, G2 L81B1 , B2 L41G1 L61 C13,; B11 ~ x L61G ~ L71 G3; L82B5 L61G9, G10, G11, G12, G13, G35 L37G20 L42C3, G6, G11 L43C5 L45B25, B26, B27 L85G3, G4 M11J2, 15 L87G18 M14C2 L83L7; L86Q7 L37G0 ; L37G2, G3, G7, GS, G9 L41C11 , C20, C21 , G8, G11, G12, G13, G14, G16, G19, G21, G25, G26, P4, P5, P6; M8J1, J2 L91B1, B5°, B20 L23G3 L64G39, G40, G41 L87G29, G30, G31 L41G4; M1G2, G5 L45B3, B6, B10, B11 L89C11 , C13, Gt, G2 L71G10, Gt 1, G12, G13, G14; L82G8, G9 L74G1 l.ocnl!ty Alakahi Stream, Upper Hamakua Ditch, South Kohala Mts., Hawaii Honakapoula, S. W. Hanalei, Hawaii Mt. Kahili, Kauai Hanaula, W. Maui Upper Olaa Forest Reserve, Hawaii Hanaula, W. Maui ~ Ponli Stream, Hanalei District, Kauai ~ Alexander Reservoir, Kauai ~ ;::s Halemanu, Kauai ~· Mt. Kualapa, Kauai Wailua River, Kauai ~ Kokee, Kauai ~· Mt. Kaala, Oahu -Q.. Wiliwilinui Ridge, Oahu '°"'.j Waikamoi, Maui ~ Ponli Stream, Hanalei District, Kauai ~ Mt. Kahili, Kauai ~ <::!-' .._ <=;· Mountain House, Hawaii ...... ~ Trail to Kaau Crater, Oahu s· ~ Lulumahu Falls, Oahu Mt. Kaala, Oahu Mt. Kahili, Kauai Mt. Kualapa, Kauai Upper Olaa Forest Reserve, Hawaii Upper Olaa Forest Reserve, Hawaii Kipuka #9, BM 5108, Saddle Road, Hawaii CJ S' TABLE 1-Continued ~ Specie., Metnplrnse Coll ection Number Locality ~ spectabilis sproati truncipenna? uniseriata villosipedis virgulata Antopocerus *arcuatus *villosus L77G6, GS, G10 L79C4 5R, 1D L61 G37, G38, G41; L91B6 5R, 1D L71G9 L77Q5 L78Q1 L79G14, G16 5R, 1D L94C1-3 6R L60G1 5R, 1D L42P8 Mt 1111 5R, 1D M3B22° 5R, 1D L87B13-14 5R, 1D L91B0 W. of Puu Ohu, S. Kohala Mts., Hawaii ~ Polulu Stream, N. Kohala Mts., Hawaii ~ Hanaula, W. Maui ~ Upper Olaa Forest Reserve, Hawaii ~ (1:) W . of Puu Ohu, S. Kohala Mts., Hawaii CJ Alakahi Stream, Upper Homakua Ditch, S. Kohala Mts., Hawaii ~ Polulu Stream, N. Kohala Mts., Hawaii 0 ""'( ~ W aikamoi, Maui 0 Vi Kupaua Valley, Oahu 0 Alexander Reservoir, Kauai ~ ~ Kokee, Kauai Kaulalewelewe, W . Maui ~ ~ Mt. Kaala, Ohau ~ Manaula, W . Maui ~. ;::;· ;::s • JV1etap1iase reported here for the first time. t) ° Karntyre determined from adult male. ., 0 l l\tletaphase reported previously as perkinsi from Waikamoi ancl Kipahulu Valley, Maui. (See discussion) en 2 Metaphase reported preYiously as brunne.'pennis. (See Hardy, this publication) 0 Editor 's nole: The name ni{{rifnd<'s, appearing in this Table, has now been supplanted by silvestris (see Hardy and Kaneshiro, 1968) . ~ .... . ........ PJ ....... 0 ....... On the basis of differences in salivary gland chromosomes (Carson, et al., 1970), metaphase configurations, and taxonomic features, Hardy (1969) has designated these strains as three different species. The strain from East Molokai was desig­nated as neoperkinsi and the others were described as new species, cyrtoloma from W aikamoi and Kipahulu Valley, Maui, and hanaulae from Hanaula and Kaulalewelewe, West Maui. Therefore, the previously reported metaphase con­figuration of five pairs of V's and one pair of J's is characteristic of cyrtoloma; both neoperkinsi and hanaulae have five pairs of rods and one pair of dots. It seems probable now that the apparent discrepancy between meiotic figures and larval metaphases (Carson, Clayton, and Stalker, 1966; Clayton, 1968) resulted from examination of two different species and the "triploid" condition is, in re­ality, the appearance of the bivalents of the metacentric chromosomes of cyr­toloma. Two species listed in Table 1 have question marks because a discrepancy exists between previously reported configurations from primary spermatocytes (Clay­ton, 1968) and determinations from larval preparations. The karyotype for D. melanocephala was determined as five pairs of rods and one pair of dots from primary spermatocytes. During the last year, using the small cage method, larvae were obtained for cytological study. Larval cells, spermatogonia, and primary spermatocytes from this strain (M12L2) contained five pairs of rods and one pair of V's. Similarly, the larval metaphase configuration for "truncipenna ?" does not agree with the previously reported configuration of six pairs of rods. The larvae of this species were obtained from eggs deposited in rotting Charpentiera bark which had been placed in the truncipenna cage. Three slides were examined; the first was from a triploid larva, all dividing cells containing fifteen rods and three dots, (figure 6). The other larvae were diploid, metaphases consisting of five pairs of rods and one pair of dots. Further analysis of both larval and meiotic figures for these species will be attempted. H eterochromatin During the earlier phase of the investigation on the metaphases of Hawaiian Drosophilidae, the prime consideration was determination of the basic karyotypes and no special effort was made to compare chromosome lengths or sizes unless some striking difference was noted. However, a preliminary comparison has in­dicated that variation exists, not only between species, but between localities within a single species. It has been noted previously (Clayton, 1966, 1968) that karyotype changes among the Hawaiian Drosophila have resulted from fusion of chromosomes with loss of a centromere or by addition of heterochromatin. The present study has shown that added heterochromatin has altered chromosomes in several ways: (1) change from a dot to a rod; (2) formation of a V-or J-shaped chromosome from a rod; ( 3) increase in rod length, (4) increase in the size of a dot; and (5) development of an extremely large chromosome, in which the heterochromatic portion is almost equal to the total mass of the remaining chro­mosomes of the set. Table 2 summarizes the alterations in chromosome shape and size resulting from added heterochromatin and the photographs (figures 1-17) represent some TABLE 2 Summary of chromosome alterations resulting from addition of heterochromatin ('j Type of change Heference Remarks ~ ALTERED KARYOTYPE Six rods (6R): asketostoma conjectura limitata mimica truncipenna uniseriata villi tibia Five rods, one V (5R, 1 V): melanocephala? Fourrods,oneJ,onedot (4·R, 1J, tD): n. sp. "immigrans-Iike" Five V's, one J (5V, 1J): cyrtoloma CHROMOSOME LENGTH Double-length rods: basimacula? basisetae disjuncta hanaulae kauluai ocellata UT6615 UT6615 UT6818 UT6615 UT6818 UT6818 UT6818 UT6818 UT6818 UT6615 UT6818 PNAS;UT6818 UT6615 ~ 0 ~ ~ ~ ~ Determined from adult male. (See discussion). ~ One pair of rods double-length. ~ ('j ~ '"'I 0 Determined from larval, spermatogonial, spermatocyte metaphases. (See dis­0 3 Vi cussion and figure 2) . 0 J-shaped chromosomes extremely large. (See figure 4) . 3 ~ Published as perkinsi ?. (See discussion and figure 3). ~ ::i:: ~ ~ One pair double-length rods. ~. E;. All configurations with one double-length pair; second pair either double­ ~ length or one double-length rod paired with rod of normal length. (See tj figure 12). ~ One pair double-length rods in Kipahulu Valley, Maui collections only; 0 IJl 0 second pair either double-length or one double-length rod paired with rod ~ ....... of normal length. One pair of rods slightly longer than normal but no .......... double-length rods from other collecting areas. (See figures 10 and 11) . o::i One do-:lble-length rod paired with rod of normal length. One male larvae had one double-length paired with rod of normal length. One pair rods double-length. (See figure 14) . ....... 0 w ....... ..f:>.. TABLE 2-Continued Type of change Hefc1·enc.:P Remarks ochrobasis orphnopeza prolaticilia silvarentis spectabilis truncipenna? Large dots: crucigera fuscoamoeba hiritipalpus ornata pectinitarsus recticilia silvarentis UT6818 UT6818 UT6818, PNAS UT6818 UT6615, PNAS UT6818 UT6818 UT6818 UT6615 UT6818 UT6818 Variable within collection; double-length pair, one double-length paired with normal, double-length rods absent. (See figure 13) . ~ One pair rods double-length, second pair longer than normal in preparations ~ from Waikamoi, Maui only. ~ Unpaired rod double-length. (See discussion and figure 17). Other collections ~. had one pair double-length or one double-length paired with shorter rod. ~ One long rod paired with normal-length rod. (See figure 15) . ~. One pair double-length rods from Hanaula, W. Maui. Three pairs almost ~ double-length from Waikamoi, Maui. .Q.. One pair of rods extremely large. (See discussion and figures 5 and 6). ""-3 Large dots in collection (G58E3) from Mauna Kapu, Oahu. Variable in col­~ lections from Kaau Crater Trail, Oahu. '"ti $::: \::l"" ......... Large dots in collection from Hanaula, W. Maui; absent in Waikamoi, Maui t=:;· ~ collections. ...... s· ;::1 Dots larger than usual; not as large as in other species listed here. (See figure 15) . of the variations observed among the Drosophila species. Seven species have had their karyotypes altered from the primitive five pairs of rods and one pair of dots by addition of heterochromatin to the dot, resulting in a diploid configuration of six pairs of rods. D. zmiseriata (figure 1) is an example of a species of this type; in addition, the uniseriata metaphase has one pair of double-length rods. The hap­loid configuration of "melanocephala ?",consisting of five rods and one V-shaped chromosome, probably resulted from addition of heterochromatin to the dot pro­ducing a rod, and conversion of a rod into a V by the addition of a heterochromatic arm. Evidence for this conclusion may be seen from the configuration shown in figure 2. The heterochromatin is darker in this photograph; there are three darker rods, including the centrally located chromosome and its homologue which is al­most superimposed on one arm of the V-chromosome. The third dark rod, which seems to be paired with a lighter rod, may represent the Y-chromosome. One arm of each Vis also darker in appearance, probably representing the heterochromatic arm of the chromosome. The metaphase of D. cyrtoloma (figure 3) represents the most extreme altera­tion of karyotype observed in the Hawaiian Drosophila. Heterochromatin has been added to all chromosomes, resulting in five pairs of V's and one pair of J's. One rod of the new species "immigrans-like" has been altered by addition of a large amount of heterochromatin, resulting in an extremely large J-chromosome (figure 4); in anaphase figures, the short arm of the J appe3rs tJ be equal in length to the other rods. The difference in staining properties may be observed in the photograph, the heterochromatic J's appearing much darker than the rods. Thus, ten of the 111 Drosophila species examined have karyotypes altered by the addition of heterochromatin. Among species retaining the primitive configu­ration, the addition of heterochromatin has brought about changes in size and length of chromosomes in at least nineteen species. The summary given in Table 2 is a preliminary report of changes noted to date. Detailed analysis of variations is still in progress. This summary includes twelve species in which at least one rod is twice the length of the normal set of rods and seven species in which dots are larger than usual. The diploid set of chromosomes of D. truncipenna is shown in figure 5. One pair or rods is extremely large and the nature of the large rod may be seen more clearly in figure 6. The metaphase from the triploid larva in the latter figure shows the double nature of the long rods, three of them visible in the upper por­tion of the metaphase plate. The dots are not visible in this photograph. Figures 7, 8, and 9 are included to illustrate the occasional variations which appear in some species of the Hawaiian Drosophila. The normal configuration for D. neoperkinsi (figure 7) consists of five pairs of rods and one pair of dots, all of the rods being approximately the same length. However, the metaphases from one larva (figure 8) consisted of four pairs of rods, one larger unpaired rod, and one pair of dots. Ai: present, this is considered an abnormal condition, unrelated to the normal sex determining mechanism, since many other slides from male larvae have b2en examined and all had five pairs of rods and one pair of dots. One tetra­ploid larva was among those examined; metaphases consisted of twenty rods and four dots (figure 9) . The photographs in figur2s 10 through 15 illustrate variations present among species with the primitive karyotype of five pairs of rods and one pair of dots. Although retaining the primitive karyotype, these species have rods that are double-length or enlarged dots, presumably through the addition of hetero­chromatin. Metaphases of D. disjuncta are variable with double-length rods present in material from Kipahulu Valley, Maui only. Seventeen slides have been examined from collections at Waikamoi, Keanae Valley, Kaulalewelewe, and Hanaula, Maui and in these preparations one pair of rods was only slightly longer than the remaining rods of the metaphase configuration. However, in 35 slides from Kipahulu Valley collections, metaphases had either one pair of double­length rods or three double-length rods. These are illustrated in figures 10 and 11; in those individuals with three longer rods, one double-length chromosome was paired with a shorter rod. Metaphases of D. basisetae contained either three double-length rods or four longer chromosomes. The type with two pairs of double-length rods is shown in figure 12. This was consistent in the two populations examined, Upper Olaa Forest and Mountain House, Hawaii. In D. ochrobasis, variation in the length of the rods was found, not only among the different populations, but within a single locality. Eight slides were examined from larvae from Saddle Road, Hawaii col­lections. In seven of the slides, there were either one pair of double-length rods or one long rod paired with a rod of normal length. In one slide, however, no long rods could be found. Two slides from a collection at Hanakapoula, Hawaii con­tained paired double-length rods. The photograph (figure 13) illustrates the type with a single long rod paired with a shorter rod. D. spectabilis larvae, from Hanaula, West Maui had one pair of double-length rods whereas those specimens from W aikamoi, Maui had three pairs of rods al­most twice the length of the remaining rods in the metaphase. The configuration of D. truncipenna, already discussed earlier, may also be classified with this group FIG. 1. D. uniseriata (C144.5A), Kupaua Valley, Oahu. Six pairs of rods, one pair of double­length rods indicated by arrow. FIG. 2. D. melanocephala (M12L2), Waikamoi, Maui. One paid of V's, five pairs of rods; one rod is superimposed over arm of V-shaped chromosome. FIG. 3. D. cyrtoloma (L13G9-11 ~ ~ X L48G~ 3 ), Kipahulu Valley, Maui females, Waika­moi, Maui males. Five pairs of V-shaped chromosomes, one pair of J-shaped chromosomes with satellites indicated by arrow. FIG. 4. D. n. sp. "immigrans-like" (L19B43), Mountain House, Hawaii. Large heterochro­matic J-shaped chromosome with satellite indicated by arrow. FIG. 5. D. truncipenna (L94C1-3), Waikamoi, Maui. Diploid set of five pairs of rods, one pair of dots. Dots are not visible. FIG. 6. D truncipenna (L94C1-3), Waikamoi, Maui. Cell from triploid larva containing twelve normal rods, three extremely long rods, three dots. Double nature of large rod evident at arrow. Dots are not visible. FIG. 7. D. neoperkinsi (L98B9-12), Nawaihulili Stream, East Molokai. Diploid set con­sisting of five pairs of rods and one pair of dots. FIG. 8. D. neoperkinsi (L98B9-12) , Nawaihulili Stream, East Molokai. Metaphase with nine rods, one dot; unpaired rod is longer than other rods. FIG. 9. D. neoperkinsi (L98G6), Nawaihulili Stream, East Molokai. Tetraploid metaphase with twenty rods and four dots. Dots not visible. (All photographs taken at an original magnification of 1212X using phase contrast microscopy.) The University of Texas Publication of species since one pair of rods is at least twice the size of the remaining rods. D. ocellata (figure 14), basimacula ?, and orphnopeza from Waikamoi, Maui are other example:; of species with one pair of double-length rods. D. silvarentis (figure 15) has a pair of large dots and, as may be seen in the photograph, one longer rod. Large dots have also been observed in six other species, D. fuscoamoeba, ornata, hirtipa!pus, setosimentum, and crucigera. Large dots were present in metaphases of hirtipalpus from Hanaula, West Maui but w~re absent in those from W aikamoi, Maui. Very large dots were recorded for setosimentum from Pawaina, Hawaii but were absent in a larva from Mountain House, Hawaii. D. crucigera metaphases have been examined from many locali­ties on Oahu and large dots have been observed from only two collecting areas. One stock (G58E3) from Mauna Kapu, Oahu contained large dots in every meta­phase examined; collections from the Kaau Crater Trail were variable, with large dots present in one specimen but absent in two others. As mentioned earlier, this report is preliminary. The variation that exists be­tween populations and within single localities makes it necessary to examine the chromosomes of a number of individuals for each locality. The data included here were obtained from slides prepared for the purpose of karyotype determination. Little information is available on the sex of the larvae examined, and although it appears that some of the double-length rods may be X or Y chromosomes, no conclusion may be reached at the present time. For example, one slide of D. hanaulae had one double-length rod paired with a shorter rod but the sex of the larva. was not determined. When it has been possible to associate a configuration with the larval sex, some X and Y chromosomes have been identified. One larva, identified as male, was obtained from D. kauluai; one double-length rod was paired with a shorter rod. The metaphase of D. infuscata was determined from two slides, one in which there were five pairs of rods and one pair of dots; in the second slide there was a rod-shaped chromosome paired with a I-chromosome (figure 16). The sex of the latter was identified as male, so the conclusion may be reached that, in this species, the Y is a J-shaped chromosome. The presence of an unpaired rod (9R, 2D) in the metaphases of D. prolaticilia (figure 17) may FIG. 10. D. disjuncta (L1G66) , Kipahulu Valley, Maui. One pair of double-length rods present. FIG. 11. D. disjuncta (L1G49), Kipahulu Valley, Maui. One pair of double-length rods and one long rod paired with shorter rod. FIG. 12. D. basisetae (L19B33), Mountain House, Hawaii. Two pairs of double-length rods present. FIG. 13. D. ochrobasis (K33G1), Saddle Road, Hawaii. One double-length rod paired with rod of normal length. FIG. 14. D. ocellata (L45G7), Mt. Kualapa, Kauai. One pair of double-leng:h rods present. FIG. 15. D. silvarentis (K47G2), Humuula Saddle, Hawaii. One pair of large dots present and one long rod paired with shorter rod. FIG. 16. D. infuscata (L19G17), Mountain House, Hawaii. J-shaped Y chromosome is indi­cated by the arrow. FIG. 17. D. prolaticilia (L19B5), Mo:..:.ntai:1 Hous(), H:iwa:i. Four pairs of rods, one pair of dots, and one unpaired, long roJ. (All photographs taken at an original macnification of 1212X using phase contr::!st microscopy.) be related to the method of sex determination, but this configuration may also be an irregular condition such as mentioned earlier for D. neoperkinsi. Although remarkably stable in the retention of the primitive karyotype, the chromosomes of some species of the Hawaiian Drosophila have been altered, most frequently through the addition of heterochromatin. This has brought about a change in karyotypes of ten species, and, among the 97 species retaining the primitive configuration, added heterochromatin is evident in twelve species through the presence of double-length rods and in seven species by the presence of large dots. ACKNOWLEDGMENTS The author is indebted to all participants in the Hawaiian project for their co­operation in making this cytological study possible, and especially to H. L. Car­son, D. E. Hardy, W. B. Heed, K. Y. Kaneshiro, J. P. Murphy, K. Resch, H. T. Spieth, and the late W. S. Stone. The technical help of Joyce Sato, Susan Aihara, and Andrew Kuniyuki in slide preparation and rearing of the larvae is gratefully acknowledged. DEDICATION This paper is dedicated to the memory of Dr. Wilson Stone whose keen interest and enthusiasm about all aspects of the project on the Hawaiian Drosophilidae were an inspiration to all who worked with him. His leadership and encourage­ment were invaluable contributions to the investigations carried out by this au­thor. LITERATURE CITED Carson, H. L., D. E. Hardy, H. T. Spieth, and W. S. Stone. 1970. The evolutionary biology of Hawaiian Drosophilidae. Evolutionary Biology 4 (in press). Carson, H. L., F. E. Clayton, and H. D. Stalker. 1967. Karyotypic stability and speciation in Hawaiian Drosophila. Proc. Natl. Acad. Sci., Washington 57: 1280-1285. Clayton, F. E. 1966. Preliminary report on the karyotypes of Hawaiian Drosophilidae. Univ. Texas Puhl. 6615: 397-404. ----. 1968. Metaphase configurations in species of the Hawaiian Drosophilidae. Univ. Texas Puhl. 6818: 263-278. Hardy, D. E. 1965. Insects of Hawaii, vol. 12. Drosophilidae. Univ. of Hawaii Press: Hono­lulu. 814 pages. ----. 1966. Descriptions and notes on Hawaiian Drosophilidae (Diptera). Univ. Texas Puhl. 6615: 195-244. ----. 1969. Notes on Hawaiian "'idiomyias" (Drosophila). (This publication). Hardy, D. E. and K. Kaneshiro. 1968. New picture-winged Drosophila from Hawaii. Univ. Texas Puhl. 6818: 171-262. Hardy, D. E. and K. Y. Kaneshiro. 1969. Descriptions of new Hawaiian Drosophila. (This publication). Heed, W. B. 1968. The ecology of Hawaiian Drosophilidae. Univ. Texas Puhl. 6818: 387-420. Wheeler, M. R. and F. E. Clayton. 1965. A new Drosophila culture technique. Drosophila Information Service 40: 98. X. Enzyme Variation in Natural Populations of Drosophila mimica1 E. S. ROCKWOOD2 To make a thorough study of genetic variation in populations, a method must be used which can detect allele differences at a single locus. A substitution, de­letion, or addition within a gene may result in a change in the net electric charge or configuration of the gene product. After appropriate staining, the altered pro­teins may be detected by mobility differences in an electrophoretic medium. Markert and M0ller ( 1959) defined the different molecular forms of an enzyme which catalyze the same reaction as isozymes. Th~ importance of gel electrophoresis as a tool for the analysis of genetic varia­tion in natural populations is evident in the investigation of Drosophila popula­tions. Lewontin and Hubby ( 1966) studied several laboratory stocks of Dro­sophila pseudoobscura from different localities. Johnson et al. (1966) examined wild-caught light and dark Drosophila ananassae from American and Western Samoa. Since these first papers, Richardson, et al. (1966) made a qualitative comparison of the esterases of Drosophila aldrichi and Drosophila mulleri in cen­tral Texas. Stone et al. (1968) made an extensive investigation of the genetic variation in natural populations of Drosophila nasuta and Drosophila ananassae subgroups in the Pacific islands. Seasonal changes in the frequency of chromosomal polymorphisms in Dro­sophila have been established (Dobzhansky. 1943), but little work has been done on seasonal or time changes in the frequency of alleles controlling isozyme dif­ferences. Over a period of five years, Semeonoff and Robertson (1968) studied changes in the esterase allele frequencies of the Field Vole Microtus agrestis L. Changes were observed which were correlated with seasonal density fluctuations of the population combined with differential survival in the winter. The purpose of this investigation was to determine the stability of isozyme frequencies in wild populations of a species of Drosophila over a period of several months. An attempt was made to correlate the frequency changes with seasonal fluctuations in the environment. A study was undertaken to gain information on the functional value of these isozymes by the localization of the enzyme under study to particular tissues and developmental stages. MATERIALS AND METHODS The organism used for this analysis was Drosophila mzmzca, an endemic Hawaiian fly found in large numbers in Kipuka Puaulu (Bird Park) and Kipuka Ki, Kilauea, on the island of Hawaii. Wild-caught individuals were collected at 1 This work was supported by USPHS Research Grant No. GM-11609 from the National Insti­tute of General Medical Sciences. 2 Present address: Genetics Committee, University of Arizona, Tucson, Arizona, 85721. STUDIES IN GENETICS V. Univ. Texas Pub!. 6918, Sept., 1969. the first of each month from the two kipukas and mailed to the University of Texas. Two collecting sites, approximately a quarter of a mile apart, in Kipuka Puaulu and one site in Kipuka Ki were used. The first collection was made on November 1 and 2, 1967. The last collection was made on June 1 and 2, 1968. Most of the population data were obtained from wild males. The Kolmogorov­Smirnov test (Smirnov 1938; Feller 1948; and Stone et al. 1968) was used to detect significant differences in allele frequencies between localities and between monthly collections. To confirm the Mendelian inheritance of the enzymes under study, iso~e~ale lines were raised in the laboratory and the F/s were assayed for enzyme activity. Stocks of D. mimica which had been maintained for several years were used to determine the tissue and stage specificity of the enzyme activity and to investi­gate the different catalytic properties of the isozymes. Table 1 gives the identifica­tion number, collection site, collection date, number of wild females used to establish each stock, and the enzyme alleles present in each stock. A sample of Drosophila kambysellisi collected in Kipuka Puaulu was also assayed, and a comparison was made between the isozymes of this "mimica-like" spe~ies and the electrophoretic patterns of D. rnimica. Sample Preparation and Enzyme Assay The abdomen of each individual was dissected from the rest of the fly body and the two £ections were assayed separately. They were homogenized in 8 ,µ.l of deionized water. The slurry was absorbed by 4 mm X 5 mm rectangles of What­man no. 1 filter paper which were inserted in a 12% "electro-starch"* gel. Starch gel electrophoresis according to Smithie:> ( 1955) with the discontinuous buffer system described by Poulik ( 1957) was used. After electrophoresis for approxi­mately four hours at 220 volts, each gel was cut into three layers and each layer was stained separately. The abdomens were assayed for octanol dehydrogenase (ODH, after Court­right, 1966), alkaline phosphatase (APH, Beckman and Johnson, 1964a), and nicotinamide adenine dinucleotide dependent malic dehydrogenase (MDH-D, after Lewontin and Hubby, 1966). The heads and thoraces were assayed for alpha-glycerophosphate dehydrogenase (a-GPDH, after Sims, 1965), and acid phosphatase (ACPH, after Macintyre, 1966). Tissue and Stage Specificity Determination Eggs, first, second, and third instar larvae, pupae, and aged adults were assayed to determine the stage during which the enzymes were active. Larval salivary glands, brain, midgut, hindgut, fat body, hemolymph, and hypodermis; and adult antennae, proboscis, eyes, brain, foregut, midgut, hindgut, Malpighian tubules, ovaries, vagina and accessory gland, testis, paragonia, and ejaculatory bulbs were dissected and separately homogenized in 5 µ.1 deionized water and assayed for ACPH and APH activity. Induction and Substrate Specificity of Adult APH The effects of starvation and possible substrate induction of APH activity in *Electro-starch Co., P.O. Box 1294, Madison, Wisconsin, 53701. D. mzmzca adults were examined. From laboratory stocks, newly eclosed D. mimica adults were placed in vials containing paper moistened with water, 1% glucose, 1 % glucose-6-phosphate, 1 % pyrophosphate, and in vials with regular corn-meal food. These cultures were maintained for three days and individuals from each vial were assayed for APH activity. The flies remaining in the vial with water were transferred to vials containing regular corn-meal food. These flies were again assayed for APH activity. Abdomens of D. mimica males were assayed for phosphata'.:e activity with several different phosphate compounds: a:-glycerophosphate, /j-glycerophosphate, fructose-1,6-diphosphate, pho3phoserine, glucose-6-phosphate, gl ucose-1-phos­phate, adenosine-5'-diphosphate, adenosine-5' -triphosphate, pyridoxal-5-phos­phate, and 6-phosphogluconic acid by a modified Gomori method, cf. Pearse ( 1954). The staining solution was as follows: 100 ml. 0.1M Tris-HCl buffer pH 7.5 30mg. lead nitrate 25mg. phosphate substrate The gels were incubated overnight. After the gels were washed in water and im­mersed in a 1 % ammonium sulfide solution, dark bands appeared at the sites of phosphatase activity. RESULTS Inheritance of Variant Types Figures 1, 2, and 3 show examples of the phenotypes used to calculate the allele frequencies. The single banded phenotypes represented homozygotes or heterozy­gotes for electrophoretically indistinguishable types. Assuming that each allele resulted in a single band with a discrete mobility, alleles were numbered accord­ing to the decreasing order from the anode of the single band. The heterozygotes produced a triplet set of bands. The middle band of the heterozygote was a hybrid enzyme indicating that the gene product was a dimer. The phenotypes of single isolated females (isofemales) and their Fi's are given for APH (Table 2), ACPH (Table 3), and ODH (Table 4). No variants were observed for MDH-D or a-GPDH in the isofemales or their offspring. In most cases, the genotype of both parents could be derived from the phenotype of the female and her offspring. For all variable enzymes, the observed Fi ratio cor­responded closely to those expected for multiple alleles at a single autosomal locus. The results from the isofemales indicated that in most cases the females mated with one or more males of the same phenotype. However, a few produced phenotypes which were not consistent with the mating of a single pair. These isofemale lines resulted from multiple matings with males carrying electro­phoretically distinguishable alleles. General Zymogram Patterns Three zones of activity were observed for APH (Fig. 1). Zone A was nonvari­able. Variability was observed at zones B and C. Zone B was not used in the popu­lation analysis because the position of its bands appeared to be correlated with the distance of migration of the bands belonging to zone C. The faster migrating Fm. 1. APH phenotypes of wild-caught D. mimica. A 0 and B0 are zones of APH activity not used in the population analysis. bands of zone B were only present in individuals with Aph-1 or -2 of zone C. Of the APH zones, only zone C, controlled by 6 alleles, was used in the population analysis. ACPH had one major zone which was variable and was controlled by three alleles. A fourth allele (not shown in the figure) which migrated anodally, ahead of the other bands, was observed in 4 individuals. It was combined with Acph-1 in the allele frequency determination. ODH (Fig. 3) consisted of two alleles or single band types. MDH-D (Fig. 4) consisted of two zones of activity. Zone B, near the origin, was never resolved into discrete bands. Zone A consisted of a single band associated with a lighter "shadow" band. A rare variant occurred in two heterozygous individuals. Figure 4 also shows the a-GPDH phenotype. Two rare variant alleles were observed. One migrated ahead of the standard type and another migrated more slowly. These alleles were only observed in heterozygous ACPH + + '/z '/3 2 o~ % 3 I I '/z Yz AC PH PHENOTYPES OOH + ~ 2 • 0 2 3 Yz ~ + 0 I fz I 2 I I ODH PHENOTYPES FIG. 2. (above). ACPH phenotypes of wild-caught D. mimica. FIG. 3 (below). ODH phenotypes of wild-caught D. mimica. MDH-D a-GPDH FIG. 4. MDH-D and a-GPDH phenotypes of wild-caught D. mimica. individuals. The fast allele appeared in 4 individuals, and the slow allele appear~d once. Population Analysis. The expected and observed phenotypes are given for APR (Table 5) , ACPH (Table 6), and ODH (Table 7). There was no significant deviation in ob:::erved genotype frequencies in the wild-caught adults from those expected from the Hardy-Weinberg ratios. Allele frequencies calculated for the monthly collections at the separate col­lection sites are given for APH (Table 8). ACPH (Table 9), and ODH (Table 10) along with the number of individuals examined. No significant differences were found between the different collection sites. The frequencies calculated for Acph-1 and Acph-2 in December were significantly different above the 5%level from those obtained for the May collection from Kipuka Ki. There was a con­sistent reduction in the frequency of Acph-1 in Kipuka Ki over the six-month period. In June~ Acph-1 increased with a decrease in Acph-2. Although the trend was observed in the collections from site IIt it was not observed at site I. The proportion of heterozygosity per individual calculated from the five en­zyme loci picked at random in this analysis was 31.6 %for the site III population in Ki.puka Puaulu and 31.7% for the population in Kipuka Ki. Developmental Patterns Figure 5 shows the stage specificity of APH. APH activity was strongest in the second and early third instar larvae. The larvae demonstrated activity only at zone A. Zones Band C did not appear until one day after the adult had eclosed. Zone A appeared in the mid-and hindgut of the larvae and all three zones were located principally in the hindgut of the adult with occasional traces in the mid­gut. The activity was stronger in the males than in the females. The stage specificity of ACPH is shown in Figure 6. This enzyme was present in all stages of development, egg to adult. In the larvae, it was located principally in the midgut with traces in the salivaries, hypodermis, and hemolymph. In the adult, the activity vvas detected in the eyes, proboscis, brain, midgut, ovaries, and hemolymph, with traces in the Malpighian tubules. Induction and Substrate Specificity of Adult APH Traces of APR zones A, B, and C could be detected in newly eclosed flies. The activity at APH zones B and C disappeared when the flies were prevented from feeding. Altered activity remained at zone A. Two bands were present instead of one. APH bands did not appear in one-or-two-day-old flies held in vials with 1% pyrophosphate or 1% glucose-6-phosphate. All three APR zones were active in two-day-old D. mimica held in vials with regular com-meal food and with 1% glucose (Fig. 7). The APH of D. mimica was nonspecific. It was active in acid and alkaline con­ditions on all the substrates tested. The six allelic isozymes at zone C appeared to react at slightly different rates with the 6-phosphogluconic acid (Fig. 8). Aph-3,-4,-5 seemed to react more rapidly than Aph-1,-2, or-6. If Aph-3,-4, or-S was present in heterozygous individuals, the band in the 3, 4, or 5 position and APH lated 3r eggs day-old adults instar larvae ACPH -o adults pupae instar larvae FrG. 5 (above). Stage specificity of APH in D. mimica. FrG. 6. (below) . Stage specificity of ACPH in D. mimica. the hybrid band stained more darkly or reacted more rapidly than the other allelic type band. Comparison with Drosophila kambysellisi The D. mimica and kambysellisi enzyme systems which were examined were similar. However, fewer alleles were indicated in most of the segregating systems of the "mimica-like" species. The APH system (Fig. 9) was similar to that de­scribed for D. mimica. Zone A was present although it was lighter. Zone B was A B A B c newly one-day-old two-day-old eclosed D E F D F one-day-old two-day-old FIG. 7. APH activity of newly enclosed, one-and two-day old D. mimica held in vials with regular corn meal food (A) , water (B). 1% glucose-6-phosphate (D), 1% pyrophosphate (E), and 1% glucose (F); C indicates fiies transferred from B-2 and held for two days on com meal food. absent. Zone C appeared to be represented by two bands which corresponded in electrophoretic mobility with Aph-1 and -3 of D. mimica. The APH of D. kamby­sellisi was present in the hindgut, and was apparently nonspecific. Activity was detected with adenosine-5' -di phosphate. The ACPH of the kambysellisi had an electrophoretic mobility which was the same as Acph-1, however, it stained much lighter. The MDH-D and a-GPDH enzymes of the two species were identical in mobility. The ODH of the kambysellisi varied with two electrophoretic types. The fast band was identical in mobility to Odh-1 of D. mimica. The second unique band migrated two mm behind Odh-1. DISCUSSION Dehydrogenases The amount of variability at the dehydrogenase loci of D. mimica was similar to that found in other Hawaiian Drosophila examined to date. Although limited <:! ­ A B APH A A - 11 11 MIMICA-L1KE FIG. 8 (above). Comparison of the activity of APH zone C with 6-phosphogluconic acid and alpha-naphthyl acid phosphate.. FIG. 9 (below) . Comparison of the APH activity in D. mimica and "mimica-like." (kamby­sellisi). population data are available, several of the species are polymorphic for ODH. All the Hawaiian Drosophila examined to date, including D. mimica, have been monomorphic for a-GPDH. MDH-D was monomorphic. Several other species have been found to be polymorphic for this enzyme. Alkaline Phosphatase The maintenance of four Aph alleles at a frequency above 10% is relatively unusual in nature. Pacific island populations of D. ananassae (Stone, et al. 1968) have an ACPH system with seven alleles, four of which are maintained at fre­quencies above 10%. The D. ananassae were collected from five islands in two island groups. The population of D. mimica is contained in an area of a few square miles. The genetic variability may be related to the large number of host plants of the fly. D. mimica occupies successfully a large number of different niches and has been raised from several different species of plants, bracket fungus, soap­berry fruits, and Peperomia leaves (Heed, 1968). The Aph allele frequencies were stable over the eight-month period (Fig. 10), and all the common alleles were found to be present in at least one of the labora­tory stocks derived from collections over the past five years. Levene (1953) pro­posed a model for the maintenance of a balanced polymorphism when more than one ecological niche is available. The model appears to fit the D. mimica popula­tion st..-u~ture. Levene assume:; a random mating population depositing zy­gotes at random into each niche. The model requires differential mortality within each niche. The larvae feeding in leaves, fungi, or fruits are probably exposed to host specific environments and competitors. To reach a balanced poly­morphism, the survivors from the separate niches should mate at random within the entire population. Conditions are more favorable for an equilibrium if indi­viduals more fit in a particular niche preferentially breed in that niche and mate with other flies reared from the came niche. Future studies ere planned to detect these preferences if they exist. The protein structure is important to the catalytic activity of an enzyme. A ACPH APH November December February Morch May June ·~l ®~-e O~-~ (jJjJ) AQ.h_-~ 0~-§ Q 0~-§ Fm. 10. Allele frequency changes through time in Kipuka Puaulu (upper) and Kipuka Ki (lower). The map shows the approximate location of the two kipukas on the main island, Hawaii. slight change in the structure of the enzyme molecule could possibly reduce the catalytic activity with respect to one substrate or increase the activity with another. Although 6-phosphogluconic acid may not be important in the metabo­lism of the fly, the differential catalytic activity which the alleles demonstrate could be of selective significance to the fly. One mechanism for phosphate regulation in insects involves the secretion of inorganic phosphate into the midgut and the reabsorption of phosphates in the hindgut (Wyatt, 1958). The association of the nonspecific APH activity with the hindgut suggested that the enzyme may be active in phosphate regulation, ab­sorption (Rode and Varicak, 1964), or possibly phosphate transfer (Stadtman, 1967). A supply of energy was necessary before the enzyme could be produced. The energy could be supplied in the form of glucose, but not glucose-6-phosphate. This suggests that the enzyme does not function as a digestive dephosphorylator prior to absorption of nutrients. Kato ( 1959) described the induction of alkaline phosphatase activity in the small intestine of the chick embryo. Since APH zone C does not appear until after the adult fly has eclosed and begun to feed, the possibility of induction by naturally occurring substrates was considered. No evidence for-induction was found. Neither inorganic phosphate nor glucose-6-phosphate appeared to inducf' activity. Acid Phosphatase The frequency of Acph-1 was reduced during the collection period from De­cember to May (Fig. 10). The allele frequency changes were significant in Kipuka Ki but were not significant from Kipuka Puaulu, although the same general trend was evident in Kipuka Puaulu. The frequency change in Kipuka Ki resulted from a reduction in the frequency of jndividuals homozygous for Acph-1. In December, the percentage of individuals homozygous for Acph-1 was 28% while in May only 4% of the individuals were homozygous Acph-1. The per­centage of individuals heterozygous for Acph-1,2 was 39% in December, 29% in February, 39% in March, and 33% in May. The significant change in the ACPH allele frequencies could be attributed to sampling error since the number of individuals examined each month was rela­tively small, but the change followed a pattern between collections. Changes in allele frequencies are attributed to either migration, mutation, selection, or a combination of these factors. Kipuka Ki and Kipuka Puaulu have the most dense population of D. mimica in Hawaii. Only a few individuals have been collected from other sites on the island of Hawaii. A kipuka is an older area of land which has been surrounded by more recent lava flows and as such is a relatively closed community. Migration into the kipukas from other areas can be discounted. A small amount of migration between the collection sites may exist, but it is im­probable that a significant number of individuals with the same allele would migrate away from Kipuka Ki. The magnitude of the change was too great to be due to mutation if the rate of mutation in isozyme loci is about equal to that of known point mutations in Drosophila. The change could be due to selection act­ing over a period of time such as one season of the year. In general, the fitness of homozygotes is more sensitive to environmental changes than heterozygotes (Dobzhansky, et al. 1955; Dobzhansky and Levene, 1955). The Acph-1 homozygotes were greatly reduced in frequency while the Acph-1,2 heterozygotes did not change in frequency. The Acph-1 homozygote may have been more sensitive to any environmental changes. A detailed analysis of the microclimates in the kipukas was made by G. A. Smathers from January through July, 1968. Kipuka Ki offered a generally harsher environment for the fly. The top soil was drier and warmer than in Kipuka Puaulu. The temperature fluctuation was greater in Kipuka Ki. Kipuka Puaulu offered a more mesic en­vironment with few temperature fluctuations. The significant changes in Kipuka Ki could have resulted from a combination of seasonal changes and the harsher environment. There was a steady rise in temperature during the collection period at all sites, and the vegetation was changing. Most of the Sapindus fruits disap­peared before April. The change in allele frequencies could be a unique event not part of a seasonal pattern. The Kilauea Volcano began erupting in November 1967. From January through mid-April the rain pH registered 3.5-5.0. The normal range of 5.0-7.7 was obtained from April through June. The acid rain also added S04, Cl, and Fl impurities to the soil. It is almost impossible to positively correlate allele frequency changes with selective pressures of the environment. There are many undefined factors de­termining the niche of D. mimica, and any correlation between environmental change and allele frequency change could be due to chance. Acph may be in a heterozygous inversion with other genes on which selection is acting. However, a possible relation between enzyme function and environment may exist. The association of acid phosphatase activity with autolysis has been established for a wide variety of organisms (Cone and Eschenberg, 1966; Schin and Clever, 1966; Anderson, 1966). In D. mimica larvae the enzyme concentration was greatest in the organs which are to be histolyzed early during the pupal stage, and the gen­eral concentration increased during the pupal stage. The ACPH appears to func­tion in histolysis in D. mimica. The rate at which the ACPH isozymes histolyze the larval organs during the pupal period could determine the length of time spent in the pupal stage. Ifthe ACPH-1 isozyme lengthens the pupal stage, those pupae would be more subject to desiccation and acid injury in the soil. No significant deviation of observed frequencies from the expected Hardy­Weinberg equilibrium was obtained. Moderate selection pressures which could cause a change in allele frequencies might not be detected as a deviation since the sample size was relatively small. Weak selection of the survival of fertilized eggs to adults could remain undetected. Several species of Drosophila closely resembling D. mimica but taxonomically distinct occur in the kipukas. These Drosophila are possibly the product of re­cent evolution, perhaps resulting from host plant isolation. The D. kambysellisi used in this study is restricted in the selection of host plants and can only breed in Pisonia leaves. Thus, the reduced variability in the case of both APH and ACPH could be related to the host specificity of the fly. However, populations of D. mimica have been identified from Kauai, and the kipuka "mimica-like" species could have evolved from these flies which may have been polymorphic for fewer alleles. The D. mimica population offers a unique opportunity to study the interaction between isozymes or single locus variability, natural selection and speciation. SUMMARY 1. A population analysis was conducted using D. mimica males collected from Kipuka Puaulu and Kipuka Ki, Kilauea, Hawaii. The flies were assayed for alka­line phosphatase, acid phosphatase, octanol dehydrogenase, nicotinamide ade­nine dinucleotide dependent malic dehydrogenase, and alpha glycerophosphate dehydrogenase activity. The inheritance of each of these enzymes is controlled by alleles at an autosomal locus. 2. The population was monomorphic for MDH-D and a-GPDH. ODH varied with two alleles. APR varied with six alleles, four of which were maintained at frequencies above 10%. ACPH varied with three alleles. The proportion of heter­ozygosity per individual of the population was about 31 % . 3. The Odh allele frequencies were stable over the eight-month collection period. The Aph alleles appeared to be maintained in a balanced polymorphism. The frequency of Acph-1 significantly decreased with a corresponding increase in the fequency of Acph-2 be_tween the December and May collections. The fre­quency of Acph-1 increased in the June collection. 4. D. mimica feeds and breeds in a large variety of host plants. The variable APH system was only active in the adult after the fly had begun to feed. The APH activity was localized in the hindgut. Different rates of catalysis were de­tected between the APR allele isozymes with 6-phosphogluconic acid. The APR polymorphism may be maintained by niche adaptation, differential rates of catalysis, or a combination of both. 5. The effects of selection appear to be expressed at the Acph locus either di­rectly or indirectly in Kipuka Ki. The change could be the result of the harsher environment of Kipuka Ki, the lowered pH of the rain due to the eruption of the Kilauea volcano, a change in the availability of host plants such as the Bracket fungus, or Sapindus frnits, or a combination of environmental factors. An obser­vation that gene frequencies change in nature, however, may or may not be ex­plained by natural selection. 6. The allelic isozymes of D. kambysellisi were reduced in variability. The evolution of the "mimica-like" species could be a case of sympatric speciation where two populations became isolated by different substrate adaptations, or it could be a case of allopatric speciation and reintroduction into the kipukas of a species which evolved during more extensive geographic isolation. ACKNOWLEDGMENTS The author gratefully acknowledges the interest and guidance of Dr. W. S. Stone in the early planning of this project. The helpful criticisms from Drs. M. R. Wheeler and Ken-ichi Kojima in the preparation of this manuscript for a masters thesis were appreciated. Dr. M. P. Kambysellis collected the D. mimica each month and provided information concerning the vegetation and environment of the kipukas without which this study could not have been undertaken. The dis­cussions with Drs. F. M. Johnson, W. B. Heed, and H. W. Kircher and the tech­ nical assistance of Cindy McWhorter, Hazel Lindsay, Julie Wareing, Joan Sech­ler, and Chester MacDanald are thankfully acknowledged. Anderson, P. J. 1966. The effect of autolysis on the distribution of acid phosphatase in rat brain. J. Neurocliem. 12: 919-925. Beckman, L. and F. M. Johnson. 1964a. Variation in larval alkaline phosphatase controlled by Aph alleles in D. melanogaster. Genetics 49: 829. Cone, M. V. and K. M. Eschenberg. 1966. Histochemical localization of acid phosphatase in the ovary of Gerris remegisSay (Hemiptera). I. Exp. Zool. 161: 337-:352. Courtright, J. B., R. B. lmberski and H. Ursprung. 1966. The genetic control of alcohol de­hydrogenase and octanol dehydrogenase isozymes in Drosophila. Genetics 54: 1251-1260. Dobzhansky, Th. 1943. Genetics of natural populations IX. Temporal changes in the com­position of populations of Drosophila pseudoobscura. Genetics 28: 162-186. Dobzhansky, Th., and H. Levene. 1955. Genetics of natural populations XXIV. Develop­mental homeostasis in natural populations of Drosophila pseudoobscura. Genetics 40: 797­ 808. Dobzhansky, Th., 0. Pavlovsky, B. Spassky and N. Spassky. 1955. Genetics of natural popu­lations XXII. Biological role of deleterious recessives in populations of Drosophila pseudoob­scura. Genetics 40: 781-796. Feller, W. 1948. On the Kolmogorov-Smirnov limit theorems for empirical distributions. Annals of Math Stat. 19: 177-189. Heed, W. B. 1968. Ecology of Hawaiian Drosophilidae. Univ. Texas Puhl. 6818: 387-419. Johnson, F. M., C. G. Kanapi, R. H. Richardson, M. R. Wheeler and W. S. Stone. 1966. An analysis of polymorphism among isozyme loci in dark: and light D. ananassae strains from American and Western Slllnoa. P.N.A.S. 56: 199. Kambysellis, M. P., F. M. Johnson and R.H. Richardson. 1967. lsozyme variability in species of the genus Drosophila IV. Distribution of the esterases in the body tissues of D. aldrichi and D. mulleri adults. Biocliemical Genetics 1: 19-27. Kambysellis, M. P. Studies of oogenesis in natural populations of Hawaiian species II. The significance of microclimactic changes in oogenesis of D. mimica. (manuscript in prepara­tion). Kato, Y. 1959. The induction of phosphatase in various organs of the chick: embryo. Develop. Biol. 1: 477. Levene, H. 1953. Genetic equilibrium when more than one ecological niche is available. Amer. Nat. 87: 331-333. Lewontin, R. C. and J. L. Hubby. 1966. A molecular approach to the study of genetic heterozy­gosity in natural populations of Drosphila pseudoobscura. Genetics 54: 595. Macintyre, R. J. 1966. The genetics of an acid phosphatase in D. melanogaster and D. simu­lans. Genetics 53: 461. Markert, C. L. and F. MfSller. 1959. Multiple forms of enzymes: Tissue, ontogenic, and species specific patterns. P.N.A.S. 45: 753. Pearse, A. G. E. 1954. Histocliemistry: Tlieoretical and Applied. J. ant.. A. Churchill Ltd. London, 460-461. Poulik, M. D. 1957. Starch gel electrophoresis in a discontinuous system of buffers. Nature 180: 1477. Richardson, R.H., M. P. Kambysellis, and F. M. Johnson. 1968. Isozyme variability in species of the genus Drosophila V. Esterase allele frequencies in natural populations of D. aldrichi and D. mulleri. Biochemical Genetics 1: 239-247. Rode, B. A. F., and A. Varicak. 1964. The distribution of acid and alkaline phosphatase activi­ties in some organs of Cyprinus carpio L. Bull Sci. Con. Acad. R.S.F. Yougoslavie Sect. A. Sci. Natur. Tech. Med. 9: 158-159. Schin, K., and U. Clever. 1968. Lysosomal and free acid phosphatase in salivary glands of Chironomus tentans. Science 150: 1053-1055. Semeonoff, R., and R. W. Robertson. 1968. A biochemical and ecological study of plasma esterase polymorphisms in natural populations of the Field Vole Microtus agrestus L. Bio­chemical Genetics 1: 205-226. Sims, M. 1966. Methods for detection of enzymatic activity after electrophoresis on poly­acrylamide gel in Drosophila species. Nature 207: 757-758. Smathers, G. A. 1968. A report on the microclimates in two Hawaiian Kipukas. (personal communication). Smirnov, N. 1939. On the estimation of the discrepancy between empirical curves of distri­bution for two independent samples. Bulletin Mathematique de l' Universite de Moscow 2: fasc. 2. Smithies, 0. 1955. Zone electrophoresis in starch gels. Group variations in the serum proteins of normal human adults. Biochemical Journal 61: 629. Stadtman, T. C. 1959. Alkaline Phosphatases. The Enzymes 5: 69. Stone, W. S., M. R. Wheeler, F. M. Johnson, and Ken-ichi Kojima. 1968. Isozyme variation in natural island populations of members of the Drosophila nasuta and Drosophila ananassae subgroups. P.N.A.S. 59: 102-109. 1968. Isozyme variation in Drosophila island populations II. An analysis of two ananassae subgroup species in the Samoan, Fijian, and Philippine Islands. Univ. Texas Publ., 6818: 157-169. Wyatt, G. R. 1958. Phosphorus Compounds in insect Development. Proceedings of the Fourth International Congress of Biochemistry XII: 163. TABLE 1 Stock numbers, collection sites, collection dates, number of wild females used to establish each stock, and Acph, Aph, and Odh alleles present in D. mimica stocks maintained at The University of Texas. Acph Aph Odh Stock munber Collection site Collection date Isofemale Alleles Alleles Alleles WH20.1 Kipuka 12/6/63 1,2 3,5,6 Puaulu J50B50 Kipuka 12/ 66 1,2 3,4 Puaulu original Kipuka unknown unknown 1,2 4 1,2 Puaulu HH11.5 Kipuka 7/14/64 2 3,4,5,6 1,2 Ki G50E50 Kipuka 4/14/66 1,2 Z,3,4 1,2 Ki 89 Kipuka unknown unknown 1,2 2,4 Puaulu 19.2 unknown unknown unknown 1,2 Z,3,4 TABLE 2. APH phenotypes of wild-caught D. mimica females and their F1 offspring. Pheno-Number of offspring of APH phenotype Deduced genotype type 1-2 1-3 2-3 2-4 2-5 2-6 3 3-4 3-5 3-6 4 of parents 3-4 2 4 6 6 3-4 2-3 (4.5)t ( 4.5) ( 4.5) (4.5) 5-6 3 5 7 5-6 2-3 (3.75) (3.75) (3.75) 3-4 9 3 3-4 4-4 (6) (6) Z-4 4 2 2 3 Z-4 3-4 (2.75) (2.75) (Z.75) (2.75) 2-3 3 4 1 9 2-3 1-6 (4.25) ( 4.25) (4.25) (4.25) 2 11 7 2-2 3-4 (9) (9) 3-4 2 5 8 3 3-4 2-3-4* 3-5 4 17 19 3-5 2-3-4* t Expected number of F1 individuals in each phenotypic class if parents of giYen genotype. • Possible multiple matings of one female. TABLE 3 ACPH phenotypes of wild-caught D. mimica females and their F1 offspring. No. of offspring of ACPH phenotype Phenotype 1-2 1-3 2 2·-3 Parental genotype 1-2 1-2 1-2 2 2 1-2 1-2 1-2 2 6 6 12 (9) 6 14 17 5 43 (51 )t 6 7 13 (17) 3 (4) 7 8 (9) 11 12 24 6 5 58 (51) 4 5 9 (9) 6 (4) 4 4 1-2 1-2 1-2 2-2 2-2 1-2 1-2 1-2 2-3 2-2 2-2 2-2 1-2 1-2 1-2 1-2 2-3 1-2-3* +Expected number of Ft individuals in each phenotypic class if parents of gh·en genotype. • Possible multiple matings of one female. TABLE4 ODH phenotypes of wild-caught D. mimica females and their F1 offspring. No. of offspring of ODH phenotype Phenotype 1 1-2 ~ Parental genotype 1-2 5 7 1-2 1-1 1-2 6 5 1-2 1-1 1-2 22 19 1-2 1-1 1-2 7 9 1-2 1-1 40 40 (40)t (40) 1-2 5 4 3 1-2 1-2 1-2 2 12 5 1-2 1-2 7 16 8 (7.5) (16) (7.5) 21 1-1 1-1 7 1-1 1-1 6 1-1 1-1 34 (34) 8 1-1 2-2 (8) 2 9 9 2-2 1-2 (9) (9) +Expected number of Ft indi,·iduals in each phenotypic class if parents of giwn genotype. TABLE 5 Observed (and expected) numbers of indiYiduals in APH phenotypic classes. Phenoty-pic da::::;es Colle..·tion ,;ite Date 2-) 2-4 3-3 3-.J 3-5 +-.J +-5 . 0thers Total Kipuka 11/1-2 9 8 H· 32 6 19 14 18 120 Puaulu, (9) (11) (1+) (34) (11) (21) (13) (7) site I 12/ 1-2 + 4 4 22 10 20 10 11 85 (2) (+) (7) (22) (7) (19) (11) (5) 2/1-2 1 2 1 4 3 2 + 3 20 (1) (1) (1) (4) (2) (4) (3) (4) 3/1-2 5 7 + 11 2 4 1 4 38 (5) (5) (5) (10) (1) (5) (1) (6) 6/1-2 + 1 5 5 3 6 1 3 28 (4) (3) (4) (8) (2) (+) (1) (2) Kipuka 11/1-2 3 8 12 30 12 11 10 17 103 Puaulu. (7) (7) (13) (25) (12) (12) (12) (18) site III 12/ 1-2 + 6 9 17 6 15 10 14 81 (6) (7) (7) (20) (7) ( 13) (9) (12) 2/1-2 + 13 12 24 5 16 13 5 92 (7) (10) (9) (25) (6) (18) (9) (8) 3/ 1-2 4 7 4 18 6 12 6 12 69 (4) (6) (5) (15 ) (6) ( 11) (8) (14) 5/ 1-2* 2 2 5 8 3 5 3 7 35 (3) (3) (4) (10) (3) (4) (3) (5) 6/ 1-2 6 3 12 18 2 5 8 8 62 (5) (4) (11) (18) (5) (7) (4) (8) Kipuka 11/1-2 8 8 9 14 8 12 7 7 73 Ki (8) (8) (8) (18) (6) (10) (7) (8) 12/ 1-2 7 5 11 13 3 5 3 9 56 (6) (4) (10) (14) (5) (5) (4) (8) 2/1-2 10 5 14 14 5 11 8 20 87 (10) (9) (11) (20) (6) (9) (5) (17) 3/ 1-2 9 5 7 30 5 13 7 17 93 (6) (7) (11) (25) (6) (14) (7) (17) 5/ 1-2 8 3 8 17 6 13 6 16 77 (6) (7) (8) (18) (7) (10) (7) (14) 6/ 1-2 12 6 12 19 6 18 9 16 98 (7) (9) (11) (26) (6) (15) (7) (17) • Collections for ;ite:; I and III combined. TABLE 6 Observed (and expected) numbers of individuals in ACPH phenotypic classes. Phenotypic classes Collection site Date 1-1 1-2 1-3 2-2 2-3 3-3 Total Kipuka 12/1-2 14 27 0 40 3 0 84 Puaulu, (9) (36) (36) (1) site I 3/1-2 8 16 2 9 4 0 39 (8) (17) (9) (3) 6/1-2 5 9 10 2 0 27 (4) (11) (9) (1) Kipuka 12/1-2 9 37 2 26 2 2 78 Puaulu, (10) (33) (27) (5) site III 2/1-2 15 24 2 35 4 81 (10) (34) (29) (5) 3/1-2 6 26 2 31 2 0 67 (6) (27) (30) (3) 5/ 1-2* 1 11 12 1 26 (2) (8) (13) (2) 6/ 1-2 8 19 3 23 4 2 59 (6) (22) (21) (6) Kipuka Ki 12/ 1-2 15 21 15 1 0 53 (13) (25) (13) (1) 2/1-2 13 25 2 39 3 2 84 (9) (34) (33) (5) 3/ 1-2 6 37 3 40 7 94 (8) (35) (41) (7) 5/ 1-2 4 29 3 47 5 0 88 (4) (29) (47) (6) 6/1-2 12 36 2 33 4 0 87 (11) (38) (32) (3) • Collections for sites I and III romb '.ned. TABLE 7 Observed (and expected) numbers of individuals in ODH phenotypic classes. Phenotypic classes Collection site Date 1-1 1-2 2-2 Total Kipuka 11 / 1-2 69 50 8 127 Puaulu, (70) (49) (8) site I 12/1-2 43 24 6 73 (42) (27) (4) 2/ 1-2 10 9 0 19 (11) (7) (1) 3/ 1-2 17 20 1 38 (19) (16) (3) 6/ 1-2 14 9 0 23 (15) (7) (1) Kipuka 11/1-2 51 37 7 95 Puaulu, (51) (37) (7) site III 12/1-2 23 16 1 40 (24) (14) (2) 2/ 1-2 32 19 3 54 (32) (19) (3) 3/ 1-2 44 20 4 68 (43) (22) (3) . 5/1-2 16 11 1 28 (17) (10) (1) 6/ 1-2 32 23 2 57 (33) (21) (3) Kipuka Ki 11/1-2 44 22 5 71 (43) (25) (3) 12/ 1-2 26 20 6 52 (25) (22) (5) 2/1-2 50 25 7 82 (48) (30) (4) 3/ 1-2 52 36 4 92 (53) (34) (5) 5/ 1-2 52 26 3 81 (52) (26) (3) 6/ 1-2 50 46 3 99 (54) (38) (7) • Collections for sites I and III combined . TABLE 8 Aph allele frequencies of D. mimica. Allele frequencies Collection site Date T otal 2 3 4 G Kipuka 11/1-2 120 .004 .108 .325 .404 .125 .033 Puaulu, 12/ 1-2 85 .012 .053 .277 .471 .147 .041 site I 2/ 1-2 20 .025 .075 .250 .425 .175 .050 3/ 1-2 38 .013 .184 .368 .368 .040 .026 6/1-2 28 0 .161 .393 .357 .071 .018 Kipuka 11 / 1-2 103 .015 .102 .350 .345 .165 .024 Puaulu, 12/ 1-2 81 .006 .117 .296 .407 .136 .037 site III 2/ 1-2 92 .016 .120 .310 .446 .109 0 3/ 1-2 69 .014 .109 .275 .406 .145 .051 5/1-2* 35 .029 .114 .357 .357 .114 .029 6/ 1-2 62 .008 .099 .419 .339 .097 .040 Kipuka Ki 11/1-2 73 0 .147 .327 .359 .122 .045 12/ 1-2 56 0 .134 .411 .295 .107 .054 2/1-2 87 0 .161 .356 .316 .092 .075 3/ 1-2 93 .011 .102 .344 .392 .091 .059 5/1-2 77 .020 .123 .325 .357 .130 .046 6/ 1-2 98 .015 .112 .337 .388 .087 .061 • Collections for sites I and III combined. TABL E 9 Acph allele frequencies of D. mimica. Collection site Date Total Allele frequencies 2 3 Kipuka Puaulu, site I 12/ 1-2 3/ 1-2 6/1-2 84 39 27 .327 .436 .370 .655 .487 .574 .018 .077 .037 Kipuka 12/1-2 78 .365 .583 .051 Puaulu, 2/ 1-2 81 .346 .605 .049 site III 3/1-2 67 .299 .672 .030 5/1-2* 26 .250 .692 .058 6/ 1-2 59 .322 .585 .093 Kipuka Ki 12/ 1-2 53 .491 .491 .019 2/ 1-2 84 .316 .631 .054 3/ 1-2 94 .277 .660 .064 5/ 1-2' 88 .227 .727 .046 6/ 1-2 87 .356 .609 .034 * Collections for sites I and III combined. +Frequencies significantly different from those calculated for collections made on 12/1-2 from the same site. The University of Texas Publication TABLE 10. Odh allele frequencies of D. mimica. Collection site Date Total Allele frequencies 2 Kipuka Puaulu, site I 11/1-2 12/1-2 2/1-2 3/1-2 6/1-2 127 73 19 38 23 .74-0 .753 .763 .711 .804 .260 .247 .237 .290 .196 Kipuka 11/1-2 95 .732 .268 Puaulu, 12/1-2 40 .775 .225 site III 2/1-2 54 .769 .232 3/1-2 68 .794 .206 5/1-2* 28 . 768 .232 . 6/1-2 57 .763 .237 Kipuka Ki 11/1-2 71 .775 .225 12/1-2 52 .692 .308 2/1-2 82 .762 .238 3/1-2 92 .761 .239 5/1-2 81 .803 .198 6/1-2 99 .737 .263 • Collections for sites I and III combined. XI. Studies on Interspecific Hybridization Within the Picture-Winged Group of Endemic Hawaiian Drosophila1 HEI YUNG YANG AND MARSHALL R. WHEELER.2 INTRODUCTION Harland (1933) stated, "It is increasingly recognized that the study of species hybrids from a genetical standpoint could throw great light on the evolutionary process. But at the same time there has been little experimental work of a con­clusive or even enlightening nature since the inter-sterility of Drosophila species prevented the application of the magnificent experimental technique of that school of workers to the question of species hybrids." Similar views were ex­pressed by Dobzhansky (1937), who wrote, "The genus Drosophila is remarkable in that most of its numerous species refuse to intercross and to produce hybrids." At that time he was able to cite only four instances of interspecific hybridization in Drosophila, and he concluded that "... although further cases of interspecific hybridization in the genus Drosophila will, no doubt, be discovered by future studies, the conclusion is justified that in this genus the isolation of species from each other is more thorough than in many other animal and particularly plant genera." The earlier attempts to produce species hybrids in Drosophila all involved species of the subgenus Sophophora, the first report being that of Sturtevant (1919) involving D. melanogaster ~ X D. simulans t>. He pointed out, however, that the "unisexual broods" of Drosophila described by Quackenbush (1910) were probably hybrids between melanogaster and simulans since this cross nor­mally produces only female offspring. Hybridization between species of the large subgenus Drosophila was first re­ported by Spencer (1938), using D. virilis and D. americana. During the follow­ing twenty years, many new reports were published, mostly involving species of this subgenus. Wheeler (1958) summarized the data up to that time; of 2.17 inter­specific crosses which had been reported, 88 of which were reciprocal crosses, 111 had produced offspring showing some degree of fertility. Patterson and Stone (1952.) had listed earlier only 156 crosses (with 55 reciprocal), including crosses 1 This investigation was supported by USPHS research grant No. GM-11609 from the Na­tional Institutes of Health. 2 The authors are especially grateful to the late Professor Wilson S. Stone who was largely responsible for the success of the project studying the Biology and Evolution of the endemic Hawaiian Drosophilidae, and who arranged to have the studies on hybridization carried out at the University of Texas. Much appreciation is also due to Mr. Johnnie Murphy who did much of the field collecting and species identifications of the material used in this study, to Kathleen Resch for advice and help on rearing endemic Hawaiian Drosophila, and to Mrs. Florence Wilson, Mrs. Margaret Harget, Mrs. Virginia Gerstenberg, Miss Hazel Lindsay, and Mr. Ken­neth Kaneshiro for their assistance at various times on this project. STUDIES IN GENETICS V. Uni,·. Texas Puhl. 6918, Sept., 1969. involving subspecies. All of the data referred to above point clearly to the con­clusion that hybridization has so far been possible only between species of the same species group (species complex), that is, between species which are most aljke genetically and morphologically. One of the most characteristic results of interspecific hybridization is the ap­pearance of sterility in the hybrid offspring. Darwin (1859), discussing hybridism in The Origin of Species, stated, "The view commonly entertained by naturalists is that species, when intercrossed, have been specially endowed with sterility, in order to prevent their confusion ... for species living together could hardly have been kept distinct had they been capable of freely crossing." He then showed~ with appropriate examples, that all possible degrees of sterility may be found, depending on which forms are being crossed. One reason for this, he felt, was that ... "The fertility of first crosses, and of the hybrids produced from them, is largely governed by their systematic affinity. This is clearly shown by hybrids never having been raised between species ranked by systematists in distinct families; and on the other hand, by very closely allied species generally uniting with facility. But the correspondence between systematic affinity and the facility of crossing is by no means strict." Modern studies of Drosophila hybrids, summar­ized above, show that these remarks apply quite well even today. Hybrids may be produced intentionally by man or they may occur spontane­ously in nature. A distinction is commonly made, therefore, between artificial and natural hybridization. However, the characteristics of the hybrids are es­sentially the same, whether they occur naturally or are induced by the experi­menter. As we have shown above, laboratory-induced hybridizations have been obtained rather commonly in Drosophila, but natural hybridization seems to be exceedingly rare. Patterson and Stone (1952), for example, list only four known instances. Various isolating mechanisms must, therefore, be operating to prevent or restrict crossing in the wild. To obtain artificial hybrids in Drosophila. it is necessary to modify or eliminate one or more of the isolating factors. The simplest technique, judging from the extensive data on Drosophila hybridization, is placing individuals of the two species together in the unnatural environment of a laboratory culture vial; chances of hybridization are increased if there is no choice of mates (the so-called "no-choice" experiment). Occasionally an experimenter is able to modify conditions in some specific, meaningful way. Mayr (1950) removed the antennae from females of D. pseudo­obscura before mating them to persimilis males; where heterogamic mating took place 2.5% of the time under normal conditions (no-choice conditions), it was increased to 26.8 % when the females lacked antennae. These experiments were repeated by Ehrman (1959) with similar results; however, when the technique was tried with species of the willistoni species group, fewer, rather than more, matings occurred. Spieth ( 1952) showed that the different species and speciPs groups of Drosophila utilize different courtship behavior patterns. It is logical, then, to assume that removal of the antennae would not necessarily produce identical results in different groups of species. Gottschewski ( 193 7) attempted to circumvent the courtship behavior aspect (ethological isolation) by using artificial insemination. He believed that in this way hybrids might be produced from non-crossable species in which the lack of hybridization was due to sexual or mechanical isolation. Although he reported a satisfactory level of success when the sperm for injection were taken from the vaginae of other females 0-30 minutes following a normal copulation, his tech­nique seems not to have been used by other workers. Kambysellis ( 1968a, 1968b). using some modifications of the technique per­fected by Kinsey (1966), used ovarian transplantation between species to per­mit hybridization without altering the normal mating behavior. For the first time, hybrids (larvae only) were obtained between species of different species groups: gibberosa X virilis and pavani x mulleri. In the hybridization experiments with endemic Hawaiian Drosophila reported here, individuals of the two species being tested were simply placed together in a food container, in a no-choice situation. Dr. Michael Kambysellis (personal com­munication), however, has used the ovarian transplantation technique, and his results, to be published elsewhere, agree in general with those reported here. The endemic Hawaiian Drosophila fauna. The unusual nature of the Dro­sophila fauna of the Hawaiian Islands was first pointed out by Grimshaw (1901, 1902) and Perkins (1910) but the true magnitude of the species diversity was shown most dramatically by Hardy (1965, 1966) and Hardy and Kaneshiro ( 1968). Hardy ( 1965) listed 400 named species of Drosophilidae from Hawaii, only 18 of them being introduced (immigrant) species. Hardy (1966) added an additional 20 endemic species, and Hardy and Kaneshiro ( 1968) described 35 more species. Other new species await description; Hardy (1966) states that "It is now obvious that at least 600 species of Drosophilidae occur in the Hawaiian Islands." The number of species known from the entire Nearctic region is only 175 (Wheeler, 1965). A majority of the end€mic Hawaiian Drosophila species have unusual or unique morphological traits when compared with the continental species. Spieth ( 1966, 1968) has found that the modified forelegs of males and the unusual pro­ tuberances of the proboscis are often involved in courtship and mating behavior. In strong contrast to their great morphological diversity. the species are remark­ ably uniform in many of the internal characters (spermathecae, for example; see Throckmorton, 1966), in the basic chromosome numbers (Clayton, 1966, 1968), and even in the banding patterns of the polytene chromosomes (Carson and Stalker, 1968a, b, c). A further unusual feature of the Hawaiian fauna involves the larval food and pupation habits. Heed (1968) lists the species which have been reared from various native plants, primarily species of Cheirodendron. Clermontia and others. He estimates that the breeding sites for about one-third of the species are now known. Mature larvae of many of these species crawl or "skip" some distance from the food source and then burrow into soil to pupate (Wheeler and Clayton, 1965). A majority of the species, however, have not been successfully raised in laboratory culture. Some of the reasons appear to be nutritional (see Robertson, et al., 1968). In view of all of the many interesting features of the endemic Hawaiina Dro­sophila species, it seemed especially desirable to attempt interspecific hybridiza­tion in the laboratory. Such experiments might show whether the ethological iso­lation was as great as one might expect from the great morphological specializa­tion; they might show whether inter-island crosses behaved any differently from intra-island crosses, and whether sympatric species were more or less easily hybridized than allopatric species. The hybrids and their backcrosses might also provide additional clues to the phylogenetic relationships. Finally, such experi­ments should show whether these island populations, which have evolved quite rapidly in a relatively short time (Throckmorton, 1966), are more or less hy­bridizable than the well-studied "mainland" species which have had a very dif­ferent evolutionary history. MATERIALS AND METHODS Stocks. Among Hawaiian drosophilids. the picture-winged species group was selected for this study of interspeci.fic hybridization. Most of the stocks used in this investigation were started from adults collected in the wild by baiting and sweep­ing in the field. However, for a few sµecies, stocks were started from adults reared from a variety of fruits and other plant materials. As far as possible, each wild­caught inseminated female was isolated in the laboratory and cultured to make an "iso-female" stock. For hybridization experiments, stocks from the collections of the Genetics Foundation of The University of Texas and The University of Hawaii were used. Five of these species ( clavisetae~ hemipeza. silvestris, neoperkinsi, and planitibia) were originally placed in the genus Idiomyia Grimshaw; however, Carson, Clayton. and Stalker ( 1967) have shown that I diomyia is clearly a syno­nym of Drosovhila. The geographical distribution, collection number, collect­ing locality, collector, and date of collection of the stocks used in this investigation are given in Table 1. During the course of the investigation. freshly-caught stocks were received in the laboratory at irregular intervals. A number of crosses could not be attempted only because the two species involved did not happen to be available in the labora­tory at the same time. Species identifications were usually made by the field col­lectors in Hawaii; in general, we assumed that the identifications were correct. Culture. Most Hawaiian picture-winged species are relatively large, being 3 to 5 mm or more in body length. Their life cycle is longer than most other Dro­sophila species, and females must "age" 15 to 18 days before they begin laying eggs. Until they were 15 days old, about 15 pairs of flies were kept in culture vials 4 cm in diameter on a medium consisting of cornmeal, Brewer's yeast, molasses, and agar, with 0.5%propionic acid as a mold inhibitor. Vials were prepared as follows. A piece of tissue (Tomac) folded three or four times into a rectangle was placed on one side of the vial in contact with the sur­face of the cornmeal food. Bacto-yeast extract was diluted with a quantity of distilled water sufficient to make the pH 7 .0 and sterilized in an autoclave. Two or three drops of sterilized bacto-yeast water and sterile distilled water were placed on the tissue paper in the vial, and the paper was smoothed. The bacto­yeast water and distilled water served the purpose of providing humidity and nourishment for adult flies and larvae, and assured that the maximum number of eggs would be laid on the paper by the mature flies. The flies were transferred to fresh cornmeal food vials at five-or six-day in­tervals. When female flies were mature enough to start laying eggs, they were transferred to vials containing "Wheeler-Clayton" food (Wheeler and Clayton, 1965), which has a high protein content. Because "Wheeler-Clayton" food is softer than cornmeal food, flies were transferred at four-to five-day intervals to prevent them from sticking to the food. All vials were kept for at least two weeks at room temperature (20-22°C) to check for the presence of eggs and larvae. Most female flies laid eggs on the tissue inside the vial, and young larvae crawled into the food three to five days after emerging from the fertile eggs. Unlike most other Drosophila species, the Hawaiian species pupate in sand (Wheeler and Clayton, 1965); hence, bottles containing moist sand were pre- TABLE 1 Stocks Used in Crosses. I= grimshawi subgroup; II= punalua subgroup; III= planitibia subgroup. Species & subgroup Distri­bution Stock Code No. Type of line Collecting locality Collector Date l .(I) balio12tera Maui J51B8 J88Bl6 J88Bl7 ·iso iso iso · Maui Maui Maui Kaneshiro Kaneshiro Kaneshiro 12/66 6/67 6/67 2. (I) bostrycha Molokai J97N2 iso Molokai Muraoka 6/67 J58Ml2 iso Molokai Iwamoto 1/67 3. (I) crucigera Oahu, Kauai C53. ll HS.l WH34.2 J53C9 pair multiple iso ¥ iso Oahu Oahu Oahu Kauai Carson Spieth Wheeler Murphy 6/63 11/64 5/64 12/66 4. (I) discreta Maui L4G3 Lllp24 K75G31 iso iso iso Maui Maui Maui Carson Kambysellis Carson 10/67 8/67 10/67 5. (I) disjuncta Maui L4Gl7 J51Bl3 J51Bl4 J51Bl6 iso iso iso iso Maui Maui Maui Maui Carson Kaneshiro Kaneshiro Kaneshiro 8/ 67 12/67 12/67 12/67 6. (I) engyochracea Hawaii G5.1U J36C2 J39C3 iso iso iso Hawaii Hawaii Hawaii Kaneshiro · Murphy Murphy 8/65 11/66 11/ 66 7 ~I) fasciculisetae Maui Molokai K75P71-78 multiple Maui K90Pl4A iso 1' Maui Kambysellis Kambysellis 10/67 11/67 8.(I) gradata Oahu G55C5 G55Cl3 iso iso !i' ~ Oahu Oahu Murphy Murphy 5/66 5/ 66 9.(I) grimshawi Hawaii Maui Molokai Lanai "None" pK-9 J66Cl2 J99C4 J88Bl8 J78Ml multiple Molokai iso ~ Molokai iso ¥ Maui iso !i' Maui iso !i' Maui iso ~ Lanai Wheeler Carson Murphy Murphy Kaneshiro Iwamoto 9/65 7/ 64 2/ 67 6/67 6/67 4/67 10. (I) hawaiiensis Hawaii Oahu G25Bll Jl4B8 K44G6 K44G9 iso iso iso iso Haw a Haw a Hawa Haw a Kaneshiro Kaneshiro C:1rson Carson 2/66 7/66 9/67 9/67 11. ( IIIlemipeza Oahu JlG2 G55Bl0 G57A4 iso iso iso Oahu Oahu Oahu Carson Kaneshiro Heed 7/66 5/66 5/66 12. (I) .hirtipalpus Maui Molokai K75P82 iso Maui Kambysellis 10/ 67 13.(III~elanocephala Maui L25Pl2 multi r:ici Maui Kambysellis 1/68 The University of Texas Publication TABLE 1 (continued) Species & Distri-Stock Type Collecting Collector Date subgroup bution Code No. of line locality 14_( III heoEerkinsi Maui L35P21 wild o'o· Maui Kambysellis 3/68 Molokai L35P65 wild oo· Maui Kambysellis 3/68 L25Pl6 K83Cool wild o·o wild do Maui Maui Kambysellis Murphy 1/68 8/67 K75Pl25 Maui Kambysellis . 10/67 K75P46 Maui Kambysellis 10/67 15.( I) ochracea Hawaii Ll9B91-93 Hawaii Kaneshiro K78G Hawaii Carson 10/67 16.( I) orEhnoeeza Maui K75Pl38 · Maui Kambyse11is 10/67 17 .(I) orthofascia Maui JlOBS iso Maui Kaneshiro 7/66 JlOG25 iso Maui Carson 7/66 18.( II) eaucieuncta Hawaii J6Gl3 iso Hawaii Carson 7/66 J6Gl8 iso Hawaii Carson 7/66 19 _(III kicticornis Kauai None multi Unknown Unknown Unknown 20.(I) eilimana Hawaii, C53.3 iso Oahu Carson 6/63 Maui, Oahu WH44. 3 iso Oahu Wheeler Unknown Molokai, Kauai, Lanai? ·21.< III )elanitibia Maui K83C002 wild oo· Maui Murphy 8/67 K90Pl7 iso :; Maui Kambysellis 11/67 L25Pl5 iso ~ Maui Kambysellis 1/68 22 {II) eunalua Hawaii, L23Gl Oahu Carson 1/68 Oahu K53G2 iso Oahu Carson 1/67 K53B5 iso Oahu Kaneshiro 9/67 23{ I) recticilia Maui K75Pl37 Maui Kambysellis 10/67 24{1) silvarentis Hawaii K47G2 iso Hawaii Carson 9/67 Kl8M3 iso Hawaii Iwamoto 7/67 25SIII)silvestris Hawaii K66P2 iso Hawaii Kambysellis 10/67 K46G6 iso Hawaii Carson 9/67 K33G3 iso Hawaii Carson 8/67 26fI) seroati Hawaii K79Cl2 iso ~ Hawaii Murphy 8/67 K79C001 wild 00 Hawaii Murphy 8/67 K78G wild def Hawaii Carson 10/67 K80G wild o·o Hawaii Carson 10/67 27 {II) uniseriata Oahu Cl44.5A iso Oahu Carson 8/65 28{ I) vi llosiEedis Kauai Gl.lC iso Kauai Kaneshiro 8/65 pared before the third instar larvae began to craw1. Coarse sand was washed and sterilized by autoclaving for 10 to 15 minutes at 120°C. Quart bottles were filled to one-quarter with sterile sand, which was then moistened with distilled water. Two or three vials with crawling third instar larvae were placed in a bottle, which was then covered with a perforated cap lined wtih cheese cloth to prevent the larvae from crawling out. Larvae pupated two to three days after cra".Vling into the sand. The pupal stage lasted from 10 to 15 days. Young imagos were collected with a vacuum pump every 24 hours and placed in vials with cornmeal food and moistened paper. Because the Hawaiian species live in cool and humid areas in nature, young and mature flies were kept in humidifier boxes at a tem­perature of 16°C with 55% humidity. Crosses. Because small mass matings were more effective in obtaining hybrids than were pair matings, all experiments involved mass matings. Newly emerged individuals of each line of flies were collected from a sand bottle, and males and females were separated immediately after collection. From 10 to 20 flies were isolated by sex and kept in a cornmeal food vial until they were 20 days old. Mass hybrid matings were performed with 10 to 20 males of one species and 10 to 20 females of the other species. When possible, several sets of mass matings were made for each hybrid cross. Each mass mating was placed in "Wheeler-Clayton" food vials, ana the flies were changed to fresh food every four or five days. After the flies were transferred to a new vial, the evacuated vials were kept two to three weeks in the laboratory and periodically examined for the presence of larvae. If there was no sign of larvae, the vials were then discarded. In vials containing larvae, two or three drops of distilled water were placed on the tissue every three to four days to keep the tissue moist. Dead embryos could be detected by a mottled dark brown color characteristic of egg cases in which partial development, followed by death and decomposition, had occurred. As further evidence that the mottled egg cases contained dead em­bryos, after one month the P1 females were dissected and their reproductive tracts were examined for the presence of sperm. Some larvae died in the first instar ,others in the third instar, and some off­spring died in pupal stage. Some eggs developed completely and produced imagos with intermediate phenotypes. When hybrid third instar larvae were crawling up the paper in a vial, two or three of the larvae were removed, and the salivary gland polytene chromosomes were examined to verify that the larvae were hy­brids. The remaining third instar larvae were put into sand bottles and allowed to pupate in the sand and continue development to imagos. Within 24 hours after emergence, newly produced F1 hybrid flies were col­lected and separated by sex. About 20 flies of each sex were kept separately in cornmeal food vials until they were 20 days old. A part of the F1 hybrid flies were used for F1 hybrid brother-sister matings; also, both backcrosses and reciprocal crosses were attempted with both parental species whenever possible. All hybrid tests were carried out in the "Wheeler-Clayton" food vials. All of these tests were followed until the flies became old and began laying fewer eggs. Most of these late eggs were unfertilized. Some females from infertile matings were dissected and their reproductive tracts examined for the presence of sperm. The reproduc­tive organs of old hybrid males were examined to determine whether motile sperm was present, following the method used by Patterson ( 1942) and Dobzhan­sky and Mather (1961). Preparation of salivary gland chromosomes from hybrid larvae. Third instar hybrid larvae were placed in a few drops of Drosophila physiological Ringer's solution for dissection. The salivary glands were removed, cleared of fat, and placed in 1N HCI for 10 seconds. The glands were then removed and stained with lacto-acetic orcein for one-and-one-half to two minutes. The stained salivary glands were mounted in 45 %acetic acid, covered with a cover slip, and squashed. Finally, the slides were examined and photographed with a phase microscope. RESULTS Table 2 summarizes the number of hybridization tests which were attempted, TABLE 2 The Result of Interspecific Crosses with 28 Picture-winged Species of Hawaiian Endemic Drosophila. L =larva; P =pupa; I= imago; N =no hybrids produced; S =presence of sperm in pa­rental female reproductive organs; Open Squares = not tested; Small thick squares ·= F1 6 6 fertile; Large thick squares= chromosomal subgroups . the hybrid developmental stages observed, and the presence or absence of sperm in the sperm receptacles of the parental females in those instances where no hy­brids were produced. Among the 318 tests in this table, 30 cases of hybrid imagos, 8 cases of hybrid pupal stages, and 12 cases of hybrid larvae were obtained. There were five crosses in which sperm were seen in the receptacles of the parental fe­males, usually non-motile or barely motile, and from which no hybrid larvae were obtained. These crosses are: D. paucipuncta ~ ~ X D. discreta 6 6, D. pauci­puncta S? ~ x D. pilimana 6 6 , D. picticornis ~ ~ X D. discreta 6 6, D . .pilimana S? S? X D. discreta 6 J, and D. punalua ¥ ~ X D. discreta 6 J. It is remarkable that discreta males were involved in all but one of these crosses; they were, how­ever, hybridized successfully with D. hawaiiensis (F1 imagos produced). Table 3 shows the results of the 39 interspecific hybridizations which resulted in offspring. Eleven of these are reciprocal crosses and 28 are one-way crosses. Thus, of the total of 50 such crosses, 30, or 60%~ produced some Fi imagos. In five cases only male offspring were produced, and in five cases only females were produced. In some instances, the hybrid imagos were too weak or died too soon to be tested for fertility. The hybrid offspring of 28 crosses were tested for fer­tility, either by inbreeding or by backcrossing, or both. Some degree of fertility in one or both sexes was found in 17 of these 28 crosses. TABLE 3 Results of Successful Crosses Between Hawaiian Drosophila Species. (Approximate numbers of offspring are shown as follows: +++=more than 50; ++ = 20 to 40; +=less than 10.) Cross DeveloEmental Stage Fertility Remarks Female Male Larva Pupa Adult of Hybrids 1. balio12tera x bostrycha + + Died in late larval and pupal stages. 2. b"lli-:-£-f:"c'ra Y.: cr•lcigera + + + Imagos weak, not tested for fertility. crucigera x lxal.i.o~i.;era + 3. balio12tera x discreta ++ ++ 4. balio12tera x disjuncta + + disjuncta x balio12tera + + 5. balio12tera x qradata + Died in first instar. 6. balio12tera x grimshawi ++ ++ ++ ~~ fertile Only 4 cfci, weak, died. grimshawi x balioEtera ++ ++ ++ fertile Adults normal in ? " co sterile structure and num­bers. 7. balioEtera x hirtiEal12us ++ ++ ++ ~~ sterile No males produced. 8 . balioetera x villosieedis ++ ++ 9. bostrycha x crucigera ++ Larvae died in 2nd instar. crucigera x bostrycha +++ ++ ++ rfcf semi-Many larvae died; no sterile 'i''i' produced. j). bostrycha x disjuncta ++ ++ ++ ~~ fertile Both reciprocal cfrf sterile crosses produced many oo and 'i''i'. disjuncta x bostrycha ++ ++ ++ i 'i' fertile o'o" sterile 11. crucigera x disjuncta ++ Many larvae which died. 12. crucigera x disjuncta + + + 'i' 'i' sterile Few imagos produced, cfcf not tested for fertility. disjuncta x crucigera + 13. crucigera x grimshawi +++ +++ +++ ~ 'i' fertile Both reciprocal oo semi-crosses produced a sterile large number of oo and ~~­ grimshawi x crucigera +++ +++ +++ 'i''i' fertile 00 semi- sterile TABLE 3 (continued) Cross Develoemental Stage Fertility Remarks Female Male Larva Pupa Adult of Hybrids 14. crucigera x uniseriata + 15. cruci9era x villosieedis + + 16. disjuncta x grimshawi + ·+ grimshawi x disjuncta + + + S?S? fertile Too few ses Dewlopmental Stage F1 X F1 of: Lan· a Pupa Adult 1. (grimshawi ~ X balioptera o ) 2. (bostrycha 9 X disjuncta c3 ) (disjunc:a ~ X bostrycha o) 3. ( cruc:gera 9 X grimshawi o ) + + + (grimshawi 9 X crucigera o) ++ ++ ++ 4. (engyochracea 9 X villosipedis 6 ) 5. (grimshawi 9 X bostrycha 6) 6. ( ha.waiiensis ~ X gradata o ) + ++ + ++ + ++ 7. (ha:vaiiensis ~ X silvarentis 6) 8. (hau:aiiensis 9 X hirtipalpus o ) 9. (hemipeza 9 X silvestris 6' ) + + + 10. (hemipeza 9 X planitibia c3) 11. (orthofascia 9 X villosipedis o ) 12. (paucipuncta 9 X uniseriata 6 ) (uniseriata 9 X paucipuncta o ) 13. (picticornis !? X crucigera 6 ) 14. (pilimana !? X grimshawi o ) + inbreeding or backcross tests could be performed. The reciprocal cross, using the same stocks, produced a few larvae which died in the larval stage. D. balioptera 9 9 (Maui) crossed to D. discreta t t (Maui) gave a large num­ber of hybrid larvae, most of which pupated in the sand. No hybrid adults were obtained from them, however. The reciprocal cross was completely negative. When D. balioptera 9 9 (20; Maui) were crossed to D. disiuncta t t (20; Maui) both larvae and pupae were produced but they died at this stage. The same re­sult was obtained in the reciprocal cross. D. balioptera 9 9 (15; Maui) crossed to D. gradata t t (15; Oahu) yielded a few larvae which died in the first instar. The reciprocal cross was negative. D. balioptera 9 9 (20; Maui) crossed to D. grimshawi t t (20; Molokai) yielded many viable hybrid females but only four very weak, abnormal males which died within a week. Consequently, a backcross was performed only one­way. the hybrid females being crossed with both types of parental males. The F1 (balioptera 9 X grimshawi t) ~ 9 crossed to D. balioptera t t produced a few larvae and a few imagos, while the cross to D. grimshawi t t produced a large number of male and female offspring; these resembled D. grimshawi in appearance more than they did D. balioptera. The male offspring were com­pletely sterile, while the females were fertile or semi-fertile. The reciprocal hybrid cross, D. grimshawi 9 ~ X D. balioptera t t, produced a number of Fi hybrids of both sexes. The inbreeding test, Fi X Fh failed to pro­duce offspring, while the backcross, F1 (grimshawi ~ X baliontera t) ~ ~ crossed to D. balioptera t t, produced offspring which were more like D. balioptera than D. grimshawi. The alternate backcross, to D. grimshawi t t, produced offspring which were more like D. grimshawi than D. balioptera. When the two parental female types, D. balioptera and D. grimshawi, were crossed to the Fi hybrid (grimshawi ~ X balioptera t ) t t simultaneously, no offspring were produced. Thus the inbreeding and backcross tests indicate that the hybrid females were fertile and that the hybrid males were sterile. D. balioptera 9 ~ (20; Maui) crossed to D. hirtipalpus t t (20; Maui) yielded a large number of female hybrids but no males. Some larvae died in each larval stage. A great many backcross tests were attempted with both parental males; the F, (balioptera 9 X hirtipalpus t) ~ ~ were crossed to both D. balioptera t t and D. hirtipalpus t t, but no offspring were obtained from them. The backcross to female parental types produced, uncommonly, Fi hybrid females that were completely sterile. No F1 hybrids were obtained from the reciprocal cross, D. hirtipalpus ~ ~ X D. balioptera t t. D. balioptera 9 ~ (Maui) crossed to D. villosipedis t t (Kauai) yielded many hybrid larvae, some reaching the pupal stage. However, no F1 imagos were ob­tained from the pupae. The reciprocal cross failed to produce hybrids. D. bostrycha 9 ~ (15~ Molokai) crossed to D. crucigera t t (15; Oahu) pro­duced larvae, but they died in the second instar. No pupae or imagos were pro·· duced. The reciprocal cross yielded only male hybrids; they were backcrossed only to D. crucigera since no D. bostrycha were available at that time. D. cruci­gera ~ ~ crossed to F 1 ( crucigera ~ X bostrycha t ) t t produced very few larvae, which died in the early larval stage. In addition to the cross tests for F 1 hybrid fertility, dissections of the hybrid male reproductive organs showed motile sperm. Yang and Wheeler: Hybridization in Hawaiian Drosophila 147 The non-viable larvae produced in this backcross test suggested that the Fi hybrid males were semi-fertile. D. bostrycha ~ ~ (20; Molokai) crossed to D. disiuncta J J (15; Maui), and the reciprocal cross (20 J, 20 9), yielded both male and female hybrids. The F1 hybrid inbreeding tests of both Fi (bostrycha 9 X disiuncta J) 9 9 crossed to Fi (bostrycha ~ X disiuncta J) J J, and Fi (disiuncta ~ X bostrycha J) ~ ~ crossed to Fi (disiuncta 9 X bostrycha J) J J failed to produce offspring. The backcrosses, D. bostrycha ~ ~ crossed to F1 (bostrycha ~ x disiuncta J) J J, D. disiuncta ~ 9 crossed to F1 (disiuncta ~ X bostrycha J) J J, and D. disiuncta ~ ~ crossed to Fi (bostrycha ~ X disiuncta J ) J J, failed to produce offspring. On the other hand, hybrid females crossed to parental males produced offspring much like the parental types. Fi (bostrycha ~ X disiuncta J) ~ ~ crossed to D. bostrycha J J produced hybrids more like D. hostrycha than D. disiuncta, and F1 (disiuncta 9 X bostrycha J) ~ 9 crossed to D. disiuncta J J produced hybrids more like D. dis;uncta than D. bostrycha. Cross tests showed both female hybrids to be fertile, whereas both male hybrids were completely sterile. D. bostrycha and D. disiuncta are homosequential, and they are very similar to each other. D. crucigera ~ 9 (Oahu) crossed to D. discreta J J (Maui) yielded a number of non-viable larvae. The reciprocal cross failed to produce hybrids. From D. crucigera ~ 9 (Oahu) crossed to D. disiuncta J J (Maui), only a few F1 hybrids were obtained, and only a one-way backcross was attempted. F1 (crucigera 9 X disiuncta J) 9 9 crossed to D. disiuncta J J failed to produce offspring. The re­ciprocal cross, D. disiuncta 9 9 (20; Maui) crossed to D. crucigera J J (20; Kauai) produced hybrid larvae which died in the early instars. However, D. disiuncta 9 9 (Maui) crossed to J J of D. crucigera from Oahu failed to produce any hybrids. In crosses between D. crucigera and D. grimshawi, Clayton (cited in Carson, 1966) reported the F1 hybrid males to be sterile. In the crosses reported here, considerable fertility of such hybrid males was observed. It is possible that this is due to genetic differences in the strains used. D. crucigera ~ 9 (20; Oahu; and 20; Kauai) crossed to D. grimshawi J J (20; Molokai), and the reciprocal cross, produced a large number of male and female hybrids. The inbreeding test, Fi (crucigera 9 X grimshawi J) 9 9 crossed to F1 (crucigera 9 X grimshawi J ) J J , resulted in few offspring~ all of which were more like D. crucigera than D. grimshawi. The other inbreeding test, Fi (grim­shawi 9 X crucigera J) 9 9 crossed to Fi (grimshawi 9 X crucigera J) J J, pro­duced many offspring which were more like D. grimshawi than D. crucigera. The backcross test, F1 (crucigera ~ X grimshawi J) 9 9 crossed to D. crucigera J J, yielded a larger number of offspring which were more like D. crucigera than D. grimshawi. The F1 (crucigera 9 X grimshawi J) 9 9 crossed to D. grimshawi J J also yielded a large number of D. grimshawi-like offspring. The F1 (grim­shawi 9 X crucigera J) 9 9 crossed to D. crucigera J J yielded a large number of D. crucigera-like offspring, and the F1 (grimshawi 9 X crucigera J) 9 9 crossed to D. grimshawi J J yielded a large number of D. grimshawi-like off­spring. However, D. grimshawi 9 9 crossed to the Fi (grimshawi 9 X crucigera J) J J failed to produce offspring, as did D. crucigera 9 ~ crossed to F1 (grim­shawi 9 X crucigera J) J J. Cross tests showed Fi hybrid females to be com­ pletely fertile, while F1 hybrid males were semi-fertile, because most hybrid males had no motile sperm, but a very few hybrid males did produce motile sperm. D. crucigera 2 2 (20; Oahu) crossed to D. zmiseriata ~ ~ (20; Oahu) produced a few weak larvae that died early. However, female D. crucigera from Kauai crossed to these males failed to produce any hybrids. The reciprocal cross, D. uniseriata 2 2 crossed to D. crucigera J ~ from both Oahu and Kauai failed to produce hybrids. D. crucigera 2 2 (Oahu; Kauai) crossed to D. villosipedis ~ ~ (20; Kauai) produced pupae but no hybrid imagos emerged from them. Some of the larvae died in the late larval stages. The reciprocal cross failed to produce hybrids. When D. disiuncta 2 2 (20; Maui) were crossed to D. grimshawi ~ ~ (20; Maui), hybrid larvae developed to the third instar, and a few reached the pupal stage. No hybrid imagos were produced. The reciprocal cross produced a few hybrid imagos, and these were used in a one-way backcross test. The F1 (grim­shawi 2 X disiuncta J) 2 2 crossed to D. grimshawi J J yielded a large number of offspring which were more like D. grimshawi than D. disiuncta. Even though these two species are chromosomally homosequential and morphologically very close to each other, it was difficult to get hybrids from these two species. D. engyochracea 2 2 (20; Hawaii) crossed to D. disiuncta ~ J (20; Maui) yielded a few larvae, but these died in the early stages. The reciprocal cross failed to produce hybrids. D. engyochracea 2 2 (20; Hawaii) crossed to D. grimshawi J J (20; Molokai) produced a few morphologically intermediate hybrid males and females; most of the hybrid larvae died in the second and third instars. No hybrid offspring were produced, however, when males from Maui were used in this cross. Only a one­way backcross test was done. D. eng~rochracea 2 2 crossed to F 1 ( engyochracea 2 X grimshawi J) J J produced no offspring. The reciprocal cross, D. grimshawi 2 2 crossed to D. engyochracea J J, failed to produce hybrids. D. engyochracea 2 2 (20; Hawaii) crossed to D. villosipedis J J (20; Kauai) yielded a good number of hybrid females and males, which were phenotypically intermediate between the parental types. The hybrid cross tests were performed by inbreeding and backcrossing the F1 hybrids with both parental types. The F, (engyochracea 2 X villosipedis J ) 2 2 crossed to their sibling F 1 males produced no offspring. Offspring-of F, (engyochracea 2 x villosipedis J ) 2 2 crossed to the two types of parental males were similar to each parental type, respectively. Another hybrid backcross test using two types of parental females crossed to the F1 hybrid (engyochracea 2 X villosipedis J) J J did not produce any offspring. Cross tests showed that the F1 hybrid females were all fertile, but the Fi hybrid males were completely sterile. On the other hand, the reciprocal cross, D. vil­losipedis 2 2 crossed to D. engyochracea J J, did not produce hybrid flies. D. gradata 2 ·2 (20; Oahu) crossed to D. hawaiiensis J J (20; Hawaii), and the reciprocal cross, produced Fi hybrid females and males. The F1 hybrid in­breeding test, F 1 ( hawaiiensis 2 X gradata J ) 2 2 crossed to F1 ( hawaiiensis 2 X gradata J) J J, produced a number of offspring. The backcross test, Fi (ha­waiiensis 2 X gradata J ) 2 2 crossed to D. hawaiiensis J J , produced off spring, but the other backcross test, Fi (hawaiiensis 2 X gradata J) 2 2 crossed to D. gradata J J , did not. D. hawaiiensis 2 2 crossed to F 1 ( hawaiiensis 2 X gradata J) J J did not produce offspring, while D. gradata 2 2 crossed to F1 (hawaiiensis 2 X g_radata J) 6 6 produced few offspring. The Fi hybrids (gradata 2 X ha­waiiensis 6) were not tested. These backcross and inbreeding tests showed that hybrid females were fertile, whereas hybrid males were semi-fertile. A large number of hybrids from D. grimshawi 2 2 (20; Maui) crossed to D. bostrycha 6 J (20; Molokai) was obtained, and inbreeding and backcross tests were carried out. The hybrid inbreeding test, Fi (grimshawi 2 x bostrycha 6) 2 2 crossed to sibling 6 6 , yielded very few larvae and imagos. The backcross test, F1 (grimshawi 2 X bostrycha 6) 2 2 crossed to D. bostrycha 6 6, yielded a few D. bostrycha-like offspring. The backcross test, F1 (grimshawi 2 x bostrycha 6) 2 2 crossed to D. grimshawi 6 6, yielded D. grimshawi-like offspring. D. grimshawi 2 2 crossed to Fi (grimshawi 2 X bostrycha 6) 6 6, produced a large number of offspring. D. grimshawi and D. bostrycha are very closely related species, and their hybrid males and females were all fertile. The reciprocal cross, D. bostrycha 2 2 crossed to D. grimshawi 6 6, failed to produce hybrids. D. grimshawi 2 2 (15; Molokai) crossed to D. hemipeza 6 J (15; Oahu) pro­duced only one larva which developed into the third larval instar. Because only one larva was obtained, it was dissected so that its chromosomal sequences could be studied. There was, therefore, no way of knowing how far the hybrid would have developed. However, D. grimshawi (Maui) 2 2 crossed to D. hemipeza (Oahu) 6 6 failed to produce hybrids. The reciprocal cross, D. hemipeza 2 2 (Oahu) crossed to two strains of D. grimshawi 6 6, failed to produce hybrids. D. grimshawi 2 2 (20; Maui) crossed to D. orthofascia 6 6 (20; Maui) pro­duced only male hybrids. Many offspring died in the larval stages. However, D. grimslzawi (Molokai) 2 2 crossed to D. orthofascia (Maui) 6 6 failed to produce hybrids. The F1 hybrid males, F1 (grimshawi 2 x orthofascia 6 ), were crossed to D. grimslzawi 2 2, but no offspring were obtained. The reciprocal cross, D. orthofascia (Maui) 2 2 crossed to D. grimshawi (Molokai) c 6, produced a few non-viable larvae which died in early larval stage. D. orthofascia (Maui) 2 2 crossed to D. grimshawi (Maui) 6 6 failed to produce hybrids. D. grimshawi 2 2 (Molokai) crossed to D. villosipedis 6 6 (Kauai) yielded many larvae most of which died in the second or third instar larval stage. Only a few weak hybrid males and females were produced, and all of the males died before they matured. The one-way backcross, Fi (grimshawi 2 x villosipedis 6) 2 2 crossed to D. grimshawi 6 6, failed to produce offspring. The reciprocal cross, D. villosipedis 2 2 crossed to D. grimshawi 6 6, failed to produce hybrids. D. hawaiiensis 2 2 (Hawaii) crossed to D. discreta 6 6 (Maui) yielded a few hybrid larvae, most of which died in larval and pupal stages, with only one fe­male and two male hybrid imagos being produced. The hybrid cross test was not performed. The reciprocal cross, D. discreta 2 2 crossed to D. hawaiiensis 6 6, failed to produce hybrids. D. hawaiiensis 2 2 (20; Hawaii) crossed to D. disiuncta 6 fl (20; Maui) yielded non-viable larvae, all of which died in the early larval stage. The re­ciprocal cross failed to produce hybrids. D. lzawaiiensis 2 2 (15; Hawaii) crossed to D. silvarentis 6 6 (15; Hawaii) produced a few adult Fi hybrid females and males. The hybrid inbreeding test, Fi X Fi, produced no offspring. The reciprocal cross, D. silvarentis 2 2 crossed to D. hawaiiensis 6 6, was attempted but no Fi hybrids were obtained. D. hawaiiensis 2 2 (Hawaii) crossed to D. hirtipalpus t t (Maui) produced many hybrid larvae, all of which pupated. However, few hybrid imagos were obtained from these pupae; the reciprocal cross failed to produce offspring. No offspring were obtained from the Fi hybrid inbreeding test. D. hawaiiensis 2 2 (20; Hawaii) crossed to D. villosipedis t t (20; Kauai) produced hybrid larvae most of which died in second and third larval instars. Only three adult males were obtained from the many hybrid larvae. No hybrids were obtained from the reciprocal cross. The hybrid cross test was not made. D. hemipeza 2 2 (15; Oahu) crossed to D. silvestris t 6 (15; Hawaii) pro­duced a large number of hybrid females and males. The hybrid inbreeding test, F1 X F1, yielded many non-viable larvae, and only a few F2 females were ob­tained. The backcross test, F1 (hemipeza 2 X silvestris 6) 2 2 crossed to D. hemi­peza 6 6, resulted in a large number of pupae, but no offspring emerged. D. hemipeza 2 2 crossed to F 1 ( hemipeza 2 X silvestris t ) 6 t , yielded a large num­ber of pupae and adults. The backcross test was not done with D. silvestris fe­males. Results of cross testing showed hybrid males and females to be fertile. When the reciprocal cross was attempted, no hybrid flies were obtained. D. hemipeza 2 2 (15; Oahu) crossed to D. planitibia 6 6 (15; Maui) yielded a large number of hybrid females and males. The hybrid inbreeding test, F1 X Fi, produced no offspring. Both backcross tests, Fi (hemipeza 2 X planitibia 6) 2 2 crossed to D. hemipeza t t, and to D. planitibia 6 t, produced offspring which were like their parental types, respectively. However, D. hemipeza 2 2 crossed to Fi (hemipeza 2 X planitibia 6) 6 t did not produce offspring. Hybrid cross test­ing showed all hybrid females to be fertile and hybrid males to be completely sterile. No reciprocal cross was made because of a shortage of D. planitibia fe­males. D. orthofascia 2 2 (20; Maui) crossed to D. villosipedis 6 6 (20; Kauai) pro­duced hybrid males and females. The hybrid inbreeding test, F1 X F1 , failed to produce off spring. The back cross tests, Fi (orthof ascia 2 X villosipedis 6 ) 2 2 crossed to D. orthofascia 6 6. produced non-viable larvae and pupae, so that no imagos were obtained. The other backcross, Fi 2 2 crossed to D. villosipedis t t, failed to produce offspring. D. villosipedis 2 2 crossed to Fi (orthofascia 2 X vil­losipedis 6 ) 6 6 did not produce any offspring. Hybrid cross testing showed all male hybrids to be sterile, and female hybrids to be semi-fertile. No hybrids were obtained from the reciprocal cross, D. villosipedis 2 2 crossed to D. orthofascia 6 6. D. paucipuncta 2 2 (20-25; Hawaii) crossed to D. uniseriata 6 6 (20-25; Oahu) and the reciprocal cross produced hybrid females and males. However, in the reciprocal cross, a larger number of more vigorous hybrids was produced than in the cross, D. paucipuncta 2 2 crossed to D. uniseriata 6 6. Both inbreeding tests. Fi males to Fi females from both original crosses, failed to produce offspring. The backcross tests~ Fi (paucipuncta 2 X uniseriata t) 2 2 crossed to D. pauci­puncta 6 6 and to D. uniseriata 6 6, F1 (uniseriata 2 X paucipuncta t) 2 2 crossed to D. paucipuncta 6 t, and to D. uniseriata 6 6, all produced offspring. D. paucipuncta 2 2 and D. zmiseriata 2 2 crossed to Fi ( uniseriata 2 X pauci­ puncta J) J J, failed to produce offspring. The results of hybrid cross tests in­dicated that all hybrid females were fertile, while all hybrid males were com­pletely sterile. D. picticornis 9 9 (Kauai) crossed to D. balioptera J J (Maui) produced a few non-viable larvae, all of which died in an early stage of development. No hybrids were obtained from the reciprocal cross. D. picticornis 9 9 (20; Kauai) crossed to D. crucigera J J (20; Oahu) pro­duced only eight male and eight female hybrids. The hybrid inbreeding test, Fife­males crossed to Fi males, produced no offspring. The reciprocal cross, D. cruci­gera 9 9 crossed to D. picticornis J J, did not succeed. As can be seen on Tabl~ 2, this was the only cross between members of different species subgroups which resulted in hybrid imagos. Although the Fi X Fi inbreeding test produced no offspring, the backcrosses, which might have shown whether the apparent sterility was real or not, was not carried out. D. picticornis 9 9 (20; Kauai) crossed to D. uniseriata J J (20; Oahu) pro­duced very few larvae, and, except for one hybrid larva which pupated, they a:U died in the larval stage. However, this pupa was non-viable and no imago emerged from it. The reciprocal cross, D. uniseriata 9 9 crossed to D. picticornis J J, pro­duced no hybrid offspring. D. pilimana 9 9 (20; Oahu) crossed to D. disfuncta J J (20; Maui) yielded many non-via1ble larvae, and only three hybrid females were produced. The hy­brid cross tests were not made. The reciprocal cross, D. disjuncta 9 9 crossed to D. pilimana J J, produced no hybrids. D. pilimana 9 9 (20; Oahu) crossed to D. grimshawi J J (20; Molokai) pro­duced a large number of both hybrid males and females, but D. pilimana 9 9 (Oahu) crossed to D. grimshawi J J (Maui) failed to produce hybrids. Hybrids were tested by inbreeding and backcrossing. The Fi X Fi cross yielded non-viable larvae which died in the early larval stages. The backcross tests, Fi (pilimana 9 X grimshawi J) 9 9 crossed to both parental males, D. grimshawi and D. pilimana, yielded offspring which were like their parental types, respectively. However, when the two types of parental females, D. grimshawi and D. pilimana, were crossed to the Fi (pilimana 9 x grimshawi J) J J, no offspring were produced. This hybrid cross test showed that the hybrid females were all fertile, whereas some hybrid males were semi-fertile. From the reciprocal cross, D. grimshawi 9 9 from Maui and Molokai crossed to D. pilimana J J, no Fi hybrids were obtained. D. pilimana 9 9 (20; Oahu) crossed to D. villosipedis J J (20; Kauai) yielded a few non-viable larvae which died in the second instar. No hybrid adults were obtained from this cross. The reciprocal cross produced no hybrids. Cytological observations on hybrid salivary gland chromosomes. The chromo­somal relationships among the Hawaiian picture-winged Drosophila species were analyzed, using D. grimshawi as the theoretical standard species, by Carson, Clayton and Stalker (1967) and by Carson and Stalker (1968a, b, c). They have pointed out that most of the species which have been studied had a metaphase configuration of five long rods and one dot, with a few species having six rods. They also noted the existence of three groups of homosequential species in which the banding sequences of the polytene chromosomes are apparently identical. Each homosequential group forms the nucleus of a species cluster, or complex, which contains other species which differ from the group sequences by only one or a few inversions. By appropriate analysis of the various banding patterns, they have been able to show the number of inversion differences which have accumu­lated between these species clusters. In the present study, the polytene chromosomes of the third instar larvae were examined. The salivary gland polytene chromosomes in hybrid larvae were usually incompletely synapsed, even though the banding sequence was the same. Figure 1 shows some of the hybrid salivary polytene chromosomes which are unpaired in certain portions and are split to some degree at the ends. Most of the dot chromosomes were not synapsed at all. In the descriptions of hybrid larval chromosomes which follows, the relative distinctiveness of the different species is suggested by citing the chromosome in­version differences reported by Carson, Clayton and Stalker ( 1967) and by Car­son and Stalker ( 1968a, b, c). The hybrid larvae from D. balioptera ~ ~ crossed to D. disiuncta oo, and their reciprocal cross, D. disiuncta ~ ~ crossed to D. balioptera oo, had almost un­synapsed chromosomes on each of the five rods and completely unpaired dot chromosomes. There were seven inversion differences between these two species, as is shown by Carson and Stalker (1968a). The hybrid larvae from D. balioptera ~ ~ crossed to D. hirtipalpus o o had five entirely unsynapsed rod chromosomes and two haploid completely unsynapsed rod chromosomes. There were six in­version differences between these two species. The hybrid larvae from D. baliop­tera ~ ~ crossed to D. villosipedis o o had five partially unsynapsed rod chromo­somes which were split at the ends, and the dot chromosome was paired. There were two inversion differences between these two species. The hybrid larvae from D. crucigera ~ ~ crossed with D. disiuncta o o had five partially unsynapsed rod chromosomes and two completely unsynapsed haploid dot chromosomes. There were eight inversion differences between these two species. The hybrid larvae from a cross between D. crucigera ~ ~ and D. grim­shawi 6 6. and the reciprocal cross, D. grimshawi 9 9 crossed to D. crucigera oo, usually showed five unsynapsed rod chromosomes. The dot chromosomes were not paired at all. There were four inversion differences between the3e two species. A hybrid larva from D. crucigera 9 ~ crossed to D. villosipedis o6 showed five mostly unsynapsed rod chromosomes, and the one dot chromosome was not synapsed at all. There 'vere three inversion differences between these two species. The hybrid larvae from D. engyochracea 9 9 crossed to D. grimshawi o o had five unsynapsed rod chromosomes. The one dot chromosome was not paired at all. There were four inversion differences between these two species. The hybrid larvae from D. engyochracea ~ 9 crossed to D. villosipedis o o had five partially unsynapsed rod chromosomes. The one dot chromosome was completely unpaired. There were three inversion differences between these two species. The cross between D. gradata 9 9 and D. hawaiiensis o o , and the reciprocal cross, showed five partially unsynapsed hybrid rod chromosomes and two un­paired haploid dot chromosomes. These two species have two homozygous inver­sions in the same position on the second and third chromosomes, so that there were six inversion differences between them. Fm. 1. Representative salivary gland chromosomes of hybrid 3rd instar larvae. 1. D pilimana ~ X D. grimshawi 8. 4. D. grimshawi ~ X D. hemipeza 8 . 2. D. crucigera ~ X D. grimshawi 8 . 5. D. balioptera ~ X D. villosipedis 8 . 3. D. crucigera ~ X D. villosipedis 8. 6. D. paucipuncta ~ X D. uniseriata o. 7. D. balioptera ~ X D. hirtipalpus 6. 10. D. hawaiiensis ~ X D. silvarentis 6. 8. D. hemipeza ~ X D. silvestris 6. 11. D. hawaiiensis ~ X D. hirtipalpus 6. 9. D. crucigera ~ X D. discreta 6. 12. D. hawaiiensis ~ X D. villosipedis 6. D. grimshawi ~ ~ crossed to D. hemipeza 6 6 produced only one F1 male larva; it had four entirely unsynapsed rod chromosomes which had many inversions on the D. hemipeza side of the chromosomes. The dot chromosomes and the X and Y chromosomes did not pair at all. There were 14 inversion differences between these two species. The hybrid larvae from D. grimshawi ~ ~ crossed to D. ortho­fascia 6 6 showed five partially unsynapsed rod chromosomes and a completely unsynapsed pair of haploid dot chromosomes. There were four inversion differ­ences between these two species. The hybrid larvae from D. hawaiiensis ~ ~ crossed to D. hirtipalpus ~ti had five partially unsynapsed rod chromosomes, all of which were split at the ends. The dot chromosome was not synapsed at all. Because these two species share the same inversion on the second chromosome, there were ten inversion differences between the two species. The hybrid larvae from D. hawaiiensis ~ ~ crossed to D. silvarentis ~ J showed five mostly synapsed rod chromosomes and unsynapsed dot chromosomes. These two species have the same fixed homozygous inversions on the X, second, third, and fourth chromosomes, and one heterozygous inversion on the third and fifth chromosomes. D. hawaiiensis had one heterozygous inver­sion on the fourth chromosome, while D. silvarentis had one heterozygous inver­sion on the fifth chromosome. The hybrid larvae from D. hawaiiensis ~ ~ crossed to D. villosipedis ~ti had five partially unpaired banding sequences in rod chro­mosomes and a completely unsynapsed dot chromosome. There were eight inver­sion difference3 between these two species. The hybrid larvae between D. hemipeza ~ ~ crossed to D. planitibia ~ J had five incompletely synapsed rod chromosomes and a completely unpaired dot chromosome. Because these two species are closely related in their chromosomal sequences, they shared the same four fixed inversions on the X chromosome, one fixed inversion on the third and fourth chromosomes, and two heterozygous in­versions on the X chromosome. The only inversion differences between them were one additional fixed homozygous Xc2 inversion and heterozygous Xr/+ inversion on D. hemipeza, with D. planitibia having one fixed homozygous in­version, Xr. There was only one inversion difference between these two species. The hybrid larvae from D. hemipeza ~ ~ crossed to D. silvestris ~ J had five partially unsynapsed rod chromosomes and an entirely unpaired dot chromosome. There were eight inversion differences between these two species. The hybrid larvae from D. orthofascia ~ ~ crossed to D. villosipedis J ~ had five partially unsynapsed rod chromosomes and a completely unsynapsed dot chromosome. There were three inversion differences between these two species. Hybrid larvae from a cross between D. paucipuncta ~ ~ and D. uniseriata ti J, and the reciprocal cross had four partially unsynapsed chromosomes, which were split at the ends. The X, dot, and one rod chromosomes were not synapsed at all. Even though these two species are homosequential, and there are no major chro­mosomal differences between them, they showed incomplete synapsis between the chromosome pairs. Both species shared two fixed inversions on the fourth chromosome and one homozygous inversion on the X and third chromosomes. D. paucipuncta had five rods and one dot, while D. uniseriata had six rods. The hybrid larval polytene chromosomes from D. picticornis ~ ~ crossed to D. crucigera ti J, members of different subgroups, had five entirely unsynapsed rod chromosomes. The dot chromosomes were not paired at all. There were 16 in­version differences between these two species. The hybrid larvae from D. pilimana ~ ~ crossed to D. disjuncta J J had five partially unsynapsed rod chromosomes and a completely unsynapsed dot chro­mosome. There were six inversion differences between these two species. A cross between D. pilimana ~ ~ and D. grimshawi ti J had five unsynapsed rod chromo­somes. The dot chromosome did not pair at a11. There were two inversion differ­ences between these two species. The Fi hybrid phenotypes. Most of the F1 hybrids were intermediate pheno­typically between their parents, but some hybrids resembled one of their parental phenotypes more strongly. Figures 2 and 3 show comparisons between the hybrid wing patterns and shapes and those of the parental types. The cross between D. balioptera and D. crucigera produced hybrids which were more similar to D. crucigera. These two species are phenotypically (body pig­mentation) very similar, with the only distinctive differentiating characters being the wing patterns. The hybrid m cross vein was concaved posteriorly, more like that of D. balioptera than D. crucigera. The hyaline spots on the wings were 3 4 5 6 8 . Frn. 2. The wing patterns of parental Hawaiian picture-winged Drosophila species. 1. D. balioptera. 5. D. crucigera. 2. D. grimshawi. 6. D. orthofascia. 3. D. bostrycha. 7. D. disjuncta. 4. D. hawaiiensis. 8. D. villnsipedis. Yang and Wheeler: Hybridization in Hawaiian Drosophila 157 FIGURE ~Continued 9. D. planitibia. 13. D. hemipeza. 10. D. paucipuncta. 14. D. uniseriata. 11. D. engrochracea. 15. D. picticornis. 12. D. gradata. 16. D. pilinurna. intermediate; that is, the hyaline spots around the outside of a small dark brown spot in the second posterior cell were more like those of D. crucigera, while two connected hyaline spots contained within the submarginal and first posterior cells near the middle of the cells were more like those of D. balioptera. The hybrids from D. balioptera ~ ~ crossed to D. grimshawi ) 2 2 crossed to D. villosipedis 1i 1i, and the hybrids from the reciprocal cross, hybrid (grimshawi 2 X crucigera 1i ) 2 2 crossed to D. villosipedis t> 1i, produced a very large number of larvae, but many of them died in the larval stage and only a few adults were produced. The com­plex offspring were stronger and in better condition than the hybrids of D. grim­shawi 2 2 crossed to D. villosipedis 1i t> . These complex imagos were more like D. grimshawi and D. crucigera in body color and mesonotum, but their wings more closely resembled those of D. villosipedis. D. crucigera 2 2 crossed to D. villosipedis 1i J failed to produce hybrid offspring. Female (grimshawi2 X crucigerat>) hybrids crossed to D. disiuncta 1i t> pro­duced offspring, and their phenotypic characters were a combination of the three species. The body color and picture-winged pattern were similar to those of D. grimshawi and D. disiuncta, but the small dark brown dot in the second posterior cell was inherited from D. crucigera. The complex hybrid [ ( crucigera 2 X grimshawi J ) 2 X engyochracea t> J 2 2 FIG. 6. The wing patterns of hybrids from complex crosses. 1. [ (D. grimshawi ~ X D. crucigera 3) ~ X D. grimshawi 6] ~ X D. disjuncta d. 2. (D. grimshawi ~ X D. crucigera 3) ~ X D. villosipedis 6. 3. (D. balioptera ~ X D. grimshawi 6) ~ X D. crucigera 6. +.D. picticornis ~ X [ (D. crucigera ~ X D. grimshawi 6) ~ X D. crucigera 6] 6. 5. [ (D. grimshawi ~ X D. crucigera 3) ~ X D. crucigera d J~ X D. engyochracea d. 6. 6. [ (D. crucigera ~ X D. grimshawi 6) ~ X D. engyochracea 6] ~ X D. crucigera 6. 7. (D. grimshawi ~ X D. crucigera d) ~ X D. disjuncta 3. 8. (D. crucigera ~ X D. grimshawi 6) ~ X D. engyochracea 6. crossed to D. crucigera d 6 yielded a number of offspring, and they were much like the hybrids from (crucigera~ X grimshawi 6) ~ ~ crossed with D. engyo­chracea. The offspring had a darker brown pigment in the wing than is present in D. engyochracea. The complex hybrid [ (grimshawi ~ X crucigera 6) ~ X crucigera 6 J ~ ~ crossed to D. engyochracea 6 6 yielded a number of offspring, and they were much like the hybrids from (crucigera ~ X grimshawi iS) ~ ~ crossed with D. engyochracea. The offspring had a darker browrt pigment in the wing than is present in D. engyochracea. The complex hybrid [ (grimshawi~ x crucigeraoN Where I* =number of atoms activated (radioisotope atoms formed) per cm3 of sample per second; cf> =neutron flux, neutrons per cm'.! per sec.; u = reaction cross section, cm2 ; N =number of target atoms per cm3 in the sample. For any activation time, t, the number of atoms activated is I*t. But, after forma­tion, the atoms of the radioisotope may decay. So there is a competition between production and decay processes. The rate of decay is given by: R = N*,\ where R =number of atoms decaying per cm3 sample per second~ ,\=decay constant, 0.693/half-life (in seconds) of the radioisotope; N* = number of radioisotope atoms per cm3 of sample. After a given time of activation, t, the radioactivity of the sample ( disintegra­tions per second per cm3 ) from the target element may be expressed as: A= Ncf>u [1 -exp(-,\t)] Thus, the effective stable isotope tags, from an activation standpoint, are those with a large cross section and large fractional isotopic abundance, and those which form radioisotopes having moderately short half-lives. Also, the analysis is much simpler if the photopeaks or X-ray peaks of the ingested tag do not coincide with the natural photopeak background for Drosophila. Elements Examined A survey was conducted :in order to obtain several elements and their com­pounds which could be fed to Drosophila at substantial intake levels without any lethal effects, and which could 1be detected by neutron activation. The rare earths were selected as the first group of elements to examine. One of these, Dy164, has the most sensitive neutron activation assay of any element. No data could be found in the literature concerning the biological effects of this or others of the group of elements on insects. The selection of elements from this group which were non-toxic to insects had to be established by laboratory tests. The second group of elements which were analyzed were those which would become highly radioactive in a short activation period, namely, those with large nuclear cross sections and short half-lives. A secondary consideration was whether the radioisotopes exhibited an easily identifiable gamma photopeak of fairly low energy. For example, this group included manganese, cobalt, aluminum, and silver. When one element was selected as a possible tag for Drosophila, a check of the chemical stockroom inventory was made to see if a soluble salt of the element was available. For a cation tag, chloride and acetate salts generally were used, while a sodium salt (eg. NaBr for bromine tag) was generally used for the anion tag. METHODS Initially, Drosophila melanogaster was chosen as the test species for analysis of potential labels, but D. aldrichi and D. mimica were included in later experi­ments. The dilution medium for the tag compounds usually was a 2% (w/v) sucrose solution. The early experiments involved ion concentrations in terms of parts per million, but this system proved to be awkward and was rejected for the much simpler concentration system based on molarity of the label ion. Hence, prepara­tion of the proper tag dilutions began by establishing, for example, a 0.1M solu­tion and then achieving lower concentrations by dilution. A single piece of filter paper was inserted in a 3.5 X 10 cm. glass vial for feed­ing. The paper was moistened with the proper solution; 10 drops with a micro­pipette were usually sufficient. Equal numbers of males and females (usually from 5-7 of each sex) were placed in each test vial and were left for a feeding period of several hours, the particular time depending upon the experiment. Two vials were prepared for each dilution. One vial contained flies to be analyzed by activa­tion analysis for presence of the experimental tag. These ultimately were trans­ferred to the polyethylene or gelatin capsules for activation analysis. Flies in the second vial were transferred to food vials and maintained in the la'boratory for toxicity observation. A pair of control vials was prepared for each experiment, containing flies fed under the same experimental conditions of tag feeding but fed only 2% sucrose. Death rate, egg production, and hatchability were noted and compared for each compound and for the sucrose control. The flies to be activated were sealed in test grade polyethylene capsules or gelatin medicine capsules. These containers had little activation response to neu­tron bombardment and made manipulation of samples convenient. Calibration of the multichannel analyzer prior to activation was accomplished by using lab­oratory sources of known decay energies or a sample of label material. In most cases a cursory examination of the analyzer output after analysis is adequate to determine if label element is associated with the sample of flies. Activation of the samples was achieved either by lowering them into a nuclear reactor via special protective plastic vials and a detacha•ble aluminum tool, or by pneumatic transfer in a C02 -filled system. Steady state power levels of from 10...:. 250 thermal kilowatts were possible with the TRIGA Mark 1 Nuclear Reactor stationed on the U.T. Austin campus. For early studies on activation~ 10 minutes at 10 Kw (or its equivalent dosage) was chosen as the maximum activation time which would be feasible for irradiating mass samples of adult Drosophila without killing them. The power level of 10 Kw produces a thermal neutron flux of 1011 n/cm2/sec. Upon activation of the flies for 10 minutes, they were removed from the reactor and transferred to the analyzer system with a delay of about 3-0 sec­onds. For mass irradiation it was realized that a higher power level and shorter activation time were much more expedient, and, in fact, the increased efficiency was necessary to complete the analysis of a large number of samples within a reasonable time. Most tests performed at 250 Kw required only 10 seconds to one minute activation time. The natural background spectrum for Drosophila consists of moderate amounts of Na and a trace of Mn (Fig. I, top). Often the lower energy range (.01 to .7 Mev), however, has relatively little background pattern. Thus there is a wide range of energy levels open to unambiguous identification of additional elements. It is, nevertheless, of great importance to compare controls of each species and collection to be certain no complicating patterns are present. RESULTS Dysprosium Dysprosium is a rare earth with an oxidation state of +3 and an atomic weight of 162.5. It is activated by neutron capture in one of the naturally occurring isotopes, Dy164, to produce both Dy165 (2.32 hour half-life) and Dy165 m (75 second half-life). Dy165 emits prominent gamma rays at .095, .280 and .362Mev.Dy 165m, an isomeric state of Dy165, decays to Dy165 by emission of a .108 Mev gamma ray. In the photospectrum obtained with the scintillation detector, the .095 and .108 photopeaks are not resolved and appear as a single broad peak at .098 Mev. For short decay times following irradiation, the major contribution to this peak is due to the shorter-lived Dy165m. This activity decreases very rapidly and within a few minutes the contribution from Dy165 is predominant. The photospectrum also exhibits X-ray excitation peaks at about .048 Mev and a broad backscatter peak from the .095 and .108 Mev gamma rays at about .068 Mev. Gamma spectroscopy techniques. Several salts of dysprosium were examined with an initial emphasis on the compound's solubility in water. The first of these salts, DyC13, was highly soluble in water, and concentrations of 3.3M were easily obtained. Dy(Acetate) 3 was not quite as soluble, but .03M concentrations were easily achieved. Dy(N03) 3 and Dy2 (S04) also were examined, and both were 3 sufficiently soluble to test. Analysis of dysprosium-fed Drosophila immediately after feeding for 12 hours gave strong labeling, reflected in large photopeaks for all four salts., The most effective concentrations for dysprosium intake were found to be about 10-2M for both DyC13 and Dy(Acetate) 3, the two compounds which were selected for further investigation. Background activity for flies maintained on laboratory culture media is very low in the dysprosium energy range. Drosophila fed or captured after feeding on some fruits show a small amount of activity between the Dy decay peaks. The change of background is not sufficient, however, to confuse the presence of the dysprosium tag. w I- z ~ ~ 0:: w Cl.. (/) I- z ~ 0 u (.9 0 _J CONTROL 0 mimico .... .·....:.:... . ... ·' .· .... .. . . .... . . ... . . . . ·. \ .· ~ ,,· .··. . . . . ··.· , . . '· .·. \ . · ..·., .. :;.· ···....~..... .. . ..... . .. .·.· ... . :· ·..·.:. ·.'··:·.·. .. .· . .·· . ., .· .. .068 .098 .277 .­ ,,.,.. STAN OARD .. DyCl3 + YEAST .._: ....,.......,..,....,....\.. .!' ........ . \ .: .. ....... -v ..,.--·/•..,,,. •"'·-i··.·.., ...•-~·........:,,­ ... ·., .....'"'......._... .068 .098 .277 GAMMA ENERGY (Mev) Fm. 1. Top: gamma spectrograph of Drosophila mimica grown in the laboratory (on Wheeler­Clayton food), which is very similar to the spectrograph of all species examined; bottom: gamma spectrograph of DyC13 mixed with yeast in nutrient broth, the substrate used in some experiments. The DyC13 or other Dy salts alone had essentially the same pattern. Analysis was done on eight individual flies to determine variations in the amount of DyCl3 among the flies. Four males and four females were all fed on .03 M DyCl3 in the same feeding vial for the same period of time. After activation, the amount of radioactivity produced by the dysprosium was measured by sum­ming the counts under the gamma photopeak indicated by the multichannel ana­lyzer. These values are given in Table 1. Counting time was 80 seconds after 10 minutes irradiation at 10 Kw. The melanogaster female marked "g" showed over three times the activity of the male marked "b", indicating that the amount of tag in the female "g" is three times that in male "b". This female also showed more than twice the activity of the male marked "a" and the female marked "e". Hence it appears to be difficult, if not impossible, to predict the number of tagged flies by noting the amount of dysprosium present, even when the flies are fed under identical conditions. To determine the exact number of tagged flies present in a large sample, one must analyze each fly individually. A rough estimate is possible but not always accurate. This also indicates that it is impossible to determine with a group of tagged flies the time lag between tagging and capture based on the amount of dysprosium present. This discrepancy in quantity of dysprosium consumed was noted many times in analyses of groups of 10 flies fed on different days and in different vials but on the same tag concentrations and under the same conditions. However, in practice one may analyze a sample of about 10 flies for presence of label, and if present, reanalyze each of the flies individually. When the frequency of labeled flies is small, most rnmples will have none present and can be discarded with one analysis; whereas, when one is found, it may be identified by only a slight loss of motion through the initial analysis of the group. The detection of a single labeled fly in a sample is very reliable. The analysis is ordinarily sensitive enough to clearly detect the labeled individual, but even the suggestion of the label's presence may be considered sufficient justification to analyze each fly. With individual analysis, a labeled fly is easily detected. The quantitative determination of the elemental composition requires the use of counts under a peak, but for purposes of semi-quantitative analysis, an index is about as descriptive and patterns of labeling more easily visualized. The index is obtained by first averaging the 6 count totals recorded in 3 channels on both sides of the peak for an estimate of the base-line, and averaging the counts recorded in 3 channels at the expected position of the mode. Then the ratio of mode to base­line is used to represent the amount of label present. This value may be viewed TABLE 1 Variation among flies in amount of Dy immediately after feeding D. melanogaster on DyCl3 .03M (Dy). Individuals male control (sucrose) male a male b male c maled female e female f female g female h Area 240 1530 1340 2750 2870 1790 2360 4170 1860 Corrected area (minus control area) 0 1290 1100 2510 2630 1550 2120 3930 1620 as the relative height of the mode to the base-line. Generally, any value above about 1.3 is a clear indication of the label's presence, a value between 1.2 and 1.3 a certain indication if there is good control measurement, and for most cases 1.1 to 1.2 is suspicious but undependable for unambiguous detection of the label. Of course, the higher count rate associated with labeled flies helps to increase the statistical reliability of the mode and base-line estimates and has a striking effect on making a decision of prernnce or absence of the label for index values associated with increased activity. Use of 250 Kw reactor power greatly facilitates detection of the label in this fashion (although ordinarily power greater than 10 Kw is used to reduce the analysis time by reducing activation time) . The use of dysprosium as a chloride or acetate markedly affected the residual properties in D. melanogaster. As shown in Figure 2, the flies maintained in the absence of labeled food, after having been fed Dy(Ac) 3 overnight, remained clearly labeled for at least two weeks as compared to those fed DyCl3, which were barely labeled after less than one week. The initial amount of labeling was about the same for the two salts but rapidly declined in the case of DyCl3, so that Dy was hardly detectable after 5 days. Considering that the data in Figure 2 is taken from a composite sample of 10 flies, it is not clear whether the remaining label is evenly distributed among the flies, or that it is restricted to only a part of the sample. Flies in this test were not analyzed separately. Nevertheless, it is clear that for usual population studies where the adults are fed the label, the acetate is the more efficient of the two salts. Table 2 presents similar data on label retention for D. aldrichi. The pattern was not nearly as uniform as that shown in the case of D. melanogaster, but generally flies at all stages of testing were labeled. In some instances the label was almost as apparent in the second week after having fed on a dysprosium salt as in the TABLE 2 Degree ofresiduallabeling of D. aldrichi (Index= (mode)/(baseline)) after having fed on 10-3 M (Dy) for 8 hours. Days after feeding 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Sample size 10 10 10 10 10 10 10 10 4 6 10 5 2 4 DyCl3 Dy (Ac) 3 Sample Index size Index 2.68 10 4.15 6.10 10 3.47 2.61 10 4.39 1.98 10 2.89 1.13 10 1.54 1.43 10 1.63 2.35 10 2.99 1.24 10 1.95 1.20 10 1.72 2.59 3 2.17 1.88 5 2.33 1.62 5 2.37 3.73 10 1.92 2.33 9 1.31 3.91 7 1.25 0 O> ~ >­ ~ ~ < ::;; ::;; ~ ·~· ·-·~\:. --~~-.(.. 3 n · •" !j3d s_'l:no:i !>01 \_ ":\:,.. ,,.•' .,.:;):.l ,,-------· \ \ ....,,. "' .·! " {.,.,,.··.-,....-·'"' . '•.\,, ··~. ~"' . > <___ ..-··· ......, 3 _0 :. d3c $_\ "O:J !JO- Fie. 2. Patterns of residual labeling for Drosophila melanogaster fed either the chloride or acetate salts of Dy. Control plot is the gamma spectrograph of flies not fed label, but only sucrose solution or corn meal food. first week. Assuming variation in degree of labeling among individuals similar to that for melanogaster, any single fly would have detectable label associated with it. However, due to the marginal composite degree of labeling for the sample tested 5 days after having fed on DyCl3, some reservations for this case seem in order. There may be some individuals in this sample with insufficient label for an unambiguous decision. Table 3 presents data on D. mimica similar to those on D. aldrichi in Table 2. TABLE 3 Degree of residual labeling of D. mimica (Index= (mode)/(baseline)) after having fed on 10-3 (Dy) for 8 hours. DyCl2 Dyi, .-\r '1~ Days after S;-mple Sinn pie feed:ng size Index s'ze Index 2 5 5.87 4 4.30 3 5 7.84 3 7.51 4 5 7.29 5 6.61 5 5 7.71 5 5.37 6 5 6.30 4 3.35 7 2 2.20 3 7.50 8 3 5.04 3 4.58 9 6 4.18 4 3.92 10 2 3.74 5 3.87 11 3 2.00 5 1.70 12 4 1.69 4 1.66 13 5 1.42 5 2.00 14 4 1.52 3 1.88 15 5 1.77 5 1.78 Again, the results clearly indicate residual label in all flies, even after two weeks removal from the label. The results for chloride and acetate salts are equivalent, which perhaps is not very surprising since the salts were fed in nutrient broth inoculated with yeast. In all cases the detection of a fly which has previously fed on Dy is unambiguous. Since the flies are larger, one might suppose that detec­tion was easier as a result of more dysprosium being present. However, size of fly should not be taken as a general indicator of amount of label. Differences in the feeding habits and, possibly, biochemistry of the label for different species may enter the picture for determining the effectiveness of the label. Day vs. night feeding. D. aldrichi was fed both during the day and during the night in an effort to determine if the feeding habits differed so as to affect the uptake of the label. The night feedings were from 6:00 p.m. until 8:00 a.m., a period of 14 hours. The day feeding was from 9: 00 a.m. until 6: 00 p.m., a period of 8 hours. The results are given in Table 4. There is no difference between day and night feedings, in spite of the differences in feeding time. It may be that the accumulation of label is incremented very little after a time less than 8 hours, and one would observe but small changes between 8 and 14 hours of feeding, and thus not detect differences by such a technique. Such would be the case if the major amount of labeling were surface contamination rather than feeding. It TABLE 4 Degree of labeling for day or night feeding; D. aldrichi after having fed on 1Q-3 M (Dy). 9 AM-6 P:\I 6 PM-8 Al\1 9 AM-6 PM 6 PM-8 AM Days after Sample Sample Sample Sample feeding Dy size Index size Index size Index size Index 10 7.24 10 6.38 10 5.49 10 6.67 2 10 4.39 10 6.34 10 6.69 10 6.18 3 10 4.15 10 4.95 10 4.71 10 3.84 average 5.26 5.89 5.63 5.56 remains to be seen what the mechanism of labeling involves, but all of the resi­dual labeling, as well as the pattern of feeding which affects the uptake of label, may be explained by surface contamination rather than internal deposition of the label (see following section for alternative). Residual through pupation. By periodically adding a few drops of label solution to the larval food, the larvae become highly labeled. However, unless the internal label is reincorporated into tissue materials during pupation, it will likely be excreted, and the adult ultimately will be free of the label. Figure 3 presents a 2.5 8 Dy(Ac) 3 DyCl 3 8 .OlM O .OlM 0 .OOlM • .001 M + x 2.0 0 0 0 w 0 z 1.5 0 0 0 00 • ooO •• ~o 1.0 • 0.5 _...._._____.___.__......................_......_....._....._....._................, 0 10 20 30 DAYS AFTER ECLOSION FIG. 3. Pattern of residual label of adults of various ages, fed the label in the larval stage, washed as third-instar larvae or pupae, and cultured thereafter on unlabeled food. Points in this experiment between 1.1 and 1.3 are considered ambiguous labeling, since variation among individuals is not known. set of survey data where it can be seen that some label is carried through pupation and retained for over three weeks of adult life. Depending upon the sample, age of adults, label concentration, and salt used, the amount of label in the adults varied from sizeable to undetectable. In the first few days after eclosion, the adults were clearly labeled. As seen in Table 5, each individual in a sample of adults 6 days after eclosion was clearly labeled, and, considering that the salt was chloride, they were labeled to a much Individual labeling of adults 6 days after eclosion, when fed the label as larvae (DyC13, 0.01 M, index of composite sample= 2.43) Index of individuals AYe. Index Males Females 1.67 2.05 1.93 1.84 2.12 2.59 1.80 1.93 1.67 1.69 1.84 2.02 greater degree than adults examined 6 days after having fed on the same label (Table 2). Note, however, that the concentration in larval feeding experiments was 10-fold greater. There was not as great a range in labeling among individuals as one might have expected from feeding the label, but there was sufficient varia­tion to require analysis of individual fly samples for accurate counts of labeled flies. There is a suggestion of an effect of both salt and concentration on the amount of label found in the adults. The acetate seems slightly more effective, and higher concentrations tend to label better. The .001M concentrations were ineffective as labels. The 0.01 M concentration of the chloride was highly effective soon after eclosion, but was of doubtful use for long residual labeling. For reliable use, the performances of even the 0.01 M acetate salt needs refinement. It is not known how much variation among individuals exists at longer periods after eclosion, but at the low levels of la1beling, almost any variation would render the label unre­liable. Tab!e 5 presents the results of individual fly analysis for DyCl3in adults 6 days after edosion. They were fed the label in the larval stage, and were the most ef­fectively labeled sample in Fig. 3. It is interesting to note that there was remark­ably little variation in amount of label, compared to labeled adults fed the label in the adult stage (Table 1). Whereas adults fed the label had about a 260% range of differences in amount of label, the larva-fed adults had just over a 55% range of differences. The uniformity likely represents the elimination of surface labeling. and the label found in flies represented in Table 5 was totally internal. Although much more refined and quantitative experiments are needed to de­termine the location of the label, there is a strong suggestion that internal depo­sition of the label is taking place when fed to larvae. Autoradiographic technique. Since radioisotopes are produced by neutron bombardment, exposure of photographic film to the activated sample results in gamma exposure of the film. The dysprosium-labeled flies were considerably more raidoactive after neutron activation than were the corresponding unlabeled controls. Samples of flies taken from overnight feeding of 0.03 M DyC13 were stuck to "Magic Mending" tape about 1 cm. apart. The tape and flies then were activated for 5 minutes at 10 Kw. Two-minute exposure of Ilford Type G indus­trial X-ray film immediately after activation resulted in a qualitative difference in film exposure of labeled and control samples of flies. The control flies resulted in very slight exposure, comparable to the tape, while the labeled flies gave dis­tinctly darkened spots on the film. If the autoradiographic technique were perfected, at least in the case for dy­ sprosium, it could greatly increase the efficiency of detecting labeled flies. Instead of an average of about one fly analyzed per minute, ten to a hundred flies could be examined in this time. No attempts have been made to detect other labels by this technique, but even the detection of multiple labeled individuals seems pos-· sible if the labels have markedly different half-lives. An application of autoradiographic scoring of pollen from pine and sugar beet plants with an elevated concentrat!ion of manganese (Fendrik and Glubrecht 1966) was recently brought to our attention. The plants either were injected with MnS04 solution or fed additional Mn through the roots. Pollen was collected in the vicinity of injected plants, and successfully analyzed autoradiographically after neutron activation. Dysprosium-salt solutions were tried, but toxicity limited their use in plants. Secondary labeling. Only a single test was made to determine if a labeled in­dividual could label, in turn, other individuals. In this test flies fed DyC13 were transferred to unlabeled food vials for 3 days, and unlabeled flies added to these vials immediately after discarding the labeled flies. The index of the unlabeled flies, tested after 2 days, was 0.98, indicating no detectable secondary labeling. Circumstantial evidence to the same conclusion is available from the examina­tion of the distribution of labeled flies after field collection (Richardson, 1969). Even though many flies were kept together for several hours after capture, only rarely was there more than one or two labeled flies found in a group from the same vial. Ifthere were secondary labeling, one would expect clusters of labeled individuals from those in the same vial. Other Elements Tested Hafnium. Hf178 constitutes about 30%.of the natural stable Hf isotopes. When a single neutron is a•bsorbed, producing Hf179, there is a decay with emission of a gamma ray with energy 0.215 Mev. Hf179 has a half-life of 19 seconds. Other radioisotopes form a negligible fraction of the photospectrum after brief activation periods, since they have considerably longer half-lives and also are produced with a lower frequency. Hafnium oxychloride (HfOCl2 • 8H20) was mixed in 2% sucrose solution and fed to D. aldrichi in a 0.1 M concentration. Table 6 presents the results of residual labeling. Although only a preliminary test, the element was a very effective label, TABLE 6 Residual labeling of D. aldrichi fed 0.1 M HfOCl., for 6 hours and maintained afterward on unlabeled banana food. (Index= (mode)/(baseline)); all samples were 10 flies. Days after feedJig Index (proportion of control) Control (unlabeled) 1.24 1 6.48 5.23 3 1.71 1.40 4 2.54 2.05 7 1.81 1.46 9 1.84 1.48 detectable 9 days after feeding label with interim feeding on regular culture me­dium. This element shows promise of performing as effectively as dysprosium. Manganese. Activation properties for Mn are appropriate for a quick and sensi­tive assay. The half-life of the principal isotope, Mn56, is 2.58 hours, with the principal peak at 0.845 Mev. The neutron cross section of the stable isotope, Mn55, is only 13.1 barns, but is sufficient for use with a high neutron flux. Typical stock­room salts were ineffective in labeling with manganese, except for only a few hours after feeding. For ease of scoring, one must at least double the normal amount of manganese in the fly, and such was not the case for longer periods after feeding flies the chloride or acetate salts. However, the citrate and gluconate forms were much more effective. In the case of manganese gluconate, there was easily detectable residual label after 5 days. More tests will be necessary to de­limit the labeling properties of these compounds. Samarium. Sm152 and Sm154, with neutron capture cross-sections of 220 and 5 barns, are promising isotopes for activation analysis. The energy levels most use­ful for detection of the element result from the decay of Sm153, which decays with a half-life of 47.1 hours with a gamma peak at 0.103 Mev, and Sm155, which de­cays with a half-life of 22 minutes, with a peak at .105 Mev. The label was present the day following feeding samarium chloride (0.1 M Sm) to D. aldrichi, but there is no doubt that a different chemical form will result in a more effective label. It remains to be seen in greater detail what residual properties this, or other forms of the label, may have. Aluminum. A1 27 activates to A1 28 , which has a half-life of 2.3 minutes and a gamma decay peak at 1.78 Mev. The neutron cross-section of A1 27 is very small, only 0.23 barns. But again, with a high neutron flux, it is a potential label ele­ment. Analysis of flies immediately after feeding the chloride salt indicated that the element could be easily detected when associated with the flies. However, residual effects of the label were not sufficient for detection. Probably the proper chemical form, allowing an accumulation and retention of higher concentrations of the element, will allow its use as a label. Bromine. Br'9 (50.5% natural abundance) has a neutron cross-section of 8.5 barns. Br80 has a half-life of 18 minutes and decays with 0.62 Mev gamma ray. It was tried as a label element, fed as NaBr (0.03 M Br) overnight. Immediately after feeding, the label was easily detectable, but there was no residual labeling. Bromo Phenol Blue was also tried, but without success. Again, selection of the proper chemical form may allow the use of bromine as a label. Other elements. AuCl3, CoCl3, Co (Ac) a, La:! (S04)3, GdCl, ~N03, AgMn04, and CsCl (prospective label element underlined) were unsuccessful in attempts to find~ore label elements. For various reasons of analysis and availability these particular compounds were tried, but the label element was not detectable in any case, even in the assay of flies taken directly from the food containing the label. Some chemical problems occurred which may partly explain the failure of some tests. In the case of gold, the hydroxide formed quickly after making an aqueous solution of the chloride and precipitated the gold from the solution. This change was reduced by adding dilute HCl until the solution was mildly acid. A complex of Tris-ethylamine cobaltic chloride was an attempt to arrive at a better form of cobalt for feeding, but this combination was not successful. The silver compounds were not toxic, as expected, but there may have been no appreciable degree of feeding on these compounds. Also AgMn04 tended to decompose during the feed­ing period, but the relevance to the results of labeling is not known. Further studies of this type are underway, and the use of chelating or complexing mole­cules with the label element offers considerable promise in improving the labeling properties of these elements. APPLICATIONS OF LABELING TECHNIQUE Feeding a stable isotope prevents the radio-contamination of the environment, yet does not require previous capture for marking. In contrast to the release of lab-cultured organisms, marked and released in the field, this labeling technique is easily used with the populations normally residing in the study area. Also dis­turbances resulting from capture and marking of wild individuals are avoided, not to mention the labor involved in the process. Thus, the general utility of the improved technique for studying dispersal behavior in natural populations is ap­parent. Several other aspects of this technique are not so apparent, but offer prom­ise of making possible detailed studies of behavior previously impractical. By using several labels simultaneously in an area, but at slightly different posi­tions, one may use data from one label to efficiently detect and measure non­random behavior. As discussed in detail elsewhere (Wright, 1968a,b; Richard­son, 1969b), the major impact of non-random behavior is in modification of fr~­quencies in dispersal distribution "tails", or at positions several standard devia­tions from the mode. Few individuals from a label source reach the areas more remote to the label, or release site. Experimental studies of effects in the "tails" of the dispersal distribution using only a single la•bel source are highly inefficient, and thus, render the evaluation of behavioral modifications impractical. Suppose rate of movement declines with age. Two labels may be used to detect such an effect. If one label, A, is continuously present, it will produce a steady­state dispersal distribution. If another label, B, is placed around the label site for A, but at a distance taking the flies an average time of, say, four days to traverse, one can examine the component of the dispersal distribution of A, involving, on the average, individuals four days older. This component distribution would be that formed by individuals having both labels. Since any given fly having both labels could have acquired them either A then B, or B then A, the complete analysis of the observed distribution is complicated. (Likely Fourier analysis techniques would be useful in this connection.) Never­theless, detection rather than accurate measurement of behavioral changes seems relatively simple. Figure 4 may be used to illustrate the idea. Distributions A, B1 , and B2 all have the same variance. However, individuals of B1 and B2cannot be distinguished so the composite distribution of B1 and B2 is observed. Individuals bearing both labels form distributions (AB1), (B1A+B2A). Again, a composite distribution would be observed, and for proper analysis should be considered. However, by proper placement of label sites, the contributions of "tails" from adjacent distributions may be decreased and subsequently ignored. Then a comparison of variances of distribution (AB1), (B1A+B2A), and (AB2) with that of A yields information concerning changes in movement with age. For Label Site Label Site Label Site B A 8 FIG. 4. Hypothetical dispersal distributions of labeled individuals. Two labels, A and B, are used at sites indicated. The abscissa represents distance and the ordinate represents frequency of labeled individuals. Note that the order of letters in naming distributions of doubly labeled individuals represents the order of acquisition of labels. the case of reduced movement with increasing age, the variance of A (and also B1 and B2, which are not easily estimated) would be larger than that of the other distributions. Nonrandom movement may likewise be detected by combinations of labels, each used at a unique site but near the other labels. In this case, tendencies to move in certain directions would be reflected in skewed distributions, the princi­pal direction of movement being that of the skew. It seems probable that ecologi­cal interfaces may now be defined ·by direct observation of the animals' behavior, rather than by indirect ways, such as by changes in population density. The ef­fects on gene distribution by behavioral characteristics at ecological interfaces has been analyzed and effects on dispersal distributions illustrated by Richardson ( 1969b). For effects on genetic structure of the population and ecological isola­tion, behavioral characteristics are extremely important, but usually must be in­ferred from indirect evidence. A number of other applications in studies of population dynamics are apparent, but further developments of technique are needed. One application where the label is pulsed in a population by only short exposure is particularly intriguing. By periodically changing labels at a common site, one could measure short-term changes in dispersal. In addition, analytical techniques for the data obtained are highly developed for use in studies of analogous physical phenomena. ACKNOWLEDGMENTS Professor Wilson S. Stone produced the initial impetus for the senior author to consider stable isotopes for labeling Drosophila. At one time, during a postdoctoral tenure, this author presented to Professor Stone a plan to study dispersal with radioisotope labeling, in particular with P32. In his usual direct and explicit fash­ion, Professor Stone indicated he would not support such environmental pollution. He maintained that broad ramifications of experimental technique must be thor­oughly considered, especially in field work, and furthermore few experiments The U niz.:ersity of Texas Publication required such hazardous procedures of poorly predictable impact as the relatively uncontrolled used of radioisotopes. His intuition and firm belief in a priority for the protection of populations bore fruit in the development of the techniques presented here. Among others " ·ho contributed to the early incubation of ideas, foremost was Professor W. R. Lee. His experiences \vith labeling honeybees offered consider­able insight into the problems of labeling insects. LITERATURE CITED Fendrik. I. and H. Glubrecht. 1967. lnwstigations of the propagation of plant pollen by an indicator activation method. In Activation Analysis in the Life Sciences, International Atomic Energy Agency. Vienna; pp. 325-333. Richardson, R. H. 1968. Migration, and isozyme polymorphisms in natural populations of Drosophila. Proc. XII lnternat. Cong. Gen.. Tokyo II (modifications in vol. III). Richardson, R. H. 1969a. Migration, and enzyme polymorphism in natural populations of Drosophila. Jap. J. Gen. -14, Suppl. 1: 162.-169. Richardson, R. H. 1969b. Models. and analyses of dispersal patterns. In Mathematical Topics in Population Genetics, ed. by K. Kojima. Springer-Verlag. N.Y. (in press). " 'right. S. 1968a. Evolution and the Genetics of Populations. Vol. I. Genetics and Biometric Foundations. Chicago: Uniwrsity of Chicago Press. "'right. S. 1968b. Dispersion of Drosophila pseudoobscura. Am. Nat. 102: 81-94. XIII. Isozyme variation in Drosophila island populations. II. An Analysis of Drosophila ananassae Populations in the Samoan, Fijian and Philippine Islands.1 F. M. JOHNSON, K. KOJIMA AND M. R. WHEELER INTRODUCTION In the first paper of this series (Stone et al, 1968) natural populations of a species of the Drosophila nasuta subgroup in Samoa and Fiji were examined for genetic variations in enzymes. Polymorphism was found in a large proportion of the enzyme systems examined. In general, however, a regional similarity in allele frequencies within Samoa and within Fiji was discovered, but significant di­vergence existed between Fiji and Samoa populations. By investigating allele frequency differences within and among populations it may be possible to determine the importance of the variations in the adaptation of species in different populations. In order to examine the generality of the ob­served pattern of polymorphic enzymes, several species and enzyme systems should be studied in a variety of localities. For this reason it was of interest to in­vestigate some additional populations of D. ananassae in addition to those pre­viously studied. (Some of this work has appeared in another publication-Stone, Wheeler, Johnson and Kojima, 1968) MATERIALS AND METHODS Two forms, or races, of Drosophila ananassae were studied: one, the dark form from Samoa and Fiji, and another, of the usual light color, from Fiji and the Philippines. Futch ( 1962) showed that the dark phase is limited to certain of the Polynesian Islands but, judging from hybridization studies and comparative cy­tology, it is conspecific with the usual lighter colored form found in most areas of the world. However, in Samoa and in at least certain localities in Fiji, a light form also occurs but it seems to be completely reproductively isolated, in nature, from the dark form, and is probably to be considered an undescribed species. In the present study this distinctive "light" species was not used, the expression "light form" or "light race" meaning only the light colored form of the species, ananassae. The collection localities are shown on Figure 1. Collections were made in April, 1967 (Philippines) and in July and August, 1967 (Samoa and Fiji). The dark form was sampled from the Samoa localities ( 1), (2) and (3) and from one Fiji 1 Supported by PHS Research Grants No. GM-11609 and GM-15769 and Training Grant No. GM-00337-07 from the National Institute of General Medical Science. The authors gratefully acknowledge the support and constant encouragement of Professor Wilson S. Stone whose interest in the problem of evolution in island populations led to the inception of this series of studies. STUDIES IN GENETICS V. Univ. Texas Puhl. 6918, Sept., 1969. The University of Texas Publication ~ ~ w ·}? ~ w (,, !::: if FIG. 1. Map showing the two locations in the Philippine Islands from which collections were made, and the localities in Samoa and Fiji which were reported earlier. locality ( 4) . The light form was collected at Fiji locality ( 6) and from two lo­calities in the Philippines: (7) Cabuyao, Laguna, Luzon, from fallen mango, pineapples and melons, collected by Dr. Mercedes Delfinado; and (8) Bongao Island, at the southwestern tip of the Sulu Archipelago, from papaya and pine­apple, collected by Dr. Delfinado. Collections in Samoa and Fiji were all made from fallen papayas or from papaya baits when no fallen fruits were available (4). Newly collected flies were sorted by sexes as soon as possible, and females were placed individually in small shipping vials; males were returned in small masses per vial. Shipment by air to the laboratory in Austin was made as soon as possible _after collection. For esterase (EST), acid phosphatase (ACPH) and alcoholjoctanol dehydro­genase (A/ODH) (cf. the previous paper of this series for references to electro­phoretic and staining procedures) in the dark form (localities 1-4), wild-caught female flies were analyzed soon after their shipment to Austin. Before they were killed, however, lines were established from single isolated flies for additional analysis and other purposes. Flies from similar isofemale line cultures, a few generations after they were established, were used in the assay of EST, ACPH and A/ODH in the light form. For the other enzymes, only isofemale line flies were used. Malic dehydro­genase requiring DPN as cofactor (MDH D) and leucine aminopeptidase (LAP) were examined in pupae, and alkaline phosphatase ( APH) in larvae. The assay of MDH D was done according to the method for A/ODH in the previous paper except that 25 mg DL-malic acid was substituted for alcoholjoctanol as substrate. Staining procedures for LAP and APH have been published elsewhere (Beckman and Johnson 1964a, 1964b) . Three individuals from each line were examined for the material from Samoa and Fiji. From the lines originating in the Philip­pines, four individuals were analyzed. A partial study of inheritance patterns was obtained from the F1 generation of wild-caught females. RESULTS Figures 2 and 3 show the phenotypes, in diagrammatic form, that were scored in this investigation. The first column in the drawings depicts the total banding pattern obt?ined including a common homozygous type in the system scored for variation. The typical appearance of starch gels after staining is shown in the photographs; Figure 4 (EST)~ Figure 5 (ACPH), Figure 6a (ADH), Figure 6b (ODH), Figure 7 (MDHD), Figure 8 (LAP) and Figure 9 (APH). Genotypes were in general, assumed to be determinable directly from the phenotypes. Thus the phenotypes (and genotypes) were designated by an arbitrary numbering system according to mobility in decreasing order from the anode. Single banded types were assumed to be homozygotes while heterozygotes displayed either a double banded pattern (EST, LAP) or a pattern of three bands (ADH, ODH, ACPH, APH, MDH D). In addition, the ADH system contained extra bands so that it appeared as a set varying as a unit. MDHD also has extra bands but they are more faintly staining than the ADH 'bands and do not appear on the photo­graph (Figure 7) or drawing (Figure 3). FIG. 2. Drawings representing all observed phenotypic variants in the EST C, ACPH, ADH, and ODH systems in the Samoan, Fijian and Philippine D. ananassae. Numbers below the drawings describe the phenotypes. All zones generally observed in a given stain are shown in the first columns while subsequent columns show only the systems that were scored. * = vari­able bands with unclear or unclassifiable patterns. •=non-variable bands. The extent of vari­ability in um\'Orked bands has not been determined. Some inheritance data were obtained from the analysis of F1 individuals of isolated wild females supplementing other similar data obtained previously (Johnson, 1966). Parental genotypes consistent with the phenotypes of the fe­male parents and offspring could be determined for almost all enzymes in all the broods. The female phenotypes, the numbers of offspring and their phenotypes, and the deduced genotypes of female and male parents are presented in Table 1 (EST C), Table 2 (ACPH) , Table 3 (ADH) and Table 4 (ODH) . Only one dis­crepancy between kinds of offspring and parental genotypes expected on the APH LAP MOHD MDHT_ 4 MOHD_ 4 4 Frn. 3. Diagram showing the phenotypic variants observed in the APH, LAP and MDHD systems. Markings and designations follow the same pattern as those in Figure 2. basis of one-locus inheritance was found. This was in the Fi group resulting from female no. 4FDK-90 where five EST C types were found in the offspring (Table 1). Possibly accidental contamination of the Fi generation with a few F2 indi­viduals might have occurred. Ratios in segregating groups of offspring appear somewhat discrepant from expectation. There are, however, a variety of explana­tions, including effects of hackground loci and multiple mating, having nothing whatever to do wih the loci considered. The observed numbers of the different phenotypes detected in the wild-caught dark ananassae from Samoa and Fiji are presented in Table 5 (EST C), Table 6 (ACPH), Table 7 (ADH) and Table 8 (ODH). Allele frequencies determined from the phenotypes of the wild flies are presented in Tables 9, 10, 11 and 12. 192 The University of Texas Publication + .. FIG. 5. 3 2h 3 213 % 3 ACPH PHENOTYES Photograph showing the typical appearance of a 2/3 % 3 gel stained for acid phosphatase activity and some of the variation in the ACPH system in adult D. ananassae. For EST C and ACPH, expected numbers of phenotypes were calculated accord­ing to Hardy-Weinberg proportions using the allele frequencies. The expected values are presented along with those observed in Tables 5 and 6. No significant difference exists between the numbers found and those expected in these two systems. Variant types in the ADH and ODH systems were very infrequent and (a.) (b) + + FIG. 6. (a) Photograph of a gel stained for alcohol and octanol dehydrogenase showing variation in the ADH system in adults. (b) Photograph of a similar gel showing an ODH variant. + 3 % 3 y3 1 3 MOHD PHENOTYPES Frn. 7. Photograph showing examples of some of the malic dehydrogenase (MDHD) pheno­types found in D. anaruzssae. The University of Texas Publication + ~~~~~~~~ 4 % 4 4/s 4 2 4 LAP PHENOTYPES FIG. 8. Photograph showing the typical appearance of leucine aminopeptidase (LAP) pheno­types in D. ananassae pupae. + \_____}\__/\__J 4 3 l 3 1/3 3 % 3 % APH PHENOTYPES F1G. 9. Photograph of a gel containing D. ananassae larvae stained for alkaline phosphatase and showing s:lme of the observed APH phenotypes. thus expected phenotypic values are not presented. The allele frequencies in the M dh D and Lap systems (determined in pupae) and the Aph system (determined in larvae) were obtained from the phenotypes in is of emale lines of both the dark and light anarzassae, and the phenotype frequencies in nature are not known. Tables 13, 14 and 15 show the frequencies for the Mdh D, Lap and Aph loci, re­spectively, in the localities sampled. The arzarzassae populations have some similarity in distribution patterns of allele frequencies to those found in the nasuta specie:; (cf. the previous paper). There is often one allele (a majority allele) of a system that is more frequent than other alleles in all regions sampled. In the ADH, ODH and MDH D systems, the majority allele frequency exceeds 0.95 as shown in Tables 11, 12 and 13, respectively, in all localities. The3e systems should not be considered polymorphic in natural populations according to the definition of balanced polymorphism by Ford (1945) . All other systems are typically polymorphic. In many localities there are two or more alleles occuring in intermediate frequencies without an obvious majority allele. One outstanding feature in these polymorphic systems is that there are three or more alleles in all populations with one exception of the ACPH system in the dark form obtained from Fiji. In some populations, as many as seven alleles were segregating as seen in Tables 10 and 15. Due to this diversity in the polymorphic enzymes, the Kolmogorov-Smirnov test (Walker and Lew, 1953, p. 428) was used to detect the significant hetero­geneity in allele frequency distributions over the different geographic locations. The first step to apply this test was to construct "cumulative frequency dis­tributions" of allele frequencies for individual locations .Since alleles in each sys­tem were numbered according to the e]ectrophoretic mobility of their correspond­ing isozymes, allele designations provided a natural ranking for the construction of cumnlative distributions. Thus, allele frequencies were added successively starting from the frequency of the lowest numbered. The test was set up in a hierarchal manner; comparisons were made by pairing island populations within each region with respect to a given race, then the popu­lations of the same race within regions were pooled and comparisons were made between regions for a given race, and finally between-race comparisons were made within Fiji where the two races coexisted (Table 16). Except the APH system (cf. Table 15), the pair-wise comparisons of island populations within regions were not significantly different at the 0.1 probability level. It is assumed that the probability of the type II error is small because the hypotheses were accepted above 0.1 probability level and the number of observa­tions (i.e., the no. of alleles) were large in all cases. In other words, the probabil­ity of accepting the hypothesis of no differences among islands within regions when there really are differences, is assumed small. Hence it follows that there is good statistical evidence to pool frequency distributions of islands in regions for such cases. However, a justification is needed for pooling the APH distribu­tions in the dark race from the Samoan islands because of the existence of sig­nificant heterogeneity within the region ( c.f. Table 16). The main purpose of the statistical test employed is to detect relative genetic divergence of populations as they are correlated with geographic distance. It is natural to assume that, for example, the Samoan flies are more closely related to each other than to flies from the Fiji islands. Thus, the heterogeneity among the Samoan populations is ex­pected to be less than the degree of divergence between the Samoa and Fiji is­lands, and this provides a reasonable basis for pooling within regions. The results of the test conducted on the pooled data verified this expectation by giving far more significant value of test criterion for Samoa vs Fiji than any computed for Samoan island comparisons. D1scuss10N AND SuMMARY The analysis of ananassae data strengthens the conclusions drawn from the earlier analysis of nasuta data. The major points are (i) that there is a high de­gree of polymorphism in a number of enzyme systems, (ii) that there are greater differences in allele frequency distributions associated with greater geographic distances as far as polymorphic loci are concerned, and (iii) consequently that there must be a high degree of migration among islands in the same regions. The type of study conducted does not render itself directly to a critical evaluation of adaptive differences which may exist within and between island populations. However, the data are not inconsistent with the existence of selectively balanced polymorphisms. There is one important point in the present ananassae data which was not nec­essarily obvious in the nasuta data. This is the finding of a large number of multi­ple alleles in some systems. Kimura and Crow ( 1964) calculated the average number of alleles maintained at a selectively neutral locus to be about 4NeM+1 where Ne is the effective size of a population and Mis the rate of mutation. The observed average number of alleles per population per enzyme system is 3.93 for the dark race and 3.62 for the light race. These estimates must be somewhat smaller than the actual average. If one accepts the mutation rate of 10-5 , then Ne must be larger than 10,000. Ne is reduced to a few thousands if the mtuation rate of 10-4 is accepted. Thus, either mutation rate is higher for this class of alleles than for the classical type of alleles, or the effective size of populations must be considerably greater than ordinarily assumed for a Drosophila population. One may argue that the idea of selective neutrality is at fault, and such an argument is probably a valid one. But there are still some difficulties. For ex­ample, there are a number of delicate mathematical relations among selective values which must hold true for a balanced polymorphism of many alleles (Ki­mura, 1956) under the constant selective value model. It is rather difficult to find a set of fitness values which satisfy Kimura's conditions and the observed allele frequency distributions. Another possibility is frequency-dependent selection as reported by Kojima and Yarbrough (1967) and Kojima and Tobari (t969) for the viability of D. melanogaster with respect to the esterase 6 locus and al­cohol dehydrogenase locus. With this mechanism, viability is enhanced for al­leles existing with sub-equilibrium frequencies, and reduced for alleles with over­equilibrium frequencies. When allele frequencies are near this equilibrium, al­leles are essentially neutral with respect to one another. In reviewing all the aspects which may account for a large number of alleles per locus, the most likely possibility is either high mutation rate, frequency-dependent selection or the combination of these two mechanisms. LITERATURE CITED Beckman, L., and F. M. Johnson. 1964a. Variations in larval alkaline phosphatase controlled by Aph alleles in Drosophila melanogaster. Genetics 49: 829. ----. 1964b. Esterase variations in Drosophila melanogaster. Hereditas 51: 212. Ford, E. B. 1945. Polymorphism. Biol. Rev., 20-88. Futch, D. G. 1966. A study of speciation in South Pacific Populations of Drosophila ananassae. Univ. Texas Puhl. 6615: 79-120. Johnson, F. M. 1966. Drosophila melanogaster: Inheritance of a deficiency of alkaline phos­phatase in larvae. Science 152: 361. Kimura, M. 1956. Rules for testing stability of a selective polymorphism. Proc. Natl. Acad. Sci. 42: 336-340. -----,and J. F. Crow. 1964. The number of alleles that can be maintained in a finite population. Genetics 49: 725-738. Kojima, K. and Y. N. Tobari. 1969. The pattern of viability changes associated with genotype frequency at the alcohol dehydrogenase locus in a population of Drosophila melanogaster. Genetics 61: 201-209. -----, and K. M. Yarbrough. 1967. Frequency-dependent selection at the Esterase 6 locus in Drosophila melanogaster. Proc. Natl. Acad. Sci. 57: 645-649. Stone, W. S., M. R. Wheeler, F. M. Johnson and K. Kojima. 1968. Genetic variation in natural populations of members of the Drosophila nasuta and Drosophila ananassae subgroups. Proc. Natl. Acad. Sci. 59: 102-109. Stone, W. S., F. M. Johnson, K. Kojima and M. R. Wheeler. 1968. Isozyme variation in island populations of Drosophila. I. An analysis of a species of the nasuta complex in Samoa and Fiji. Univ. Texas Puhl. 6818: 157-170. Walker, H. M., and J. Lew. 1953. Statistical Inference. Holt, Rinehart and Winston, Inc. TABLE 1 EST C Phenotypes in F1 Offspring from Wild-Caught Dark ananassae Females. No. offspring. EST C Phenot>·pe ';( Ident. 9¥ D2dured Parental No.2 Phenol>'Pe> 3.4 ~-') 4 4.·) Genol»pes 1BDK-24 3 50 3,3 x 3,3 2CDK-137 3 63 3,3 x 3,3 2CDK-62 3 25 33 x 3,3 3DDK-83 3 23 3,3 x 3,3 1EDK-74 3 52 3,3 x 3,3 213 2CDK-7 3,4 14 8 7 3 4 x 3,4 1EDK-14 3,4 14 9 3 3,4 x 3,4 4FDK-13 3,4 7 22 8 3,4 x 3.4 4FDK-101 3,4 13 20 11 3,4 x 3,4 48 59 29 4FDK-17 3,4 28 16 3,4 x 3,3 4FDK-18 3,4 22 18 3,4 x 4.4 1ADK-8 3 14 23 3,3 x 3,5 1ADK-7 3,5 42 22 3,5 x 3,3 2CDK-170 3,5 23 14 3,5 x 3,3 4FDK-104 3,5 25 25 3,5 x 3,3 104 84 4FDK-901 4,5 21 8 5 9 7 4,5 x 3,4 1 The extra phenotype possibly resulted from contamination of the Fgeneration with F~ imh·iduals. ,·ee text. 1 2 The first numeral represents the collecting locality, Fig. 1. TABLE 2 ACPH Phenotypes in F 1 Offspring from Wild-Caught Dark ananassae Females. No. offspring, ACPJI Pheno1ype <; Jdent. 9 0 0 ~ <( 0 0 ~ 0 0 0 FIG. 2. Distribution of species of the nasuta subgroup. 1. Drosophi.la albomicans; 2. D. pal­lidifrons: n.sp.; 3. D. pulaua; n.sp.; 4. D. kepulauana, n .sp.; 5. D. kohkoa, n.sp. From each of the repeated collections made in the Hawaiian Islands from 1966­1968, iso-female lines have been checked for mating behavior and taxonomic identity; and several sorts of cytological and genetic tests have been made, show­ing all such lines to be the same. This Hawaiian form, in which no inversions have been observed, has been used as our "standard" for genetic tests as well as cytological comparisons. The immigrans group The immigrans group, or complex, contains an estimated 70 nominal species, without including the 25 species of African Zaprionus and 7 species of Samoaia, which are fairly certainly related to the complex. Duda (1923) established the subgenus Spinulophila for some of these species, and later (1925) changed the name to Acanthophila (which is preoccupied by Acanthophila Hein, 1870, in Lepidoptera). Other species have been placed in Chaetodrosophilella Duda (at times considered a genus, at others a subgenus of Drosophila) or in Phorticella Duda (sometimes considered a genus, at other times a subgenus of Zaprionus). Judging from the original descriptions, and from published keys, the 70 ± species seem to be assignable to five discrete subgroups, with a few species being so im­perfectly known or so aberrant that they cannot be assigned as yet even on a ten­tative basis. As a "first approximation," then, the following subgroup:; may be distinguished: Subgroup I. The "typical" immigrans types, the males with some degree of modification of the fore tarsi. The nominal species which are presumed to be­long here1 are: 1. brouni Hutton 1901 9. monochaeta Sturtevant 1927 2. cilifemur Villeneuve 1923 10. ruberrima de Meijere 1911 3. curviceps Okada & Kurokawa 11. rubra Sturtevant 1927 4. fiexipilosa Pipkin 1964 12. signata Duda 1923 5. formosana Duda 1926 13. subfasciata de Meijere 1914 6. formosana Sturtevant 1927 14. synpanishi Okada 1964 7. immigrans Sturtevant 1921 15. unicolor de Meijere 1914 8. metallescens Malloch 1934 16. ustulata de Meijere 1908 Subgroup II. The "nasuta" subgroup; males lack tarsal ornamentation, but usually have the frons silvery-whitish, all or in part. This section of the immi­grans group is discussed in greater detail in the following pages. 1. albomicans Duda 1924 9. pallidifrons Wheeler, n. sp. 2. albovittata Duda 1926 10. pulaua Wheeler, n. sp. 3. bilimbata Bezzi 1928 11. setifemur Malloch 1924 4. kepulauana Wheeler, n. sp. 12. spinofemora Patterson & Wheeler 5. kohkoa Wheeler, n. sp. 1942 6. komaii Kikkawa & Peng 1938 13. sulfurigaster Duda 1923 7. nasuta Lamb 1914 14. willowsi Curran 1936 8. nixifrons Tan, Hsu &Sheng 1949 1 No attempt is being made to indicate synonyms; homonyms, etc. at this time. The University of Texas Publication Subgroup III. The "quadrilineata" subgroup. The:::e species all have a pale mesonotum with a prominent series of darker stripes. 1. annulipes Duda 1924 8. notostriata Okada 1966 2. circumdata Duda 1926 9. pentastriata Okada 1966 3. crockeri Curran 1936 10. pseudotetrachaeta Angus 1967 4. hexastriata Tan, Hsu & Sheng 11. quadrilineata de Meijere 1911 1949 12. solennis Walker 1860 5. lineata van der Wulp 1881 13. tetrachaeta Angus 1964 6. lineolata de Meijere 1914 14. virgata Tan, Hsu &Sheng 1949 7. nigrilineata Angus 1967 Subgroup IV. The"lineosa" subgroup. These species are similar to Zaprionus, s.s., in having prominent silvery to chalky-white longitudinal stripes on the me­sonotum. 1. albicornis (Enderlein) 1922 7. lineata (de Meijere) 1911 (Zaprionus) (Stegana) 2. albostriata Malloch 1924 8. lineosa (Walker) 1860 3. argentostriata Bock 1966 (N otiphila) 4. bakeri (Sturtevant) 1927 q. multistriata Duda 1923 (Zaprionus) 10. multistriata (Sturtevant) 1927 5. bistriata de Meijere 1911 (Zaprionus) 6. fenestrata (Duda) 1923 11. obscuricornis (de Meijere) 1915 (P horticella) (Stegana) 12. silvistriata Bock &Baimai 1967 Subgroup V. The "hypocausta" subgroup. In this subgroup there is usually a strong sexual dimorphism in body color, males being much darker; further, the comb-like bristle row on the inner side of the first femur is absent or poorly de­veloped. 1. calceolata Duda 1926 t;, pararubida Mather 1961 2. hypo~austa Osten Sacken 1882 6. rubida Mather 1960 3. hypopygialis Malloch 1934 7. xanthogaster Duda 1924 4. nasutoides Okada 1964 Subgroup VI. The following species are listed in this strictly artificial group since their true affinities are unknown or they seem to be quite aberrant with respect to groups I-V. 1. balneorum Sturtevant 1927 4. fuscicostata Okada 1966 2. coei (Okada) 1966 5. trichaeta Angus 1967 (Chaetodrosophilella) 6. trilimbata Bezzi 1928 3. cubivicittata Okada 1966 The nasuta subgroup In addition to discussing each of the ten "names" which have been used in this subgroup, we are also naming four new species and one new subspecies which were discovered in the course of the cytological and hybridization studies. In all, we are recognizing eight valid species, as follows: nasuta (sensu strictu) and nixifrons, neither of which has been available in laboratory culture; sulfuri­gaster and pulaua, with silvery-white frontal orbits on males; albomicans, kepu­lauana and kohkoa, in which the entire male frons is silvery to whitish; and pallidifrons, the only member lacking frontal pollinosity. Table 2 and Figures 1 and 2 show the localities from which we have had labora­tory cultures. Other reported distributions are not shown since we have had no way to verify the species identifications. This problem is especially acute since the species of this complex are so remarkably similar in appearance and, as a result, nearly all of them have been placed at one time or another as synonyms of nasuta (see, for example, Wheeler and Takada, 1964: 180). Type specimens of the new forms described here are located in the Drosophila Type and Reference Collection, University of Texas, Austin. 1. Drosophila albomicans Duda 1923:43, 47, 48; 1924a:209; 1924b:245, Fig. 70; 1926:83, 88-89; 1940:23. D. komaiiis a probable synonym. Duda (1923:47) stated that this new species was being described in another publication from specimens in the Berlin Entomological Museum from Paroe, Formosa. He also reported numerous specimens in the Budapest Museum from Formosa and New Guinea. The other publication never appeared, however, and the name is a nomen nudum here. Hennig ( 1941) reported that the Berlin museum had 8 "Typen" plus 9 other specimens from Paroe, and 47 additional specimens from three other localities on Formosa. Duda (1923:48), however, in a brief discussion of sulfurigaster~ compared albomicans with it, mentioning several traits, so that it is necessary to cite the name as valid at this date and page. The males of albomicans, when viewed "head on," show a silvery to chalky white coloration over the entire frons2• The face, especially the carina, and the third antenna! segments are a little darkened, and there is a noticeable dark longitudinal band on the pleura, reaching back to the vving base. Distribution: Okinawa, Taiwan (Formosa), Pescadores Islands, Thailand. See Table 2 and Figure 2. 2. Drosophila albovittata Duda 1926: 8 7. Duda clearly stated that he was proposing this as a new name for sulfurigaster (based on a single male from Madang, New Guinea) since he now had better material (from Sumatra) that showed a more distinctive characteristic. Such name changes are not allowable, however, as stated in Article 18a of the Interna­tional Rules of Zoological Nomenclature (... a name, once established, cannot afterwards be reiected, even by its own author, because of inappropriateness), and therefore, as an unjustified new name, albovittata is an absolute synonym of sulfurigaster. The taxon represented by Duda's Sumatra material is, according 2 The m)rphological terms to be used here are those currently used by most dipterists, the frons or front being the area between the compound eyes and above the antennae; the face being that part below the antenna! baseline. Ferris (1950), however, stated that the entire area from the ocelli to the ventral margin of the cranium consitutes the frons; the term postfrons was applied to the area between the ptilinal suture and the ocelli, and the term prefrons was applied to the area anterior to, that is, below, the ptilinal suture. The University of Texas Publication 214 .5 .7 F1G. 3. Drawings of male and female genitalia of D. pallidifrons. 3.1 ventral view, .2 dorsal view and .3 lateral view of the ~ internal genitalia; .4 lateral view of the left half of the ~ ex­ternal genitalia; .5 bridge which connects the claspers; .6 spermatheca; .7 lateral view of ovi­positor. Labels on this and other figures are: a) bristled area of ventral side of penis; b) dorsal cylindrical process of penis; c) setae on hypandrium; d) setae on novasternum; e) medioventral fan-shaped flap; £) basal introvert of spermatheca; g) apical indentation of spermatheca. to our present studies, not the same as that from New Guinea, and thus requires description and naming. We are proposing albostrigata, as Drosophila sulfuri­gaster albostrigata Wheeler, new subspecies (see later). 3. Drosophila bilimbata Bezzi 1928: 159. This was described from the Fiji Islands, the localities being listed as follows: from the Lautoka Mountains; at Loloti, "bred from the fruits of Kawika, Eugenia malaccensis"; and "in thousands on decaying pomelos, Citrus decumana, and on decaying fruits of Spondias dulcis." In our own collections, we have found the species to be common and wide­spread in Fiji. It is clearly the same species as that present on most islands of Polynesia, and, judging from the cytological and hybridization data, represents one of three "races" or "subspecies" of a very widespread species, sulfurigaster. Accordingly, we are using the combination, sulfurigaster bilimbata for this Polynesian type; spinofemora Patterson and Wheeler, which came originally from Hawaii, is a synonym. The known distributions of the three types of sulfurigaster are shown on Figure 1 and listed in Table 2. 4. Drosophila komaii Kikkawa and Peng 1938: 525. The types were originally placed in the Zoological Institute of Kyoto Imperial University, but have now been moved to Tokyo (Dr. T. Okada, personal com­munication); they came from Amami-Osima and Isiga Kizima, of the Ryukyu Island group, and from Taihoku [Taipei3 ] and Sintik.u [Hsin-chu], Taiwan. In most respects it seems to agree with albomicans Duda, also known principally from Okinawa and Taiwan. A major discrepancy is cytological-komaii was described as having chromosomes of the "D" type, i.e., three pairs of rods and one pair of V's; we have not found this configuration in any stocks from Okinawa or Taiwan and feel fairly confident that the earlier report was in error. Accordingly, we are placing komaii as a synonym of albomicans. 5. Drosophila nasuta Lamb 1914:346; Fig. 30; Pl. 20, Fig. 32. The specimens came from the Seychelles Islands, from several localities vary­ing from 800 to 2000 feet. Males have the entire frons silvery; this was not only recorded by Lamb but was also seen by Duda ( 1940) who wrote: I received from Mr. Lamb a 3 and ~ from the Seychelles Expedition which match exactly my species [i.e., albomicans]. His remarks should mean only that the two were much alike, and in view of the great similarities between the species of this complex and the remarkable degree of speciation which has taken place in south­east Asia, it seems more reasonable to assume that the true nasuta, more than 3000 miles from the closest known albomicans, is still another species. 6. Drosophila nixifrons Tan, Hsu and Sheng 1949:202. The culture which was used to describe this species came from Meitan, China, a village south of Chungking, about 800 miles from the nearest seacoast. We cannot recognize the species among the stocks which are now available; the chromosome constitution, especially, is not at all like any we have seen: one pair of V's, three pairs of rods (both X and Y are rods, the Y being longer), and one pair of dots. We are considering nixifrons to be a valid, but unrecognized species with silvery orbits. 7. Drosophila set ifemur Malloch 1924: 351; redescription, Clark 195 7: 221. This Australian species is a member of the sulfurigaster complex; it also 3 Modern place-names for Taiwan were furnished by Mr. Fei-Jann Lin. .5 Fm. 4. Drawings of male and female genitalia of D. sulfurigaster bilimbata. The sequence of drawings is as in Fig. 3. occurs on New Guinea (see Figure 1, and Table 2), and forms one of the three "subspecies" of sulfurigaster. Nomenclaturally we are placing it as a synonym of D. s. sulfurigaster (type from New Guinea). 8. Drosophila spinofemora Patterson and Wheeler 1942: 104. The type-culture came from Honolulu, Hawaii; this fly is not only very common and widespread in the Hawaiian Islands but is the same as those found on nearly all Polynesian islands, as shown on Figure 1 and in Table 2. D. sulfurigaster bilimbata appears to be the best name for these populations, and spinofemora is a synonym. 9. Drosophila sulfurigaster Duda 1923:48; 1926: 83, 87-89. Duda's original material was one male from the Budapest Museum, labelled "N. Guinea, 1896, Friedrich-Wilhemshafen"; this locality is now known as Madang. He characterized it as having only the frontal orbits whitish rather than the entire front as in albomicans. He later (1926) reported on numerous specimens from Sumatra which he considered to be the same sp2cies, and since the whitish orbits of the males seemed to be so characteristic, he changed the name to albovittata, and gave a redescription (see remarks under albovittata) . Our work on comparative cytology and hybridization indicates that the New Guinea-Australian type forms a race or subspecies of sulfurigaster, the Malay­sian, Indonesian, Philippine populations belong to a different subspecies, while the Polynesian group represents a third type. Wakahama (personal communication) has good evidence that some form of sulfurigaster occurs on Taiwan; we have not seen cultures of it from this area, however, and cannot say which of the three subtypes it represents. Figure 1 and Table 2 show the distributions of the three subspecies of sulfurigaster, based on the present study. D. sulfurigaster sulfurigaster (with synonyms, setifemur and willowsi) occurs only in the Australia-New Guinea region. D. sulfurigaster bilimbata (synonym: spinofemora) is widespread, rang­ing from Hawaii and Polynesia to Guam. The third form is as follows: 10. Drosophila sulfurigaster alhostrigata "Wheeler, n. subsp. Type locality: Semongok Forest Reserve, 12.5 miles south of Kuching, Sara­wak, Borneo, Malaysia. Type culture: U. T. No. 3121.2-~ 33, collected March 17, 1968 by Drs. D. E. Hardy and M. Delfinado from bananas and fruits of Lanzium domesticum. Additional collections, all from southeastern Asia, are shown on Figure 1 and in Table 2 and include the Philippine Islands, continental Malaysia, Cambodia and Thailand. We cannot find any satisfactory distinguishing morphological characters to separate this form from the other subspecies of sulfurigaster. It differs greatly, however, in the number and distribution of chromosomal inversions (Table 4) and in its behavior in crosses to the other subspecies (Table 9). 11. Drosophila willowsi Curran 1936:42. This species was based on two males from the Santa Cruz Islands, about midway between the New Hebrides and the Solomon Islands. Since only the orbits have a whitish pollinosity, this is probably the same form as sulfurigaster on New Guinea and New Ireland: D. s. sulfurigaster. 12. Drosophila kohkoa Wheeler, n. sp. Type locality: Ari Ksatr, a village across the Mekong River from Phnom Penh, Cambodia. Type culture: U. T. No. 3057.3, collected May 1967 by M. Delfinado from fallen mango fruits and banana bait. Additional material from: Semongok Forest Reserve, Sarawak, Malaysia; Bon Chakkrarat, northeast of Bangkok, Thailand; Balsahan River, Iwahig, Palwan, Philippine Is.; Manilos, Belait Dist., Brunei, Borneo. The University of Texas Publication 5 FIG. 5. Photographs of male and female genitalia of D. pallidifrons. The name is compounded from the Cambodian Koh, meaning island, and from Thai Ko, also meaning island. The chromosome complement and polytene chromosome patterns are described in the following pages; the courtship be­havior is discussed by Spieth in the next article. The wing is shown in Figure 8 and some details of the male genitalia are shown in Figure 9. In this species the face, and especially the carina and third antenna! segment, is rather darker than in the other species. 13. Drosophila pulaua Wheeler, n. sp. Type locality: S::mongok Forest Reserve, 12.5 miles south of Kuching, Sara­wak, Malaysia. Type culture: U. T. No. 3121.5, descended from a single female from the above locality, collected in March 1968 by D. E. Hardy and M. Delfi­nado. There nre no additional records. The name is from Malaysian, meaning island. The chromosome complement and polytene chromosome patterns are described in the following pages; the courtship behavior is discussed by Spieth in the next article; the wing is shown in Figure 8 (}nd some G.etails of the male genitalia are shown in Figure 9. 14. Drosophila kepulauana Wheeler, n. sp. Type locality: Manilos, Belait Dist., about 65 miles southwest of Brunei Town, Brunei, Borneo. Type culture: U. T. No. 3122.3, collected March 1968 by D. E. Hardy and M. Delfinado. Additional material: a culture from Panitian, Gungnan, southwestern Palawan, Philippine Is. The name is Indonesian, mean­ing archipelago. The chromosome complement and polytene chromosome patterns are described in the following pages; the courtship behavior is discussed by Spieth in the next article. The wing is shown in Figure 8 and some details of the male genitalia are shown in Figure 9. 15. Drosophila pallidifrons Wheeler, n. sp. Type locality: Kolonia, Ponape Island, eastern Caroline Islands. Type culture: U. T. No. 2535.4 collected in July and August, 1959, by M. R. Wheeler, W. S. Stone and M. Wasserman, mostly from fallen breadfruit. Fm. 6. Photographs of male and female genitalia of D. sulfurigaster bilimbata (.1 -.6) ;.7 spermatheca of D. albomicans. The University of Texas Publication Frn. 7. Photographs of internal male genitalia of: .1-.2 D. albomicans (from Taiwan); .3-.4 D. sulfurigaster sulfurigaster (from Madang, New Guinea) ; .5-.6 D. sulfurigaster bilimbata (from Samoa); .7-.8 D. sulfurigaster albostrigata (from Kuala Lumpur, Malaysia) . Additional material: U. T. Stock No. 3131.2, from the same general area on Ponape, April 1968, H. T. Spieth, coll. This is the most distinctive species of the subgroup; not only do males lack the silvery pollinosity of the frons but there are also decided differences in both male and female genitalia, illustrated in Figures 3 and 5. The wing is shown in Figure 8. The chromosome complement and polytene chromosome patterns are described in the following pages, and Spieth discusses the courtship behavior in the next article. Notes on the male and female genitalia of the nasuta subgroup The members of the nasuta subgroup are remarkably similar in their syste­matic characters. For example, the egg and puparium of albomicans, figured by Okada ( 1968, as "nasuta?"), are essentially identical in all the species. The wings, shown in Figure 8, are similarly alike, with only a few minor differences being evident. The male and female genitalia, although showing the same gross morphology for all the species, reveal some detectable differences in most in­stances. D. pallidifrons, from Ponape, is the only highly distinct species. Fig-. ures 3 and 5 illustrate the special features: numerous short, strong bristles on FIG. 8. Photographs of the wings of: .1 sul/urigaster bilimbata; .2 pulaua; .3 albomicans; .4 kepulauana; .5 kohkoa; .6 pallidifrons. All photographs were taken at the same magnification and developing and printing were identical; therefore size and pigmentation differences are real. FIG. 9. Photographs of parts of the internal male genitalia of: .1 and inset, pulaua; .2 kepu­lauana; .3 kohkoa. the ventral side of the pen:s; reduced size of the dorsal cylindrical expansion of the penis; two large sensilla and numerous setae on the novasternum; extremely small medioventral fan-shaped flap; numerous small setae on the hypandrium (see Okada, 1964, for terminology) ; a secondary row of stout bristles on the ovipositor; absence of an apical indentation on the spermatheca; and rougher surface structure of the spermatheca. All other species are very similar, as is shown in Figures 4, 6 and 7. A careful study of the spermathecae might reveal statistically significant differences in proportions, such as width: length ratio, or in ratios of parts, as indicated in Figure 4.6. Here the ratio of length of basal introvert (I): length of apical indentation (II): overall length (III) was found to be 1: 1: 2.3 for albomicans, and 1: 1.5: 2.5 for sulfurigaster (Malaysian stock). These two also differ slightly in the width:length ratio-1: 1 for albomicans, and 1: 1.2 for Malaysian sul­furigaster. The figures of the male genitalia of sulfurigaster bilimbata are representative of all the species except pallidifrons, and as shown in Figures 4, 6 and 7, these species are all remarkably similar. The medioventral fan-shaped flap appears to be the most useful gen~talic character, but only in pulaua is it obviously different, as shown in Figure 9 .1. A more detailed study of the male genitalia, including variation within populations, is underway and will be reported later. MATERIALS AND METHODS The first series of tests in this investigation was made with flies from random stocks maintained at the University of Texas laboratory with the majority of them being subspecies of D. sulfurigaster. The D. s. bilimbata stocks were from collections made primarily by Stone and Wheeler in the South Pacific islands from 1955-1965 and included samples of populations from Hawaii, Palmyra, Tutuila, Savaii, Upolu, Niue, and Tonga. D. s. sulfurigaster material consisted of three probable iso-female lines from New Guinea and one from New Ireland collected by Marvin Wasserman in 1961, and one from Queensland, Australia received from Wharton B. Mather in 1955. The three D. s. albostrigata iso-female lines from continental Malaysia were received from Dr. Wasserman in 1962. Other laboratory stocks used include D. pallidifrons from Ponape collected by Wheeler, Stone, and Wasserman in 1959, and a stock of albomicans from Oki­nawa, which was given to us by Drs. Toyohi Okada and Osamu Kitagawa in 1966. In 1966 and 1967 Stone and \7\Theeler collected more extensively in the islands of the South Pacific, providing additional strains from Hawaii, Palmyra, Tutuila, Savaii, Upolu, and Fiji. In 1967, Drosophila collections by Dr. Mercedes Del­ finado yielded material from Palawan and Luzon, Philippine Islands, and from Cambodia; and Dr. L. H. Throckmorton sent us population samples from Taiwan and the Pescadores Islands collected by him and Mr. Fei-Jann Lin. In 1968 addi­ tional cultures were received from Drs. H. T. Spieth and H. L. Carson who col­ lected on Ponape and Guam, from Throckmorton and Lin who made further collections in Taiwan and the Philippine Islands, and from Dr. D. Elmo Hardy and Dr. Delfinado who collected in Cambodia, Thailand, Sarawak, and Sabah in Malaysia, Brunei, Borneo and in the Philippine Islands. Additional flies from Luzon, P. I., were received from Dr. Carmen Kanapi. See Table 2 for a complete listing of materials. Females from the wild were isolated individually when they were received. Where possible males and females were separated upon collecting; therefore, the majority of iso-female lines established were from matings in the wild and not from matings in vials during shipment. Larval brain squashes were made in the laboratory of each iso-female line and random stock to check the metaphase karyotype. Salivary gland preparations were made of each strain to determine heterozygosity for inversions; from thirty to fifty slides were made of random stocks (established from several to many iso­female lines) and at least ten slides were made from each iso-female line (al­though not necessarily of Fi larvae). Representative lines were then selected from each locality for a complete rnries of crosses to detect sexual isolation and cytological variation. Reciprocal crosses were made between flies from each of the different geograph­ic locations. No attempt was made to ascertain percent fertility from pair matings at this time because of some degree of sexual isolation in many instances between members of the same species, and a high degree of sexual isolation in many inter­specific crosses. Consequently, all matings were mass matings. They were carried out in large vials on cornmeal medium. Virgin females and males were collected twice each day and aged five to seven days before mating. In intraspecific crosses 15-20 pairs of flies per vial were used; in interspecific crosses 30 pairs per vial were used. Initially, five such mass matings were made for intraspecific crosses, and as many as ten to fifteen for interspec~fic crosses; where crosses were sterile, many of them were repeated in an attempt to get Fi larvae for cytological anal­ysis. Flies were transferred to fresh food every five days and at the end of thirty days were called sterile if no larvae were visible by that time. From sterile Pi crosses the females were dissected in many cases to check for sperm in their spermathecae or feminal receptacles. When Fi flies were obtained, sex ratio counts were noted and Fi~~ XFi J J crosses were made to test for fertility. If F1XFi were sterile, a sample of both males and females was dissected to check for the presence of sperm in the testes or sperm storage organs. In addition, Fi fl~e> were sometimes backcrossed to both p3rent strains to s~e if either F1 males or Fi females were then fertile. After analyzing the random stocks and iso-female lines as indicated above, salivary gland slides were prepared from the larvae of all fertile crosses to the standard Hawaiian strain and examined for fixed inversion differences, and photomicrographs were made. Also, a representative photographic record was kept of all inversion differences noted in subsequent Pi crosses. A more extensive series of photographs will be published at a later date than are shown here. Metaphase brain smears from each random or iso-female line were prepared by removing the ganglia of male larvae in physiological saline and allowing them to swell in a hypotonic solution of 1%sodium citrate for ten minutes. The ganglia were then transferred to aceto-orcein stain for five minutes and mounted in 45% acetic acid solution. Pressure was applied to the cover glass with a dissecting needle to spread the cells before squashing with thumb pressure; the cover glass was then sealed with clear nail polish. to TABLE 2 ~ Strains of the D. nasuta subgroup species used in genetic tests and cytological investigations. U. T. slnrk Type of stock; nmnhcr 11tm1hcr Lornlily collcclcd of lines d1crkcd c:ylrilop;ically Dale collPclcd and c:ollcctor(s) Linc(s) usccl in crosses D. sulfurigaster D. s. bilimbata 3045.1 Tantalus, Oahu, Hawaii random stock 1966; J. Gross field 3045.1 ~ ~ 3045.12 Hilo-Kona, Hawaii random 1966; W. S. Stone & D. E. Hardy ~ 3045.13 Maui, Hawaii random 1966; W. S. Stone & H. T . Spieth ~ 3074.1 Kipuka 9, Saddle Rd., Hawaii, Hawaii 16 iso-~ lines 1967; M. Kambyscllis ~ ~· 3074.2 Bonsey garden, Maui, Hawaii 20 iso-~ lines 1967; H. T. Spieth ~ 3074.3 Kipahulu, Maui, Hawaii 14 iso-~ lines 1967; M. Kambysellis ~· 3074.4 Bonsey garden, Maui, Hawaii 15 iso-~ lines 1967; H. T. Spieth ..._ a 3074.5 Kapaa, Kauai, Hawaii 18 iso-~ lines 1967; C. Kanapi 3074.6 Manoa Valley, Oahu, Hawaii 22 iso-~ lines 1967; S. Rockwood ~ 3074.7 Wiliwilinui, Oahu, Hawaii 6 iso-~ lines 1967; C. Kana pi "\:j 3045.5 Palmyra I., Line Is. random 1962; W. S. Stone & M. R. Wheeler i:::: <::3-' ..._ 3045.6 Palmyra I., Line Is. random 1962; W. S. Stone & M. R. Wheeler 3045.6 -2a· · 3045.2 Pago Pago, Tutuila, Am. Samoa random 1962; W. S. Stone & M. R. Wheeler 3045.2 3045.4 Taputimu, Tutuila, Am. Samoa random 1965; W. S. Stone & C. P. Oliver ~ 3045.7 Aopo, Savaii, Samoa random 1965; W. S. Stone & C. P. Oliver 3045.7 3061.2 Aopo, Savaii, Samoa random 1967; W. S. Stone & M. R. Wheeler 3045.8 Apia, Upolu, Samoa random 1965; W. S. Stone & C. P. Oliver 3045.8 3045.9 Niue I. random 1965; W . S. Stone & M. R. Wheeler 3045.9 3045.10 Tongatapu, Tonga Is. random 1963; W. S. Stone & M. R. Wheeler 3045.10 3044.3 Coloisuva, Viti Levu, Fiji random 1966; M. R. Wheeler 3044.3 3063.2 Nadarivatu, Viti Levu, Fiji random 1967; W . S. Stone & M. R. Wheeler TABLE 2-(Continued) Strains of the D. nasuta subgroup species used in genetic tests and cytological investigations. U. T. stock Type of stock; number number Locality collected of lines checked cytologically Date collected and collector(s) Line(s) used in cros•:es 3100.2 Korolevu, Viti Levu, Fiji 80 iso­~ lines 1968; M . R. Wheeler 3071.6 Guam, Marianas Is. 60 iso-~ lines 1968; H . L. Carson 3071.6 ~ D. s. sulfurigaster ~ 3017.4 3019.8 Kavieng, New Ireland Wau, New Guinea iso-~ line iso-~ line 1961; M. Wasserman 1961; M. Wasserman 3017.4 3019.8 0 ~ ro- 3020.2 Brown River, Papua, New Guinea iso-2 line 1961; M. Wasserman 3020.2 O.l !""""' 3016.2 Madang, New Guinea iso-~ line 1961; M. Wasserman 3016.2 t:l 2372.16 Queensland, Australia random 1955; W . B. Mather 2372.16 >-1 0 (f) 0 D. s. albostrigata 3054.2 Cabuyao, Laguna, Luzon, Philippines iso-~ line 1966; M. Delfinado 3054.2 "'dg-: P.i"' 3138.2 Tagaytay, Luzon, Philippines iso-~ line 1966; L. H. Throckmorton & F. J. Lin 3138.2 ~ O.l 3139.2 Luzon, Philippines 11 iso-~ lines 1968; C. Kanapi 3139.2 ~ 2, ~ 3 CJ> ~- 3126.2 Baguio, Luzon, Philippines 2 iso-~ lines 1968; D. E. Hardy & M. Delfinado 3126.2 ~a O.l ""' 3056.2 Panitian, Palawan, Philippines 8 iso-~ lines 1967; M . Delfinado 3056.2 ~ 3, ~ 6 ~ ~ 3146.2 3057.2 3120.2 Los Banos, Luzon, Philippines Ari Ksatr, Cambodia Siem Reap, Cambodia 5 iso-~ lines 27 iso-~ lines 3 iso-~ lines 1968; L. H. Throckmorton & F. J. Lin 1967; M. Delfinado 1968; D. E. Hardy & M. Delfimdo 3146.2 ~43 3057.2 ~ 9, ~ 10, 212 3120.2 ~ b, ~ c ~ ""j 0 ~ ""l;;j 3119.2 Semongok Forest Reserve, Sarawak 3 iso-~ lines 1968; D. E. Hardy & M . Delfinado 3119.2 ~a, ~c 3121.2 Semongok Forest Reserve, Sarawak 49 iso-~ lines 1968; D. E. Hardy & M. Delfinado 3121.2 ~ 33, ~ 52 3122.2 Brunei, Borneo 72 iso-~ lines 1968; D. E. Hardy & M . Delfinado 3122.2 ~ 44, ~ 60 3124.2 Jesselton, Sabah, Sarawak 7 iso-~ lines 1968; D. E. Hardy & M . Delfinado 3124.2 ~ 2, ~ 5 3033.17­ Kuala Lumpur, .18-.49 Malaysia 3 iso-~ lines 1962; M. Wasserman 3033.17 to to (Jl ~ TABLE 2-(Continued) O') Strains of the D. nasula subgroup species used in genetic tests and cytological investigations. U. T. stork Type of stock; number number Locality collected of lines checked cytologically Date collected and collector(s) Line(s) used in crosses 3116.2 Bon Chakkrarat, Thailand 43 iso-9 lines 1968; D. E. Hardy & M . Delfinado 3116.2 9 21 , 9 29 3117.2 Bon Phra, Thailand 3 iso-9 lines 1968; D. E. Hardy & M. Delfina­ ~ ~ -·-----· ('\) + ~ ;::::! ~· ('\) + ""I+ ~. ....... + ~ + 0 ........... + ~ ~ c,, Al-I-, Cl+ ~ "ti Al+ ~ Al-1-,CI+ ......... ...... ""' Al+ , c;+ (") ~ ....... c1+ ...... 0 A,CI+ ;::::! Al+, Bl+, Cl + + + + TABLE 4---Continucd C:hromosomes Species & Strains x 2H 2J. D. pulaua D. pallidifrons D. kepulauana Pa1awan, P. I. Brunei, Borneo D. albomicans Okinawa Taiwan Pescadores Islands Thailand D. kohkoa Palawan, P. I. Brunei, Borneo Sarawak, Malaysia Cambodia Thailand + + A, B, G, I,J B,G, I,J B, G, I, J B,G, I, J B, G, I, J Al+, B, G, I, J B, C, I, J B, C, I, J B, C, I, J B, C, E, I, J. B, C, E, I, J. + + B,C B,C c c c Cl+ C,D C,D C,D C,D C,D + + F Al+,F Al+,F,HI+ Al+,F,HI+ F,HI+ Al+, F, HI+, II-!-, JI+ Al-f-, El +,F Al+,El+,F Al+,El+,F El+,F Al+,El -!-,F + + B,G,O,Q A, B, C/+, G, 0, Q B,F,G, R B,Fl +,G,R B,F,G,R Al+,Bl+,Fl+,G,KI+ A,D,Hl +,R A, DI+, El+, HI-!-, II+, JI+, Kl+, R A, D, E/-!-, 1/-!-, Kl+, L/+, O/+, R A, D, HI+, II-!-, R A,D,R ~ ......... "> 0 ;::s ('t) -Q.) .......... t1 "'i 0 (f) 0 ""O ~ ..... Si" t::::l Q.) -(J) ~ Q.) "> !::::: \:)"" (Jt) ""t 0 !::::: "t:j + indicates the standard chromosomal sequence. Capital letters alone indicate fixed inversion differences from the standard. Capital letters/+ indicate inversion heterozygosity. to w Ul D. pallidifrons and D. pulaua are homosequential with the standard. No in­versions were found in the 1959 laboratory stock of D. pallidifrons nor in the 12 iso-female lines collected in 1968, and none is present in the iso-female line of D. pulaun. Of the 39 recorded inversions, 30 are found to be heterozygous in one or more of the four other species; 18 of these are found only as heterozygotes with no populations being fixed for the inverted sequence. Nineteen inversions are fixed in one or more localities and/or species, with 9 of these always fixed. The three subspecies of D. sulfurigaster share none of the fourteen inversions found in this widespread species .All of these inversions exist heterozygous, but three of them are fixed in all strains from two different populations. D. s. bilim­bata strains from twenty-one localities in the South Pacific islands have been extensively checked and each has been found to possess only the same sequence as the Hawaiian standard. Population samples of D. s. sulfurigaster are limited to one line each from five localities in the Australia-New Guinea area, but from larger samples, Mather (1962) and Clark (1957) reported no inversions. We have found two heterozygous inversions in this subspecies: there is one on the X chromosome of the Madang, New Guinea strain, XD, and one on the 2L of the New Ireland strain, 2LD; neither has been seen again in any of the other species of the subgroup. The third subspecies, D. s. albostrigata, is quite polymorphic in all localities whe~e it has been collected. A total of twelve heterozygous inversions distributed among all four long euchromatic arms have been analyzed, with three of them being fixed in certain continental populations. The two inversions on the 2R have not been found to exist in the same population. 2RA is seen only in this subspecies and only in the Thailand and Cambodia lines where it is heterozygous when the Y chromosome is a V but is homozygous in those lines where the Y chromosome is a J. 2RB is heterozygous in most of the island populations of FIG. 15. Fixed inversion differences on chromosome 3 between: (a) D. albomicans X D. s. albostrigata; (b) D. kohkoa X D. albomicans; (c) D. kepulauana X D. albomicans. D. s. albostrigata with Malaysia being the only continental area having it. Of the four heterozygous inversions on 2L, only 2LA is observed from all localities and is the only 2L inversion shared with other species. Moreover, it is found in all three other species which possess inversions. 2LC is common to the island members and continental Malaysia, 2G is found in strains from Cambodia and Thailand, and 2LB belongs only to the Thailand strains. Two of the three heterozygous inversions of the third chromosome, 3A and 3C, are widespread in this subspecies. They are seen in other species of the subgroup as is 3B, but Cambodia is the only locality where 3B has been observed in D. s. albostrigata. 3A of continental Malaysia is the only fixed inversion difference on the 3 from the standard. XC is found throughout the island populations and in a few conti­nental areas; it is fixed in continental Malaysia. XB is limited to the island strains of D. s. albostrigata and both XB and XC are found as fixed inversions in one or more of the other species. XF, seen only in this subspecies, is present in lines from the Philippine Islands and Cambodia. The remaining 25 analyzed inversions of the D. nasuta subgroup (Table 4) have been observed only in one or more of the three species characterized by more extensive silvery markings on the frons: D. kepulauana, D. albomicans, and D. kohkoa. Eighteen of the fixed 19 inversion differences found throughout the subgroup exist in these species, and they share one or more of the fixed inversion differences on three of the chromosome arms: 2RC, 2LF, and XB, XI, and XJ. Each of the three species has a set of fixed simple inversions or inversion FIG. 16. Some inversion differences from the standard on chromosome 2L. (a) 2LA and 2RA of D. s. albostrigata. 2LA is polym::>rphic in D. sulfurigaster, D. kepulauana, D. kohkoa and D. albomicans. (b) Three polymorphisms of D. albostrigata. This complex is made up of 2LA, 2LB and 2LC. (c) 2LA and 2LC form this included configuration. (d) Inversion differences be­tween D. albomicans and D. albostrigata include 2LA, 2LC and 2LF. (e) Two of the polymor­phisms found in D. albomicans (Thailand). 2LI and 2LA form an overlapping configuration. FIG. 17. Fixed inversion differences on the X chromosome between: (a) D. albomicans and D. s. bilimbata; (b) D. kohkoa and D. kepulauana; (c) D. albomicans and D. kohkoa. complex differences, a few of which are shared by one of the two other species, but not by both of them. Six inversions which are polymorphic in D. s. albostri­gata are also found in at least one of these species, where all but 2LA tend to be fixed. Thus, 31 of the 39 sequences which differ from the standard have been observed in at least one strain of one of these three species. The population samples of D. kepulauana have at least ten fixed inversion differences from the standard. One of these inversions, XA, is found in the iso­female line from Palawan, Philippines (3056.8), and it has not been observed in any of the lines from Brunei, Borneo. Throughout this species the characteristic "loop" between chromosomes 2L and 2R was absent. Strains of D. albomicans from Taiwan, Okinawa, and the Pescadores Islands are each fixed for the same four simple paracentric inversions, 2RC, 2LF, 3G, and XB, and for inversion complexes on the 3 and the X. In the strains from Thailand some of these inversions are found to be fixed while the others are heterozygous. Additional heterozygous inversions are observed in the Thailand population that are not found from any of the other localities where D. albo­micans has been collected. In D. kohkoa eight fixed simple paracentric inversions and two fixed inversion complexes are found in all lines of the species with the exception of the large XE which is fixed only in the continental population samples. Two inversions on the 2L, 2LA, and 2LF, are the only heterozygous inversions widely distributed throughout the range of D. kohkoa. However, in strains from Brunei, Borneo, and Malaysia seven heterozygous inversions have been observed on chromosome 3 which so far are limited to these lines. Because of the chromosomal variability within D. sulfurigaster, D. kepulauana, D. kohkoa, and D. albomicans, Table 4 breaks down each species into areas of collection to show the intraspecific as well as the interspecific inversions of the D. nasuta subgroup, which have been analyzed at this time. Some of the inver­sions, inversion complexes, and crosses of the D. nasuta subgroup are shown in Figures 15-20. HYBRIDIZATION EXPERIMENTS A series of reciprocal crosses was made among laboratory stocks available in 1966. These stocks were primarily D. sulfurigaster including eight D. s. bilim­ Frn. 18. Inversion differences in crosses between (a) D. s. bilimbata standard and D. kepu­lauana; (b) D. kohkoa and D. albomicans; (c) D. albomicans and D. s. albostrigata; (d) D. s. bilimbata standard and D. albomicans; (e) D. kepulauana and D. albomicans; (£) D. kohkoa and D. kt?pulauna. bata, five D. s. sulfurigaster, and one D. s. albostrigata. The other random stocks used were one D. pallidifrons and two D. albomicans (Table 5). The complete series of crosses of D. sulfurigaster is presented to show the gradation of F1 sterility between different sub-species of this species. Where D. s. bilimbata is the female of crosses to D. s. albostrigata or D. s. sulfurigaster, all crosses are fertile through F1 • Where D. s. bilimbata is the male of the crosses, the results agree with Clark (1957); when D. s. sulfurigaster lines from Australia (2372.16) or Madang, N. G., are the males of these crosses, F1 X Fi is sterile. However, when D. s. sulfurigaster (3019.8) from Wau, N. G., and No. 3020.2 from Papua, N. G., are the males of these crosses, Fi X F1 are consistently slightly fertile, and when the New Ireland ( 3017. 4) line is used, all F 1 X F1 are very fertile. New Ireland is not reproductively isolated from either D. s. bilimbata nor D. s. sulfurigaster. F1 X F1 from crosses with the D. s. albostrigata line from Malaysia (3033.18) behaved the same in crosses as when the D. s. sulfurigaster lines from Australia were used. New material received from collections made in 1967 and early 1968 provided strains of D. s. bilimbata from Guam (3017.6) and population samples of D. s. albostrigata from Luzon, P. I., Palawan, P. I., and from Cambodia. The lines from Palawan and Cambodia were found to have both V-and J-shaped Y chromosomes; strains with each metaphase configuration type were selected to cross to some of the stocks used in crosses of Table 5. The latter strains consist of our standard, D. s. bilimbata Hawaii (3045.1 ), D. s. sulfurigaster Australia Q) rt) rt) 0 rt) ~ g 0 g? ~ 0 Cl) () 0 ~ ~ c Ci) 0 u c::s ~ ...... c::: "' () ~ ~ ~ It:) ~ cs ~ It> rt> "'> N cP Q 0 c 0 Q.. ~ T F -F } ..... 2­ £ j_ -£F s } £ F 2. I-F - 2. - N - It> cD v v 0 c::: "' c;:, rt) 0 rt) -~ 0 ~ • c ~ () 0 0 ~ c It:) ~ 0 0 ~ c::s ·-....__ FF II .£.£ .£..£. F F FF ,___ ,___ .£.£. F F ~~ ; I .£..£. F F FF FF II ISi [fil [fil] ~ ISi [ill] ffi ~ ~ TABLE 5. Results of 1966 crosses between the subspecies of Drosophila sulfurigaster, D. pal­lidifrons and D. albomicans. Symbols are as follows: Capital F within black square: P 1 X P1 & F1 X F1 fertile; Lower-case fin black square: P1 X P1 fertile, F 1 X F 1 slightly fertile; Capital F alone: P 1 X P 1 fertile, F 1 X F 1 sterile; Lower case f alone: P1 X P1 slightly fertile; F1 XF1 not tested; S: P1 X P1 sterile. (2372.16), and New Ireland (3017.4); these D. s. sulfurigaster lines show the two extremes in F1 fertility and sterility between D. s. bilimbata and D. s. sul­furigaster. The D. s. sulfurigaster, Madang line, (3016.2), was also chosen to check further on a male: female sex ratio difference which was observed in crosses where the Madang stock furnished the male of the cross to strains from Malaysia (3033.17). The Malaysia strain was used to represent D. s. albostrigata (Table 6). Results of tests with the strain from Guam placed it with the Polynesian D. s. bilimbata; genetically, it is like the Hawaiian standard, and in the two crosses to D. s. albostrigata males from Cambodia, the F1 male: female sex ratio distortion was present. The D. s. albostrigata lines from Palawan, Philippine Islands, and Cambodia and the single iso-female line from Luzon, Philippine Islands, showed the same Cont. Malavsia 3033.18 Luzon 3054.2 Palawan 3056.2 # 3 Palawan 3056.2 # 6 Cambodia 3057.2 #9 Cambodia 3057.2 # 10 FF F F F F FF F FF F F F F £.£. F F F Cambodia 3057.2 # 12 _£_.£. F F F 0. pollidifrons l'\FFFFFF Fl\FFFFF FF"'FFFF F F Ff'\. F F F FFFFl'\.FF FFFFFl'\.F FFFFFFI'.. I £ £ F .£_ F Ponaoe 2535.4 ITill lslslsl !sit slslsls sl IS) []] [fil [§] 0. kepulauano Palawan 3056.8 ~ lslt Isl Isis slslslt sl ~ IS) [] [§] 0. olbomicons Okinawa 3045.11 [ill] It I tit I ltlt tit It It ti [TI II) (SJ [] 0. kohkoo Cambodia 3057.3 #2 [fil] lsltlsl !sis tlslsls sl [ID [[] [f] [SJ TABLE 6. Results of crosses ( 1967-May, 1968) between additional strains of D. s. albostrigata to the "standard" D. s. bilimbata, and to D. s. sulfurigaster, D. pallidifrons, D. kepulauana, and D. albomicans. Fertility-sterility symbols are as in Table 5. F 1 sterility that the limited material from continental Malaysia had shown. Seven crosses to D. s. sulfurigaster males had the disturbance of male:female sex ratio. Since these crosses were mass matings there was no test of the percent of fertile pairs, but the crosses of D. s. bilimbata and D. s. sulfurigaster to D. s. albostrigata strains were slow in showing fertility and in several instances had to be repeated. Dissections of females to check the spermathecae and seminal receptacles for sperm showed that failure to mate was high (Table 12). Drosophila collections in 1968 enabled us to extend the boundaries of D. s. albostrigata to include Sarawak, Malaysia; Sabah, Malaysia; Brunei, Borneo; and Thailand. In studying individual iso-female lines it was discovered that twelve lines from Sarawak, Malaysia had a sex ratio imbalance which was pre­dominantly female. All collection sites and karyotypes are represented in the reciprocal crosses of these strains to the standard Hawaiian strain (Table 7). Additional crosses among D. s. albostrigata lines were made totaling 100 in all; >­ ~ O' ~ ~ ~Cl> 0 Q. .§ .J: >­ a.­ ~ 9 og, -... Cl') <1>o ~~ c::::i D. sulfuriqoster bilimboto I Hawaiian Standard 3045.1 D. s. sulfuriqoster New Ireland t 3017.4 /\ Australia 2372.6 D. s. olbostriqoto I Cabuyao Laguna, Luzon 3054.2 Tagaytay, Luzon 3138.2 /\ /\ Luzon 3139.2 #2 Baguio Benquet t 3126.20 Baguio Benquet /\ t 3126.2b Los Banos 3146.2 #43 Polawan /\ 3056.2 # 3 PaIowan t 3056.2 #6 c: 0 N :J ~ -' 0 "'­ c: ""O ~ ~ ~ 0 0 ~ c: ~ Cl') c: N ~ -~ :> ~.Q ·N• ~o 6'N u;0 ..... CN U> -'N "gv c:N 0 . N -~ .f: cu • cu . 0 . Cl') -"'­ -'V ON c:: iri CDU) _f'­ >:cri cri CDU> (:) olO oV ~ :..=I'­ cu- Or<) r<') ON ON Or<) oO ·-o ~o ~ c: ...... ·:;;;; 'err> %;;; ·5;;; ~r<') '=> .=N or<"> ~ )::r<') ):: Ill C7' C7' N .Q C7' v) 0 :> 0 cu 0 :J vi 0 ~ ~ CD 2 v N 111N w U> lgw c: l{') c: l{') oV oo oo CD- r<') ~r<') ~r<') Ill ~ 0 0 0 ....J a.. a.. [SJ IF IF IF IF IF IF IF IF I ~ tB ~ F F F F F F F F F F F F F F F F F .2!__ F F* ...:k_ F ~ F ....:k_ F I--­ F ...:k_ F ._A_ F F F F * F F * * F F F F * F F F F F F "" F ../ F ../ F ../ F ../ F ../ F ../ F ../ j "' F ../ j j j j j F ../ F ../ ~ F ../ F ../ F ../ F ../ F ../ F ../ F ../ F ../ "" F ../ F ../ F ../ F ../ F ../ F ./ F ./ F ../ F ../ F ../ F ../ F ../ F ./ F ../ F ./ F ./ ""' F ../ F ../ F ./ F ../ F ../ F ./ F ../ ""'F ../ F ../ ~ ~ ~ ~ ~ ~ j""' "' TABLE 7. Results of crosses of strains of D. s. albostrigata from the Philippines, to the "standard" D. s. bilimbata and to D. s. sulfurigaster. Fertility-sterility symbols are as in Table 5. Stan-ed squares= sex ratio distortion was observed in hybrids; checked squares= normal sex ratio in hybrids. a complete series was not felt necessary, because no deviations were found from previous crosses made within this subspecies. The earlier analysis of the single line from Luzon, P. I., was supported by genetically similar lines from four other locations in Luzon (Table 7). Among crosses with these lines to D. s. bilimbata males, some of the more extreme examples of the unusual male: female Fi sex ratio are seen; for example, 3138.2~ X Hawaiian standard male produced 198 Fi~ : 2 Fi~, and in 3126.2a ~ X Hawaiian ~, there were 280 Fi~ : 52 Fi~. Only three iso-female D. s. albostrigata lines were established from the 1968 collection at Siem Reap, Cambodia. Table 8 shows that they are like the earlier lines from Ari Ksatr, Cambodia, and again the abnormal Fi sex ratio is seen. Examples of the Fi sex ratio differences which have been observed in the crosses between members of the subspecies of D. sulfurigaster are listed in Table 9. In each of these crosses the number of Fi males is significantly higher than the number of Fi females. There were both abnormal males and females as TABLE 8. Results of crosses of 1967-1968 strains of D. s. albostrigata originating from Cam­bodia. Fertility-sterility symbols are as in Table 5; starred squares = sex ratio distortion present. well as some intersex flies in several of the crosses. Further data on different aspects of the sex ratio factor will be published at a later date. The species from Ponape, D. pallidifrons, was the only Pacific Island popula­tion sample from the Stone-Wheeler collections of 1955-1966 which did not fit into the D. s. bilimbata subspecies of D. sulfurigaster, and from none of the ensuing collections was this type seen from any locality but Ponape. The original strain, 2535.4, was crossed to 26 lines of the different species of the subgroup (Tables 5 and 6) with at least fifteen mass matings per cross. P1 fertility was very low, and dissections of females for the presence of sperm showed almost complete sexual isolation between D. pallidifrons and each of the other species. From the 500 or more pairs per cross no more than two F 1 flies were recovered from any one cross. Stock No. 3056.8 was the only iso-female strain from Palawan, P. I., with silvery markings all over the frons from the 1967 collection. Reciprocal crosses to D. albomicans from Okinawa (3045.11) went readily and were fertile through the F'.? showing the two types to be closely related. However, they did not behave the same in crosses to the other species (Tabl es 6 and 10) . The 1969 lines from Brunei, Borneo, were placed in the same group with the Palawan 3056.8 strain after cytological and genetic tests. The Borneo lines used in the crosses were 3122.3 Q No. 1 and Q No. 5. When crossed to D. albomicans they differed from 3056.3 in that the Fi X Fi were sterile. This type has now been named D. kepulauana. In 1967 the only strains collected of D. kohkoa were Cambodia. 3057.3 Q No. 2 was the line chosen to use in crosses to the material available at the time (Table 6). P1 sexual isolation was seen between this D. kohkoa strain and D. sulfurigaster~ D. pallidifrons~ and D. kepulauana. In 1968 D. kohkoa from Sarawak~ Malaysia, Brunei, Borneo, Palawan, Philippine Islands, and Thailand TABLE 9 Sex ratio distortion among F 1 hybrids from crosses within D. sulfurigaster. No. of !\o. of % oo of No. of No. of % oo of I\ Crnss F1<;:9 F1 :"c F1 progeny P1 Cross Fi~~ F1 oo F1 progeny D. s. albostrigata Q Q D. s. albostrigata Q ~ X D. s. bilimbata 6 <5 X D . s. bilimbata c c 3616.2 # 29 x 3045.1 51 172 92.0 3126.Za X 3045.1 52 280 84.3 3116.2 # 21 x 3045.1 17 171 91.0 3033.18 x 3045.1 91 199 68.6 3117.2n X 3045.1 159 387 70.9 3054.2 x 3045.1 4 15 78.9 3117.2c X 3045.1 49 165 77.1 3120.2a X 3045.1 5 85 94.4 3119.2c X 3045.1 29 130 81.8 3120.2b.x 3045.t 18 69 79.3 3121.2 # 52 x 3043.1 90 226 71.5 3057.2 # 10 x 3045.1 47 92 66.2 3122.2 # 60 x 30-1-5.1 27 64 70.3 3057.2 # 7 x 3045.1 158 342 68.4 3122.2 # 44 x 3045.1 59 134 69.4 D. s. albostrigata Q Q 3124.2b x 3045.1 64 174 73.1 X D. s. sulfurigaster c c 3124.2 # 2 x 3045.1 42 245 85.4 3192.2 # 2 x 3017.4 112 227 67.0 3124.2 # 5 x 30+5.1 56 183 76.6 3139.2 # 2 x 2372.16 207 541 72.3 3139.2 # 2 x 3045.1 60 187 75.7 3126.2b x 3017.4 82 148 64.3 3138.2 ?< 3045.1 2 198 99.0 3033.18 x 3016.2 5 242 98.0 3138.2 x 2372.16 5 72 93.5 was collected and representative strains from each locality were crossed to the other species of the subgroup (Table 10). Sexual isolation was not so great at the P1 level as with the Cambodia line, but there was Fi sterility in all interspecific crosses. D. albomicans from Okinawa ( 3045.11) and Taiwan ( 3046.2) were crossed extensively to D. sulfurigaster (Table 5). P1 D. albomicans females were slightly fertile in these crosses if enough mass matings were attempted. The few F 1 recovered were backcrossed to both parent strains. F1 females backcrossed were fertile; F1 males backcrossed were sterile. The reciprocal matings, using D. albomicans males to D. sulfurigaster females were fertile in all cases but with Fi sterility.* Backcrosses were fertile using Fi females; with F1 males of D. sulfuri­gaster they were sterile, and with Fi males of D. albomicans they were slightly fertile (Table 12). D. albomicans was also found in the Pescadores Islands and all tests showed it to be like the lines from Okinawa and Taiwan. In crosses to 0. su/luriaaster bilimboto Hawaii 3045.1 ISJ Isl sit I IF]FIFlsltlFI !£"lI] 0. kepulouono Polawon 3056.8 Ffffff 1~ f f f s f f Borneo 3122.3 #5 Borneo 3122.3#1 f f f s f f D. kohkoo 'FT f f s l'\FFFFF Palawan 3129. 3 Sarawak 3121.3 # 7 f s s F"s F"'FFFF Sarawak 3121.3 #13 f s s s f Cambodia 3057.3 #2 FF"'FFF F F F K F F f F f "FT f f f FF F Fl\ F Thailand 3116.3 #55 F S Sarawak 3122.3 #6 FF FF F~ f f s F"s _....._ D. albomicons Thailand 3116.3 #46 F f f f f f Okinawa 3045.11 f f f f f s TABLE 10. Results of crosses between D. kepulauana, D. kohkoa, and D. albomicans and the "standard" D. s. bilimbata. Fertility-sterility symbols are as in Table 5. * In these crosses, where D. albomicans was the male, a sex ratio difference was also seen, but it was the opposite of that seen between D. sulfurigaster subspecies crosses. Here the females were in excess of the males, but there is insufficient data to be presented at this time. collections from 1967 and 1968, the Okinawa stock ( 3045.11) was used to repre­sent the species (Tables 6 and 10), and the 1968 D. albomicans strain from Thailand (3116.3 9 No. 46) was also tested in the latter series of crosses. Our population sample of D. pulaua consisted of one iso-female line from the Semongok Forest Reserve in Sarawak, Malaysia (3121.5). It was crossed to members of each of the other species (Table 11). All crosses showed P1 repro­ductive isolation with the exception of those in which D. pulaua was the male in crosses to the subspecies of D. sulfurigaster, and F1 hybrids of these crosses were fertile. In crosses to D. s. sulfurigaster males, the sex ratio was normal, but D. pulaua females X D. s. albostrigata males showed an imbalance of males over females. In the four crosses made, the percent of F 1 hybrid males ranged from 80.3% to97.7% ofthetotalprogenyof each cross. In initial crosses if no offspring (larvae, etc.) were obtained, a sample of the females was dissected from some of the crosses and their spermathecae and seminal receptacles checked for sperm. In most of the cases P 1 crosses that failed to produce offspring were the result of sexual isolation, i.e., the females had apparently not mated since there were no stored sperm. This was true for several hundred females dissected from various cross combinations. These tests showed that the D. pallidifrons did not mate often with any of the other species. TABLE 11 Crosses of D. pulaua to other members of the subgroup, illustrating the high degree of reproductive isolation. Fertility Fertility P1 Cross pl F1 P1 Cross pl F1 sterile- D. s. bilimbata.. 9 D. albomicans 9 slightly all F1 flies X D.pulaua ~ fertile fertile X D.pulaua ~ fertile abnormal D. pulaua 9 D. pulaua 9 slightly X D. s. bilimbata ~ sterile none X D. albomicans ~ fertile sterile D. s. sulfurigaster 9 D. kepulauana 9 X D. pulaua ~ fertile* fertile X D. pulaua ~ sterile none D. pulaua 9 D. pulaua 9 X D. s. sulfurigaster ~ sterile none X D. kepulauana ~ sterile none sterile to D. s. albostrigata 9 D. kohkoa 9 slightly too few X D. pulaua ~ fertile* fertile X D.pulaua ~ fertile to test sterile to D. pulau.a 9 D. paulau 9 slightly too few X D.s. albostrigata t sterile none X D. kohkoa ~ fertile to test D. pallidifrons 9 slightly too few X D. pulaua t fertile to test D. pulaua 9 slightly too few X D. pallidifrons ~ fertile to test • Crosses where sex ratio distortion was observed: 3054.2'¥ X D. pulaua 3121.50 13 F1'fl, 98 F1d' 3017.4'f>X D. pulaua 3121.50 157 F1'fl, 191 F1o" 3119.2a'f>XD. pulaua 3121.So" 52 F1~, 212F1d' 311i.2a'jlxD. pulaua 3121.50 6 F8, 35 F1o" 3122.2 # 60'jl X D. pulaua 3121.So" 1 F1~, 42 Fo" 1 Most females without offspring in crosses of D. albomicans (Okinawa) females to D. sulfurigaster males were unmated; one female in 32 was inseminated. Also 2 in 17 females were fertilized in a cross between D. albomicans (Okinawa) and D. -pallidifrons. The D. albomicans species was not only isolated by sexual isolation, but by Fi sterility as well. This conclusion was also reached by Wakahama et al. (1968). D. kepulauana is almost completely sexually isolated from all other species except D. albomicans; when 156 females of crosses to D. sulfurigaster were dissected, there were no sperm, and 46 females of crosses to D. kohkoa showed no sperm. In crosses between the subspecies of D. sulfuri­gaster and the two Cambodian lines of D. kohkoa (3057.3 ~ No. 2 and No. 24) 226 females were dissected, and there were no sperm. An occasional offspring was obtained in some of these crosses, but sterile crosses usually resulted from sexual isolation. When Fi hybrids were obtained from crosses and found to be sterile when inbred, some of these flies were dissected. In crosses of D. kohkoa using strain 3057.3 ~ No. 2 and No. 24 to D. albomicans (Okinawa), 33 F1 females had no sperm although 31 Fi males with them had non-motile sperm. In crosses between D. albomicans and D. s. bilimbata, 41 F1 females were unfertilized although 7of13 Fi males had motile sperm. This is one of the series of crosses where some of the back crosses were fertile when F1 X F1 did not produce off spring (Table 12) . TABLE 12 Results of F1 backcrosses from reciprocal crosses between D. sulfurigaster and D. albomicans. P1 cross: sulf urigaster ~ ~ X albomicans ~ ~ F1 backcrosses to F1 backcrosses to P1 stocks crossed maternal type paternal type ~ ~ ~ ~ ~ ~ ~ ~ Hawaii X Okinawa 51 F2 sF3 F Savaii X Okinawa s F sF F Luzon X Okinawa s F sF F Madang X Okinawa s F sF F Malaysia X Okinawa s F sF F Australia X Okinawa s F sF F P1 cross: albomicans ~ ~ X sulfurigaster ~ c! 4 F1 backcrosses to F1 backcrosses to P1 stocks crossed maternal type paternal type Okinawa X Savaii s F s F Okinawa X New Ireland s F s Okinawa X Madang s F s F 1 Sterile. 2 Fertile. 3 Slightly fertile. 4 These crosses went very poorly; the ones listed are the only ones in which there were enough F1 to make a backcross test. D. albomicans was represented in all cases by stock No. 3045.1 t. DISCUSSION Population samples of the D. nasuta subgroup of the D. immigrans group of species collected from tvventy-seven localities in the South Pacific and Southeast Asia have proved of cons:derable variability and interest. This report primarily presents the results of taxonomical comparisons, genetic tests and cytological analysis that led to our classifications within this subgroup. Together with the findings on sexual behavior by Spieth (this Bulletin), each line of investigation supports the other in dividing this divergent material into six separate species. After compiling information on all related D. nasuta forms described since the description of D. nasuta by Lamb, 1914, and studying the above mentioned data. we have classified the six species here reported on as D. sulfurigaster, D. albomicans, D. pulaua, D. pallidifrons~ D. kepulauana, and D. kohkoa, the latter four being new species. The division of D. sulfurigaster into three subspecies, D. s. bilimbata, D. s. sulfurigaster and D. s. albostrigata was made primarily on cytogenetic analysis; the patterns of F1 reproductive isolation, F1 sex ratio distortion and chromosomal rearrangements compliment one another in supporting this decision. It was also compatible with the geographic isolation and distribution of members of this species. Spieths' observations on sexual behavior within the species fall into three types which overlap the subspecies as they are seen from cytogenetic tests. However, differences between subspecies are not distinct in all respects as they evolve toward becoming separate !::pecies. Characteristically subspecies remain open systems during the process of speciation, but it is doubtful that there is an opportunity for the isolated island populations of D. sulfuri~aster to have more than rare, if any. interbreed.ng. Of the many strains te>ted of D. sulfurigaster from twenty-four localities, only the one fnm the island of New Ireland shows no distinct sexual isolation from any strain of the whole species. Although it has not become reproductively isolated. this strajn dces show the ~ex ratio d:stortion characteristically found among F1 hybrids from crosses between the different subspecies of D. sulfurigaster. Kanapi (1967) studied isozyme differences in three species of th2 subgroup, using starch gel electrophoresis. Although the species terminology was not known at that time. she was able to divide the 24 strains into three "groups"; these correspond to our species identificafons as follows: Group I= pallidifrons (one strain used); Group II = albomicans (two strains); Group III = sulfurigaster (five strains of s. sulfurigaster. two strains of s. albostrigata, and 14 strains of s. bilimbata). She found pallidifrons to b2 the most distinct, with a unique Esterase F form. an absence of activity at the Leucine Aminopeptidase (LAP A) zone, and an Acid Phosphatase (ACPH B) band which, although classed as Acph B1 consistently sho"1,·ed reduced activity in the faster bands of the Acph B1 triplet. The strains of albomicans '"'ere alike electrophoretically. Their unique male-specific alcohol-inhibi~ed. esterase, Est C', set them apart from all other forms. The>-also showed two Esterase F alleles (Est F3 and Est F5) which were not seen in any other strains. All strains of su!furigaster were essentially in­distinguishable7 no ev~denc2s of geographic subspeciation being noted. The elec­trophoretic pattern for alcohol dehydrogenase in this species was found to differ from that of the other two. In view of the interesting differences which were found by Kanapi, it seems especially desirable to try this or similar techniques on the other members of the nasuta subgroup. The overlapping geographical distribution of the D. nasuta subgroup species is of interest. The populations of the small isolated islands of the South Pacific are each of one species per island. But in several other areas the unusual occurrence is seen of two and sometimes three species, so closely related that they could be called sibling species, co-existing together. D. s. albostrigata and D. kohkoa were found together in the majority of the localities where either was collected as in Brunei, Borneo, and Sarawak, Malaysia, and Palawan, Philippine Islands, and Cambodia and Thailand. Only in the very small sample from continental Malaysia and in the limited lines from Luzon, P. I., was D. s. albostrigata found without D. kohkoa. D. kohkoa was never collected without D. s. albostrigata. In three of the collections which included these two species, a third species was also found. In the Semongok Forest Reserve, Sarawak, Malaysia, D. pulaua was present; in Bon Chakkrarat, Thailand, D. albomicans was also found; and in Brunei, Borneo, they co-existed with D. kepulauana. In Palawan, P. I., D. s. albostrigata, D. kohkoa and D. kepulauana were all collected, but the latter was not present in the same population sample. Wherever collected, D. kohkoa and D. kepulauana were found to co-exist sympatrically with at least one other species of the subgroup, and in none of the instances of two or three species occupying the same niche was there any evidence that there was any gene flow between them. Over 800 crosses were attempted during this investigation to determine intra­and interspecific reproductive isolation. Small mass matings of at least 100 pairs of flies were made in intraspecific crosses and from 300 to 1,000 pairs in inter­specific crosses. The data is summarized in Tables 5 through 11 showing that the six species demonstrate a range from partial to almost complete reproductive isolation between one another. Among the Fi hybrids obtained, two different sex ratio distortions were observed. In crosses between the subspecies of D. sulfurigaster the percentage of Fi males was much greater than Fi females; in interspecific crosses to D. albomicans males, more Fi hybrid females were ob­tained. Both of these conditions are being investigated. In different respects this scattered, isolated complex of species resembles several subgroups in the D. repleta group of species (Wasserman, 1963), the D. virilis group of species (Patterson and Stone, 1952; Stone, Guest and Wilson, 1960) or the D. paulistorum group (Dobzhansky and Pavlovsky, 1967; Kastritsis, 1966). Some species of D. nasuta are also comparable to those of the D. ananassae complex of species found in many of these areas (Stone, Wheeler, Wilson, et al., 1966; Spieth, 1966; and Futch, 1966). These isolated island popu­lations receive few migrants and, even on large islands, are often decimated by the effects of typhoons. The latter could be the case in the islands of Yap and Palau where Carson collected in 1968 and expected to, but did not find any representative of the D. nasuta complex. Stone has pointed out that such facton contribute in a major way to the rapid divergent evolution of populations. For example, on Ponape the new species, D. pallidifrons, was collected and has been found only on this isolated island, and it is almost completely genetically isolated from the other subgroup species. The D. repleta group of species (Wasserman, 1963) consists of many desert adapted populations that often exist in small numbers. Some of these forms occur in other areas also. The cytological differences between species are often small and many species have not been demonstrated to have heterozygous inversion systems. The Pacific island populations of D. s. bilimbata, D. pallidifrons, and D. pulaua of the D. nasuta subgroup are similar to these. Another similarity to the D. repleta group is the presence of a chromosomal polymorphism in several descendent species. The inversion 2LA is found as a heterozy.gote in D. kohkoa, D. kepulauana, D. albomicans, and D. s. albostrigata; 3A is heterozygous in D. s. albostrigata and in D. albomicans from Thailand. These are exceptions to the general rule as seen in studies by Carson (1959) where descendent species are not expested to retain an inversion polymorphism. D. nasuta subgroup populations of the Philippine Island-Sarawak area and Southeast Asia, which have 28 known heterozygous inversions, can be compared to the D. virilis group and the D. paulistorum group (Kastritsis, 1967) which have considerable inversion polymorphism. All of these groups have species with vary­ing degrees of isolation, and all utilize various types of isolation mechanisms. Between the species of the D. nasuta subgroup, isolation is both genetic and be­havioral. The D. repleta group and the D. immigrans group (Mather, 1962) have cases of added heterochromatin and chromosomal fusion. In the D. nasuta subgroup added heterochromatin to the dot chromosome ( 4) of the metaphase karyotype is a fixed characteristic of D. kepulauana and D. albomicans. A small amount has been added in D. kepulauana making the dot appear larger; in strains of D. albo­micans from Thailand, it appears as a short rod; and in all strains of D. albo­micans from Okinawa, Taiwan and the Pescadores Islands, it appears a-; a heavy longer rod. Lines of D. kohkoa often have a small amount of heterochromatin added to the dot giving it a comma-shaped appearance. D. albomicans is unusual in having a fusion which combines three of the five primitive rod elements of the genus Drosophila into one chromosome. In many cases (Patterson and Stone, 1952) it has been observed in Drosophila that the dot chromosome has fused with or has been transferred bodily to an autosome or the sex chromosomes. Were this to happen in one of the D. albomicans island populations and become established, it would produce a Drosophila with only two pairs of chromosomes. The characteristics of at least one of the species of the D. nasuta subgroup support the view of Carson (1955, 1959) and DaCunha, Dobzhansky, Pavlovsky, and Spassky, ( 1959) regarding the extent of inversion polymorphism being great­er toward the center of the population distribution while marginal populations tend to be chromosomally homozygous. In D. sulfurigaster all population samples of the subspecies albostrigata from Sarawak, which seems to be the central area of geographical distribution of this subspecies, show a high degree of polymorph­ism. There are ten heterozygous inversions plus a heterozygous complex on the X chromosome. In the marginal South Pacific islands, all D. s. bilimbata popu­lations are monomorphic, and the Australia-New Guinea area has very slightly polymorphic populations; one inversion has been seen in the strain from New Ireland and one in the line from Madang, New Guinea. Mather ( 1962) has re­ported no polymorphisms in collections from this area. In the continental locali­ties from which there are lines of D. s. albostrigata, there is a slight decline in polymorphism. Collections have been extensive toward the northern and eastern limits of the distribution of this subgroup, but on the continent they have been limited to the coastal area; therefore, the lines studied are not necessarily margin­al to the west. Further comment on the implications of chromosomal rearrangements in the D. nasula subgroup will he reserved for a later date. Due to the wealth of material acquired over a short period of time, the cytological complexities encountered, and some unique facets of the investigation, considerable studies remain in pro­gress or are yet to be carried out. It is hoped it will be possible to coordinate a chromosomal phylogeny compatible with the independent investigations on genitalia differences, and patterns of courtship and mating behavior, because all indications are that this species cluster ranks with that limited number of groups of species with a sufficient range of presently existing and available material necessary to determine the phylogenetic relationships. ACKNOWLEDGMENTS Since this project began several years ago, a great many people have aided in its completion, both in field collecting and in routine laboratory work. We are es­pecially grateful to the following for collecting and sending to Austin Drosophila stocks from various Pacific areas: Drs. D. E. Hardy and M. Delfinado, Honolulu, for material from many southeast Asian localities: Drs. T. Okada, 0. Kl.tagawa and H. Kurokawa, Tokyo, Japan; Dr. H. Takada, Sapporo, Japan; Dr. Ken-ichi Wakahama, Matsue, Japan; Dr. L. H. Throckmorton and Mr. Fei-Jann Lin for material from Taiwan; Dr. Carmen Kanapi, Manila, Philippines; Drs. H. L. Carson and M. Wasserman for material from New Guinea, and Dr. H. Spieth for material from Micronesia. We also wish to thank many individuals for their help and advice in the laboratory phases of the work, especially Mrs. Virginia Gerstenherg, Miss Hazel Lindsay, Mrs. Darlene Voelker, Miss Kathleen Resch, Mr. Kenneth Kaneshiro, Mr. Fei-Jann Lin, Mr. Shiu-Lan Huang, and Mr. Robert Voelker. Their help has been appreciated. Many kind individuals in many lands aided our own collectors through the years; these have been mentioned earlier (Studies in Genetics III, 1966, pp. v-vi) but we wish to repeat our extreme gratitude for the many courtesies which were extended to us. We also wish to thank once again those people in various govern­mental agencies which have supported our work: The National Science Founda­tion, The National Institutes of Health, the Atomic Energy Commission, and JSPS-the Japan Society for the Promotion of Science. REFERENCES CITED Bezzi, M. 1928. Diptera Brachycera and Athericera of the Fiji Islands based on material in the British Museum (N.H.) London, British Museum, 220 pp. Carson, H. L. 1955. The genetic characteristics of marginal populations of Drosophila. Cold Spring Harbor Symp. quant. Biol. 20: 276-287. 1959. Genetic conditions which promote or retard the formation of species. Cold Spring Harbor Symp. quant. Biol. 24: 87-105. Clark, A. M. 1957. Hybridization between Drosophila setifemur and D. spinofemora. Austral­ian J. Zool. 5: 216-222. Curran. C. H . 1936. The Templeton Crocker Expedition to Western Polynesian and Mela­nesian Islands, 1933. No. 30, Diptera. Proc. Calif. Acad. Sci., 4th ser., 22: 1-66. DaCunha, A. B., Th. Dobzhansky, 0. Pavlovsky, and B. Spassky. 1959. Supplementary data on the chromosomal polymorphism in Drosophila willistoni in its relation to the environ­ment. Evolution 13: 289-404. Dobzhansky. Th. and 0. Pavlovsky. 1967. Experiments on the incipient species of the Dro­sophila paulistorum complex. Genetics 55: 141-156. Duda, 0. 1923. Die orientalischen und australischen Drosophiliden-Arten des unagrischen National-Museums zu Budapest. Ann. Mus. Nat. Hung. 20: 24--59. 1924a. Beitrag zur S:vstematik der Drosophiliden unter besonderer Beriicks icht­igung der palaarktischen u. orientalischen Arten. Arch. f. Naturg. 90(A3): 172-234. 1924b. Die Drosophiliden des Deutschen Entomologischen Institutes der Kaiser Wilhelm-Gesellschaft aus H. Sauter's Ausbeute. Arch. f. Naturg. 90(A3): 235-249. 1926. Fauna sumatrensis. Beitrag 26, Drosophilidae (Dipt.). Suppl. Ent. Berlin 14: 42-116. 194-0. Revision der afrikanischen Drosophiliden (Diptera) . II. Ann. Mus. Nat. Hung. 33: 19-53. Fenis. G. F. Cha pt. 5. External morphology of the adult [Drosophila]. in Biology of Dro­sophila. M. Demerec, Ed. Wiley & Sons, 632 pp. (cited reference, pp. 368-419). Freire-Maia. N .. I. F. Banardini and A. Freire-Maia. 1953. Chromosome variation in Dro­sophila immigrans. Dusenia 4: 303-311. Futch, D. G. 1966. A study of speciation in South Pacific populations of Drosophila ananassae. Univ. Texas Puhl. 6615: 79-120. Hennig, W. 1941. Verzeichnis der Dipteren von Formosa. Ent. Beihefte. Bd. 8: 1-239; 35 figs. Kanapi, C. G. 1967. Isozyme variability in the Drosophila nasuta complex. Ph.D. Disserta­ tion. University of Texas. Kastritsis. C. D. 1966. A comparative study of the incipient species of Drosophila paulistorum complex. Chromos:ima (Berl.) 19: 208-222. 1967. A comparative study of the chromosomal polymorphs in the incipient species of the Drosophila paulistorum complex. Chromosoma (Berl.) 23: 180-202. Kikkawa. H. a!1d F. T. Peng. 1938. Drosophila species of Japan and adjacent localities. Jap. Jour. Zool. 7: 507-552. Lamb. C. G. 1914. Diptera: Heteroneuridae, Ortalidae, Trypetidae, Sepsidae, Micropezidae, Drosophilidae, Geomyzidae, Milichiidae of the Seychelles. Trans. Linn. Soc. London 16: 307-372. Malloch, J. R. 1924. Notes on Australian Diptera. IV. Proc. Linn. Soc. N. S. Wales 49: 348­ 359. Mather, Wharton B. 1962. Patterns of chromosomal evolution in the immigrans group of Drosophila. Evol. 16: 20-26. Okada, T. 1968. Systematic study of the early stages of Drosophilidae. Bunka Zugeisha Co., Tokyo. 188 pp. Patterson, J. T. and M. R. Wheeler. 1942. Description of new species of the subgenera Hirtodrosophila and Drosophila. Univ. Texas Puhl. 4213: 67-109. -----and W. S. Stone. 1952. Evolution in the Genus Drosophila. MacMillan Co., N. Y., 610pp. Spieth, H. T. 1966. Mating behavior of D. ananassae and ananassce-like flies from the Pacific. Univ. Texas Puhl. 6615: 113-146. Stone, W. S., W. C. Guest and F. D. Wilson. 1960. The evolutionary implications of the cyto­logical polymorphism and phylogeny of the virilis group of Drosophila. Proc. Natl. Acad. Sci. 46: 350-361. -----1962. The dominance of natural selection and the reality of superspecies (species groups) in the evolution of Drosophila. Univ. Texas Puhl. 6205: 506-537. -----, M. R. Wheeler, F. D. Wilson, V. L. Gerstenberg and H. Yang. 1966. Genetic studies of natural populations of Drosophila. II. Pacific island populations. Univ. Texas Puhl. 6615: 1-36. Tan, C. C., T. C. Hsu and T. C. Sheng. 1949. Known Drosophila species in China with descrip­tions of twelve new species. Univ. Texas Puhl. 4920: 196-206. Wakahama, K., 0. Kitagawa and C. Kastritsis. 1968. Chromosomal variation and the sexual isolation in Drosophila nasuta complex. Proceedings, XII Int. Congress of Genetics, Vol. I (Tokyo): 17.3.5. Wasserman, M. 1963. Cytology and phylogeny of Drosophila. Amer. Nat. 97: 333-353. Wheeler, M. R. and H. Takada. 1964. Insects of Micronesia. Diptera: Drosophilidae. B. P. Bishop Museum, Insects of Micronesia 14 (6): 163-242. XV. Courtship and Mating Behavior of the Drosophila nasuta Subgroup of Species1 HERMAN T. SPIETH2 The nasuta subgroup is a widespread tropical and subtropical complex of species which extends from the Seychelles in the Indian Ocean to Hawaii in the Pacific. One or more species of the subgroup have been recorded from South­eastern Asia, Australia and various islands of the Pacific. The present study considers 52 stocks of nasuta type flies from the Pacific Basin, Australia and Southeast Asia which exhibit six different but closely related types of courtship behavior. Simultaneous studies by Wilson et al. (this Bulletin) par­allel these findings and confirm that each of these different courtship patterns belongs to a separate species. MATING BEHAVIORS OF MALES Basic pattern Males of the six species of the subgroup that were observed display a basic mating pattern in common. It consists in the male orienting upon the female, approaching her posterior and tapping her abdomen. The male typically stands diagonally or transversely with reference to the median plane of the female's body when he taps (Fig. 7). Only rarely does he stand directly behind the female. After tapping, he engages in wing displays. Excepting males of D. kohkoa, all the others circle to the front of the female and wing-display either in front of her or as they move past her face in their circling actions. The male behavior during copu­lation and also the terminal dismounting actions are identical for flies of all the six species. Thus the qualitative interspecific differences are essentially restricted to circling movements, wing actions and frontal displays of the males. Drosophila sulfurigaster Duda This species has an extensive distribution ranging from Hawaii southward to Australia and westward to Thailand and the Philippines. Within the various stocks of D. sulfurigaster, three intergrading subtypes of male display are to be found. 1 This study was supported in part by PHS Research Grant No. GM-11609 from the Na­tional Institute of General Medical Science. The majority of the stocks studied were made avail­able by the Genetics Foundation, Department of Zoology, University of Texas. My sincere thanks are given to the late Wilson S. Stone, to Marshall R. Wheeler and to Florence D. Wilson for their invaluable assistance and help in making this investigation possible. 2 Department of Zoology, The University of California, Davis. STUDIES IN GENETICS Y. Uni,-. Texas Puhl. 6918. Sept., 1969. 256 Subtype A* After tapping, the male circles to the front and places himself directly in front of and in close proximity to the female's face. He then spreads both wings out­ward about 45° and upward 5°-10° (Fig. 4). Holding the wings in this extended position, he then arcs back and forth in front of the female's face-moving through an arc of about 25°-30°, i.e., approximately 15° on either side of the female's median plane. As the male arcs back and forth, he twists his body along the longi­tudinal axis, thus elevating one wing slightly above the opposite one. Simultane­ously the tip of his abdomen is curled slightly downward and the anterior end of his body is depressed so that his face is positioned lower than that of the female. Typically he positions himself so close to the female that he is in actual contact with her and his depressed head is actually thrust under the female's face and mouthparts. Some individual males position themselves far enough from the females, i.e. 1-2 mm., that there is no physical contact between the two sexes but usually contact does exist and the female fends against the male with her forelegs. After a period of 2-4 seconds and arcing back and forth, the male returns his '· 2. 4. 5. 6. *Editor's note: The letters A, B and C used here for subtypes are not intended to represent the same subtypes A, B and C in Wilson et al. (this Bulletin). wings to the normal position and engages in either of two actions: ( 1) he may circle to the rear of the female and tap her abdomen, or (2) more typically he moves backwards 0.5-1 cm. from the female and, standing diagonally in front of her and directly in front of one of her eyes, he rapidly "paws" with one of the forelegs, using the legs alternately. After a burst of either tapping or pawing, the male returns to arcing position and engages in another bout of display. As the male moves about the female between bouts of arcing, he may infre­quently bob his body up and down in small amplitude. This occurs more often when he has tapped the female's abdomen and is returning (circling) to the front position. Not all individuals display bobbing and a given individual may not bob consistently. If the female does not decamp or if other individuals do not intrude into the immediate vicinity of the pair, the male will continue to court for pro­longed periods of time. A courting male is constantly alert to the intrusion of other individuals, be they males or females, into the immediate area of the courting pair. Invariably he will cease courtship to fend off and attempt to drive away any intruding individual. Often this action results in the breakup of the courtship since the distraction of the courting male by another individual results in his 1becoming separated from the female. Subtype A stocks: Australia (2372.16); Madang, New Guinea (3016.2); New Ireland (3017.4); Hawaii (3045.1); Savai'i, Western Samoa (3045.7); Nadari­vatu, Fiji (3064.2); Savusavuitaga, Fiji (3065.2); Baguio, Luzon, P.I. (3126.2a); and Agana, Guam (3130). Subtype B The display of the subtype B males is similar to that of subtype A males except that when arcing back and forth in front of the fem ale one wing is extended out and upward 45° and the other wing only 20°-30° (Fig. 5). As the male alternates in his movements back and forth in front of the female, the wings are alternated between the two positions. Males that are courting with great vigor and arcing back and forth at a very rapid rate appear not able completely to retract the wings between alternation of movement, and thus approach the subtype A display. Such individuals when less excited, however, slow down their rate of display and then show the typical subtype B display. After a male has courted a female for some time he often engages in pulses of wing flicking. This action involves flicking the wing nearest the female's head rapidly outward to goo. Occasionally the movement may involve vibration in that after the wing is flicked to goo it is then vibrated rapidly for about one second up and dovm in an arc of 45°-60° before being returned to the resting position. The male may perform these movements when standing at the rear or in front of the female. These movements of wing flicking and wing flicking plus vibration appear identical to the typ~cal courting action of D. kohkoa males (see below) but with the subtype B males they are relatively infrequent subsidiary types of movement and occur only after a male has persistently courted one or more nonresponsive females for a period of time. Subtype B stocks: Malaya (3033.17); Palawan Island (3056.2-6 and 3056.2­ The University of Texas Publication TABLE 1 U. ofT. Duration of Copulations stockt No. obs. No. oo No. th D. albomicans and D. kepulauana also appear to have been derived from D. kclzkoa like ancestors. In b::>th species the occurrence of wing action has been displaced from the rearward position to the frontal position. It should be recalled that the kohko2 male, like those of all the other species, initiates courtship by tapping and positioning himself at the diagonal rear of the female. Then after one or more bouts of wing movements, he circles about the female and assumes a mirror image position on the opposite side of the female and again engages in wing action before again circling back to the opposite side. During each courtship the male repeatedly circles back and forth. D. albomicans also initiates courtship in a similar manner but does not vibrate at the rearward position, rather he defers his wing vibration until he is in front of the female as he circles past her face. D. kepulauana also defers wing vibration until he circles to the front of the female, faces her, and then engages in the unique body throwing-wing flicking action. The D. pallidifrons male also displays a deferred type of wing action. After tapping and positioning himself at the rear he circles toward the front of the female with a crablike movement and then turns and backs in front of the female's face and engages in bouts of wing vibration. Interestingly the kohkoa male also tends to move with a crablike movement as he circles about the female. Significantly, however, a pallidifrons male, which never faces the female in his courtship, lacks silvery markings on the vertex, while kohkoa, albomicans and kepulauana males all have the entire vertex marked with silver. D. pallidifrons is also unique in that it is restricted to the small volcanic island of Ponape. Despite intensive collecting on this island, pallidifrons is the only species of the nasuta subgroup that has been found there. The conclusion seems inescapable that pallidifrons evolved on Ponape and that as yet it appears not to have successfully escaped and become established anywhere else. It therefore represents a prime example of the evolution of behavioral characters under con­ditions of geographical isolation. In sum it would appear that two distinct evolutionary lines are represented within these six species of the nasuta subgroup. One line is repre:::ented by a cluster of four species, i.e. , the primitive kohkoa, the unique pallidifrons, plus albomicans and kepulauana. The other line is repre.::ented by the widespread sulfurigaster with its three subtypes plus pulaua which currently is known from only a sir..gle stock from Sarawak. REFERENCES CITED Grossfield, J. 1966. The influence of light on the mating behavior of Drosophila. Univ. of Texas Puhl. 6615: 147-176. Spieth, H. T. and T. C. Hsu. 1950. The influence of light on the mating behavior of seven species of the Drosophila melanogaster group. Evolution 4: 316-325. Spieth, H. T. 1952. Mating behavior within the genus Drosophila (Diptera). Bull. Amer. Mus. Nat. Hist. 99: 399-474. XVI. Polymorphism in Esterases and Hemoglobin in Wild Populations of the House Mouse (Mus musculus)1 ROBERT K. SELANDER, SUH Y. YANG, AND W. GRAINGER HUNT INTRODUCTION The technique of gel electrophoresis, combined with histochemical staining (Hunter and Markert, 1957), provides a powerful tool for investigating genetic variation in natural populations. Using this technique, we have amassed a con­siderable volume of data on allele frequencies at numerous loci controlling enzymatic and other proteins in wild populations of the house mouse (Mus musculus), a species which, in its domesticated form, has been prominent in research on the genetic basis of protein polymorphism (Lush, 1967). Our re­search on the house mouse has to some degree paralleled that dealing with protein polymorphism in humans (Harris, 1966, 1967; Harris et al., 1968; Livingstone, 1967) and in Drosophila populations (Hubby and Lewontin, 1966; Lewontin and Hubby, 1966; Johnson et al., 1966; Hubby and Throckmorton, 1968; O'Brien and Macintyre, 1969; Prakash et al., 1969; Stone et al., 1968). Major objectives of our research program on electrophoretic protein variation in wild house mice have been ( 1) to estimate the proportion of polymorphic loci in the gene pools of wild populations by analyzing variation in a large number of proteins, on the assumption that the controlling loci are random samples of the total genomes (Selander and Yang, 1969b; Selander et al., 1969); (2) to compare nominal subspecies and subspecies groups of the house mouse with respect to total genetic character (Selander et al., 1969); (3) to determine the effects of population structure on patterns of genetic variation (Selander and Yang, 1969a; Selander, 1969); and (4) to measure interpopulation variation in allele frequen­cies at various loci, and, by describing geographic patterns of variation, to provide a basis for future investigations of possible selective factors maintaining the protein polymorphisms. Previous work by others on biochemical polymorphism in wild populations of the house mouse may be summarized as follows. Petras (1967) reported allele frequencies at the Es-1, Es-2, and Hbb loci in several hundred mice collected on five neighboring farms near Ann Arbor, Michigan; Petras and Biddle (1967) determined allele frequencies at the Es-5 locus in samples from seven buildings on two farms near Windsor, Ontario; Shows and Ruddle (1968) reported the existence of polymorphism at the LDH B regulator locus (Ldr-1) in several samples of wild populations from North America; and several other workers concerned primarily with variation among domestic strains have mentioned the occurrence of particular alleles in wild populations, without, however, presenting data on allele frequencies. Our own extensive work has revealed numerous polymorphic proteins in wild 1 Research supported by NIH grant GM-15769 and NSF grant GB-6662. STUDIES IN GENETICS V. Univ. Texas Pub!. 6918, Sept., 1969. populations. In a survey of electrophoretic variation in 36 proteins (controlled by 41 genetic loci) in seven regional populations in California and in Denmark, 16 or 44% of the proteins proved to be polymorphic in one or more populations. In terms of controlling loci, 17 or 41 % were polymorphic, segregating for two or more alleles. We have estimated that, on the average, regional populations of the house mouse are polymorphic at 26% of their loci and individuals are heterozy­gous at 8.5% of their loci (Selander et al., 1969). In the present paper we have summarized our findings on geographic variation in the five proteins most intensively studied, namely Esterase 1, Esterase 2, Esterase 3, Esterare 5, and hemoglobin. The controlling loci are designated, re­spectively, Es-1, Es-2, Es-3, Es-5, and Hbb. Phenotypic proportions and allele frequencies are presented for our larger samples, most of which were collected in the United States, and, particularly, in Texas. METHODS AND MATERIALS Samp!ing pro~edures. All but a few of our s!'mples were colle~ted in farm buildings, particularly in barns in which laying hens were housed.. Occasionally we have attempted to obtain samples from agricultural fields, but in this habitat we generally have been unable to trap sufficient numbers of mice for studies of genetic variation. Mice were captured with Sherman single or Ketch-All multiple live-traps set in a grid pattern throughout a barn at intervals of approximately 10 feet. In barns housing large mouse populations, samples were collected in a single night's trapping, but trapping over a period of two or three days was necessary to obtain adequate samples of small populations. Mice were housed in plastic laboratory cages on a diet of Purina Laboratory Chow and water for from 1 to 10 days before processing. For the purpose of studying geographic variation in allele frequencies, a collec­tion of mice from a single barn (or field) was the unit of sampling. Because of marked interbarn heterogeneity in allele frequencies (Selander, 1969), even within single farms, samples from different barns were segregated. Genotypes of all mice collected, male and female, adult and immature, were used in calculating allele frequencies, with the exception that those judged to be younger than four weeks were not used in determining frequencies at the Es-2 locus, since the patterns of Esterase 2 may not be stabilized until mice reach this age (see Petras, 1963, and Pantelouris arid Amason, 1966). Approximately 90% of the mice processed were adult. The variation we have analyzed is independent of sex. Laboratory procedures. The protein polymorphisms considered in this report were demonstrated electrophoretically in blood plasma or hemolysate. Approxi­mately 1.0 ml of blood was obtained from each mouse by cutting the throat and allowing blood to drip into a micro-beaker containing crystalline heparin, or by puncturing the suborbital canthal sinus with a heparinized pipette. The whole blood was mixed with an equal volume of 0.85% saline and centrifuged at 2,500 g for 10 minutes at 4° C, after which the diluted plasma was removed and stored at -20° C. The erythrocytes were then resuspended and washed three times in 10 volumes of saline and lysed by shaking for 60 seconds with a volume of deionized water equal to that of the whole blood and a volume of toluene equal to one-half that of the whole blood. Stroma and lipids were removed by centri­fuging the hemolysates at 49,500 g for 30 minutes at 0° C, and the extracted hemolysates were stored at 4° C for not more than 24 hours before elec­trnphoresis. Fer routine determination of phenotypes, horizontal starch-gel electrophoresis was employed, using a system described by Poulik (1957), as modified by Beck­man and Johnson (1964). S3mples of hemolysate or plasma were absorbed on 9 X 6 mm pieces of Whatman 3 MM filter paper, which were then inserted in slits cut in gels of Electro-Starch (Otto Hiller, Madison, Wisconsin) prepared at a concentration of 12.5 g/100 ml buffer and poured in 9 X 190 x 210 mm lucite molds. Following electrophoresis, 3-mm sliceJ were cut from the gel and incu­bated in appropriate staining solutions. For electrophoresis of hemolysates, we used a 0.01 M Tris-HCl gel buffer (pH 8.5) and a 0.30 M borate (boric acid and sodium hydroxide) tray buffer (pH 8.2). Gels were run for two hours at a gradient of 25 volts/ cm. Substrat~s med for demonstrating erythrocyte esterases included a-naphthyl acetate, /3-naphthyl acetate, and a-naphthyl propionate. The sharpest esterase bands were obtained with a-naphthyl acetate, but the darkest bands were produced with a-naphthyl propionate. The erythrocyte esterase staining solution was composed of 50 ml of a 0.2 M monobasic sodium phosphate solution (pH 4.4) , 10 ml of a 0.2 M dibasic sodium phosphate solution (pH 8.7), 40 ml of deionized water, 1.5 ml of a 1% solution ( w /v) of substrate in a 1: 1 ( v / v) mixture of acetone and de­ionized water, and 40 mg of Fast Garnet GBC salt. After staining for two hours at 37° C, the gel slices were fixed in a 1:5:5 solution (v/v/v) of acetic acid, methanol, and water. Hemoglobin was stained on a second slice of the hemolysate gel placed in fixing solution saturated with naphthol blue black (Buffalo Black NBR) for 15 minutes at 20° C. The slice was destained with clear fixing solution. For plasma esterases, we used a lithium hydroxide gel buffer composed of a 1: 9 mixture of a solution of 1.2 g monohydrate lithium hydroxide and 11.89 g boric acid/ I wate... (pH 8.1) and a solution of 1.6 g monohydrate citric acid and 6.2 g Tris/ I water (pH 8.4). The former solution was also used as the tray buffer. Gels were run for four hours at a gradient of 30 volts/ cm. The staining solution for plasma esterases was composed of 4 ml of a 0.1 M monobasic sodium phosphate solution (pH 4.4), 4 ml of a 0.1 M dibasic sodium phosphate solution (pH 8.7), 90 ml of water, 2 ml of a 1 % solution (v/ v) of a-naphthyl butyrate in acetone, and 50 mg of Fast Blue RR salt. The gels were incubated for one hour at 37° C. For certain purpo~es, the vertical starch-gel electrophoresis procedure of Smithies ( 1959) was employed, with the same gel and tray buffers used for horizontal electrophoresis. Vertical gels were run at room temperature for five hours at a gradient of 10 volts/ cm, with a fan directed at the gel to dissipate heat. Storage of hemolysate and plasma samples at -20° C for periods up to 12 months did not appreciably affect the activity or mobility of the esterases studied. However, homozygous and heterozygous diffuse hemoglobins (Hbbd/ Hbbd and Hbbd/H?b8) were difficult to distinguish in hemolysates frozen for even a few days before electrophoresis. Statistical procedures. Expected proportions of genotypes and phenotypes were calculated using Levene's (1949) exact formula for small samples. Where allele frequencies were estimated from the proportion of a single phenotype, Haldane's ( 1956) formula was employed to correct for bias in small samples. In samples in which the presence of a "silent" ("null") allele was known (by the occurrence of one or more "silent" homozygotes) or suspected (from marked heterozygote deficiency), allele frequencies were calculated by a maximum likelihood method developed by R. H. Richardson. For the purpose of describing geographic variation, sample localities were grouped into sample regions, as shown in Figures 1 and 2. For each region, pre­sumed equilibrium allele frequencies were estimated by computing unweighted means for all barns and fields represented by samples of 10 or more mice. Because of interbarn and interfarm heterogeneity in allele frequencies (see Selander, 1969, and tables in the present paper), weighted means obtained by pooling samples within regions provide less accurate estimates of equilibrium frequencies. However, the use of pooled estimates would not seriously alter the patterns of geographic variation we have described, since, for most regions, un­weighted and weighted means are closely similar. FIG. 1. Sample localities and regions in North America. 1, Utah; 2, Central California; 3, Southern California; 4, Phoenix, Arizona; 5, Tucson, Arizona; 6, Minnesota-Wisconsin; 7, Illinois; 8, Ohio; 9, Kansas-Missouri; 10, Florida; 11, Jamaica. For distribution of localities within regions in Texas (indioated by circles), see Fig. 2. Frn. 2. Sample localities and regions in Texas. BIOCHEMICAL POLYMORPHISM Alleles at four esterase loci (Es-1, Es-2, Es-3, and Es-5) and at the /3-chain hemoglobin locus (Hbb) are listed in Table 1, and various phenotypes are shown in Figure 3. For these loci we have determined the phenotype for each of 8,000 wild house mice collected in 185 barns, largely in the United States but also in parts of Denmark, Venezuela, and Jamaica. Additional data for these loci in Hawaiian and Danish populations will be presented elsewhere by L. Wheeler (unpublished data), and W. G. Hunt (unpublished data), respectively. Unweighted mean allele frequencies for barns (and fields) are given for regions other than those in Texas in Tables 2 and 3. For Texas, mean frequencies for regions indicated in Figure 2 are presented in Tables 4 and 5. Phenotypic proportions and allele frequencies for individual samples of 10 or more mice2 are given for the Es-1 locus in Table 6, the Es-2 locus in Tables 7 and 8, the Es-3 locus in Tables 9 and 10, the Es-5 locus in Tables 11 and 12, and the Hbb locus in Tables 13 and 14. In describing variation at the five loci, we will first consider the distribution of alleles in inbred strains and in wild populations, and then discuss in greater detail the patterns of geographic variation in Texas. 2 Data for smaller samples will be supplied, upon request, by the authors. ESTER-ESTERASE 2 ASE 1 (+) bib clc did ble blc bid cld ala alb ale old ~ ala bib alb Cf) --=---= ­---­ _J C1l r.z.:;..~ c-~ CG.!': • ·,.. ­ 0 -...., ~ w :r: origin -{-) -­ FIG. 3. Representative phenotypes of Esterase 1, Esterase 2, Esterase 3, and hemoglobin in the house mouse. Not illustrated are bands corresponding to the following alleles known from wild populations: Es-Jd, Es-2f, Es-2g, Es-3e, and Es-3.f. Diagram prepared by L. Wheeler. Esterase 1 Distribution of alleles (Tables 2 and 6). Four alleles have been demonstrated at this locus, two of which (Es-ta and Es-Jb) were originally described from .. n­bred strains (Popp and Popp, 1962; Popp, 1965). Throughout most of the United States and in Jamaica and Venezuela, wild populations are homozygous for Es-Jb, which is the allele found in all inbred strains except C57. But Es-ta occurs in low frequency in southern California (0.05) and southern Arizona (0.01), and the two alleles are in approximately equal frequency in the Ha­waiian Islands (L. Wheeler, unpublished data). On the Jutland Peninsula of Denmark, Es-ta is fixed in most populations of the northern subspecies M. m. musculus, while Es-Jb is fixed in thm:e of the southern subspecies M. m. domesti­cus (see Table 2 and additional data presented by Selander et al., 1969). The "silent" allele Es-tc is known only from Hallowell Farm, southern Cali­fornia, where a mean frequency of 0.143 was recorded in a series of eight barns sampled in December, 1968, and from one farm in northern Denmark. Es-Jd has been detected in heterozygous state with Es-Jb in two individuals from Barns 1and2 at Hallowell Farm. Variation in Texas. All Texas populations are fixed for Es-Jb. Discussion. It is difficult to interpret the geographic distribution of alleles at the Es-1 locus in relation to climatic or other environmental variables. For the species as a whole, Es-Jb is the predom·nant allele, judging from studies of both wild populations and domestic strains. The presence of Es-ta in high frequency in such diverse climatic and biotic regions as northern Denmark and the Ha­waiian Islands suggests that a s~mple explanation of the distribution of alleles in terms of environmental adaptation will not be forthcoming. The alleles Es-JC and Es-Jd are so localized in distribution that it is unnecessary to invoke selective mechanisms to account for their presence. In the United States, Es-tc probably arose by mutation in a founder population at Hallowell F~T'YY'l. southern California, and reached moderate frequency by genetic drift. Similarly, the occurrence of Es-11 at HaPowell Farm may be attributed to muta­tion. Because wild house mouse populations are finely subdivided into small tribal units in which close inbreeding occurs, and populations of single barns or fields may be founded by a few dispersing individuals, occasional rapid increases in frequencies of mutant alleles through genetic drift are to be expected in single barns or fields (see review by Selander, 1969) . As noted beyond, alleles that presumably arose by mutation and increased in frequency through drift in local populations have been detected at other loci. We are less inclined to invoke mutation and drift as an explanation for the occurrence of the Es-ta allele in the United States, since it apparently is widely distributed, albeit rare, having been recorded at three farms at two localities (Yucaipa and Ramona) in California and at two farms in Tucson, Arizona. Esterase 2 Distribution of alleles (Tables 2, 4, 7, and 8 and Figures 4 and 5). Seven alleles are represented at the Es-2 locus, which controls an esterase that is generally "scored" from plasma, although the system stains more intensely in liver and kidney extracts, where it appears in the V esterase zone designated by Ruddle and Roderick (1966). The Es-2b allele is fixed in all inbred strains thus far ex­amined, with the exceptions of PL/J, which is homozygous for Es-2c (Ruddle et al., 1969), and RFM/Un and RF/AL, which are homozygous for Es-2a (Popp, 1967). Es-2b is also the commonest allele in most wild populations. In the United States, the frequency of Es-2b exceeds 0.80 in Florida and in the region from the Midwestern states southwest through Utah and northern and central Texas.3 In the region from southern Texas west through New Mexico and Arizona to California, frequencies average lower, with extremes of 0.43 in far western Texas and 0.59 to 0.61 in California. On the Jutland Peninsula, Denmark, Es-2b apparently is fixed in the subspecies M. m. domesticus, but Es-2c predominates or is fixed in populations of M. m. musculus. The "silent" allele Es-2a4 is widespread in both North America and Europe but displays no recognizable geographic pattern of variation in frequency. In the United States, the highest frequencies are recorded in central California (0.39) and in eastern Texas (0.40 at Houston). The Es-2c allele has been recorded in low to moderate frequency in central and southern Texas but is rare or absent in our samples from elsewhere in the United States. Es-2d is moderately -common in southern California, southern Arizona, and the Rio Grande Valley of Texas, all of which are regions of sub­tropical climate. However, its absence from Hawaii, Jamaica, and Venezuela, and its presence in low frequency (0.04) in Ohio, indicate that it is not exclusively a 3 Petras (1967) reported frequencies of 0.70 and 0.30 for the Es-2b and Es-2a alleles, respec­tively, in a pooled sample of 296 mice taken over a three-year period on five adjacent farms in Washtenaw County, southeastern Michigan. Because of the close proximity of these farms, this sample probably does not provide a good basis for estimating equilibrium frequencies for this region. 4 Es-2a is, technically, not a "silent" allele. On darkly stained gels, plasma samples from mice homozygous for Es-2a often show a faint esterase band intermediate in mobility between those pro­duced by the Es-2b and Es-2c alleles, but this band is not detectable in heterozygous condition. FIG. 4. Regional mean allele frequencies at Es-2 locus in North America. Indicated areas of circles are proportional to frequencies. Diagrams for Texas present unweighted means of several adjac~nt sample regions; see Fig. 5. "tropical allele." Es-2e, recorded in the United States in a single barn at Mayhard Farm, near Prosper, Texas, is presumed to have arisen there as a mutant and to have increased to a frequency of 0.08 through genetic drift. Es-2f has been found only in low frequency locally in Denmark (W. G. Hunt, unpublished data), and Es-29 is known only from the Island of Hawaii (L. Wheeler, unpublished data). Additional data on the distribution of alleles at the Es-2 locus are provided by Ruddle et al. (1969) , as follows: In.25 mice from Alberta, Canada, and in .24 mice from North Carolina (precise localities unspecified), Es-2b and Es-2c were repre­sented, with the former more common; and all of 35 mice from Vermont were homozygous for Es-2b. Variation in Texas (Table 8 and Figure 5). The overall pattern of variation at the Es-2 locus in Texas is one in which populations in certain major physiographic and/or climatic regions are characterized by distinctive allele frequencies, with clinal variation between adjacent regions. Thus, the frequency of Es-2b exceeds 0.90 in the High Plains and Lower Plains of the Panhandle and in the Blackland Prairie region extending from the northeastern part of the state south in a narrow strip to the Austin-San Antonio region. South in the northern plains region to the latitude of Dallas, the only other allele represented in significant frequency is Es-2a, but Es-2c appears in low to moderate frequency south of this line. Fig. 5. Regional mean allele frequencies at Es-2 locus in Texas. Small diagrams represent small samples. From the Temple-Austin region eastward, the frequency of Es-2a increases clinally from 0.02 at Austin to 0.08 at Bryan, 0.24 at Nacogdoches, and 0.40 at Houston. The small El Paso sample shows an unusually high frequency of Es-2a. South of San Antonio, the Es-2d allele appears at Corpus Christi (0.02) and increases to a frequency of 0.18 at Brownsville. This allele also occurs, at a fre­quency of 0.12, in the Ensinal and Alpine regions. Discussion. Too few regions in the United States have been sampled to permit an analysis of patterns of variation for all alleles at the Es-2 locus. According to our information, Es-2c and Es-2d reach significant frequencies only in the south­western states. As a consequence, most regions in the southwestern United States have three or four alleles represented in reasonable frequencies, while those in the northern part of the country have only two, Es-2b and Es-2a. Esterase 3 Distribution of alleles (Tables 3, 4, 9, and 10 and Figures 6 and 7). Two promi­nent esterase systems appear on Tris-HCI gels (pH 8.5). The faster migrating system, designated Esterase 3, is only weakly inhibited by eserine or eserine sulfate, while the slower migrating system is almost completely inhibited by these compounds. In hemolysates of Pallid and Swiss albino inbred strains, Hunter and FIG. 6. Regional mean allele frequencies at Es-3 locus in North America. Strachan (1961) found two prominent esterase bands in gels stained with a-naphthyl butyrate, the slower migrating one of which reportedly was inhibited by 5 x 10-5 M eserine. By duplicating their technique with hemolysates from inbred strains and wild mice, we have determined that their second band, run­ning just ahead of the hemoglobin, is the Es-3 system. A similar reversal in posi­tion of the Es-3 and the slower systems is obtained when hemolysates are run on gels prepared with a metaborate buffer. The esterase patterns of hemolysates of inbred strains described by Pelzer ( 1965) do not correspond to those we have observed. Ruddle and Roderick ( 1966) probably are correct in suggesting that Pelzer's Ee-ta and Ee-tb bands are identi­cal with the Es-ta and Es-tb plasma esterases described by Popp and Popp (1962). Because these esterases are not demonstrable in the hemolysates we have proc­essed, we suspect that Pelzer's hemolysates were contaminated with plasma. Of the six alleles which have been identified at this locus, only two, Es-3b and Es-Y~ occur commonly in wild populations. In the United States, regional fre­quencies for Es-3b generally exceed 0.80, with lower values being recorded only in Texas (see beyond), central California (0.55), and Florida (0.72). The Es-JC allele is present in all regions for which adequate samples are available. At Ca­racas~ Venezuela, Es-3b and Es-JC have frequencies of 0.55 and 0.45, respectively. In the Danish subspecies, Es-Y is the more common allele, with mean fre­ Frn. 7. Regional mean allele frequencies at Es-3 locus in Texas. quencies of 0.65 in samples of M. m. domesticus and 0.56 in those of M. m. mus­ culus (Selander et al., 1969). Three other alleles known from wild populations are uncommon or rare and are localized in distribution. Es-Jd has a mean frequency of 0.05 to 0.07-in Cali­fornia and 0.02 in the Brownsville region, but elsewhere in the United States it has been recorded only in a single mouse from Illinois. Recently, L. Wheeler (un­published data) has found it in low frequency on the Island of Hawaii. Es-Je, a "silent" allele detectable only in homozygous condition, is known only from two wild mice, one collected at Norco Ranch, Austin, Texas, and another trapped in Fiji. From the latter individual, a laboratory line homozygous for Es-Je was established by L. Wheeler (unpublished data). The Es-3a allele, de­scribed from several inbred strains, is as yet unrecorded in wild populations. Finally, Es-3f occurs rarely in populations in the region of the zone of hybridiza­tion of M. m. musculus and M. m. domesticus in Denmark (W. G. Hunt, unpub­lished data). Variation in Texas (Table 10 and Figure 7). The pattern of geographic varia­tion in allele frequencies at the Es-3 locus in Texas is relatively complex. In all regions except Brownsville, only two alleles, Es-Jb and Es-JC, are represented. In the Panhandle, represented by the Amarillo and Lubbock regions, the frequency of Es-Jb is about 0.70. In central Texas west to the Alpine region, east to the The University of Texas Publication Austin region, north to the Dallas region, and south to the Ensinal region, the fre­quency of this allele is considerably lower, averaging approximately 0.55, and there is considerable variation among regions. The frequency of Es-3b decreases to about 0.40 in the Corpus Christi and Victoria regions of the Gulf Coastal Plain. Eastward from the Austin region of central Texas through the Caldwell and Bryan regions to far eastern Texas, there is a clinal increase in frequency of the Es-3b allele from 0.40 to 0.90. There is relatively little interregional variation in allele frequencies in eastern Texas. Discussion. Apart from the fact that the Es-3c allele has a somewhat lower fre­quency in the northeastern United States than in the southwestern states and in Florida, there is no particular pattern of variation in North America. The occur­rence of the Es-Jd allele in appreciable frequencies only in regions of subtropical or mild temperate climate (Brownsville, Texas, and southern and central Cali­fornia) is suggestive of a causal relationship, but we note also the absence of this allele in Florida, Jamaica, and the Caracas region, and the fact of its low frequency in the Hawaiian Islands, where it apparently is confined to the Island of Hawaii. Esterase 5 Distribution of alleles (Tables 3, 5, 11, and 12 and Figures 8 and 9). Petras and Biddle (1967) described polymorphic variation in serum esterase band VII, as originally designated by Petras ( 1963). Breeding experiments indicate that the presence of this band is due to a dominant allele, Es-5b, while its absence re­sults from homozygosity of an alternate allele, Es-5a. Eight inbred strains examined by Petras and Biddle ( 1967) were homozygous for Es-5b, but, in wild populations in North America and elsewhere, Es-5a is the predominant allele. However, Es-)b is widespread, and polymorphism has been found in all but one region of the range of the species represented by large sam­ples. In M. m. musculus of the northern Jutland Peninsula, Denmark, Es-5a ap­parently is fixed, but the frequency of this allele drops to 0.56 in M. m. domesticus of the southern Jutland Peninsula. In North America, Petras and Biddle (1967) reported frequencies of 0.58 and 0.80, respectively, for two farms sampled near Windsor, Ontario. The fact that the mean frequency for the Windsor area (0.69) is similar to the mean regional frequency (0.65) for the Minnesota-Wisconsin region suggests that northern populations are characterized by a relatively low frequency of this allele. In most other parts of North America, including much of Texas (see beyond) but excluding Florida, frequencies of the Es-5a allele exceed 0.85. This allele occurs in very high frequency in California, and was fixed in our samples from Jamaica and the Caracas region of Venezuela. Variation in Texas (Table 12 and Figure 9). A notable feature of geographic variation in Texas is the unusually low frequency (0.38) of the Es-5a allele in the Mt. Pleasant region in the extreme northeastern part of the state. In the adjacent Dallas and Nacogdoches regions to the west and south, respectively, the frequency is about 0. 79. Southwestward from the Houston and Bryan regions of south­eastern Texas, the frequency decreases to about 0.60 at Seguin and Victoria, then increases to 0.94 in extreme southern Texas (Corpus Christi and Brownsville re­gions). Over most of central and western Texas, regional frequencies of 0.90 or Fm. 8. Regional mean allele frequencies at Es-5 locus in North America. higher are recorded, but there is a clinal decrease to 0. 76 from central Texas north to Amarillo in the Panhandle. Discussion. The sudden change from a frequency of 0.79 to one of 0.38 in mov­ing from central-eastern Texas into the Mt. Pleasant region is perhaps the most surprising feature of geographic variation observed in our material. At no other locus do we have evidence of such a drastic shift in allele frequency over such a short distance. Possibly the observed pattern can be attributed to sampling error in the Mt. Pleasant region, since the estimate of 0.38 is based on mice from only five barns, four at Pilgrim Farm, near Pittsburg, and one at Richardson Farm, near Mt. Pleasant. At Richardson Farm, samples of three mice each from two other barns yielded frequencies of 0.62, suggesting a relatively low frequency in these barns also; and a sample of six mice from Hall Farm, near Mt. Pleasant, also yielded a low frequency, 0.45. However, at other farms in the Mt. Pleasant region, we have evidence from small samples that the frequency of Es-5a is much higher: Berry Farm, near Gilmer, Barn 1, 2 mice, frequency= 0.75; Barn 2, 7 mice, fre­quency= 0.85; Bradley Farm, near Jefferson, 2 mice, frequency= 0.75; and Wood Farm, also near Jefferson, 1 mouse homozygous for Es-5a. Further collecting in northeastern Texas will be necessary to assess this situa­tion. Present evidence suggests an unusual degree of interfarm heterogeneity of allele frequencies at the Es-5 locus in this region, so that our sample may not be adequate for an accurate estimate of the mean regional frequency. Es-5a-()-Es-5b Frn. 9. Regional mean allele frequencies at Es-5 locus in Texas. HEMOGLOBIN Distribution of alleles (Tables 3, 5, 13, and 14 and Figures 10 and 11). Three alleles at the Hbb or fi-chain hemoglobin locus (Ranney and Gluecksohn­Waelsch~ 1955; Popp and St. Amand, 1960; Hutton et al., 1962) have been identi­fied. These are Hbb8 and Hbbd, both of which occur in inbred strains and in wild populations, and HbbP, as yet recorded only in the strain Au/Ss (Morton, 1962, 1966; Russell et al., 1968) . Electrophoretically demonstrable variation has yet to be recorded for the a-chain Hba locus (Russell and Bernstein, 1966). Both Hbb·~ and Hbbd alleles have been recorded in all regions sampled ade­quately, and, in the majority of these, the Hbb8 allele predominates. In the United States, frequencies higher than 0.30 for the Hbbd allele occur only in California, Arizona, and southern Texas. Elsewhere, high frequencies of the Hbbd allele have been recorded only in Venezuela and in the Hawaiian Islands (L. Wheeler, un­published dat 9 size F s Silent Silent F M s Utah 1 49 1.000 1.000 Central California 3 73 1.000 0.390 0.587 0.020 Southern California 9 182 0.046 0.826 0.127 0.101 0.608 0.291 Arizona: Phoenix 5 1.000 1.000 Arizona: Tucson 7 363 0.013 0.987 0.136 0.796 0.067 Minnesota-Wisconsin 4 136 1.000 0.082 0.910 0.008 Illinois 5 275 1.000 0.104 0.896 Ohio 39 1.000 0.910 0.051 0.039 Kansas 2 42 1.000 0.190 0.810 Florida 3 57 1.000 0.040 0.900 0.060 Jamaica 2 29 1.000 1.000 Venezuela 2 76 1.000 1.000 Northern Denmark 2 64 1.000 1.000 Central Denmark (Hybrid zone) 2 59 0.211 0.789 0.270 0.730 Southern Denmark 2 64 1.000 1.000 TABLE3 Mean Allele Frequencies at Es-3, Es-5, and Hbb Loci in Regions other than Texas l\Iean allele frequency Esterase 3 Esterase 5 Hemoglobin Number of P0oled sample; sample Es-3' Es-Jc Es-3'1 Es-)a Es-5b Hbbd Hbb• Region withn > 9 size :\I s F A p D s Utah 1 49 0.949 0.051 1.000 Central California 3 73 0.547 0.407 0.047 0.990 0.010 0.447 0.553 Southern California 9 182 0.835 0.093 0.073 0.922 0.078 0.565 0.435 Arizona: Phoenix 1 5 0.100 0.900 1.000 1.000 Arizona: Tucson 7 363 0.937 0.063 0.847 0.153 0.391 0.609 Minnesota-Wisconsin 4 136 0.918 0.082 0.648 0.352 0.058 0.942 Illinois 5 275 0.912 0.086 0.002 0.902 0.098 0.092 0.908 Ohio 39 0.808 0.192 0.848 0.152 0.064 0.936 Kansas 2 42 0.815 0.185 0.875 0.125 0.230 0.770 Florida 3 57 0.720 0.280 0.593 0.407 0.197 0.803 Jamaica 2 29 1.000 1.000 0.065 0.935 Venezuela 2 76 0.545 0.455 1.000 0.460 0.540 Northern Denmark 2 64 0.406 0.594 1.000 0.329 0.671 Central Denmark (Hybrid zone) 2 59 0.497 0.503 0.846 0.154 0.039 0.961 Southern Denmark 2 64 0.597 0.403 0.560 0.340 0.153 0.847 TABLE4 Mean Allele Frequencies at Es-2 and Es-3 Loci in Texas Mean allele frequency Esterase 2 Esterase 3 )l'urnber of Pooled samples sample Es-2a Es-2b Es-2c Es-2" Es-3b Es-3c Es-3" Region with n > 9 size Silent F M s M s F Dallas 71 184 0.023 0.961 0.004 0.551 0.449 Temple 1 34 0.941 0.059 0.529 0.471 Austin 15 22452 0.016 0.926 0.058 0.397 0.603 Fredericksburg 14 0.786 0.214 0.500 0.500 Abilene 1 32 0.120 0.860 0.020 0.484 0.516 San Angelo 5 1533 0.794 0.206 0.588 0.412 Seguin 8 257 0.929 0.071 0.586 0.414 Houston 3 32 0.403 0.580 0.017 0.920 0.080 Ensinal 13 765 0.020 0.766 0.093 0.119 0.524 0.476 Brownsville 12 5154 0.728 0.090 0.182 0.709 0.267 0.024 Alpine 3 116 0.047 0.693 0.143 0.117 0.583 0.417 Victoria 3 62 0.087 0.857 0.057 0.453 0.547 Corpus Christi 6 147 0.838 0.147 0.017 0.363 0.637 Nacogdoches 7 123 0.241 0.759 0.901 0.099 Bryan 7 488 0.083 0.781 0.136 0.760 0.240 Huntsville 2 32 0.715 0.285 1.000 Caldwell 2 84 0.900 0.100 o.5on 0.500 Mt. Pleasant 5 235 0.036 0.962 0.002 0.900 0.100 Lamesa 2 37 0.900 0.100 0.655 0.345 Lubbock 9 588 0.031 0.969 0.712 0.288 Amarillo 5 188 0.040 0.960 0.724 0.276 El Paso 2 20 0.570 0.430 0.850 0.150 Eagle Pass 2 80 0.905 0.095 0.815 0.185 1 Frequency of Es-2• = 0.01 1. 2 2188 for Es-3. a 157 for Es-3. • Es-2a present in one sample of 7 from Jackson \Varehouse, Harlingen, with a frequency of 0.4.~. TABLE5 Mean Allele Frequencies at Es-5 and Hbb Loci in Texas Mean allele frequency Region Number of sampl es with n > 9 Pooled sample size Esterase 5 Es-5° Es-5b A p Hemoglobin Hbb1 llbb• D s Dallas 71 1842 0.792 0_208 0.204 0.796 Temple 1 34 0.955 0.045 0.290 0.710 Austin 153 21894 0.921 0.079 0.144 0.856 Fredericksburg 1 145 0.921 0.079 1.000 Abilene 1 32 0.885 0.115 0.047 0.953 San Angelo 5 1576 0.930 0.070 0.094 0.906 Seguin 8 257 0.639 0.361 0.124 0.876 Houston 3 32 0.787 0.213 0.017 0.983 Ensinal 137 765 8 0.788 0.212 0.145 0.855 Brownsville 129 51510 0.937 0.063 0.500 0.500 Alpine 3 116 0.976 0.024 0.137 0.863 Victoria 3 62 0.570 0.430 0.060 0.940 Corpus Christi 6 147 0.935 0.065 0.155 0.845 Nacogdoches 7 123 0.804 0.196 0.003 0.997 Bryan 7 488 0.794 0.206 0.259 0.741 Huntsville 2 32 0.600 0.400 0.170 0.830 Caldwell 2 84 0.935 0.065 1.000 Mt. Pleasant 5 235 0.376 0.624 0.204 0.796 Lamesa 2 37 0.870 0.130 0.080 0.920 Lubbock 9 588 0.737 0.263 0.174 0.826 Amarillo 5 188 0.760 0.240 0.070 0.930 El Paso 2 20 0.873 0.127 0.200 0.800 Eagle Pass 2 8011 0.995 0.005 0.175 0.825 6 153 for Es-5. 1 6 for Es-5. 7 12 for Hbb. 2 106 for Es-5. s 744 for Hbb. 3 10 for Es-5. 0 11 for Es-5 _ 4 962 for Es-5. 10 427 for Es-5_ 5 13 for Es-5. 11 63 for Es-5 . TABLE6 Variation in Esterase 1 in Regions other than Texas Observed genotypes and phenotypes Allele frequency Es-fa/ Rs-Jc/ Number Es-Ja/-Es-Jb/-Es-fb Es-Jc Es-fa Es-Jb Es-Jc Samrile of mice F s FS Silent F s Silent Southern California: 2-4 December 1968 Yucaipa Daily Fresh Egg Farm 29 25 4 0.07 0.93 Ramona Hallowell Farm Barn 1 18 141 3 0.12 0.79 0.08 Barn2 29 261 2 0.02 0.73 0.25 Barn3 12 11 0.04 0.96 Barn4 21 20 0.76 0.24 Barn 9 29 1 27 0.04 0.85 0.11 Barn 12 17 2 13 0.09 0.60 0.30 Barn 15 17 16 0.03 0.81 0.16 Barn 16 10 10 1.00 Arizona: 12-13 December 1968 Tucson Thompson Farm Barn 1 46 44 2 0.02 0.98 Barn2 49 47 2 0.02 0.98 Barn3 19 18 0.03 0.97 Barn 6 39 38 0.01 0.99 Arizona State Poultry 25 25 1.00 Arizona Star Ranch Barn2 120 119 0.004 0.996 Barn 17 65 64 0.01 0.99 Jutland Peninsula, Denmark: 9-21 August 1968 Northern Part Hesselberg Farm 34 34 1.00 VadFarm 30 30 1.00 Central Part Sondergard Farm 32 5 10 17 0.42 0.58 RavnFarm 27 27 1.00 Southern Part 0stergard Farm 31 31 1.00 H¢jboFarm 33 33 1.00 1 One mouse, scored here as Es-Jb/-, is actually heterozygous for Es-Jb and Es-Jd. <,).) ~ TABLE 7 Variation in Esterase 2 in Regions other than Texas Observed genotypes and phenotypes Allele frequency Sample Number of mice Es-2a /Es-2a Silent Es-2b/-Es-20/­F M Es-2"/­s Es-2b/Es2c FM Es-2b/Es-21l Es-2o/Es-2" FS MS Es-2a Silent Es-2b F Es-2° M Es-2" s ~ Utah: 1 July 1967 ~ Salt Lake City ~ Adamson Ranch 49 .. 49 . . . . . . . . . . . .. 1.00 . .. . .. :::s...... !\::: Central California: 2--4 December 1968 ~ Modesto ...... ~ BlixisFarm ~ Barn2 22 16 6 . . . . . . . . . . 0.85 0.15 . .. ... Fresno ~ Guthrie,Farm 34 .. 32 1 . . 1 . . 0.10 0.86 0.03 . .. ~ Kemp Farm 17 1 15 .. . . 1 . . . . 0.22 0.75 0.03 . . . ~ Southern California: 23 November'"""8 December 1968 ~ Yucaipa Daily Fresh Egg Farm Ramona 29 1 28 .. . . . . 0.21 0.79 . .. . .. a­~ s· :::s Hallowell Farm Barn 1 18 1 6 . . 5 .. 6 . . 0.21 0.41 . .. 0.37 Barn2 29 . . 6 .. R . . 15 . . . .. 0.47 . .. 0.53 Barn3 12 . . 6 .. 4 . . 2 . . 0.22 0.46 . .. 0.32 Barn4 21 .. 9 ~ .. 10 . . . .. 0.67 . .. 0.33 Barn9 29 19 1 9 . . 0.81 0.19 Barn 12 17 10 2 5 0.08 0.68 0.24 Barn 15 17 8 5 4 0.17 0.49 . . . 0.34 Barn 16 10 5 1 4 0.02 0.69 . .. 0.30 (/.) ~ Arizona: 12-13 December 1968 S"' Tucson Thompson Farm ;::::! f} ""I Barn 1 46 46 1.00 .. .. . .. (!) ~ Barn2 Barn3 49 19 1 46 17 1 1 1 1 0.17 0.13 0.81 0.81 .. . 0.02 0.06 ll) !'"""" Barn6 39 3 33 3 0.25 0.71 . .. 0.04 ~ Arizona State Poultry 25 1 18 2 4 0.21 0.66 ... 0.13 '"­~ Arizona Star Ranch ;3 Barn2 120 5 92 5 18 0.19 0.70 ... 0.10 0 """ ~ Barn 17 65 51 2 12 0.88 ... 0.12 ~ c;;· Minnesota: 5 October 1968 ~ Anoka ..... ~ Ghostly Farm ~ Barn 1 30 . . 30 . . . 1.00 . .. s:: Barn2 Barn3 52 12 1 49 11 3 . . . 0.33 0.97 0.67 0.03 . . . .. . . .. (Fl ss:: (Fl Whitewater Wisconsin: 17-23 Octoher 1968 ("')s:: 2"' (Fl Duffin Farm Chicken House 42 .. 42 . .. 1.00 w Ul. ~ 0 O"l TABLE 7-Continued Observed genotypes and phenotypes Allele frequency Number Es-2" /Es-2" Es-2b/-Es-2c/-Es-2 ,...;. OJ !""""" ~ -.. ~ 3 0 """! (;:j ~ ~· ~ ....... ~ ~ ~ en s ~ en (') ~ 2"' en (.)..) ~ w 0 00 TABLES Variation in Esterase 2 in Texas Observed genotypes and phenotypes Allele frequency Number Es-2°'/Es-2°' Es-2b/-Es-2c/-Es-2"!-Es-2b/Es2c Es-2b/Es-2 July 1967 108 89 19 . . . 0.91 0.09 ... 3 December 1967 186 167 1 18 . . . 0.95 0.05 . . . s· Barn2 August 1967 December 1967 Barn3 108 310 2 99 287 1 5 6 18 0."14 0.06 0.83 0.90 0.03 0.04 ... ... 3:: ~ (F) s ~ August 1967 108 1 104 3 0.09 0.90 0.01 ... (F) ("') December 1967 80 . . 77 3 . . . 0.98 0.02 . .. ~ ......... ~ Barn4 (F) August 1967 108 91 3 14 0.06 0.86 0.08 December 1967 198 163 5 30 0.05 0.86 0.09 Fredericksburg Region: 26-28 April 1967 Fredericksburg Petch Farm 13 . . 7 .. 6 . .. 0.77 0.23 . .. w 0 ~ TABLE 8-(Continued) OJ -­0 Observed genotypes and phenotypes Allele frequency Number Es-2a/ Es-2a Es-2b/-Es-2°/-Es-2<1/_ Es-2b/Es2° Es-2b /Es-2<1 Es-2° /Es-2<1 Es-2a Es-2b Es-2° Es-2a Sample of m:cc Silent F M s FM FS MS Silent F M s Abilene Region: 20 February 1968 Hodges Moore Field 32 31 1 0.12 0.86 0.02 ~ San Angelo Region: 16-19 February 1968 ~ (1:) San Angelo Heinze Field c::j ~ ..... 18 February 19February Wilkie Downs Racetrack 45 55 16 34 47 9 1 2 11 7 5 ... ... .. . 0.88 0.92 0.72 0.12 0.08 0.28 . .. . . . . .. ~ (1:) "! c,,..... ~ Veribest ~ Hurst Farm 21 .. 11 4 6 ... 0.67 0.33 . . . ~ Bitner Field 16 10 1 5 .. . .. 0.78 0.22 . .. ~ ~ Seguin Seguin Region: 20-22 October 1967 c,, "t1;:::: Cotton Oil Company Milling & Storage Rooms 21 13 1 7 0.79 0.21 . .. ~ ""'..... ~ ~ Smiley Farm2 17 .. 14 3 . .. 0.91 0.09 . .. ~..... 0 ~ Farm3 Barn3 37 .. 37 .. .. . .. 1.00 Nixon Chessher Farm Regular Series Barn4 11 .. 7 4 . . . .. 0.82 0.18 Irregular Series Barn 1 42 35 7 . .. 0.92 0.08 w TABLE 8-(Continued) --to Observed genotypes and phenotypes Allele frequency Number Es-2a /Es-2a Es-2 1'/-Es-2c/-Es-2d/-F.s-2b/£;·2C Es-2 1'/Es-2<1 Es-2c/Es-2­ Observed genotypes and phenotypes Allele frequency Number Es-2a/Es­2a Es-2h/­ Es-2C /-Es-2dj - Es-2b/Es2c Es-2b/Es-2d Es-2c/Es­2d Es-2a Es-2b Es-2c Es-2d Sample of mice Silent F M s FM FS MS Silent F M s -- Rusk Hassell Farm Barn2 20 20 1.00 Barn4 12 1 11 0.32 0.68 ... . . . ~ Barn 5 15 . . 15 1.00 . .. . . . ~ ~ Bryan Region: 22 March 1968 ~ Bryan ~ ..... . ~ Reliance Farm Barn 3 89 43 10 36 0.68 0.32 . . . ~ ""C Vi...... Barn 6 45 34 1 10 . . 0.87 0.13 . . . ~ Barn 7 112 59 10 43 0.72 0.28 . . . 0 "'-1-. Bullock Farm 42 1 40 1 .. 0.15 0.84 0.01 . . . ~ Steep Hollow Farm ~ Barn 1 72 .. 65 1 6 0.05 0.90 0.05 ~ Barn 5 Barn 6 67 61 2 2 56 49 1 5 8 5 . . 0.15 0.23 0.78 0.68 0.07 0.09 . . . . . . "t1 $::'. \:t" .............. Phelps Huntsville Region: 28 March 1968 (°') i::> .......... . 0 Swearingen Farm ~ Corn Crib 22 . . 9 1 12 . . . 0.68 0.32 Huntsville White Farm 10 . . 6 1 3 .. .. . .. 0.75 0.25 Caldwell Region: 15, 17 April 1968 Paige Schindler Farm 72 57 15 .. . . . 0.90 0.1 0 Lamesa Region: 12 June 1968 Lamesa Womack Farm Barn4 23 .. 22 1 0.98 0.02 Mt. Pleasant Region: 2-5 May 1968 MI. Pleasant ~ Richardson Farm S"' Barn 1 38 1 37 0.18 0.82 ;::s ~ Pittsburg ~ ""I Pilgrim Farm ('ti .-!­ Barn 1 77 75 2 . . 0.99 o.oi ~ Barn2 Barn3 Barn 4 36 14 70 . . 36 14 70 1.00 1.00 1.00 ... "t:l 0 ....... ~ 3 Lubbock Region: 14-19 June 1968 0 ""I Littlefield ~ ~ Barton Farm i;;· Barn 1 71 1 70 0.13 0.87 s Barn3 126 .. 126 1.00 . . . ...... ;::s Barn 10 90 90 1.00 . .. ~ Barn 11 51 51 1.00 s:: (J) Barn 12 Shallowater 106 .. 106 1.00 s s:: (J) Gary Farm Barn4 53 1 52 .. . . . . 0.15 0.85 ("") E-c Barn5 42 . . 42 .. . . . . . . 1.00 (Fl Barn 12 35 .. 35 . . . . .. .. .. 1.00 Vance Farm Barn 11 14 . . 14 .. . . . . 1.00 w ~ (,]\ ()..) ...... O') TABLE8~(Continued) Observed genotypes and phenotypes Allele frequency Number Es-2" /Es-2" Es-2b/-Es-2c/-Es-2d./-Es-2b/Es2c Es-2b/Es-2" Es-2•/Es-2" F.s-2" Es-2b Es-2° Es-2d Sample of mice Silent F M S FM FS MS Silent F M s Amarillo Region: 21-22 June 1968 Panhandle Gibson Farm Barn3 10 10 1.00 . . . ... ~ Barn4 Stinnett 34 . . 34 . . 1.00 . . . . . . t:J;::s..... Hughes Farm 10 10 .. . . . . . 1.00 . .. . .. ~ ~ Pleasant .Valley ~ ..... Winkleman Farm ~ Barn 1 33 33 . . . . . 1.00 . .. ... .Q.. Barn2 101 4 97 .. 0.20 0.80 . . . . . . ~ El Paso Region: 10-16 July 1968 H ~ Clint Wilson Farm ~ Barn2 Las Cruces, New Mexico 10 1 9 . . .. 0.35 0.65 . . . . . . <::l-< """'§' Stahlman Farm t"4o s· Barns 1 and2 10 6 4 .. 0.79 0.21 . .. ;::s Eagle Pass Region: 16 July 1968 Quemado MorroFarm · Barn 1 65 54 1 10 .. . 0.91 0.09 Barn2 15 . . 12 .. 3 . . . . . 0.90 0.10 TABLE9 Variation in Esterase 3 in Regions other than Texas Observed genotypes and phenotypes Allele frequency Number Es-Jb /Es-Jb Es-Jc /Es-Jc Es-3b /Es-Jc Es-Jb Es-Jc Sample of mice M s MS M s Utah: 1 July 1967 Salt Lake City Adamson Farm 49 44 :3 0.95 0.05 Arizona: 12-13 December 1968 Tucson Thompson Farm Barn 1 46 41 4 0.93 0.07 Barn2 49 46 3 0.97 0.03 Barn3 19 17 2 0.95 0.05 Barn6 39 38 0.99 0.01 Arizona State Poultry 25 25 1.00 Arizona Star Ranch Barn2 120 82 3 35 0.83 0.17 Barn 17 65 52 12 0.89 0.11 Minnesota: 5 October 1968 Anoka Ghostly Farm Barn 1 30 29 0.98 0.02 Barn2 52 48 4 0.96 0.04 Barn3 12 10 2 0.92 0.08 Wisconsin: 17-23 October 1968 Whitewater Duffin Farm Chicken House 42 31 5 6 0.81 0.19 Ohio: 1967 15 Localities 39 27 3 9 0.81 0.19 Kansas: 27 September-2 October 1968 Lawrence Powell Farm Barn 1 20 16 4 0.90 0.10 West Feed Store 22 11 10 0.73 0.27 Florida: 25-26 July 1968 Princeton Fruman Fann Barn2 18 13 5 0.86 0.14 Barn3 15 10 4 0.80 0.20 Homestead Ashworth Farm 24 6 6 12 0.50 0.:110 Jamaica: 3 August 1968 Anchoy Salmon Farm Barn 1 20 20 1.00 TABLE 9-Continued Observed genotypes and phenotypes Allele frequency Sample Number of mire Es-3b /Es-J b Es-Jc /Es-Jc Es-Jb /Es-Jc .M S MS Es-Jb M Es-Jc S Caracas Region, Venezuela: 25-26 November 1967 Cagua Huevos de Hoy Farm Barn4 51 9 20 22 0.39 0.61 Mixed Barns 25 13 3 9 0.70 0.30 Jutland Peninsula, Denmark: 9-21August1968 Northern Part Hesselberg Farm 34 15 4 15 0.66 0.34 Vad Farm 30 21 9 0.15 0.85 Central Part Sondergard Farm 32 12 4 16 0.62 0.38 Ravn Farm 27 5 12 10 0.37 0.63 Southern Part 0stergard Farm 31 6 10 15 0.44 0.56 HS1Sjbo Farm 33 17 16 0.76 0.24 Observed genotypes and phenotypes Allele frequency Number Es-Jb/ Es-3b Es-Jc/ Es-Jc Es-Jb/ Es-Jdj Es-jc Es-l'1 Es-Jdj Es-Ja/ F.s-Jb Es-Jc Es-lb Es-3c Es-3d Sample of mice M s l\IS F F:\I FS M s F Central California: 2-4 1968 Modesto Blixis Farm Barn2 22 2 7 11 0.36 0.59 0.05 Fresno Guthrie Farm 34 18 3 13 0.72 0.28 Kemp Farm 17 4 9 2 0.56 0.35 0.09 Southern California: 23 November-8 December 1968 Yucaipa Daily Fresh Egg Fann 29 21 6 0.84 0.14 0.02 Ramona Hallowell Farm Barn 1 18 15 2 0.92 0.06 0.03 Barn2 29 18 1 10 0.81 0.02 0.17 Barn3 12 7 3 2 0.79 0.13 0.08 Barn4 21 12 4 4 0.76 0.14 0.10 Barn5 10 6 1 3 0.80 0.05 0.15 Barn9 31 26 1 3 0.90 0.05 0.05 Barn 12 17 8 6 2 0.71 0.21 0.08 Barn 15 17 16 0.97 0.03 Barn 16 10 8 0.85 0.10 0.05 Illinois: 26-28 October-1 November 1968 Tuscola Kleiss Farm 30 26 4 0.93 0.07 Atwood Schable Farm 59 39 20 0.83 0.17 Germantown Albers Farm 138 88 4 44 0.80 0.19 0.01 Millstadt Albert Farm Barn 1 38 38 1.00 Barn2 10 10 1.00 Observed genotypes and phenotypes Allele frequency Number Es-3b/Es-3b Es-JC/Es-Jc Es-3b/£s-3c F.s-Jh Es-Jc Sample of mice M s i\IS M s Dallas Region: 2-5 April 1967 (Dallas) and 26-27 June 1968 (Prosper and Weatherford) TABLE 10 Variation in Esterase 3 in Texas Dallas Littlebrook Fann 78 9 39 30 0.31 0.69 Prosper Mayhard Farm Barn3 15 8 1 6 0.73 0.27 Barn4 14 6 3 5 0.61 0.39 Barn5 20 7 2 11 0.63 0.37 Barn6 . 24 12 12 0.75 0.25 Weatherford Rockwell Fann Barn2 15 2 8 5 0.30 0.70 Barn3 18 5 4 9 0.53 0.47 Temple Region: 6-9 May 1967 Temp/,e Waskow Farm 34 11 9 14 0.53 0.47 Austin Region: 1967-1968 Bastrop Henderson Ranch 58 5 27 26 0.31 0.69 Austin Cook-Synoett Farm March 1967 105 9 50 46 0.31 0.69 June 1967 11 6 4 0.27 0.73 Garfield Farm January 1967 12 6 6 0.25 0.75 Herbert Fann May-June 1967 90 20 27 43 0.46 0.54 Orta Farm 19 5 5 9 0.50 0.50 Norco Ranch 141 6 4 4 0.57 0.43 Robinson Farm 18 5 8 5 0.42 0.58 Liberty Hill Lay Ranch March 1967 106 19 40 47 0.40 0.60 February 1968 324 65 129 130 0.40 0.60 Dripping Springs Empire Ranch March 1967 81 6 42 33 0.28 0.72 October 1967 15 5 9 0.37 0.63 Stanley Farm Barn2 45 19 6 20 0.64 0.36 Barn4 10 2 4 4 0.40 0.60 Hildreth Farm Barn 1 April 1967 54 1 37 16 0.17 0.83 July 1967 108 3 80 25 0.14 0.86 December 1967 186 10 126 50 0.19 0.81 Barns 1 & 2 April 1967 74 6 48 20 0.22 0.78 The University of Texas Publication TABLE 10-Continued Observed genotypes and phenotypes Allele frequency Number Es-Jb/Es-Jb Es-Jc/Es-Jc Es-Jb/Es-J0 Es-Jb Es-J 0 Sample of mice M S MS M s Barn2 August 1967 108 14 34 60 0.41 0.59 December 1967 310 62 133 115 0.39 0.61 Barn3 August 1967 108 20 40 48 0.41 0.59 December 1967 80 11 28 41 0.39 0.61 Barn4 August 1967 108 16 50 42 0.34 0.66 December 1967 198 33 70 95 0.41 0.59 Fredericksburg Region: 26-28 April 1967 Fredericksburg Petch Farm 13 2 2 9 0.50 0.50 Abilene Region: 20 February 1968 Hodges Moore Field 32 7 8 17 0.48 0.52 San Angelo Region: 16-19 February 1968 San Angelo Heinze Field 18 February 45 16 11 18 0.56 0.44 19 February 55 22 11 22 0.60 0.40 Wilkie Downs Racetrack 16 4 4 8 0.50 0.50 Veribest Hurst Farm 25 12 12 0.72 0.28 Bitner Field 16 6 4 6 0.56 0.44 Seguin Region: 20-22 October 1967 Seguin Cotton Oil Company Milling &Storage Rooms 21 3 6 12 0.43 0.57 Smiley Farm2 17 4 3 10 0.53 0.47 Farm3 Barn3 37 7 11 19 0.45 0.55 Nixon Chessher Farm Regular Series Barn4 11 5 6 0.73 0.27 Irregular Series Barn 1 42 23 3 16 0.74 0.26 Barn2 54 11 9 34 0.52 0.48 Barn3 24 13 2 9 0.73 0.27 Barn4 51 15 9 27 0.56 0.44 Houston Region: 29 September-1 October 1967 Houston Otte Farm Barn3 11 8 3 0.86 0.14 Baytown Barker Farm Barn5 11 10 0.95 0.05 Chandler Farm Barn6 10 9 0.95 0.05 Ensinal Region: 22 October 1967 (Riker Farm and Nieschwietz Farm Barns 1-7) and 3November1967 (Nieschwietz Farm Barns 8-11 and Burkholder Farm) Cotulla RikerFarm 21 19 2 0.95 0.05 Ensinal Nieschwietz Farm Barn 1 49 4 20 25 0.34 0.66 Barn 2 86 12 35 39 0.37 0.63 Barn3 101 22 31 48 0.45 0.55 Barn4 25 6 12 7 0.38 0.62 Barn5 50 5 23 22 0.32 0.68 Barn 7 106 21 35 50 0.43 0.57 Barn8 90 26 23 41 0.52 0.48 Barn9 31 13 3 15 0.66 0.34 Barn 10 47 5 26 16 0.28 0.72 Barn 11 31 8 13 10 0.42 0.58 Burkholder Farm Barn 1 81 51 2 28 0.80 0.20 Barn2 47 38 8 0.89 0.11 Alpine Region: 6-9 April and 15 May 1967 (Sul Ross Hog Barn, May) Alpine Sul Ross Hog Barn April 53 18 5 30 0.62 0.38 May 36 12 3 21 0.63 0.37 Sul Ross Cow Barn 11 1 3 7 0.41 0.59 NeieRanch 16 8 7 0.72 0.28 Victoria Region: 27-28 February 1968 Victoria Raines Farm 14 7 7 0.25 0.75 Guadalupe Schoener Farm 37 2 14 21 0.34 0.66 Dacosta Reges Farm 11 6 5 0.77 0.23 Corpus Christi Region: 1-4 March 1968 Bishop Felder Farm Duck House 18 6 11 0.36 0.64 Barkley Farm 18 4 7 7 0.42 0.58 Odem Naylor Farm Barn 1 20 3 12 5 0.28 0.72 Barn3 14 2 5 7 0.39 0.61 Barn4 55 3 21 '.31 0.34 0.66 Barn5 22 2 7 13 0.39 0.61 Nacogdoches Region: 12-14 March 1968 (Nacogdoches) and 1 May 1968 (Rusk) Nacogdoches Fuller Farm Barn2 18 15 2 0.89 0.11 Barn3 21 18 3 0.93 0.07 Linthicum Farm Corn Crib 1 15 10 5 0.83 0.17 Corn Crib 2 22 13 9 0.79 0.21 TABLE 10-Continued Observed genotypes and phenotypes Allele frequency Number Es-Jb /Es-Jb Es-Y/Es-Jc Es-Jb /Es-Jc Es-Jb Es-Jc Sample of lllice M S l\1S M s Rusk Hassell Farm Barn2 20 19 1 0.98 0.02 Barn4 12 10 2 0.92 0.08 Barn5 15 14 0.97 0.03 Bryan Region: 22 March 1968 Bryan Reliance Farm Barn3 89 47 8 34 0.72 0.28 Barn6 45 25 3 17 0.74 0.26 Barn 7 112 46 12 54 0.65 0.35 Bullock Farm 42 29 13 0.84 0.16 Steep Hollow Farm Barn 1 72 44 2 26 0.79 0.21 Barn5 67 42 1 24 0.81 0.19 Barn6 61 39 6 16 0.77 0.23 Huntsville Region: 28 March 1968 Phelps Swearingen Farm Corn Crib 22 22 1.00 Huntsville White Farm 10 10 1.00 Caldwell Region: 15, 17 April 1968 Paige Schindler Farm 72 8 32 .i2 0.33 0.67 Caldwell Clark Farm Brooder House 12 4 8 0.67 0.33 Mt. Pleasant Region: 2-5 May 1968 Mt. Pleasant Richardson Farm Barn 1 38 8 8 22 0.50 0.50 Pittsburg Pilgrim Farm Barn 1 77 77 1.00 Barn2 36 36 1.00 Barn 3 14 14 1.00 Barn4 70 70 1.00 Lamesa Region: 12 June 1968 Lamesa Womack Farm Barn4 23 8 2 13 0.63 0.37 Lubbock Region: 14-19 June 1968 Littlefield Barton Farm Barn 1 71 40 5 26 0.75 0.25 Barn3 126 59 13 54 0.68 0.32 Barn 10 90 51 8 31 0.74 0.26 Selander et al.: Polymorphism in Mus musculus 323 Barn 11 51 34 17 0.83 0.17 Barn 12 106 67 9 30 0.77 0.23 Shallowater Gary Farm Barn4 53 31 9 13 0.71 0.29 Barn5 42 21 8 13 0.66 0.34 Barn 12 35 18 2 15 0.73 0.27 Vance Farm Barn 11 14 5 4 5 0.54 0.46 Amarillo Region: 21-22 June 1968 Panhandle Gibson Farm Barn3 10 7 t 2 0.80 0.20 Barn4 34 8 9 17 0.49 0.51 Stinnett Hughes Farm 10 5 5 0.75 0.25 Pleasant Valley Winkleman Farm Barn 1 33 22 3 8 0.79 0.21 Barn 2 101 64 5 32 0.79 0.21 El Paso Region: 10-16 July 1968 Clint Wilson Farm Barn2 10 7 3 0.85 0.15 Las Cruces, New Mexico Stahlman Farm Barns 1 and 2 10 6 4 0.80 0.20 Eagle Pass Region: 16 July 1968 Quemado Morro Farm Barn 1 65 37 3 25 0.76 0.24 Barn2 15 11 4 0.87 0.13 Observed genotypes and phenotypes Allele frequency Es-3 1> I Es-YI Es-3''/ Es-Jd/ Es-3"/ Es-3'1/ Number Es-3 1> Es-3< £.5.)C Es-3'1 Es-31> Es-3c F..<-$'' E<-3'1 J{s.$ C Sample of mice :\I s >IS F DI FS l\I s F Brownsville Region: 20-23 April 1967 (Jackson Warehouse, Brownsville) and 19-25 August 1967 Brownsville Jackson Warehouse April 56 53 3 0.97 0.03 Mercedes Coil Farm 13 5 2 5 0.61 0.35 0.04 Alamo McNeill Store 14 8 4 0.75 0.21 0.04 Donna WatsbyFarm West Barn 16 9 4 3 0.78 0.13 0.09 East Barn 107 59 8 27 8 4 0.71 0.22 0.07 Harlingen Guillen Farm North Barn 27 18 8 0.81 0.19 Goat Barn 11 'j 4 0.82 0.18 TABLE 10-Continued Observed genotypes and phenotypes Allele frequency Es-3b/ Es-3c/ Es-3b/ Es-3df Es-3d/ Es-3d/ Number Es-3b Es-3c Es-3~ Es-3 Es-3b Es-3c Es-3b Es-Jc Es-3d Sample of mice M s MS F FM FS M s F Egg Room 10 2 5 3 0.35 0.65 Grimsell Store 5 3 0.10 0.50 0.40 Raymondville Pickard Farm Barn 1 75 17 28 30 0.43 0.57 Barn3 60 32 4 19 3 2 0.72 0.24 0.04 Barn4 60 36 2 22 0.78 0.22 Barn5 61 41 3 16 0.81 0.18 0.01 1 One mouse lacking bands and presumed to be of the genotype Es-Je / Es-JC excluded from sample of 15. Allowing for presence of F.s-3 • allele, maximum likelihood estimates of frequencies are as follows: Es-3 b =0.42, Es-3c =0.32, and Es-3• =0.26. TABLE 11 Variation in Esterase 5 in Regions other than Texas Observed genotypes and phenotypes Allele frequency Number Es-5a/Es-5a Es-5b/--Es-5a Es-5b Sample of mice Band absent Band present A p Utah: 1 July 1967 Salt Lake City Adamson Ranch 49 37 12 0.87 0.13 Central California: 2-4 December 1968 Modesto Blixis Farm Barn2 22 22 1.00 Fresno Guthrie Farm 34 34 1.00 Kemp Farm 17 16 0.97 0.03 Southern California: 23 November-8 December 1968 Yucaipa Daily Fresh Egg Farm 29 22 7 0.87 0.13 Ramona Hallowell Farm Barn 1 18 14 4 0.88 0.12 Barn2 29 24 5 0.91 0.09 Barn3 12 12 1.00 Barn4 21 20 1 0.98 0.02 Barn 9 29 26 3 0.95 0.05 Barn 12 17 12 5 0.84 0.16 Barn 15 17 16 1 0.97 0.03 Barn 16 10 8 2 0.90 0.10 Arizona: 12-13 December 1968 Tucson Thompson Farm Barn 1 46 29 17 0.80 0.20 Barn2 49 32 17 0.81 0.19 Barn3 19 15 4 0.89 0.11 Barn6 39 28 11 0.85 0.15 Arizona State Poultry 25 13 12 0.72 0.28 Arizona Star Ranch Barn2 120 102 18 0.92 0.08 Barn 17 65 58 7 0.94 0.06 Minnesota: 5 October 1968 Anoka Ghostly Farm Barn 1 30 6 24 0.45 0.55 Barn2 52 20 32 0.62 0.38 Barn3 12 4 8 0.59 0.41 Wisconsin: 17-23 October 1968 Whitewater Duffin Farm Chicken House 42 36 6 0.93 0.07 Illinois: 26-28 October-1 November 1968 Tuscola Kleiss Farm 30 25 5 0.91 0.09 Atwood Schable Farm 59 50 9 0.92 0.08 Germantown Albers Farm 138 98 40 0.84 0.16 Millstadt Albert Farm Barn 1 38 30 8 0.89 0.11 Barn2 10 9 0.95 0.05 Ohio: 1967 15 Localities 39 28 11 0.85 0.15 Kansas: 27 September-2 October 1968 Lawrence Powell Farm Barn 1 20 15 5 0.87 0.13 West Feed Store 22 17 5 0.88 0.12 Florida: 25-26 July 1968 Princeton Fruman Farm Barn2 18 5 13 0.54 0.46 Barn3 15 9 6 0.78 0.22 Homestead Ashworth Farm 24 5 19 0.47 0.53 Jamaica: 3 August 1968 Anchor Salmon Farm Barn 1 20 20 1.00 Caracas Region, Venezuela: 25-26 November 1967 Cagua Huevos de Hoy Farm Barn4 51 51 1.00 Mixed Barns 25 25 1.00 TABLE 11-Continued Observed genotypes and phenotypes Allele frequency ~umber Es-5a/Es-5a Es-5b/-Es-5a Es-5b Sample of mice Band absent Band present A p Northern Part Hesselberg Farm VadFarm Central Part Sondergard Farm RavnFarm Southern Part 0stergard Farm Hji'.SjboFarm Jutland Peninsula, Denmark: 9-21 August 1968 34 34 30 30 32 30 2 27 14 13 31 8 23 33 12 21 1.00 1.00 0.97 0.72 0.51 0.61 0.03 0.28 0.49 0.39 TABLE 12 Variation in Esterase 5 in Texas Observed genotypes and phenotypes Allele frequency >lumber Es-5"/ Es-5" Es-5b/-Es-5a Es-5b Sample of mice Band absent Band present A p Dallas Region: 26-27 June 1968 Prosper Mayhard Farm Barn3 15 7 8 0.69 0.31 Barn4 14 5 9 0.68 0.32 Barn 5 20 12 8 0.78 0.22 Barn 6 24 13 11 0.74 0.26 Weatherford Rockwell Farm Barn2 15 11 4 0.86 0.14 Barn 3 18 18 1.00 Temple Region: 6-9 May 1967 Temple Waskow Farm 34 31 3 0.95 0.05 Austin Region: 1967-1968 Austin Cook-Synoett Farm June 1967 11 11 1.00 Herbert Farm May-June 1967 65 57 8 0.94 0.06 Orta Farm 19 17 2 0.95 0.05 Dripping Springs Empire Ranch 10 4 6 0.64 0.36 Stanley Farm Barn2 +5 45 1.00 Barn 4 10 10 1.00 Hildreth Farm Barn 1 April 1967 54 50 4 0.96 0.04 December 1967 186 143 43 0.88 0.12 Barn2 August 1967 108 98 10 0.95 0.05 December 1967 310 268 42 0.93 0.07 Barn3 August 1967 108 97 11 0.95 0.05 December 1967 80 75 5 0.97 0.03 Barn4 December 1967 198 168 30 0.92 0.08 Fredericksburg Region: 26-28 April 1967 Fredencksbur g Petch Farm 13 11 2 0.92 0.08 Abilene Region: 20 February 1968 Hodges Moore Field 32 25 7 0.88 0.12 San Angelo Region: 16-19 February 1968 San Angelo Heinze Field 18 February 45 37 8 0.91 0.09 19 February 55 40 1:5 0.85 0.15 Wilkie Downs Racetrack 16 15 1 0.97 0.03 Veribest Hurst Farm 21 19 2 0.95 0.05 Bitner Field 16 15 0.97 0.03 Seguin Region: 20-22 October 1967 Seguin Cotton Oil Company Milling & Storage R ooms 21 19 2 0.95 0.05 Smiler Farm2 17 12 5 0.84 0.16 Farm3 Barn3 37 12 25 0.57 0.43 Nixon Chessher Farm Regular Series Barn 4 11 5 6 0.68 0.32 Irregular Series Barn 1 42 12 30 0.54 0.46 Barn2 54 22 32 0.64 0.36 Barn 3 24 6 18 0.51 0.49 Barn4 51 7 44 0.38 0.62 Houston Region: 29 September-1 October 1967 Houston Otte Farm Barn3 11 9 0.91 0.09 Baytown Barker Farm Barn5 11 4 7 0.61 0.39 Chandler Farm Barn 6 10 7 3 0.84 0.16 Ensinal Region: 22October1967 (Riker Farm and Nieschwietz Farm Barns 1-7) and 3 November 1967 (Nieschwietz Farm Barns 8-11 and Burkholder Farm) Cotulla Riker Farm 21 7 14 0.58 0.42 TABLE 12-Continued Observed genotypes and phenotypes Allele frequency Number Es-5a/Es-5a Es-5b/-Es-5a Es-5b Sample of mice Band absent Band present A p Ensinal Nieschwietz Fann Barn 1 49 41 8 0.92 0.08 Barn2 86 61 25 0.84 0.16 Barn3 101 79 22 0.88 0.12 Barn4 25 21 4 0.92 0.08 Barn 5 50 42 8 0.92 0.08 Barn 7 106 86 20 0.90 0.10 Barn8 90 62 28 0.83 0.17 Barn 9 31 24 7 0.88 0.12 Barn 10 47 23 24 0.70 0.30 Barn 11 31 20 11 0.80 0.20 Burkholder Fann Barn 1 81 25 56 0.56 0.44 Barn2 47 12 35 0.51 0.49 Brownsville Region: 20-23 April 1967 (Jackson Warehouse, Brownsville) and 19-25 August 1967 Brownsville Jackson Warehouse April 56 35 21 0.79 0.21 Mercedes Coil Fann 13 13 1.00 Alamo McNeill Store 14 11 3 0.89 0.11 Donna WatsbyFann 40 40 1.00 Harlingen Guillen Fann North Barn 27 21 6 0.88 0.12 Goat Barn 11 9 2 0.91 0.09 Egg Room 10 10 1.00 Raymondville Pickard Fann Barn 1 75 63 12 0.92 0.08 Barn3 60 56 4 0.97 O.o3 Barn4 60 55 5 0.96 0.04 Barn 5 61 60 0.99 0.01 Alpine Region: 6-9 April and 15 May 1967 (Sul Ross Hog Barn, May) Alpine Sul Ross Hog Barn April 53 53 1.00 May 36 24 12 0.82 0.18 Sul Ross Cow Barn 11 11 1.00 Neie Ranch 16 16 1.00 Victoria Region: 27-28 February 1968 Victoria Raines Farm 14 8 6 0.76 0.24 Guadalupe Schoener Farm 37 14 23 0.62 0.38 Dacosta Reges Farm 11 10 0.33 0.67 Bishop Felder Farm Duck House 18 18 1.00 Barkley Farm 18 13 5 0.85 0.15 Odem Naylor Farm Barn 1 20 15 5 0.87 0.13 Barn3 14 12 2 0.93 0.07 Barn4 55 51 4 0.96 0.04 Barn5 22 22 1.00 Nacogdoches Region: 12-14 March 1968 (Nacogdoches) and 1May1968 (Rusk) Nacogdoches Fuller Farm Barn2 18 11 7 0.79 0.21 Barn 3 21 15 6 0.85 0.15 Linthicum Farm Corn Crib 1 15 8 7 0.74 0.26 Corn Crib 2 22 14 8 0.80 0.20 Rusk Hassell Farm Barn2 20 16 4 0.90 0.10 Bam4 12 5 7 0.65 0.35 Barn 5 15 12 3 0.90 0.10 Bryan Region: 22 March 1968 Bryan Reliance Farm Barn3 89 42 47 0.69 0.31 Barn6 45 12 33 0.52 0.48 Bam7 112 45 67 0.63 0.37 Bullock Farm 42 37 5 0.94 0.06 Steep Hollow Farm Barn 1 72 58 14 0.90 0.10 Barn5 67 62 5 0.96 0.04 Barn6 61 52 9 0.92 0.08 Huntsville Region: 28 March 1968 Phelps Swearingen Farm Corn Crib 22 9 13 0.64 0.36 Huntsville White Farm 10 3 7 0.56 0.44 Caldwell Region: 15, 17 April 1968 Paige Schindler Farm 72 72 1.00 Caldwell Clark Farm Brooder House 12 9 3 0.87 0.13 Mt. Pleasant Region: 2-5 May 1968 Mt. Pleasant Richardson Farm Barn 1 38 8 30 0.46 0.54 TABLE 12-Continued Observed genotypes and phenotypes Allele frequency Sample Number of mice Es-5a /Es-54 Band absent Es-5b/­Band present Es-5a A Es-5b p Pittsburg Pilgrim Farm Barn 1 77 13 64 0.41 0.59 Barn2 36 6 30 0.41 0.59 Barn3 14 1 13 0.30 0.70 Barn4 70 6 64 0.30 0.70 Lamesa Region: 12 June 1968 Lamesa Womack Farm Barn4 23 15 8 0.81 0.19 Lubbock Region: 14-19 June 1968 Littlefield Barton Farm Barn 1 71 42 29 0.77 0.23 Barn3 126 77 48 0.78 0.22 Barn 10 90 46 44 0.72 0.28 Barn 11 51 25 26 0.70 0.30 Barn 12 106 41 65 0.62 0.38 Shallowater Gary Farm Barn4 53 39 14 0.86 0.14 Barn5 42 26 16 0.79 0.21 Barn 12 35 16 19 0.68 0.32 Vance Farm Barn 11 14 7 7 0.71 0~29 Amarillo Region: 21-22 June 1968 Panhandle Gibson Farm Barn4 34 22 12 0.81 0.19 Stinnett Hughes Farm 10 3 7 0.56 0.44 Pleasant Valley Winkleman Farm Barn 1 33 20 13 0.78 0.22 Barn2 101 57 44 0.75 025 I · ~ ~ : • El Paso Region: 10-16 July 1968 Clint Wilson Farm Barn2 10 8 2 0.90 0.10 Las Cruces, New Mexico Stahlman Farm Barns 1 and 2 10 7 3 0.8-4 0.16 Eagle Pass Region: 16 July 1968 Quemado Morro Farm Barn 1 48 47 0.99 0.01 Barn2 15 15 1.00 TABLE 13 Variation in Hemoglobin in Regions other than Texas Observed genotypes and phenotypes Allele frequency Hbbd/ Hbb•/ Hbbd/ Nwnber Hbbrl/-Hbbrl Hbb·• Hbb• Hbbd Hbb• Sample of mice Diffuse Strong diffuse Single Weak diffuse D s Utah: 1 July 1967 Salt Lake City Adamson Ranch 49 49 1.00 Central California: 2-4 December 1968 Modesto Blixis Farm Barn2 22 5 8 9 0.43 0.57 Fresno Guthrie Farm 34 5 3 26 0.53 0.47 Kemp Farm 17 4 8 5 0.38 0.62 Southern California: 23 November-8 December 1968 Yucaipa Daily Fresh Egg Farm 29 3 13 13 0.33 0.67 Ramona Hallowell Farm Barn 1 18 10 3 5 0.69 0.31 Barn2 29 11 17 0.67 0.33 Barn3 12 5 1 6 0.67 0.33 Barn4 21 7 3 11 0.60 0.40 Barn5 10 3 2 5 0.55 0.45 Barn9 31 10 8 13 0.53 0.47 Barn 12 17 4 3 10 0.53 0.47 Barn 15 17 3 7 7 0.38 0.62 Barn 16 10 5 4 0.70 0.30 Arizona: 12-13 December 1968 Tucson Thompson Farm Barn 1 46 5 14 27 0.40 0.60 Barn2 49 8 16 25 0.42 0.58 Barn3 19 2 11 6 0.26 0.74 Barn 6 39 6 19 14 0.33 0.67 Arizona State Poultry 25 18 6 0.16 0.84 Arizona Star Ranch Barn2 120 40 36 44 0.52 0.48 Barn 17 65 26 7 32 0.65 0.35 Minnesota: 5 October 1968 Anoka Ghostly Farm Barn 1 30 29 1 0.02 0.98 Barn2 52 44 8 0.08 0.92 Barn 3 12 12 1.00 Wisconsin: 17-23 October 1968 Whitewater Duffin Farm Chicken House 42 32 9 0.13 0.87 TABLE 13-continued Observed genotypes and phenotypes Allele frequency Hbbd/ llbb•/ /lbbd/ Number /-lbbd/-Hbbd /lbb• llbb• /lbbd /-lbb• Sample of mice Diffuse Strong diffuse Single Weak diffuse D S Illinois: 26-28 October-1 November 1968 Tuscola KleissFarm 30 30 Atwood Schable Fann 59 59 Germantown Albers Farm 138 119 19 Millstadt Albert Farm Barn 1 38 31 7 Bam2 10 5 4 Ohio: 1967 15 Localities 39 34 5 Kansas: 27 September-2 October 1968 Lawrence Powell Farm Barn 1 20 10 9 West Feed Store 22 15 6 Florida: 25-26 July 1968 Princeton Fruman Farm Barn2 18 9 7 Barn3 15 14 Homestead Ashworth Fann 24 13 10 Jamaica: 3 August 1968 Anchor Salmon Fann Barn 1 20 17 3 Caracas Region, Venezuela: 25-26 November 1967 Cagua Huevos de Hoy Fann Barn4 51 38 13 Mixed Barns 25 17 8 Jutland Peninsula. Denmark: 9-21 August 1968 Northern Part Hesselberg Farm 34 30 4 VadFarm 30 9 3 18 Central Part Sondergard Fann 32 27 5 RavnFarm 27 27 Southern Part 0stergard Fann 31 13 17 0.07 1.00 1.00 0.93 0.09 0.30 0.91 0.70 0.06 0.94 0.28 0.18 0.72 0.82 0.31 0.03 0.25 0.69 0.97 0.75 0.08 0.92 0.49 0.43 0.51 0.57 0.06 0.60 0.08 0.31 0.94 0.40 0.92 1.00 0.69 H~jboFarm 33 33 1.00 Observed genotypes and phenotypes Allele frequency HbbtL/ Hbb•/ llbbd/ Number Hbbd/-llbbd Hbb" f/bb• Hbb~ Hbb• Sample of mice Diffuse Strong diffuse Single Weak diffuse D s Dallas Region: 25 April 1967 (Dallas) and 26-27 June 1968 (Prosper and Weatherford) TABLE 14---continued TABLE 14 Variation in Hemoglobin in Texas Dallas Littlebrook Fann 78 19 59 0.13 0.87 Prosper Mayhard Fann Barn3 15 8 6 0.27 0.73 Barn4 14 10 3 0.18 0.82 Barn5 20 15 5 0.13 0.87 Barn6 24 2 13 9 0.27 0.73 Weatherford Rockwell Farm Barn2 15 11 3 0.17 0.83 Barn3 18 8 10 0.28 0.72 Temple Region: 6-9 May 1967 Temple Waskow Farm 34 17 17 0.29 0.71 Austin Region: 1967-1968 Bastrop Henderson Ranch 58 8 50 0.07 0.93 Austin Cook-Synoett Fann March 1967 105 26 79 0.13 0.87 June 1967 11 10 0.05 0.95 Garfield Fann January 1967 12 4 8 0.18 0.82 Herbert Fann May-June 1967 90 35 55 0.22 0.78 Orta Fann 19 1 18 0.03 0.97 Norco Ranch 15 6 9 0.22 0,78, Robinson Fann 18 4 14 0.12 0.88 . Liberty Hill Lay Ranch March 1967 106 61 45 0.35 0.65 February 1968 324 29 190 105 0.25 0.75 Dripping Springs Empire Ranch March 1967 81 4 77 0.03 0.97 October 1967 15 12 2 0.13 0.87 Stanley Farm Barn2 45 45 1.00 Barn4 10 10 1.00 Hildreth Fann Barn 1 April 1967 54 17 37 0.17 0.83 July 1967 108 27 81 0.13 0.87 December 1967 186 6 130 50 0.17 0.83 Observed genotypes and phenotypes Allele frequency Hbb•/ Hbbd/ Number Hbbd/-1ttbd/ llbb• lib/,. llbbd llbb• Sample of mice Diffuse Strong diffuse Single Weak diffuse D s Barns 1and2 April 1967 74 32 42 0.25 0.75 Bam2 August 1967 108 71 37 0.41 0.59 December 1967 310 41 147 122 0.33 0.67 Barn3 August 1967 108 42 66 0.22 0.78 December 1967 80 3 52 25 0.19 0.81 Barn4 August 1967 108 37 71 0.19 0.81 December 1967 198 11 122 65 0.22 0.78 Fredericksburg Region: 26-28 April 1967 Fredericksburg Petch Farm 13 13 1.00 Abilene Region: 20 February 1968 Hodges Moore Field 32 29 3 0.05 0.95 San Angelo Region: 16-19 February 1968 San Angelo Heinze Field 18 February 45 37 7 0.10 0.90 19 February 55 47 8 0.07 0.93 Wilkie Downs Racetrack 16 12 4 0.13 0.87 Veribest Hurst Farm 25 2 20 3 0.14 0.86 Bitner Field 16 15 0.03 0.97 Seguin Region: 20-22 October 1967 Seguin Cotton Oil Company Milling & Storage Rooms 21 17 4 0.10 0.90 Smiley Farm2 17 11 5 0.21 0.79 Farm3 Barn3 37 32 5 0.07 0.93 Nixon Chessher Farm Regular Series Barn4 11 10 0.05 0.95 Irregular Series Barn 1 42 2 25 15 0.23 0.77 Barn2 54 51 3 O.o3 0.97 Barn3 24 19 5 0.10 0.90 Barn4 ' 51 3 34 14 0.20 0.80 Houston Region: 29 September-1 October 1967 Houston Otte Farm Bam3 11 11 1.00 Selander et al.: Polymorphism in Mus musculus 335 Baytown Barker Farm Barn5 11 10 0.05 0.95 Chandler Farm Barn6 10 10 1.00 Ensinal Region: 22 October 1967 (Nieschwietz Fann Barns 1-7) and 3 November 1967 (Nieschwietz Farm Barns 8-11 and Burkholder Fann) Ensinal Nieschwietz Fann Barn 1 49 2 34 13 0.17 0.83 Barn2 86 3 50 33 0.23 0.77 Barn3 101 4 71 26 0.17 0.83 Barn4 25 3 17 5 0.22 0.78 Barn5 50 40 10 0.10 0.90 Barn 7 106 3 72 31 0.17 0.83 Barn8 90 2 72 16 0.11 0.89 Barn 9 31 23 8 0.13 0.87 Barn 10 47 2 28 17 0.22 0.78 Barn 11 31 25 5 0.11 0.89 Burkholder Fann Barn 1 81 71 9 0.07 0.93 Barn2 47 43 4 0.04 0.96 Brownsville Region: 20-23 April 1967 (Jackson Warehouse, Brownsville) and 19-25 August 1967 Brownsville Jackson Warehouse April 56 28 5 23 0.70 0.30 Mercedes Coil Farm 13 11 2 0.59. 0.41 Alamo McNeill Store 14 11 3 0.52 0.48 Donna Watsby Fann West Barn 16 11 5 0.43 0.57 East Barn 107 88 19 0.58 . 0.42 Harlingen Guillen Fann North Barn 27 12 15 0.25 0.75 Goat Barn 11 6 5 0.32 0.68 Egg Room 10 8 2 0.53 0.47 Raymondville Pickard Fann Barn 1 75 7 28 40 0.36 0.64 Barn3 60 13 15 32 0.48 0.52 Barn4 60 25 10 25 0.62 0.38 Barn5 61 24 10 27 0.62 0.38 Alpine Region: 6-9 April and 15May1967 (Sul Ross Hog Barn, May) Alpine Sul Ross Hog Barn April 53 4 49 O.o4 0.96 May 36 36 1.00 r Sul Ross Cow Barn 11 5 6 0.26 0.74 (• Neie Ranch 16 4 12 0.13 0.87 TABLE 14-continued Oliserved genotypes and phenotypes Allele freq~en~y Hbbdf Hbb•/ Hbb"I Number Hbbd/-Hbbd llbb• llbb• Hbbd, Hbb• Sample of mice Diffuse Strong diffuse Single Weak diffuse D S Victoria Regioi;i: 27-28 February 1968 Victoria Raines Farm 13 0.04 0.96 Guadalupe Schoener Farm 37 33 0.05 0.95 Dacosta Reges Farm 11 10 0.09 0.91 Corpus Christi Region: 1-4 March 1968 Bishop Felder Farm Duck House 18 18 1.00 Barkley Farm 18 11 6 0.22 0.78 Odem Naylor Farm Barn 1 20 17 3 0.08 0.92 Barn3 14 6 8 0.29 0.71 Barn4 55 34 17 0.23 0.77 Barn5 22 17 5 0.11 o.89 Nacogdoches Region: 12-14March1968 (Nacogdoches) and 1May1968 (Rusk) Nacogdoches Fuller Farm Barn 2 18 18 1.00 Barn 3 21 21 1.00 Linthicum Farm CornCrib 1 15 15 1.00 ComCrib2 22 22 1.00 Rusk Hassell Farm Bam2 20 19 0.02 0.98 Barn4 12 12 1.00 Barn5 15 15 1.00 Bryan Region: 22 March 1968 Bryan Reliance Farm Barn3 89 7 46 36 0.28 0.72 Bam6 45 4 25 16 0.27 0.73 Barn7 112 8 63 41 0.25 0.75 Bullock Farm 42 34 8 0.10 0.90 Steep Hollow Farm Barn 1 72 9 30 33 0.35 0.65 Barn5 ,67 5 45 17 0.20 0.80 Barn6 61 11 28 22 0.36 0.64 Huntsville Region: 28 March 1968 Phelps Swearingen Farm Corn Crib 22 10 11 0.29 0.71 Huntsville White Farm 10 9 1 0.05 0.95 Caldwell Region: 15, 17 April 1968 Paige Schindler Farm 72 72 1.00 Caldwell Clark Farm Brooder House 12 12 1.00 Lamesa Region: 12 June 1968 Lamesa VVomack Farm Barn4 23 17 6 0.13 0.87 Mt. Pleasant Region: 2--5 May 1968 Mt. Pleasant Richardson Farm Barn 1 38 31 7 0.09 0.91 Pittsburg Pilgrim Farm Barn 1 77 3 40 34 0.26 0.74 Barn2 36 16 19 0.29 0.71 Barn3 14 1 11 2 0.14 0.86 Barn4 70 4 41 25 0.24 0.76 Lubbock Region: 1~19 June 1968 Littlefield Barton Farm Barn 1 71 3 39 29 0.25 0.75 Barn3 126 2 92 32 0.14 0.86 Barn 10 90 16 34 40 0.40 0.60 Barn 11 51 9 26 16 0.33 0.67 Barn 12 106 12 64 30 0.26 0.74 Shallowater Gary Farm Barn4 53 43 10 0.09 0.91 Barn5 42 42 1.00 Barn 12 35 2 30 3 0.10 0.90 Vance Farm Barn 11 14 14 1.00 Amarillo Region: 21-22 June 1968 Panhandle Gibson Farm Barn3 10 10 1.00 Barn4 34 33 0.03 0.97 Stinnett Hughes Farm 10 8 0.15 0.85 Pleasant Valley Winkleman Farm Barn 1 33 28 5 0.08 0.92 Barn2 101 3 85 13 0.09 0.91 El Pas~ Region: 10-16 July 1968 Clint Wilson Farm Barn2 10 4 5 0.35 0.65 La.s Cruces, New Mexico Stahlman Farm Barns 1 and2 10 9 0.05 0.95 TABLE 14--continued Observed genotypes and phenotypes Allele frequency Hbbd / Hbb• / Hbbd/ Number Hbbd/-HbbJ Hbb• Hbb• Hbbd Hbb • Sample of mice Diffuse Strong diffuse Single Weak diffuse D S Eagle Pass Region: 16 July 1968 Quemado Morro Farm Barn 1 65 44 19 0.18 0.82 Barn2 15 11 3 0.17 0.83 TABLE 15 Partial Correlation Coefficients of Mean Allele Frequencies at Four Loci in 12 Regions Outside of Texas (Upper Matrix) and in 23 Regions in Texas (Lower Matrix) 1 Locus and allele Es-2b Es-3b Es-5a Hbbd Es-2b x - 0.097 0.130 0.732* Es-3b -0.502* x -0.247 (}.084 Es-5a -0.386 - 0.509* x - 0.173 Hbbd -0.065 ·­ 0.165 -0.124 x 1 Degrees of freedom= 8 for upper matrix, 19 for lower matrix. * indicates significance at the 5% level. W. S. Stone, your work is done And you are now resting a/Jove. It was the hope of your country and the world over For your orator's staff not to be severed; But alas our fate is destined for us By God our Father and Redeemer, So your service given freely for your country and more Shall never be forgotten the world over, Especially Samoa. . .. traditional Samoan Chief's eulogy. Contributed by friends in Apia, Upolu, Western Samoa.