Bulletins of the Texas Memorial Museum 

No. 1. 	Mylohyus nasutus, Long-nosed Peccary of the Texas Pleistocene, by E. L. Lundelius, Jr., 1960 
No. 2. 	The Friesenhahn Cave (Part I), by Glen L. Evans; The Saber-toothed at, Dinobastis serus (Part II),by Grayson E. Meade, 1950 
No. 3. 	A Bibliography of Recent Texas Mammals, byGerald G. Raun, 1962 
No. 4. 	Handbook of Texas Archeology: Type Descriptions,
edited by Dee Ann Suhm and Edward B. Jelks. Pub­lished jointly by the Texas Memorial Museum and the Texas Archeological Society (reprints available) 
No. 5. 	Salvage Archeology of Canyon Reservoir: The Wunderlich, Footbridge, and Oblate Sites, by Leßoy Johnson, Jr., Dee Ann Suhm, and Curtis Tunnell, 1952 
No. 6. 	The Ethnography and Ethnology of Franz Boaz, byLeslie A. White, 1963 
No. 7. 	Fossil Vertebrates from Miller's Cave, Llano County, Texas, by Thomas Patton, 1963 
No. 8. 	Interactions Between a Bisexual Fish Species and its Gynogenetic Sexual Parasite, by Clark Hubbs, 1963 
No. 9. 	Oedaleops campi (Reptilia: Pelycosauria) a new genusand species from the Lower Permian of New Mexico, and the family Eothyrididae, by Wann Langston, Jr., 1965 
No. 10. 	Blancan Mammalian Fauna and Pleistocene Formations, Hudspeth County, Texas, by William Samuel Strain, 1956 
No. 11. 	A Population of Woodrats (Neotoma micropus), by Gerald 
G. Raun, 1966 
No. 12. 	Toward a Statistical Overview of the Archaic Cultures of Central and Southwestern Texas, by Leßoy Johnson, Jr., 
No. 13. 	Geographic Variations in Survival of Hybrids Between Etheostomatine Fishes, by Clark Hubbs, 1967 
No. 14. 	A Lipan Apache Mission, San Lorenzo de la Santa Cruz, 1762-1771, by Curtis Tunnell and W. W. Newcomb, Jr., 1969 
No. 15. 	Pliocene Carnivores of the Coffee Ranch, by-Walter W. Dalquest, 1969 
$l.OO 
1.00 
1.00 
9.00 
2.00 
2.00 
2.00 
2.00 
1,00 
2.00 
2.00 1967 2.00 
2.00 
3.00 
2.00 
BULLETIN 
OF THE TEXAS MEMORIAL MUSEUM 16 
July, 1970 
EXCAVATIONS AT BAKER CAVE, 
VAL VERDE COUNTY, TEXAS 

PART I 
THE ARCHEOLOGICAL INVESTIGATION 

James H. Word PART II ANALYSIS OF THE FAUNAL REMAINS Charles L. Douglas 
The Texas Memorial Museum 
24th & Trinity, Austin, Texas 78705 
The University of Texas at Austin 


PART I 
EXCAVATIONS AT BAKER CAVE, VAL VERDE COUNTY, TEXAS 
James H. Word 

ABSTRACT 
Intermittent excavations in a dry shelter in Val Verde County, Texas from April 1962 to July 1966 revealed well stratified deposits. The earliest deposits yielded late Paleo-Indian projectile points of the Plainview Golondrina variety, corroborated by two charcoal samples radiocarbon-dated at 8910± B.P. and 9030 l B.P. Higher deposits produced projectile points from Early Archaic to Late Archaic. In addition to lithic materials, perishable materials from Mid-Archaic through Late Archaic were recovered. Features associated with the Archaic were recorded. Arrow points were recovered from the surface stratum but of insufficient quantity to indicate an intensive Neo-American occupation. 
i 
TABLE OF CONTENTS 
PART I 
Page Abstract i Introduction 1 Acknowledgments 1 Previous Research in Trans-Pecos 2 Environment 3 Site Description 4 Excavation Techniques 6 Description of Occupational Units 10 Features 14 Chipped Stone Tools 20 Smoothed, Pecked, Ground, Scratched and Unmodified Stone Objects 59 Articles of Wood and Fiber 65 Articles of Bone 91 Articles of Leather and Sinew 97 Summary 98 Bibliography 106 
ILLUSTRATIONS 
PART I 
Figure 	Page 
1. 	Geographic location of Baker Cave and surrounding area 5 
2. View of Baker Cave looking south 	6 
3. Floor plan of Baker Cave and areas tested 	7 
4. Stratigraphy in early testing of Unit 1 	9 
5. Profile of east wall 	11 
5. Stylized hearth 	15 
7. Feature 2, antler under rock 	16 
8. Feature 3, tied prickly pear internodes 	17 
9. Feature 4, relation of "stakes" to roof spall 19 
10. 	Dart points. Almagre, Baker I, Baker 11, Bandy, Castroville, Catan 22 
11. 	Dart points. Dari, Desmuke, Early Barbed I, Early Barbed 11, Early Barbed 111, EarlyBarbed IV, Early Barbed V, Ensor I, Ensor 11, Fairland 24 
12. 	Dart points. Frio, Langtry I, Langtry 11, Langtry 111, Langtry IV 27 
13. 	Dart points. Langtry V, Langtry VI, Marcos, Marshall, Matamoros, Montell 30 
14. 	Dart points. Nolan, Palmillas, Pandale, Pan­dora, Pedernales, Plainview Golondrina, Plainview (?), Shumla 32 15. Dart points. Tortugas Travis, Triangular
/ 
Concave, Triangular Convex, Uvalde, Zorra, 
Miscellaneous I, 11, 111, IV. Arrow points. 
Fresno, Harrell, Perdiz, Scallorn, Coryell, 
Toyah, Miscellaneous 35 

16. Knives. Types I and II 	40 
17. Knives. Types 111, IV, and V 	42 
18. Knives. Types VI, VII, VIII, and IX 	43 
19. 	Knives. Types X, Xla, Xlb, Xlc, XII, and XIII 45 
20. 	Knives. Type XIV. Drills. Types I, 11, and 111. Choppers 48 
21. Chisel choppers, hammerstones, fist axe, core 50 
22. Scrapers. Types I, 11, 111, IV, and V 	52 
23. 	Utilized flakes. Types I, 11, 111, and IV. Spokeshaves. Types I and II 54 
24. 	Serrated flakes, burins, angle burins, flaked burins, burin-graver, gravers 57 
25. Pendant, pigments, grinding slabs 	60 
26. Manos. Types I, 11, and 111 	62 
27. Scratched river pebbles 	64 
28. Sandals. Types A, B, C, and D 	66 
29. Netting. Knotless and knotted 	68 
30. Detail of knotless netting 	69 
31. Cordage 	70 
32. 	Knots. Square, overhand, granny, sheet bend, twisted, slipknot, larkshead, double overhand, figure eight, clove hitch, half hitch, two half hitches, bowknot, running square knot 72 
iv 
33. 
Strands of lechuguilla joined by square knots 	75 

34. 	
Common knots used by prehistoric occupants 77 


of Baker 	Cave 
35. 	
Tied lechuguilla, twill woven basket fragment, coiled basket fragments, wicker woven bas­ket fragment, and checker woven matting 79 

36. 	
Twill woven matting, tump strap, plaited ob­jects 81 

37. 	
Fiber loops, split leaf objects, tied loops,twisted lechuguilla, tied bundles, foreshaft with hafted point, tenon, cupped end of atlatl shaft, arrow nocks 83 

38. 	
Wooden scoop, pegs, stake, pointed stick, wooden arrow point, thorn needle, rabbit stick, and lashed sticks 85 

39. 
Thorn needle and mesquite thorn 	87 

40. 	
Firedrills, fire hearths, cut sticks, chewed stick, wooden artifacts, fish hook cactus spines, shaved stick 89 

41. 	
Prickly pear skin, toy bow, bone artifacts, flaking tools 93 

42. 	
Objects of bone, leather, sinew-wrapped dart points, and shell 96 



INTRODUCTION 
Early in February 1962 the author learned of an undisturbed shelter located on the Walter Baker Ranch. Jim Baker, a son of the ranch owner, graciously offered to allow archeological research in the shelter. Deposits were intact because of the shelter's isolation and the desire of the Baker family to protect it from vandals. In April 1962 the shelter was visited by a small crew and preliminary work was begun. The depth and richness of the deposits prompted a more thorough test of the cave. The author made eight visits of from two to six days' duration in the next four years. The Texas Archeological Research Laboratory of The University of Texas at Austin designated the shelter 41 W 213. 
ACKNOWLEDGMENTS 
This report would not have been possible without the assis­tance of a great number of friends and archeologists, both profes­sional and amateur; it is with great pleasure that their efforts are acknowledged here. 
A major contribution was made by the Baker family, who allowed our crossing their property when conditions permitted.Their interest in the "goings on" gave a boost to morale that was welcomed, and their hospitality as dinner hosts is appreciated.Stories of earlier times were always the highlight of each visit. 
Dr. E. Mott Davis and Dr. T. N. Campbell of The University of Texas Department of Anthropology, Mr. Curtis Tunnell, then of the Texas Memorial Museum, Mr. John W. Greer of Austin, Dr. David 
R. Kelley, then of Texas Technological College, Lubbock, and Mrs. Anne Fox of the Witte Memorial Museum of San Antonio were generous in sharing their professional skills and advice. 
Amateur archeologists who contributed their time were L. D. Britton, Dr. 0. R. Mclntosh, Jimmy Owens, Mike Bishop, Roland Adams, and Ronnie Fullerton of Floydada; Johnny Ellison of Peters­burg; F. A. Runkles and Emmett Shedd of Post; Charles Bucy, Ed Bucy, Kim Bucy, Claude Brown, Major Forrest Fenn, and Jim Strader of Lubbock; Charles Bandy, Helen Bandy, Charlene Bandy, and Vickie Hill of Brownfield; and W. E. Garrett of San Antonio. 
My three daughters, Ann Carr, Mary Garrett, and Beth Arnold, and my son-in-law, Charles Carr, were devoted workers. The 
patience and understanding of my wife, Dorothy, is of such magni­tude that mere words of appreciation will not suffice. 
Dr. Charles L. Douglas, Mr. Dudley M. Varner, and Dr. W. W. Newcomb, Jr. of the Texas Memorial Museum and Dr. David S. Dibble of the Texas Archeological Research Laboratory criticized this report and made appreciated suggestions. Mrs. Willena C. Adams of the Texas Memorial Museum edited the report. 
PREVIOUS ARCHEOLOGICAL RESEARCH IN THE TRANS-PECOS REGION 
Previous archeological research began in the Trans-Pecos with excavations in 1932 at Fate Bell Shelter (Pearce and Jackson,1933), closely followed by research in Eagle Canyon (H. C. Taylor,1949b). In 1933 the Shumla Caves were dug (Martin, 1933) and the Smithsonian Institution (Setzler, 1934) tested two caves. In 1936 The University of Texas dug at Horseshoe Cave (Butler, 1948).
W. C. Holden (1937) excavated Murrah Cave in 1937 and Eagle Cave was worked by Davenport in 1938. This was the last field work in Val Verde County in the 1930'5. Mrs. Mardith K. Schuetz, Curator of Anthropology at the Witte Museum in San Antonio, later re­analyzed Val Verde cave material in the Witte collections (Schuetz,1956, 1961, 1963). 
These early excavations recorded little on the stratification of the investigated sites, and few attempts were made to correlate the various cultural strata. Prehistoric inhabitants of the Trans-Pecos and Big Bend areas were referred to as "Big Bend Basket Makers" (Martin, 1933: 86) or "Texas Cave Dweller" (Sayles, 1935),which was divided into "Pecos River Cave Dweller", "Big Bend Cave Dweller", and "Hueco Cave Dweller." J. Charles Kelley (Kelley,Campbell, and Lehmer, 1940) further developed Sayles's "Texas Cave Dweller." Under Kelley's refinement, "Pecos River Cave Dwel­ler" became the "Pecos River Focus" and "Big Bend Cave Dweller" became the "Chisos Focus." 
With the advent of World War II there was a hiatus in archeo­logical research in the Trans-Pecos which continued until 1958; but after plans for construction of the Amistad dam and reservoir were approved, archeological investigations were resumed. Graham and Davis (1958) extensively surveyed the Amistad Reservoir area, and various sites were recommended for excavation. As a result. Centipede and Damp caves (Epstein, 1963), Devil's Mouth Site 
(Johnson, 1964), Coontail Spin, Zopilote Cave, Mosquito Cave, and the Doss Site (Nunley, Duffield, and Jelks, 1965), Eagle Cave 
2 
(Ross, 1965), Fate Bell Shelter (Parsons, 1965), and Bonfire Shel­ter (Dibble and Lorrain, 1968) were investigated and research is continuing. 
The post-World War II research has contributed additional detailed information of the area. One major contribution of this research has been the definite establishment of a late Paleo-Indian occupation. Another contribution has been the setting upof tentative point type sequences. Archaic projectile point types have been divided into the Parida Series, Devils Series, and Rio Bravo Series (Nunley, Duffield, and Jelks, 1965). 
A preliminary study of local ecology (Story and Bryant,1966) made from data provided by the salvage program is one of the more significant results of archeological research in the Amistad Reservoir Area. 
ENVIRONMENT 
Val Verde County includes portions of the western edge of the Edwards Plateau and an easterly segment of the Stockton Pla­teau (Texas Almanac, 1956-57: 172-173). The surface is under­lain by Cretaceous limestone of the Comanchean Series (Graham and Davis, 1958) which is cut by many deep ravines. The tops of the hills are covered by scant soils produced from weathered lime­stone and vegetal matter. The bottoms of some wider valleys con­tain soils of alluvial and colluvial origin. 
The Walter Baker Ranch locale contains a wide variety of animal life. Deer, peccary, raccoons, skunks, armadillos, squir­rels, beaver, rabbits, rats, and mice are the principal mammals. In addition to these, Mr. Walter Baker, who has lived in the Devils River area for 80 years, remembers the presence of wolf, coyote, black bear, and panther. He recalls hearing his father mention bison in the immediate vicinity in still earlier times. Reptiles including turtles, horned lizards, and snakes are common. In permanent streams are catfish, sunfish, black bass, and gar.The few types of amphibians include various frogs. Turkeys,eagles, hawks, buzzards, scaled quail, doves, and roadrunners in­habit the area, as well as a large assortment of smaller birds. 
The wide variety of plant life includes trees such as pecan, walnut, liveoak, willow, sycamore, and persimmon. Ocotillo, cat-claw, cenizo, guahillo, and a great variety of cacti are common 
in the highlands as are lechuguilla, sotol, sacahuisti, and yuc­cas 
. 
Thus we have an area in which the typical flora and fauna of the Edwards Plateau merges with desert flora and fauna typicalof the Trans-Pecos region and the Stockton Plateau to the west. 
Surface water is scarce except in the Rio Grande, Pecos, and Devils rivers. Potholes fill with water during periods of rain­fall and for some time retain it. A permanent reservoir of water was once supplied by springs a short distance east of the site (Walter Baker, personal communication). 
SITE DESCRIPTION 
Baker Cave lies on the property of Walter H. Baker, J. S. Baker, and S. P. Baker about equidistant between Juno and Comstock in north-central Val Verde County (Fig. 1). It is on the eastern edge of the Baker Ranch and on the west bank of Phillips Creek, a dry tributary of the Devils River. 
Baker Cave is high in the face of a northeasterly facingbluff (Fig. 2). Of intermediate size as classified by Graham and Davis (1958: 49-53), it is 120 feet long and 56 feet deep. The ceiling height above the deposits varies from 18 feet near the mouth to only a few feet at the back. In outline, the shelter is semicircular. In spite of the evidence of extensive human occupa­tion, the roof is unblackened. This is possibly due to two factors 
(1) rapid spalling of the roof, (2) updrafting winds which tend to vent smoke and dust to the outside of the shelter. The depositsof the shelter slope gently from the back to the front. Large segments of the roof had fallen (Fig. 3), and presented a problemin the preliminary testing. The overhang of Baker Cave follows, in general, the outer margin of the floor, which drops abruptlydown the face of the bluff. An extensive talus slope consists of fire-cracked rock immediately in front of and below the shel­ter. No pictographs are evident. 
The surface of the deposits is dry and in low places sheepdung has accumulated to a depth of two to three inches. In "blow­out" areas of the shelter, burned rock, flint debris, and fiber are evident. There is, however, very little surface evidence of occupation. 
-
Fig. 1. Geographic location of Baker Cave and surrounding area 
-
Fig. 2. View of Baker Cave looking south 
EXCAVATION TECHNIQUES 
A datum point was established on a large roof spall in the northern part of the shelter. This datum point was used to layout the shelter into five-foot units (Fig. 3). The base line for the grid system was laid out on an axis running the length of the shelter, and while it was not exactly oriented to any of the car­dinal directions, it was designated as the east-west line. Lines were established parallel to the base line at five-foot intervals; these were intersected by lines running perpendicular to the base 
-
Fig. 3. Floor plan of Baker Cave and areas tested 
line, also at five-foot intervals, that were called north-south lines. Approximately three-fourths of the shelter was thus divid­ed in five-foot excavation units. The shelter was mapped (Fig. 3),recording the outlines of the shelter and the large roof spallsexposed on the surface. 
The first unit excavated was a test for the purpose of de­termining the depth of the deposits, the nature of the occupation­al deposits, and to establish the stratification of the deposits so that future units could be excavated by strata. All sediments were screened through quarter-inch mesh screen. All artifacts were kept in sacks designating the unit and level of provenience.All flint, bone, shell, and wood was kept; samples of vegetal matter were saved from each stratum since it was impractical to keep all of this material. Profiles were recorded of occupational strata on three sides of the test unit, but the fourth side was not recorded due to the necessity of providing a means of egressfor any animal that might fall into the pit. 
It soon became evident that other units had to be excavated to prevent cave-ins. Unit 2, opened next, was south of Unit 1. Unit 3, begun next at the request of the ranch owner, allowed es­cape of any sheep accidentally falling into the test pit. Ad­ditional units were opened and given the numerical number in the sequence of excavation. 
Unit 6 lies at the extreme rear of the shelter and is 15 feet east and 35 feet south of the southwest corner of Unit 1. This unit was begun in order to determine whether occupational use varied from that in the front of the shelter. It will be noted from an examination of Fig. 3 that the designation "unit" does not necessarily imply a five-foot unit. 
Unit 1 was first excavated in six-inch levels. It soon be­came apparent (Fig. 4) that it would be more informative to ex­cavate by occupational strata since occupational deposits were distinct and easily followed. The strata were given numbers as they were discovered and this system of stratum numbering was con­tinued for the entire period of research. Unfortunately, strata were not continuous throughout the various excavation units. Lenses were commonly encountered and strata frequently merged to become one, or split to become two strata in adjacent units. 
Unit1 
of

testing

early

in


-Stratigraphy

4. 
Fig. 

DESCRIPTION OF OCCUPATIONAL UNITS 
Some strata were combined into zones. A "zone" as used herein refers to a major division distinguishable from other soil units on the basis of rather distinct changes in soil color, content, and texture. Divisions within the zone, termed strata, exhibit more subtle differences than those used in separating zones. A "lens" refers to a soil unit which does not extend over a great distance and generally has distinct characterics different from the zone or stratum in which it is contained. 
Strata between cultural deposits are not given numerical designations. These strata are of a decomposed limestone con­sisting of finely divided particles and roof spall of varyingsize. Artifacts occurred in these strata, but it is felt that they were intrusive and not initially deposited within these 
areas. Lumps of charcoal, flint chips, and an occasional fibrous artifact were recovered, but the deposit did not have the colora­tion or content of the strata above or below. Probably this material was intruded into the decomposed limestone deposits as 
a result of human and animal activity. Materials recovered from the unnumbered deposits were assigned to the overlying occupa­
tional deposit. 
Zone 1_ (Fig. 5). Zone I is comprised of two strata toward the front of the shelter resting on a finely divided limestone deposit varying from one to 24 inches thick. Midway toward the back of the shelter. Zone I consists of three strata. The two or three strata are indistinguishable in character but are sep­arated by sterile deposits. The soil in Zone I has a grayishcast and contains specks of charcoal with an occasional lump to 
3.0 cm. A few fire-cracked hearthstones are present. Flint debris was quite common and of large size; the lithic artifact yield, however, was comparatively low. Roof spalls varied from approximately 20.0 cm. to as small as 1.0 cm., but the majority were small. Rodent bones were most common; some cracked deer bone was charred. The deposit was damp. It was from the upperof the two strata near the front and the upper two from the middle of the shelter that late Paleo-Indian points were re­covered. 
Within Zone I in Units 6, 9, 10, and 14, lenses of tan-colored soil held greater concentrations of charcoal than the strata in which they were contained, varying from finely divided particles to fragmented lumps. If these lenses were fire hearths. 
10 

wall 
east 

of

-Profile

5. 

Fig. 

they were very generalized and there was little or no formal con­struction. A slight increase in soot-stained rock was of insuf­ficient quantity to reveal the placement of a definite hearth. 
Zone II (Fig. 5). Zone II is more complex than Zone I. It appears as a single stratum in Unit 1 and consists of three strata and a lens in Units 7, 8, 9, 10, and 14. Zone II in Unit 6 con­sists of two strata. A principal characteristic is the presenceof decayed fiber and other woody material; less decay is present toward the upper part of the zone. In the lower parts the stratum is tan and contains much ash, indicating the possibility of a smoldering fire; however, some unburned and partly burned fiber is found within the zone. In the upper parts evidence of a gener­alized fire is not present but the fiber is very punky, suggestingdecay as a result of percolation of moisture upward from the rock base of the shelter. The only vegetal matter are mescal beans and hulls, and hulls of pecans, walnuts, and acorns. Lumps of charcoal of varying size are common. Flint chips are common but of smaller size than in Zone I. There is a slight increase in large animal bones such as deer, and a decrease in the relative frequency of rodent bones. Artifacts of fiber were found; a few badly decomposed sandals, whittled sticks, fragmentary portions of woven mats, and netting were present but not common. It is possible that rodent activity, the results of which were evident, lowered the wooden and fibrous artifacts into the zone. There is a noticeable increase in fire-fractured limestone. Lithic arti­facts increased in number. 
Lenses consisting of white ash occurred in Units 7 and 9 and a definite firepit was visible in the profile of Unit 14. This firepit, containing an extremely dark deposit of nearly purecharcoal, appears to have been a prepared, shallow, saucer-like pit 
Zone 111 (Fig. 5). Here, there is a definite increase in fiber and a decrease in moisture. The perishable material is slightly decomposed but less so than in Zone 11. Woody material does not reflect the degree of decomposition of the fibrous ma­terial. An increase in small roof spalls is noted. Fire-cracked hearth rocks are common throughout the deposits. Flint scrap is very common, much of it of small size. Quids, prickly pear leaves, sticks, mescal beans, pecans, walnuts, and acorns are present.Reeds and yucca stalks show evidence of slight decay. Knotted, split leaves of sotol, sacahuisti, and lechuguilla were observed. Lithic artifacts appeared throughout. Fragments of fibrous arti­facts and bone artifacts were present. Some scraps show an in­crease of rodents and birds and a decrease of deer bones. Some 
12 
snail shells appear. 
A long, relatively thin lens of white ash that occurred in Units 6, 9, and 10 probably resulted from a large fire. Fragments of charred sticks, and the butt portions of lechuguilla support this conclusion. The artifact content is similar to the principal part of Zone 111. A second lens of tan ash and fiber is confined to Unit 6 and underlies a part of the first lens noted. This area indicates greater heat intensity and probably accounts for the lack of fiber, but the lithic artifactual material is the same as the rest of Zone 111. 
Zone IV (Fig. 5). In Unit I three levels constitute Zone 
IV. The upper and lower strata consist of intense concentrations of fire-cracked rock and ash. These two strata have much the same appearance as burned rock middens of the Edwards Plateau to the east. It is estimated that 75 percent of these deposits consists of fire-fractured rock. Between the rock is fine sediment con­taining charcoal particles and ash. The soil of these two strata has a greasy feeling. Flint chips are not as prevalent as in Zone 
111. Bone scrap is very common; deer, rodents, and some fish re­mains are present. Partly burned twigs and fibrous material are noted. Many lithic artifacts are fire-fractured. The stratum be­tween the upper and lower strata is a tan deposit impregnated with ash, flint chips, some fiber, bone, and an occasional large snail shell and mussel shell fragments. The soil from this stratum is fine and contains a small amount of roof spall. In Units 7 and 8, Zone IV consists of two strata. The upper stratum is the same as the second stratum in Unit 1, and the lower stratum is the same as the upper and lower strata in Unit 1. In Units 6, 9, 10, and 14 the lowest stratum is characterized by fiber and grass; the next higher stratum is the tan, ashy stratum noted in Units 1, 7, and 8. A lens starts in the southern part of Unit 14 and continues through Unit 10 and just into Unit 9. This large lens consists of a tan ashy soil and fibrous material in some quantity. Two additional lenses, observed in the southern part of Unit 6, are similar in nature; they consist of a white ash and small frag­ments of carbonized fibrous material. The upper is the smaller of the two. Both are probably the result of localized fires. 
Lithic artifacts, fibrous artifacts, and bone artifacts were common in Zone IV. There is an increase in fish bones. Deer, rodent, and bird bones are common. An increase in flint 
chips was noted. 
13 
Zone V (Fig. 5). This is a zone containing much fiber. It consists of three strata present in all units. Quids are the most common evidence of human occupation. Throughout this zone and intermixed with the fibrous material is flint scrap and bone. The percentage of flint scrap and bone is small relative to the total content of the strata of the zone. There is a small amount of fire-cracked rock. Mesquite beans, mescal beans and hulls, leaves, pecans, walnuts, persimmon seeds, pods of sotol, sacahuisti, and lechuguilla are common. No single animal predominates; deer, fish, bird, and rodent bones are equally represented (see Part II for a detailed analysis of the faunal remains). Much of the bone ma­terial in the middle stratum is fish bones. Flint scrap and lithic artifacts are more numerous in the lower parts and a slight in­crease in fire-cracked rock occurs in the lower part of Zone V. 
The most common artifact was knotted fibers. Other arti­facts of woven fibers were frequently found. Lithic artifacts and bone artifacts were not as common as fiber artifacts. 
Three charcoal lenses were noted in Units 6, 8, and 10. At the rear of Unit 6 a massive layer of fiber and grass constituted another lens. 
Surface (Fig. 5). Overlying the deposits of the Baker Cave was a limestone dust deposit containing flint chips, a few artifacts, isolated charcoal lumps, and fiber. Roof spalls were observed throughout this stratum. Evidence of occupation was sparse. The surface deposits appeared to have been churned byanimal activity. Sheep and javelina frequently bedded down in the shelter, both disturbing the upper parts of the deposits.For these reasons the upper stratum does not appear to constitute a definitive occupational zone. 
FEATURES 
During excavations in Baker Cave many features were noted 
and recorded. Most of these were hearths of similar nature for which a generalized description will be given. Other features will be described in detail. No burials were found during the excavation, but one was later discovered by Jim Baker while exam­ining the excavations. Features were given arbitrary numbers and are not in the sequence found. 
The hearths (Fig. 6) were round or ovoid in outline, shallow,and bowl-like in cross section. They varied from 12.5 inches to 
14 
26.0 inches on the long axis and from 15.0 inches to 25.0 inches 
on the short axis. They averaged
5.5 inches in depth. All but one contained heavy concentra­tions of charcoal, and none was lined by rock. 
Feature No. I_. In Unit 2 and in the uppermost stratum in Zone V was a mass of grasswound about twigs of the mescal shrub (Sophora secundiflora) (Gould, 1962: 53). Measured in place, it was 4.08 inches thick, 
9.67 inches long, and 7.0 inches wide. Observed from above, it was ovoid in outline. It was composed of "sand drop seed grass" (Sporobolum cryptandrus)
(T. N. Campbell, personal commu­nication) The mescal twigs had
. 
been broken from the shrub and all the broken ends were placedtogether. There was a small rep­resentation (10 percent) of two 
-
Fig. 6. Hearth, stylized.
other plants. They were the western walnut (Juglans rupes-Charcoal filled fire pit (up­
per) ; cross section (lower).
tris) (Gould, 1962: 35) and 
sacahuisti (Nolina texana (ibid.:
31). 
.
Fea.ture No. 2_ (Fig. 7) This shed antler of a whitetail 
deer had been placed under a boulder and was found in the north­west wall of Unit 3 and in Zone IV. The curve of the antler was placed in such a way that it matched the outside contour of the boulder under which it had been placed. All the tines were broken off with no evidence of having been cut in any way. It is possible that the antler was placed under the boulder when the level on which it was found was the surface of the shelter, and was placed in such a way that it would not be underfoot. 
.
Feature No. _3 (Fig. 8) An unused pit and twelve prickly pear internodes were tied with strands of split sacahuisti. The pit is nearly round, being 25.0 inches on one axis and 26.0 inches on the other. The prickly pear internodes were neatly 
-
Fig. 7. Feature No. 2, top view of antler under rock (upper); cross section of antler under rock (lower). 
-
Fig. 8. Feature 3, tied prickly pear internodes and unused fire pit. 
stacked in three layers. On the long axis of the bundle two saca­huisti leaves were joined in a square knot and this composite
strand tied about the stacked internodes. A single sacahuisti leaf was used on the narrow axis. The internodes were thus pack­aged, much like a parcel prepared for mailing. The bundle was 
5.0 inches northwest of the unused pit and was on the bottom of Stratum 2 (Zone V). The pit extended through the sterile layerunderneath and nearly through the third stratum of this zone. It appears that the internodes were harvested and tied in preparationfor baking or roasting. Possibly a pit was dug for this purposebut apparently no fire was started in the pit and there is no evi­dence of roasting of the prickly pear internodes. 
Feature N0.4 (Fig. 9. Reconstructed from field notes). This consisted of 53 sotol or lechuguilla flower stalks driven into the floor. First found in Unit 6, a continuation of the "stakes" was later found in Units 11 and 13. Driven into the deposits in a gentle arc, the stakes were not on a true vertical but slanted to the inside of the curve. The tops of the stakes had been brok­en off. A few isolated stakes were found in Unit 12, but were of insufficient number to establish that they were a continuation of those found in Units 6, 11, and 13. Within the area enclosed bythe arc and along a line between Units 9 and 10 was a large roof spall. Only a small part of this roof spall was visible on the surface of the cave floor. On the north and west sides of the spall was a heavy deposit of human fecal matter. Grass had been laid over the first layer of coprolites; a second layer of copro­lites was covered with grass, and a third layer was on top of the previous two. It was noted in the course of excavation that the dust intermixed with the coprolites was concreted, the only such instance in the shelter. It is assumed that the stakes were used to form a base for a screen around the large roof spall, which possibly served as a back rest while the occupants of the shelter passed excreta. The concreted area could possibly be the result of passing urine. 
If the stakes constituted a base for a screen, it would be a natural assumption that some material such as twigs or grass were woven between the stakes, but such a feature was not found. This could be the result of contemporary or later occupants usingthe twigs or grass for other purposes after the toilet was no longer serving its original purpose. 
Feature No._s. A small, grass-lined pit measuring about 4.5 inches by 3.5 inches on its two axes and 2.9 inches deep, was com­pletely filled with small chips of flint and a broken leaf-shaped 
-
Fig. 9. Feature 4, relation of "stakes" to roof spall 
biface from the same nodule. This feature, its function unknown, was found in Unit 9, Zone 111. It is possible that the biface was "killed" and placed in the small, grass-lined pit along with 
the waste chips of its manufacture. 
Features No. 6_ and 7_. Two bowl-shaped, grass-lined pits were filled with fiber. The smaller had most of its outline in Unit 13 and the remainder in Unit 9. It is circular in outline with a diameter of 22.3 inches and a depth of 15.0 inches. The larger was in Unit 9. It is oval in outline with a long axis of 
21.0 inches and a short axis of 17.0 inches. It was 12.0 inches deep. The two features were only 18.0 inches apart and may have a relationship to each other due to their proximity and similarity. If they were intended to be hearths, there was no indication that they were used. The fiber fill may have been a secondary deposition. 
Feature No. 8,. The burial found by Jim Baker was described by Baker and John W. Greer, who furnished the author with a de­scription of it. The following is a composite of both accounts. The burial was found approximately two feet from the rear wall in Unit 12. The west wall of Unit 6 had slumped and revealed a large,folded twill-woven mat and two fragments of a checker-woven mat. Under the matting a whitish gray layer about six inches thick con­tained roof spalls, dust, and ash. About three feet deeper skull fragments and a clavicle were stacked together. The bones were about 2.4 feet below the surface. It appears that the burial was deposited in a used cooking pit, for the bottom of the pit con­tained 'a deposit of white ash. 
The twilled mat was tightly woven and was made from sotol or sacahuisti that had been split into quarter-inch wide strands. The checker-woven mat fragments were of sotol strands of approxi­mately 0.5 inch wide. 
The skull fragments were of an infant 1.5 to 2 years old. 
CHIPPED STONE TOOLS 
DART POINTS 
With the exception of the Paisano point (included in Ross's Rio Bravo Series, 1965: 27-47) and the Figueroa point (includedin the Nunley, Duffield and Jelks Rio Bravo Series, 1965: 32-50)the points found in Baker Cave include all of the point typesfound since this system was first proposed. Since the series of points from recently excavated sites of the Trans-Pecos are not firmly established, as witnessed by the inclusions and exclusions of the various reporters, no attempt will be made to group the Baker Cave projectile points into any of the series. 
Dart points were the third most numerous artifacts found in Baker Cave. A total of 396 complete or identifiable fragmentsof dart points was found, of which 355 are of known provenience.
These were divided into 45 separate types. Fragments of thin, bi-
facially worked artifacts are not tabulated since such statistics would be of no real value. An assignment to general cultural af­filiations will be presented at the end of the descriptive part
of this report. 
Almagre (5 specimens; Fig. 10, A-B) 
Description: Broad triangular blade. Blade edges convex 
or straight, one edge may be straighten than the other. Shoulders weak to strong, no barbs. Stems contract. Chipping crude. Length ranges from 6.7 cm. to 8.2 cm. Width varies from 4.3 cm. to 4.5 cm. 
Provenience: 1 Zone II; 2 zone 111, 2 Zone IV. 
Comment: One specimen was unfinished, giving an interestinginsight into the method of manufacture of this particu­lar specimen. The stem apparently was formed from the thin or opposite end of the percussion bulb. Percussion flaking was used to shape the point and pressure flaking was used to finish it. 
Baker 1_ (7 specimens; Fig. 10, C-D) 
Description: Blade straight to slightly convex. Shouldered to strongly barbed. Stems expand. Base bifurcated. Thick in cross section. Symmetrically made; good flak­ing. Maximum length 6.5 cm. Minimum length 4.6 cm. Width 2.6 cm. to 4.5 cm. 
Provenience: 3 Zone II; 2 Zone IV; 2 unknown. 
Baker II (9 specimens; Fig. 10, E-F) 
Description: Straight to convex blade. Most often shoul­dered from strong to weak, less commonly barbed. Barbed variety approaches Baker I in appearance. Stems con­vex, had "bowlegged" appearance. Notches in base more open, base of notch rounded. Thick lens in cross sec­tion. Fairly well made. Length 5.0 cm. to 6.5 cm.; width 2.4 cm. to 2.8 cm. 
Provenience: 4 Zone II; 1 Zone III; 3 Zone IV; 1 unknown. Comment: Appears to be earlier than Baker I. 
Bandy (6 specimens; Fig. 10, G-H) 
Description: Blade edges vary from gently convex to slight­ly recurved. Well barbed. Notches narrow or V shaped.Corner notching produces slightly expanding stem. Bases arc gently and convexly to center producing fishtail appearance. Length estimate 4.2 cm. to 5.0 cm. Width 
2.5 cm. to 3.7 cm. Thin cross section. Very well made. Provenience: 5 Zone II; 1 unknown. Comment: An early Archaic form. The smaller specimens ap­
proach arrow points in size. 
Castroville (3 specimens; Fig. 10, I-J) 
21 

-
Fig. 10. Dart points. Almagre, A-B; Baker I, C-D; Baker 11, E-F; Bandy, G-H; Castroville, I-J; Catan, K-L. 
Description: Large triangular blade, edges straight. Cor­ner notched. Prominent barbs. Stems wide and expanded. Bases straight to slightly convex. Well made by percus­sion, pressure rechipped. Length 4.9 cm. to 6.4 cm.; width 3.6 cm. to 4.2 cm. 
Provenience: 2 Zone IV; 1 Zone V. 
Catan (4 specimens; Fig. 10, K-L) 
Description: Triangular blade, straight to slightly convex. Base convex to rounded. Blades often beveled. Length from 3.6 cm. to 4.2 cm. Width from 1.67 cm. to 2.5 cm. Percussion flaked, retouched edges.
Provenience: 4 Zone V. 
Dari (3 specimens; Fig. 11, D) 
Description: Triangular, long, slender blade. Shouldered from slight to angular. Blade often beveled. Stems vary from straight to flared. Symmetrically made; aver­age flaking.
Provenience: 1 zone II; 2 unknown. 
Desmuke (1 specimen; Fig. 11, C) 
Description: Shoulderless, lozenge-shaped. Straight blade edges. Base contracts. Length unknown; width 2.5 cm. Percussion chipped. Slight retouch on edges.
Provenience: Unknown. 
Comment: The illustrated specimen does not have the pointed base but has the characteristic contraction from the lower part of the blade toward the base. It resembles points of this type I have personally examined from the Falcon Reservoir area. 
Early Barbed I_ (2 specimens: Fig. 11, D) 
Description: Convex, recurved blade forming slender point.Wide corner notching forms prominent barbs. Stem edges and base straight. Well made. Small for dart points. Only complete specimen, 3.8 cm. long and 3.0 cm. wide. 
Provenience: Zone 11. 
Comment: This point came from the lowest part of Zone 11. 
-
Fig. 11. Dart points. Dari, A, B; Desmuke, C; Early Barbed I, D; Early Barbed 11, E, F; Early Barbed 111, G; Early Barbed IV, H; Early Barbed V, I; Ensor I, J, K; Ensor 11, L-N; Fairland, 0. 
Early Barbed II (4 specimens; Fig. 11, E-F) 
Description; Straight to slightly convex beveled blades, wide corner notching forms shoulders or weak barbs. Stem short and slightly expanding. Bases deeply con­cave. Length 4.0 cm. to 8.0 cm.; width 1.5 cm. to 3.0 cm. Percussion flaked. 
Provenience: 1 Zone I; 2 Zone II; 1 Zone V. 
Early Barbed 111 (1 specimen; Fig. 11, G) 
Description; Wide, strongly convex blade, recurved near tip. Incipient barbs and expanding stem. Corner notched. Base slightly concave. Length 5.0 cm.; width 
3.5 cm. Well made. 
Provenience: Zone 11. 

Early Barbed IV (1 specimen; Fig. 11, H) 
Description; Asymmetrical blade. Short contracted stem. Base straight. Crudely percussion flaked. Length 4.4 cm.; width 2.3 cm. 
Provenience: Zone 11, 
Early Barbed V (1 specimen; Fig. 11, I) 
Description; Blade slightly convex, shoulders formed bybasal notching. Slightly expanded stem. Base thinned. Length unknown; width 2.5 cm. 
Provenience: Surface. Comment: This point is arbitrarily assigned to this classi­fication. It has the appearance of Johnson's (1964; 55) "Other Paleo-Indian Points" as illustrated in Fig. 17, L. 
Ensor 
a
The Ensor group presents difficult problem in taxonomy. 
Epstein (1963: 46) divided the group into three subdivisions. Johnson (1964: 34-35) also divided the Ensor into three subdi­visions but his subdivision combines two of Epstein's group. One of Johnson's Ensor specimens (Type III) appears to be more similar to the Frio type. Parsons (1965: 22-24) divided the group into two divisions. Epstein's basis for his division is based prin­cipally on the manner of notching. Johnson used the shape of the base for his taxonomy. Parsons based his division on blade shape. Ross (1965; 45), Nunley, Duffield, and Jelks (1965: 47-48), 
25 
and Kelly and Smith (1963: 172) made no subdivision. The Baker Cave specimens were divided into the groupings of Epstein, Johnson, Parsons, Ross, and Nunley, Duffield, and Jelks. The results were not satisfactory. The Ensor specimens were divided according to zones and subdivided again by strata within the zones. When this was done, no definite conclusions could be made; but a tenuous pattern appeared. On this basis the Ensor group was divided into two groups. 
Ensor I_ (8 specimens; Fig. 11, J-K) 
Description: Elongated, convex-bladed. Roughly triangular.Lateral notches shallow and confluent with base. Bases vary from slightly concave to slightly convex. Lengthsapproximately 2.5 times width. Lengths from 4.1 cm. to 
6.7 cm.; width from 1.8 cm. to 2.1 cm. Symmetrically
made. Well flaked. 
Provenience: 8 Zone V. 

Ensor II (30 specimens; Fig. 11, L-N) 
Description: Blades wider than Type I. Blades vary from straight to convex. Notches prominent. Stem corners vary from dull points, to slightly rounded to squared.Bases straight, slightly convex to slightly concave. Irregular edges. Poorly flaked, widest at base. Lengthapproximately twice width. Average length 5.6 cm.; average width 2.7 cm. 
Provenience: 10 Zone IV; 19 Zone V; 1 unknown. 
Fairland (1 specimen; Fig. 11, 0) 
Description: Triangular blade; shouldered, not barbed. Asymmetrical stem. Notched from side. Base nearly as wide as shoulders. Base concave. Estimated length 5.1 cm.; width 2.5 cm. 
Provenience: 1 Zone IV. 
Frio (3 specimens; Fig. 12, A) 
Description: Triangular blades, occasionally recurved. Cor­ner notched, strongly barbed. Stems expanded. Bases deeply concave. Length 6.2 to 7.1 cm.; width 2.2 cm. 
Provenience: 2 Zone V; 1 unknown. Comment: Two specimens are broken in such a way that the range in width cannot be determined. 
26 
-
Fig. 12. Dart points. Frio, A; Langtry I, B-D; Langtry 11,E-G; Langtry 111 (Val Verde), H-J: Langtry IV, K-M. 
Langtry 
Langtry points are the most common point type found in Baker Cave. This is another group that has presented taxonomic problems to reporters of archeological work in the Trans-Pecos area. Schuetz (1956: 141) was the first to subdivide this type.Epstein (1963: 36-40) refined the division into six types with alphabetical designations. Kelly and Smith (1963; 170) divided the group into five variations, and Kelly (1963: 205-207) also divided this type into five variations. Johnson (1964; 38-39)makes three divisions and a fourth catch-all division. Graham and Davis (1958: 16-18) recognized five subdivisions of the 
Langtry type in their survey of the area. Ross (1965; 35-39) attacked the Langtry problem with five divisions of the type.Nunley, Duffield, and Jelks (1965; 35-39) list three types of Langtry and set the Val Verde point apart from the Langtry group as did Schuetz (ibid.). Parsons (1965: 31-33) recovered only six specimens in his survey of the Fate Bell Shelter and does not at­tempt to subdivide such a small representation, but he does sepa­rate the Val Verde from the Langtry group. 
Nunley, Duffield, and Jelks (1965: 36) suggest that the Val Verde point is a separate type as substantiated by stratigraphic evidence. It is stated that in the Coontail Spin Site the Val Verde is "comparatively early in comparison to Langtry and Alma­gre." This condition does not prevail at Baker Cave. In fact there is an indication that a variation (ray Langtry I) is of equal age or older. Since the Val Verde occurs in Zones 111, IV, and V, as do all of the Langtry variations except one (LangtryIV), this author can find no reason to separate the Val Verde from the Langtry, and it is included within the group. 
Langtry I_ (53 specimens; Fig. 12, B-D) 
Description: Blade straight to slightly concave to slightly convex. Shouldered, not barbed. Stem short, wide, and contracting. Base straight or slightly concave. Widest at shoulders. Length 4.0 cm. to 7.8 cm.; width 2.7 cm. to 4.4 cm. Thin, well made. 
Provenience: 1 Zone II; 23 Zone III; 17 Zone IV; 7 Zone V; 5 unknown. 
Langtry II (43 specimens; Fig. 12, E-G) 
Description: Blades triangular, straight to slightly convex to slightly concave. Pronounced shoulders, occasionally 
incipiently barbed. Stems straight and contracting. Bases concave. Length 4.9 cm. to 8.8 cm.; width 3.3 cm. to 5.1 cm. Exceptionally well made and thin. 
Provenience: 1 Zone 11, 2 Zone III; 22 Zone IV; 17 Zone V; 1 unknown. 
Langtry 111 (27 specimens; Fig. 12, H-J) 
Description: Blades straight or concave and shouldered. Stems flared, beveled stems usually on right side. Bases concave and generally beveled on one face only. Well made. Length 4.4 cm. to 5.7 cm.; width 2.0 cm. to 2.4 cm. 
Provenience; 8 Zone III; 11 Zone IV; 6 Zone V; 2 unknown. Comment; Type 111 Langtry has been named Val Verde point. 
Langtry IV (12 specimens: Fig. 12, K-M) 
Description: Slightly convex to concave blades. Blades long for width. Most often shouldered, occasionallywell barbed. Stems long, tapered. Bases sharplyrounded or pointed. Length 5.0 cm. to 6.5 cm.; width 
2.0 cm. to 3.5 cm. 
Provenience; 7 Zone IV; 2 Zone V; 3 unknown. 

Langtry V (18 specimens; Fig. 13, A-C) 
Description: Blades straight to slightly concave. Shoul­ders slight to prominent. Stems flair, not beveled. Bases generally concave, occasionally straight. Length
4.0 cm. to 8.7 cm.; width 2.0 cm. to 3.2 cm. Finelymade. Similar in outline to Langtry 111. 
Provenience: 	2 Zone II; 4 Zone III; 9 Zone IV; 1 Zone V; 2 unknown. 
Langtry VI (15 specimens; Fig. 13, D-F) 
Description: Blades convex to concave. Shoulders incipient to pronounced, not barbed. Stems generally short, stubbyBases straight, some rounded. Generally poorly made and asymmetrical. Length 3.3 cm. to 6.2 cm.; width 2.0 cm. to 3.3 cm. 
Provenience: 5 Zone III; 8 Zone IV; 1 Zone V; 1 unknown. Comment: This is a catch-all group. 
Marcos (1 specimen: Fig. 13, G) 
-
Fig. 13. Dart points. Langtry V, A-C; Langtry VI, D-F; Marcos, G; Marshall, H, I; Matamoros, J; Montell, K-M. 30 
Description; Blade convex. Expanded stem and barbs formed by corner notching. Length 5.9 cm.; width 2.7 cm. Provenience; Zone 111. 
Marshall (9 specimens; Fig. 13, H-I) 
Description: Blades vary from triangular to strongly con­vex. Widest above barbs in convex specimens. Strongly barbed. Stems vary from straight to slightly expanded. Base convex. Percussion chipped, smaller specimens better made. Larger specimens could be knives. Lengthindeterminable; width 3.0 cm. to 5.0 cm. 
Provenience: 2 Zone III; 7 Zone IV. 
Comment: Larger specimens may have been used as knives. Semenov (1964: 93-94) has demonstrated by microscopic study of wear patterns that some upper Paleolithic shouldered lithic forms considered to be points were used as knives. 
Matamoros (2 specimens; Fig. 13, J) 
Description: Triangular blade, sometimes serrated. Bases convex and often thinned. Length unknown; width 2.3 cm. to 2.7 cm. 
Provenience: 2 Zone V. 
Montell (9 specimens; Fig. 13, K-M) 
Description: Triangular to slightly convex blades. General ly barbed. Stems most often expand but may contract. Basal notch varies from V-shaped to U-shaped. Well flaked and thin. Length 5.9 to 8.6 cm.; width 3.0
cm. 
cm, to 4.6 cm. 
Provenience: 1 Zone IV; 8 Zone V. 

Nolan (7 specimens; Fig. 14, A-B) 
Description: Triangular convex blade often recurved. Shoul ders usually strong. Stems strongly beveled usuallyparallel-sided, sometimes expand or contract slightly. Bases generally straight, can be convex. Percussion chipped, retouched on blade edges. Well made. Length
6.9 cm. to 10.5 cm.; width 2.3 cm. to 4.0 cm. Provenience: 2 Zone III; 1 Zone IV; 3 Zone V; 1 unknown. 
Palmillas (1 specimen; Fig. 14, C) 
-
Fig. 14. Dart points. Nolan, A, B; Palmillas, C; Pandale, D-F;
Pandora, G, H; Pedernales, I, J; Plainview Golondrina, K-M; Plain­view (?), N; Shumla, 0, P. 
Description; Strongly convex blade edge produces blunt point. Weak barbs formed by corner notching. Stem ex­pands slightly. Base rounded. Poorly made by percus­sion. Length 6.5 cm.; width 3.3 cm. 
Provenience; Zone V. 
Pandale (37 specimens; Fig. 14, D-F) 
Description; Long, slender leaf-shaped blade. Blade con­vex. Often recurved. Blade heavily beveled on left side producing a twisted appearance. Shoulders absent or poorly developed. Stems beveled on right side and expand. Bases strongly convex. Oblique flaking charac­teristic. Length 4.3 cm. to 6.1 cm.; width 2.0 cm. to 
2.6 cm. Provenience: 3 Zone II; 20 Zone III; 6 Zone IV; 4 Zone V; 4 unknown. 
Comment; 	It has been stated by Epstein (1963; 48), John­son (1964: 40), Nunley, Duffield, and Jelks (1965; 33),Parsons (1965; 37), and Ross (1965; 29-30) that the blades of pandale points are beveled to the degree that they have a twisted appearance. It would seem that the term "twisted" would be more appropriate than "beveled." This may appear to be semantic nit picking, but it is intended to be consistent with the manner in which the word "beveling" is normally used in archeological con­text. Beveling is accomplished by flaking along the same edge of an object. Such action produces a promi­nent steepness along the edge. This edge is frequentlyreferred to as "beveled", and if done on alternate edgesproduces a flattened parallelogram in cross section. The Pandale blade is not beveled in this context. On the "twisted" biface the chipping extends all the way to the center of the blade. This was done on each side of the blade and as the blade narrows toward the distal part,the edge of the blade veers slightly to the left (on a blade with a left twist). In the beveled blade this variation to the left is considerably more pronounced.In cross section the "twisted" blade is lenticular. 
Pandora (4 specimens; Fig. 14, G-H) 
Description: Triangular to leaf-shaped. Blade generally convex. Bases generally straight. Length 5.1 cm. to 
7.9 cm.; width 2.8 cm. to 3.4 cm. Well made points.Provenience; 1 Zone III; 2 Zone V; 1 unknown. 
33 
Pedernales (7 specimens; Fig. 14, I-J) 
Description: Blade straight to slightly convex. Shoulders vary from prominent to weak. Sometimes barbed. Widely corner notched. Stems usually parallel-sided. Bases indented. Often confused with Baker point. Estimated length 6.0 cm. to 6.6 cm.; width 2.5 cm, to 2.8 cm. 
Provenience: 1 Zone II; 2 Zone III; 1 Zone IV; 2 Zone V; 1 unknown. 
Plainview golondrina (6 specimens; Fig. 14, K-M) 
Description; Leaf-shaped, edges expand midway of blade. Basal edges out-flaring. Basal concavity deeper; cruder than classic Plainview. Basal and lateral smoothing.Estimated length 6.0 cm. to 7.0 cm.; width 2.0 cm. to 
2.5 cm. Provenience: 5 Zone I; 1 Zone II Comment: Two charcoal samples near the provenience of one 
of the specimens produced dates of 8910-B.P. and 9030-B.P. (Pearson, Davis, Tamers, and Johnstone, 1965: 305). A single sample from Bonfire Cave (ibid.: 304)produced a date of 10,230-B.P. There seems to be a difference in time of approximately 1,200 years between the Baker Cave golondrina Plainview and those from Bon­fire Cave. Perhaps the Plainview point persisted long­er in time in the area than was previously thought. 
Plainview-like (1 specimen: Fig. 14, N) 
Description: Leaf-shaped. Widest one-fourth up from base. Beveled on left side. Constricted below widest partthen expands to form slight ears. Base concave. Edgesand base not smoothed. Length 6.0 cm.; width 1.9 cm. 
Provenience: Zone I. 
Shumla (3 specimens; Fig. 14, 0-P) 
Description: Triangular, straight-edged blades. Prominent­ly barbed. Notched from corner. Stems parallel-sidedand rectangular. Bases convex. Thin, well made. Length
5.2 cm. to 5.9 cm.; width 2.6 cm. to 3.3 cm. Provenience: 3 Zone V. 
Tortugas (13 specimens; Fig. 15, A-C) 
-
Fig. 15. Dart points. Tortugas, A-C; Travis, D, E; Triangular Concave, F, G; Triangular Convex, H; Uvalde, I, J; Zorra, L; Mis­cellaneous I, M; Miscellaneous 11, N; Miscellaneous 111, 0; Mis­cellaneous IV, P. Arrow points. Fresno, Q, R; Harrell, S; Perdiz, T, U; Scallorn, Coryell, V, W; Toyah, X; Miscellaneous, Y. 
Description: Convex, triangular, beveled or unbeveled blades. Bases straight to slightly convex. Four have gum or resin on base. Thin, very well made. Length
3.2 cm. to 5.0 cm.; width 2.5 cm. to 3.4 cm. Provenience; 1 Zone II; 6 Zone III; 1 Zone IV; 5 unknown. 
Comment; 	Crudely made points of similar outline were found in Baker Cave. These have been arbitrarily assigned to a knife type associated with the Ensor-Frio point types. 
Travis (8 specimens; Fig. 15, D-E) 
Description: Convex blades. Shoulders vary from rounded to fairly prominent. Stems straight to slightly con­tracted. Bases straight to slightly concave. Percus­sion chipped, minimum retouch on edges. Length 6.0 to 8.8 cm.; width 1.9 cm. to 3.9 cm. 
Provenience: 1 Zone II; 2 Zone III; 2 Zone IV; 2 Zone V; 1 unknown. 
Triangular Concave (17 specimens; Fig. 15, F-G) 
Description: Subtriangular concave blades and bases. Bases unthinned. Percussion flaked, no secondary flaking in most specimens. Length indeterminable; width 3.2 cm. to 4.7 cm. 
Provenience; 8 Zone III; 4 Zone IV; 2 Zone V; 3 unknown. 
Triangular Convex (1 specimen; Fig. 15, H) 
Description: Blunt tip, blade slightly convex. Rounded base. Similar to Refugio point (Suhm, Krieger, and Jelks, 1954: 474) but shorter. Percussion flaked, no secondary flaking, no smoothing of lateral edges.Length 4.5 cm; width 1.7 cm. 
Provenience; Zone I. 
Uvalde (4 specimens; Fig. 15, I-J) 
Description: Triangular to leaf-shaped blade, edgesstraight to convex. Shouldered to barbed. Stem ex­pands strongly, sometimes as wide as shoulders. Bases deeply U-shaped. One, Fig. 15, K, aberrant form with only one barb. Length 4.0 cm. to 5.7 cm.; width 2.3 
cm. to 3.2 cm. 
Provenience; 4 Zone V. 

Zorra (3 specimens; Fig. 15, L) 
Description: Similar in outline to pandale point. Not 
obliquely flaked and lacks twisted appearance. Stems beveled on right side, two specimens, left side one specimen, one specimen beveled on right side of blade 
Average length 6.5 cm.; width 1.85 cm. to 2.1 cm. Provenience: 1 Zone II; 1 Zone III; 1 unknown. 
Miscellaneous l_ (1 specimen; Fig. 15, M) 
Description: Subtriangular, one edge straight, other con­vex. Incipient side notches form stem nearly as wide as blade. Base slightly convex. Beveled distal half of one blade edge, possibly from resharpening. Length
4.9 cm.; width 2.2 cm. 
Provenience: Zone V. 

Miscellaneous II (1 specimen; Fig. 15, N) 
Description: Subtriangular blade, one edge straight, other convex. Wide corner notching, prominent shoulders. Contracting stem. Base narrow, straight. Percussion chipped, pressure retouched on edges and shoulders. Length 4.2 cm.; width 3.1 cm. 
Provenience: Zone IV. 
Miscellaneous 111 (1 specimen; Fig. 15, 0) 
Description: Small, thin point. Convex blade, one edgeslightly recurved near tip. Wide corner notching forms prominent shoulders. Slightly expanded stem. Base rounded laterally. Deeply concave in center. Finelymade. Length 3.0 cm.; width 2.1 cm. 
Provenience: Zone V. 
Miscellaneous IV (1 specimen; Fig. 15, P) 
Description: Small triangular point, straight blade edges recurve near tip;forms fine, sharp point. Notchingfrom side forms prominent shoulders and expanded stem. Corners of base rounded. Base slightly concave.* Length 3.6 cm.; width 2.5 cm. 
Provenience: Zone V. 
37 
ARROW POINTS 
Only 13 arrow points were found in Baker Cave. Two are of unknown provenience. Stratigraphically, ten were found in the uppermost part of Zone V and the surface (two in Zone V and eight on the surface). One was found in Zone IV. Points typical of Zone IV in Baker Cave are Ensor and late Langtry point types.The recovery of a single arrow point specimen suggests that it was displaced by animal or human activities and is out of place. 
Fresno (3 specimens; Fig. 15, Q-R) 
Description: Triangular. Blade edges vary from straightand serrated to convex and unserrated to recurved. Bases slightly convex to concave to straight. One face unmodified most consistent feature. Length 2.0 cm. to 3.15 cm.; width 1.1 cm. to 2,0 cm. 
Provenience: 1 Zone V; 1 surface, 1 unknown. 
Harrell (1 specimen; Fig. 15, S) 
Description: Triangular, side notched. Notches shallow,U-shaped. Blade edges round below notches forming
rounded base. Point concavo-convex. convex side, unaltered concave side. width 1.7 cm.  Finely chippedLength 2.7 cm. ;  
Provenience:  Surface.  
Comment:  Convex base not typical of type  
Perdiz  (2 specimens; Fig. 15, T-U)  

Description: Considerable difference between two. Both widest at barbs. Smaller has triangular, slightly convex 
blade corner notched and moderately barbed. Longer specimen has triangular, finely serrated blade. Single remaining barb exaggerated, delicate, and sharplypointed. Stems contract. Base rounded. 
Provenience: 1 surface; 1 unknown. 
Comment: The perdiz, according to Nunley, Duffield, and Jelks (1965: 30) is the diagnostic point type of the Trans-Pecos region and is frequently found both 
on 
terrace sites and in shelters. 
Scallorn (2 specimens; Fig. 15/ V-W) 
Description: Eddy variant (Jelks, 1962; 28-30). Blade 
38 
triangular, concave, finely serrated. Barbed. Stems expand, bases convex. Length 2.0 cm. to 2.2 cm.; width 
1.5 cm. to 1.7 cm. 
Provenience: 1 Zone IV; 1 surface. 

Toyah (1 specimen; Fig. 15, X) 
Description: Strongly convex blade minutely serrated. U-shaped side notches. Concave base with central notch. Length 2.1 cm.; width 1.5 cm. 
Provenience; Surface. 
Miscellaneous I_ (1 specimen; Fig. 15, Y) 
Description: Large for arrow point. Blade triangular,slightly concave, serrated. Barbs extend nearly to base. Barbs rounded. Wide corner notching. Stem parallel-sided and straight. Base rounded. Length
3.6 cm.; width 3.0 cm. 
Provenience: Zone V. 

KNIVES 
Knives are defined herein as bifacially flaked implements,generally well formed. The edges are usually convex, althoughsometimes one side is straight. All are chipped by percussion and the better made specimens exhibit pressure flaking. The taxonom­ic designation of this class as knives is arbitrary and is not meant to imply a function, but it is traditional in literature of the area. 
Only complete specimens or fragments of sufficient size to give an obvious indication of the original shape were considered. One hundred sixty-eight specimens, of which 153 are of known pro­venience, have been divided into 14 types. These were sorted by provenience and by general form. 
Type I_ (22 specimens; Fig. 16, A-C) 
Description: Lanceolate outline. Distal end pointed. Base rounded. Striking platform present in some specimens.Viewed from edge, the knife edge is straight or nearly so. Percussion-chipped and secondarily pressure-flaked.Nicely made. Length 6.3 cm. to 13.6 cm.; width 3.6 cm. to 5.0 cm. 
Provenience: 6 Zone II; 8 Zone III; 4 Zone IV; 2 Zone V; 2 unknown. 
-
Fig. 16. Knives. Type I, A-C; Type 11, D-F. 
40 
Type II (18 specimens; Fig. 16, D-F) 
Description: Strongly convex blade edges and bases. Distal part rounded. Widest just above base. Viewed from edge, knife edge sinuous. Percussion chipped, no pressureflaking. Length 6.2 cm. to 9.9 cm.; width 3.5 cm. to 
6.2 	cm. 
Provenience: 	1 Zone I; 2 Zone II; 2 Zone III; 2 Zone IV; 7 Zone V; 1 surface; 5 unknown. 
Type 111 (8 specimens; Fig. 17, A-B) 
Description: Leaf-shaped. Edges very sinuous. Bases varyfrom semi-pointed to rounded. Percussion chipped; secondary chipping on some specimens near base. Length
5.8 cm. to 10.5 cm.; width 4.3 cm. to 5.7 cm. Provenience: 1 Zone II; 5 Zone IV; 1 Zone V; 1 unknown. Comment: This form is reminiscent of pre-form. 
Type IV (17 specimens; Fig. 17, C-E) 
Description: Convex blade edges taper to dull point. Bases convex. Symmetrical and thin. Broad, thin flake scars suggest billet flaking. Secondary pressure flakingalong edges. Length 6.0 cm. to 9.7 cm.; width 3.4 cm. to 5.8 cm. 
Provenience: 2 Zone II; 4 Zone IV; 10 Zone V 
Type V (23 specimens; Fig. 17, F-G) 
Description: Triangular outline. Blade edges vary from slightly convex to strongly convex. Bases convex. Per­cussion chipped; secondary retouching. Thin blades. Length 5.4 cm. to 8.2 cm.; width 2.5 cm. to 3.7 cm. 
Provenience: 1 Zone I; 1 Zone II; 1 Zone IV; 19 Zone V; 1 unknown. 
Comment: Resembles Abasolo dart point, but generally too large and more crudely made. Arbitrarily assigned to knife class. 
Type VI (23 specimens; Fig. 18, A-C) 
Description: Lanceolate in outline. Distal part sharp.Blade convex. Bases straight to convex. Percussion flaked, well made. Secondarily chipped on sides and base. Striking platform and cortex on some specimens. 
Fig. 17. Knives. Type 111, A, B; Type IV, C-E; Type V, F, G. 
Fig.  18.  - Knives.  Type VI, A-C;  Type VII, D,  E;  Type VIII,  
F,  G;  Type IX,  H.  
43  

Length 6.1 cm. to 10.2 cm.; width 2.0 cm. to 2.3 cm. Fragments suggest greater widths. 
Provenience: 2 Zone I; 4 Zone II; 9 Zone III; 4 Zone IV; 3 Zone V; 1 unknown. 
Comment: Appear associated with Pandale points. 
Type VII (6 specimens; Fig. 18, D-E) 
Description: Small lanceolate,asymmetrical, crude, thick in cross section. Bases rounded to convex to straight.Percussion chipped, retouched on blade sides. Edgessinuous. Distal part pointed. Evenly spaced chippingfrom alternate sides produced sinuous edge. Occasion­
ally to well-defined point. straight and thinned. 
ally recurved on distal part. cm.; width 2.2 cm. to 2.7 cm.  Length 5.0  cm.  to 6.2  
Provenience:  1 Zone  IV;  2  Zone V;  3  unknown.  
Type VIII (6 specimens; Fig. 18, F-G)  
Description:  Blade edges parallel near base.  Tapers gradu- 

Bases Thinness and width of flake scars suggest billet flak­ing technique. Excellently made. Length 7.4 cm. to 
8.5 cm.; width 3.4 cm. to 5.4 cm. Longer specimens 
seen by author from area. Provenience: 1 Zone III; 3 Zone IV; 2 unknown. 
Type IX (3 specimens; Fig. 18, H) 
Description: Asymmetrical, triangular form. One blade edgemore convex than other. Bases concave. Percussion flaked, edges pressure-retouched. Single measurable specimen, length 5.4 cm.; width 4.0 cm. 
Provenience: 1 Zone IV; 1 Zone V; 1 surface. 
Type X (6 specimens; Fig. 19, A) 
Description: Asymmetrical, ovate shape. Irregular edges convex, blunt point. Bases vary from straight to slightly convex. Crudely percussion flaked, pressureretouched. Thigh cross section. Length 5.7 cm. to 
7.8 cm.; width 4.0 cm. to 4.6 cm. Provenience: 2 Zone IV; 4 Zone V. 
Type XI 
-
Fig. 19. Knives. Type X, A; Type Xla, B; Type Xlb, C; Type Xlc, D-E; Type XII, F-G; Type XIII, H-K. 
45 
Four specimens of tanged knives found at Baker Cave were divided into three sub-types. 
Type XIA (1 specimen; Fig. 19, B) 
Description: Base corner tang. Blade back straight; cut­
ting edge convex. Blade heel rounded, heel notch obliquely flaked from proximal part of blade. Tang contracts more rapidly on cutting side. Finely made, secondary chipping more prominent on cutting edge. Thin lens in cross section. Length unknown; width 3.6 cm. 
Provenience: zone V. 
Type XIB (1 specimen; Fig. 19, C) 
Description: Back irregular. Strongly convex cutting edge meets back to form pointed blade. Weakly shouldered. Tang contracts slightly,base convex and thick. Percus­sion chipped. Occasional rechipping along cutting edge.Length 7.1 cm.; width 3.0 cm. 
Provenience: Zone V. 
Type XIC (2 specimens; Fig. 19, D-E) 
Description: Back irregularly straight. Blade edge straight to convex. Tangs formed by reducing blade on one edgeforming short, stubby tang. Tangs thinned on each side. Percussion-chipped and pressure-chipped on cutting edgeand back. Length 7.1 cm. to 8.1 cm.; width 3.1 cm. to 
3.7 cm. 
Provenience: 1 Zone I; 1 Zone IV. 

Type XII (2 specimens; Fig. 19, F-G) 
Description: Ovate. Made from thin flakes. Flat flake scars suggest billet flaked. Fine pressure chippingalong edges. One edge more curved and beveled. Length
7.6 cm. to 9.5 cm.; width 3.6 cm. to 5.3 cm. 
Provenience: 1 Zone V; 1 unknown. 

Type XIII (26 specimens; Fig. 19, H-K) 
Description: Small leaf-shaped, convex pointed blade. Straight base. Crudely percussion-flaked. Length
4.3 cm. to 6.6 cm.; width 2.7 cm. to 3.4 cm. Provenience: 2 Zone II; 5 Zone IV; 13 Zone V; 6 unknown. 
Comment; Arbitrarily classed as knife. Previously typed as point by Epstein (1963: 41 and 45), Johnson (1964;44-45); typed as knife form by Parsons (1965: 41-42),Nunley, Duffield, and Jelks (1965: 68), Ross (1965: 69) Seems to have a definite chronological relationship with Ensor point. 
Type XIV (4 specimens; Fig. 20, A-B) 
Description; Leaf-shaped. Sinuous edges. Crudely percus­
sion chipped. Secondary chipping along edges.8.3 cm. to 11.5 cm.; width 4.6 cm. to 5.9 cm. Provenience; 2 Zone IV; 2 Zone V.  Length  
Comment; Thickness suggests primary use as source ial for other artifacts. Possibly preforms.  of mater- 
DRILLS  

Only five drills or perforators were found in Baker Cave and these have been divided into two general classifications. 
Type I_ (2 specimens; Fig. 20, C-D) 
Description; Strongly convex bases, nearly semi-circular. Stem elongated, parallel-sided. Stem reduced to form drill shank. Shank parellogram or lens in cross section Edges of shanks worn. Drill point rounded. Length 8.9 cm.; width 2.0 cm. to 2.3 cm. 
Provenience: 1 Zone II; 1 Zone V. 
Type II (1 specimen; Fig. 20, E) 
Description; Made from Type IV Langtry. Distal part of blade reduced to form small, delicate point. Lengthestimated 7.8 cm.; width 4.9 cm. 
Provenience: Zone 111 
Type 111 (2 specimens; Fig. 20, F-G) 
Description; Formed on thin, unmodified flakes. Shank thin, wide, and gradually constricted. Reminiscent of exaggerated graver points. Possibly used as per­forators 
. 
Provenience: 1 Zone III; 1 surface. 
47 
-
Fig. 20. Knives. Type XIV, A, B. Drills. Type I, C, D; Type 11, E; Type 111, F, G. Choppers, H, I. 
48 
CHOPPERS (30 specimens; Fig. 20, H-I) 
Description: Massive core tools generally amorphous in shape. Battered bit-like sinuous working edge. Flat­tened surface opposite working edge. Possibly by-productresulting from flake production. Ideal tool for severingjoints of game or splitting and fracturing bone. Length
6.3 cm. to 15.3 cm.; width 4.7 cm. to 11.0 cm. 
Provenience: 	2 Zone I; 3 Zone II; 10 Zone III; 2 Zone IV; 7 Zone V; 1 surface; 5 unknown. 
CHISEL-CHOPPERS (8 specimens; Fig. 21, A) 
Description: Made from massive triangular and rectangularthick, heavy flakes. Straight bits with secondarypreparation. Bits vary from pointed to straight. Flat surface opposite bit not as common as in choppers. Length 7.9 cm. to 10.1 cm.; width 4.4 cm. to 6.7 cm. 
Provenience: 1 Zone III; 4 Zone V; 3 unknown. 
HAMMERSTONES (5 specimens; Fig. 21, B) 
Description: Made from heavy flint flakes or bean-shapedriver cobbles. Battered edges from repeated use. One battered convex edge on flint specimens. Two ends battered on river cobbles. Length 7.5 cm. to 10.5 cm.; width 5.5 cm. to 7.7 cm. 
Provenience: 1 Zone II; 1 Zone III; 3 Zone V. 
FIST AXES (10 specimens; Fig. 21, C) 
Description: Triangular outline. Rounded cutting edge at apex of triangle. Edges well flaked by percussion.Longitudinal cross section wedge-shaped. Lateral edgesslightly sinuous. Ideally fitted for heavy butcheringtools or chopping tree limbs. Only one complete speci­men. Length 8.6 cm.; width 6.4 cm. 
Provenience: 1 Zone II; 4 Zone IV; 5 Zone V. 
CORES (25 specimens; Fig. 21, D) 
Description: Amorphous shape from tabular flint. Spallsdetached by heavy haphazard percussion blows. Smooth cortical surface eliminates prepared platforms in most instances. Length 6.9 cm. to 9.3 cm.; width 5.6 cm. to 
8.0 	cm. 
Fig. 21.  - Chisel choppers,  A.  Hammerstones,  B.  Fist axe,  C.  
Core,  D.  
50  

Provenience: 1 Zone III; 9 Zone IV; 11 Zone V; 4 unknown. 
SCRAPERS 
After examination of the 490 artifacts classed as scrapers,it was decided to divide this group into five types. For taxon­omic purposes scrapers are described as flake tools with one or more bit-like working edges. These bit-like edges are the result of secondary and repeated pressure flaking. There does not appearto be any particular type or shape of flake used. With five ex­ceptions, the flakes used as scraping tools appear to have been unaltered except for the flaking of the bit-like edge. 
Type I_ (244 specimens; Fig. 22, A-B) 
Description: Single worked edge on end or side of flake. Made on initial cortical, cortical, and sequentialflakes. Length 3.2 cm. to 10.2 cm.; 2.6 to 6.4 cm. 
Provenience: 3 Zone I; 24 Zone II; 12 Zone III; 40 Zone IV; 124 Zone V; 10 surface; 31 unknown. 
Type II (172 specimens; Fig. 22, C-D) 
Description: Two worked edges most often on opposing sides. Also on contiguous edges. Most consistent feature is thinness. No flake selection except thin edge. Length
3.0 cm. to 11.1 cm.; width 3.1 cm. to 5.5 cm. 
Provenience: 	4 Zone I; 15 Zone II; 11 Zone III; 33 Zone IV; 84 Zone V; 8 surface; 17 unknown. 
Type 111 (44 specimens; Fig. 22, E-F) 
Description: Three worked edges. Oval or rectangularflakes preferred. Generally smaller than Types I or 
11. 	Cortical surface less common. Length 3.3 cm. to 
7.6 	cm.; width 3.2 cm. to 6.4 cm. 
Provenience: 	1 Zone I; 9 Zone II; 4 Zone III; 7 Zone IV 17 Zone V; 2 surface; 4 unknown. 
Type IV (25 specimens; Fig. 22, G-H) 
Description: All edges worked. Most commonly round or oval in shape, few rectangular. Generally only incipient remnant cortex remains. Exception, thin, waterworn wafer-like nodule with all edges trimmed. Length 3.4 cm. to 7.4 cm.; width 3.4 cm. to 6.6 cm. 
-
Fig. 23. Utilized flakes-Type I, A, B; Type 11, C, D; Type111, E, F; Type IV, G, H. Spokeshaves: Type I, I, J; Type 11, K, L. 
width 2.2 cm. to 4.6 cm. Provenience: 2 Zone I; 9 Zone II; 7 Zone III; 14 Zone IV; 32 Zone V; 6 surface; 11 unknown. 
Type IV (10 specimens; Fig. 23, G-H) 
Description: All sides used. Length 4.0 cm. to 5.9 cm.; width 2.5 cm. to 6.2 cm. Provenience: 3 Zone IV; 5 Zone V; 2 surface. 
SPOKESHAVES 
In keeping with custom, 37 unifacially-flaked artifacts hav­ing concave scraping surfaces are called spokeshaves; this term is not intended to imply a manner of usage. It is assumed that 
-
they were made and used for specialized purposes as a spoke­shave, and perhaps in the fleshing of hides. This type of arti­fact would be particularly useful in working parts of a hide that are difficult to flesh with a straight-sided or slightly convex scraping surface. By placing a hide over a rounded surface the concave scraper could be used advantageously. 
Spokeshaves or concave scrapers are divided into two groups,based on the configuration of the concave scraping surface. 
Type I (15 specimens; Fig. 23, I-J) 
Description; Concave flaked notch, occurring on scrapers,but more frequently on flakes otherwise unaltered. Notch deep and narrow. Two notches commonly found on same flake. Width of notch 0.8 to 1.8 cm.; depth 0.4 to 0.6 cm. 
Provenience; 1 Zone I; 2 Zone II; 1 Zone III; 2 Zone IV 9 Zone V; 1 unknown. 
Type II (21 specimens; Fig. 23, K-L) 
Description: Notch more open, concavity less prominent. Depth of notch in some instances barely discernible. Occur on scrapers and otherwise unmodified flakes. Working part has more tapering shouldered notch. Notch width 1.4 cm. to 3.2 cm.; depth 0.3 cm. to 0.5 cm. 
Provenience: 3 Zone II; 3 Zone III; 3 Zone IV; 10 Zone V; 2 unknown. 
SERRATED FLAKES 
(4 specimens; Fig. 24, A-C)
Description; Two made from thin flakes, flaked from one 

direction to form serrations on long axis. Serrations alternately chipped from opposite sides on third speci­men. Fourth specimen made from fragment of bifaciallyflaked blade. Possibly for specialized use such as cutting wood. Length 4.8 cm. to 5.6 cm.; width 2.3 cm. 
to 3.1 cm. Provenience; 1 Zone II; 1 Zone III; 1 Zone V; 1 unknown. 
BURINS 
A group of 34 artifacts is classed as burins or tools giv­ing evidence of having been used in a burin-like function. Ep­stein (1963; 70-72) describes burins as "pointed engraving tools 
produced by removing one or more spalls from the long or wide axis of the parent piece" which is the classic burin or the bee de flute. Johnson (1964; 67) elaborates somewhat; "... burins are thin flakes or bifacial artifacts which have had two or more long, narrow flakes knocked from their edges (as opposed to their faces), the scars of which come together to form a small bit that was seemingly used for graving or incising. These burin flake scars are generally characterized by a small concavity (negativebulb of percussion) at the point of impact, and sometimes byhinge fractures at the opposite end of the scar." 
Examination of pointed artifacts that were burins or burin­like, revealed that there were five true burins (bee de flute)and 29 artifacts with various burin-like characteristics. These are grouped with the burin but are divided into taxa according
to the fashion in which they were made. This group of 29 arti­facts and the true burin have been divided into four categories. 
Burins (5 specimens; Fig. 24, D-E) 
Description: Four made from flakes, fifth made from broken blade. Natural fracture plane utilized. Burin spallsstruck from one side or both sides. Length 3.6 cm. to 
6.1 cm.; width 2.0 cm. to 4.1 cm. 
Provenience; 1 Zone II; 1 Zone IV; 1 Zone V; 2 unknown. 
Angle Burins (6 specimens; Fig. 24, F-G) 
Description: Sturdy angularly pointed spalls used. Angle 
-
Fig. 24. Serrated flakes, A-C; Burins, D, E; Angle Burins, F, G;
Flaked Burins Type I, H, I; Flaked Burin Type 11, j; Burin-Graver,K; Gravers, L, M. 
of burin-like point varies from 20 degrees to 45 degreesAll have minute use flaking on pointed ends. Made of heavy/ sturdy flakes. Length 7.2 cm. to 4.2 cm.; width 
2.7 cm. to 4.0 cm. 
Provenience: 2 Zone IV; 2 Zone V; 2 unknown. 

Flaked Burin, Type I_ (12 specimens; Fig. 24, H-I) 
Description: Made by intersection of burin spall with re­touched edge. Flaked edge result of resharpening burin point. One or two burin spalls struck off. Massive flakes used occasionally. Made from thin scrapers.Length 6.7 cm. to 6.1 cm.; width 3.2 cm. to 5.5 cm. 
Provenience: 2 Zone II; 3 Zone III; 4 Zone V; 1 surface;2 unknown. 
Flaked Burin, Type II (6 specimens; Fig. 24, J) 
Description: Similar to Type I except no flaked edge. Pos­sibly unsharpened Type I. Length 3.9 cm. to 5.9 cm.; width 3.4 cm. to 5.6 cm. 
Provenience: 2 Zone IV; 2 Zone V; 2 unknown. 
Burin-Graver (5 specimens; Fig. 24, K) 
Description: Made from transversely broken unifacial tools. Graver point further recessed along flaked edge; sharpworking point not as sturdy as typical burin, more mas­sive then graver. Microscopic use flakes on point.Length 4.1 cm. to 5.5 cm.; width 1.6 cm. to 2.3 cm. 
Provenience: 1 Zone IV; 2 Zone V; 2 unknown. 
BURIN SPALLS 
(30 specimens; not illustrated)
With the recovery of burins it is a natural consequencethat burin spalls would be recovered. Epstein (1963: 72) states that burin spalls exhibit more wear than burins themselves. This would be a natural result if the purpose of striking off burin spalls was to resharpen a tool that had become dulled. If the burin had not been resharpened, the dulled point would still be present; this feature is noted on some Baker Cave specimens.Rafted burin spalls would be easier to use in grooving material for splitting, particularly hard substances like bone. Yet no hafted burin spalls have been found in the Trans-Pecos region. 
Description: Positive bulb. Four instances of smaller burin spalls exhibit hinge fractures resulting from successful attempts to resharpen burin. Majority tri­angular in cross section; lesser number rectangular. 
Minute pressure flakes at intersecting angles, 
Burin spalls generally dulled from use. Provenience: 3 Zone I; 4 Zone II; 4 Zone III; 8  Zone IV;  
6  Zone V;  5 unknown.  
GRAVERS  
(14 specimens; Fig. 24, L-M)Description: Small beak-like projections formed  two ways.  

or straight surface recessed by small pressure flakes. First method produces more prominent graver. Occasion­ally bifacially flaked. Graver beaks vary from 0.1 cm. to 0.3 cm. in length.
Provenience: 2 Zone II; 6 Zone IV; 5 Zone V; 1 unknown. 
SMOOTHED, PECKED, GROUND, SCRATCHED AND UNMODIFIED STONE OBJECTS 
PENDANT 
(1 specimen; Fig. 25, A)
Description: Half of an oval slate pendant recovered. Cross section flat ellipse. Smoothly finished; low magnification reveals fine, unoriented striations. Part recovered not broken accidentally, but grooved for depth of 0.3 cm. and snapped in two. Groove intersects biconically drilled hole, drilled principally from one side, in center of pendant. Minimum diameter of hole 
0.6 cm. at intersection of opposing cones. Depth of conical holes 0.3 cm. and 0.6 cm.; maximum diameter of drilled hole 0.95 cm. Estimated length of completependant 11.1 cm.; width 8.6 cm. 
Provenience: Unknown. 
PIGMENT 
Twenty-four pieces of pigment were recovered. The distribu tion within the deposits indicates more common usage in the late stages of the occupation of the shelter. 
-
Fig. 25. Pendant A; Pigment Type I, B, C; Pigment Type 11, D-F. Grinding slabs, G-H. 
60 
Type I_ (18 specimens; Fig. 25, B-C) 
Description: Unaltered amorphous hematite. One form coarse in texture. Second form fine, soft, brighter color. Single specimen of lumpy black mineral included in this type. Length 0.5 cm. to 11.1 cm.; width 0.4 cm. 
to 7.3 cm. 
Provenience: 3 Zone III; 1 Zone IV; 14 Zone V. 

Type II (6 specimens; Fig. 25, D-E) 
Description: Facets produced by rubbing against another object. Five of soft textured hematite; sixth speci­men fine textured limestone. Length 1.2 cm. to 3.5 cm.; width 1.0 cm. to 2.4 cm. 
Provenience: 1 zone II; 2 Zone III; 1 Zone IV; 2 Zone V. 
GRINDING SLABS 
(5 specimens; Fig. 25, G-H) 
Description: Small unshaped roof spalls. All exhibit minor surface alterations evidenced by smoothing. One exhibits unoriented striations on one face but concen­trated along one side; another heavily caked by deep orange pigment. Length 12.2 cm. to 14.8 cm.; width 
9.2 cm. to 10.3 cm. 
Provenience: 2 Zone IV; 3 Zone V. 

MISCELLANEOUS STONES 
(2 specimens; not illustrated) 
Description: Heavily burned spall fragment stained by pig­ment. One stained orange on one surface; other reddish-orange on both sides. Length 4.5 cm. to 7.0 cm.; width 
4.0 cm. to 5.8 cm. 
Provenience: 2 Zone V. 

MANOS 
Type (2 specimens; Fig. 26, A) 
Description: Wedge-shaped in cross section. One of dark maroon sandstone, other quartzite stream pebble. Rec­tangular to oval in outline. One face used more ex­tensively than other. Pecked marks more prominent on lesser used face. Wedge shape probably produced by 
-
Fig. 26. Manos. Type I, A; Type 11, B; Type 111, C. 
forward motion. Pigment stain on one specimen. Length 
cm. 
Provenience: 1 Zone III; 1 unknown. 

8.5 to 13.9 cm.; width 8.5 cm. to 11.2 cm. 
Type II (6 specimens; Fig. 26, B) 
Description: Two parallel grinding surfaces. Plano-convex cross section. Heterogeneous in outline. Made from river pebbles. Grinding surfaces flat or slightly con­vex. None appears to be shaped by pecking. Length
7.5 cm. to 12.2 cm.; width 5.7 cm. to 8.3 cm. Provenience: 1 Zone II; 1 Zone III; 1 Zone IV; 3 Zone V. 
Type 111 (1 specimen; Fig. 26, C) 
Description: Flattened diamond in cross section; oval in outline, one end broken. Grinding surfaces polishedfrom use. Subsequently used as awl sharpener. Eleven grooves vary from 0.3 cm. to 1.6 cm. in length. Ran­domly spaced on broken end and one side. Grooves varyfrom 0.005 cm. to 0.2 cm. in depth. Highly polishedshallow depression 0.4 cm. deep and 1.95 cm. wide originates at edge. Deepest at edge, gradually lessens in depth toward center. Length of depression 2.5 cm. 
Provenience: Zone IV. 
RIVER PEBBLES 
Forty-one waterworn pebbles of varying sizes were found in 
Baker Cave. Sixteen were altered by finely-incised or scratched 
lines and 25 were unaltered. These pebbles are divided according­
ly into two categories. 
Type IA (15 specimens; Fig. 27, B) 
Description: Indiscriminately oriented incised lines most common on both ends of river pebbles, occasionally on one end only. On flat pebbles one or both surfaces incised. Complete surface of one pebble incised longi­tudinally with occasional indiscriminate incisions of lesser density toward center. Outline of stones sub-triangular, subrectangular, oval, and round. Cross sections elliptical, thin, parallel-sided, or wedge-shaped. Length 6.9 cm. to 11.0 cm.; width 4.5 cm. to 

8.9 cm. 
-
Fig. 27. Scratched river pebbles.
Type 18, A; Type lA, B. 
Provenience: 2 Zone II; 5 Zone IV; 6 Zone V; 2 unknown. 
Type IB (1 specimen; Fig. 27, A) 
Description: Differs from Type IA in that design was at­tempted. Subrectangular stone resembling human torso and neck. Design consists of oblique lines originatingfrom sides. Some lines nearly transverse width and length of pebble. No indication of painting or that incised lines contained pigment. Length 8.9 cm.;width 6.1 cm. 
Provenience: Zone V. 
Type II (25 specimens; not illustrated) 
Description: Unaltered river pebbles. No uniformity in shape. Great variation in size. Length 2.6 cm. to 
9.6 	cm.; width 1.6 cm. to 7.1 cm. 
Provenience; 	4 Zone II; 4 Zone III; 9 Zone IV; 4 Zone V; 4 unknown. 
ARTICLES OF FIBER AND WOOD 
SANDALS 
Baker Cave produced 33 fragmentary and partially completesandals. Thirty-two can be identified with some accuracy; the remaining specimen is of doubtful type. The classification of Baker Cave sandals will follow the typology of Schuetz (1956: 129 131), who divided the Trans-Pecos sandals into four types. 
Type A, Biparallel Warp (10 specimens; Fig. 28, A) 
Description: Frame of two lechuguilla strips doubled back in center of sandal. Additional strips woven over and under across frame. Reinforcing strips originate from heel and extend to toe, bent over toe, and rewoven into frame. Contrasted to heel, toe has no loose ends. Larger specimens incomplete, smallest 24.2 cm. in length.
Provenience: 1 Zone IV; 7 Zone V; 1 surface; 1 unknown. 
Type B, Bent Frame (1 specimen; Fig. 28, B) 
Description; Two strips of lechuguilla laid parallel and bent together at toe, overlapping, forming rounded toe. Another strip loosely woven back and forth to form base for weft strands. Due to small size of fragment, iden­tification not positive. Size unknown. 
Provenience: Zone V. 
Type C, Double Frame (5 specimens; Fig. 28, C) 
Description: Strips of lechuguilla laid parallel, formingbase twice length of completed sandal. Additional strips woven back and forth. Folded in half at toe and tied. One specimen folded at side. Doubling of base eliminates reinforcing of base. Length 17.7 cm. to 26.5 cm. 
Provenience: 2 Zone IV; 2 Zone V; 1 unknown. 
Type D, Opposed Warp Frame (17 specimens; Fig. 28, D) 
Description: Lechuguilla strips placed parallel, crossed at toe. Toe ends woven back and forth in figure eightfashion toward heel. Additional strips added as needed until heel reached. Reinforcing strips added by 
-
Fig. 28. Sandals. Types A, B, C, and D 
irregularly weaving in additional strands. Length 15.2 
cm. to 25.2 cm. Provenience: 2 Zone III; 5 Zone IV; 8 Zone V; 2 unknown. Comment: The Type D sandal is the type most frequently
found in Baker Cave. It also appears to have a longerperiod of use. 
SANDAL TIES 
Schuetz (1956: 131-134) made a definitive study of the methods of securing sandal ties. Her division into three typesis used in this report. 
Tie A (10 specimens; not illustrated) 
Description: Two lechuguilla strips tied with square knot,passed through sole at toe. Ties crossed over instepsand to outside, then to center. Two strands often joined to provide proper length; 7 on Type A sandal; 
1 on Type C sandal; 2 on Type D sandal. No heel ties. 
Tie B (5 specimens; not illustrated) 
Description: Two lechuguilla strips similar to Tie A method pass through toe of sandal. Strands joined to other strands at sides of instep, down and over outside or inside of frame. Strands taken up and around descend­ing strand to heel and tied, forming heel strap. One tie on Type A sandal; 2 on Type C sandal; 2 on Type D sandal. 
Tie B (4 specimens; not illustrated) 
Description: Similar to B except strands are twisted two to four times around descending strand after beingbrought up from sides. One on Type A sandal; 1 on Type C sandal; 2 on Type D sandal. 
NETTING 
Knotted (12 specimens; Fig. 29, B-D) 
Description: Tied with sheet bend from two-strand, Z-twist­ed cord. Size of mesh 1.0 cm. by 1.0 cm. to 4.0 by 6.0 cm. Two instances of repairs: (1) 17 strands gathered and tied by overhand knot; (2) smaller cord threaded throughsheet bend, through another sheet bend, and knotted with square knot. 
Provenience: 1 Zone II; 2 Zone III; 3 Zone IV; 5 Zone V; 1 unknown. 
Knotless (3 specimens; Fig. 29, A; 30) 
Description: Cord looped about cord below. Descending cord brought over and around descending strand and up.Looped about above strand in same manner. Mesh size smaller in general than knotted type. Size approxi­mately 1.0 cm. in diameter. 
Provenience: 3 Zone V. 
Comment: Sample too small for definitive statement, but knotless netting appears to be a later manifestation than knotted variety. 
-
Fig. 29. Netting. Knotless, A; knotted, B-D. 
CORDAGE 
Fibers are the basic elements of cordage. By twistingfibers together yarns are made. Strands are made by twisting yarns together in an opposite direction. Cordage is formed when two or more strands are twisted together in an opposite di­rection to the twist of the strands. It is impossible to form cordage of any appreciable length in any other fashion. Short lengths of cordage can be made by twisting a strand tightly and then allowing it to twist back on itself. Such cords are general­ly loosely made and of limited length. The Encyclopaedia Britan­nica (1953: Vol. 19; 545) states that, "The yarn has a tendencyto untwist; to overcome this the twist of the strand is made in the opposite direction. The result is that the tendency of the yarn to untwist in one direction and the tendency of the strand to untwist in the other direction balance each other." 
Cordage was a common type of perishable artifactual material 
-
Fig. 30. Detail of knotless netting 
found in the dry deposits of Baker Cave. Previous reports on the archeological contents of dry shelters in the area record similar findings. There were 189 specimens of cordage of lengths varyingfrom 3.0 cm. to 275.3 cm. 
A cord is said to be Z-twisted (Fig. 31, A) when the direc­tion of the twist is from upper right to lower left. An S-twisted cord (Fig. 31, B) has a direction of twist from upper left to lower right. Thus, a Z-twist is formed by clockwise twisting and an S-twist is formed by counterclockwise twisting. 
Z-Twisted Cord (172 specimens; Fig. 31, A, C-F) 
Description: Two single-strand; 169 two-strand; one four-strand. Hard fibers generally used; two examples of bast-like fiber source. Length 3.0 cm. to 275.3 cm.; diameter 0.05 cm. to 0.7 cm. 
Provenience: 4 Zone II; 10 Zone III; 58 Zone IV; 79 Zone V; 5 surface; 16 unknown. 
S-Twisted Cord (17 specimens; Fig. 31, B) 
Description: Three single-strand, six two-strand, six three-strand, two four-strand. Hard fibers used. Length 5.8 cm. to 67.2 cm.; diameter 0.05 cm. to 1.4 cm. 
Provenience: 1 Zone II; 6 Zone IV; 7 Zone V; 3 unknown. 
-
Fig. 31. Cordage 
KNOTTED FIBER 
Knotted fibers (Fig. 32) are among the most commonly found evidences of man's occupation in Baker Cave. This is normal since previous investigations of dry shelters all report knotted fibrous 
70 
materials in the area. Baker Cave produced 943 knots. Not in­cluded in this number are knots used in cordage, netting, and in artifacts such as sandal ties, which have already been discussed. Fourteen different types of knots were found, which to the writer's knowledge is the greatest representation of knot variety found to date. Since in the past little attention has been given to knot­ted leaves or fibers, this type of artifact will be dealt with at some length here. Figure 34 was prepared as a convenience for future writers as an aid in identifying the more common knots found in dry shelters. 
Lechuguilla, sotol, and sacahuisti were the most common plants used. Grass was rarely used, and there was a single in­stance of a grapevine tendril. Whole leaves were occasionallyused, but more commonly leaves were split into strands. Lechu­guilla was always split. Strands varied from 0.2 cm. to 1.0 cm. in width. Knotted grasses were loosely tied, probably due to inherent weakness. There was one instance (Fig. 33) of ten strands joined by square knots wrapped loosely about a stick. 
Square (607 specimens; Fig. 32, A) 
Provenience: 7 Zone II; 37 Zone III; 64 Zone IV; 471 Zone V; 6 surface; 22 unknown. 
Overhand (156 specimens; Fig. 32, B) 
Provenience: 2 Zone II; 11 Zone III; 23 Zone IV; 112 Zone V; 4 surface; 4 unknown. 
Granny (70 specimens; Fig. 32, C) 
Provenience: 1 Zone III; 14 Zone IV; 50 Zone V; 2 surface; 3 unknown. 
Sheet Bend (39 specimens; Fig. 32, D) 
Provenience: 1 Zone II; 1 Zone III; 8 Zone IV; 25 Zone V; 3 surface; 1 unknown. 
Twisted (16 specimens; Fig. 32, E) 
Provenience; 2 Zone IV; 12 Zone V; 2 unknown. 
Comment; Overhand with more than three turns. 

-
Fig. 32. Knots. Square, A; Overhand, B; Granny, C; Sheet bend, D; Twisted, E; Slipknot, F; Larkshead, G; Double Overhand, H;Figure 8, 17 Clove Hitch, J; Half Hitch, K; Two Half Hitches, L; Bowknot, M; Running Square Knot, N. 
Slipknot (14 specimens; Fig. 32, F) Provenience: 1 zone II; 12 Zone V; 1 unknown. Larkshead (6 specimens; Fig. 32, G) Provenience: 5 Zone V; 1 surface. Double overhand (6 specimens; Fig. 32, H) Provenience: Zone V. Comment: Regular overhand with an extra turn. Figure Eight (3 specimens; Fig. 32, I) Provenience: 1 Zone IV; 2 Zone V. Clove Hitch (3 specimens; Fig. 32, J) Provenience: 2 Zone V; 1 unknown. Half Hitch (3 specimens; Fig. 32, K) Provenience: 2 Zone V; 1 unknown. Two Half-Hitches (1 specimen; Fig. 32, L) Provenience: Zone IV. Comment: Scarcity of this knot in Baker Cave probablyresults from misidentification in previous reports. If either loose end of a square knot is pulled, it slips. Frequently called two half-hitches, which are tied properly by reversing second half-hitch from first. Bow (1 specimen; Fig. 32, M) Provenience: Zone V. Running Square (20 specimens; Fig. 32, N) Provenience; 1 Zone II; 5 Zone IV; 13 Zone V; 1 surface. Comment: Running square made by tying overhand, looping one end and tying other end over it. Note: See also Figs. 33 and 34. 
TIED LECHUGUILLA 
(8 specimens: Fig. 35, A)Description: Short sections of lechuguilla tied centrallywith larkshead knot. Length of lechuguilla sections 
5.5 cm. to 10.5 cm. Provenience: 3 Zone IV; 5 Zone V. Comment: Martin (1933: 54) and Schuetz (1963: 157) de­
scribe similar objects from Shumla Caves called "fish stringers" by Martin. While such a use cannot be cate­gorically stated, these objects were all found in the uppermost parts of Baker Cave deposits containing the most fish bones. 
BASKETRY 
Only four fragments of baskets were found in Baker Cave. None is of sufficient size to give any indication of the size of the complete artifact. All were found in the upper deposits. 
Twill woven (1 specimen: Fig. 35, A) 
Description; Selvedge fragment. Formed by alternating woven elements two over and two under. Selvedge formed by bending warp element 45 degrees, over two opposingelements, bending again 45 degrees, and becoming woof elements. Made from split lechuguilla 0.2 cm. to 0.3 cm. wide. 
Provenience: Zone V. 
Coiled Type _I (1 specimen; Fig. 35, D) 
Description: Bottom only. Circular 5.5 cm. diameter. Coils 0.5 cm. thick; bundles of finely split sacahuisti or sotol. Coils are laced tightly, joined by passingstrand over coil, under second coil, and over third coil. Coil started in center of basket bottom, in­creased by additional turns laced to inner coil. Cuppedfragment suggests rounded bottom. 
Provenience: Zone IV. 
Coiled Type II (2 specimens; Fig. 35, C) 
Description; Coils of two to three elements. Method of joining coils unknown. Side reinforced by bundles of fiber and stem parts of sacahuisti. Reinforcing 6.0 cm. 
74 
-
Fig. 33. Strands of lechuguilla joined by square knots 
to 7.2 cm. long; 2.0 cm. to 2.2 cm. wide. Reinforcing bun les stitched to coils in unpatterned manner. Carry­ing strap attached by stitching; strap is diagonal checker weave of 10 elements. 
Provenience: 1 Zone IV; 1 Zone V. 
Wicker (1 specimen; Fig. 35, E) 
Description: Poorly made of small split twigs from 0.15 cm. to 0.2 cm. in diameter. Split strand woven in checker fashion. In some instances two elements crossed. Op­posing elements go at angle and are reincorporated into weave without regard for extra element. Loosely made basket. 
Provenience; Zone V. 
MATTING 
Checker Woven (25 specimens; Fig. 35, F-G) 
Description; Simply made. First strand laid over another at 90 degrees, under second. Second laid under oppos­ing strand, over next. Process repeated width of mat. Selvedge formed by strand bent 45 degrees and rewoven, thus, weave 45 degrees to lay of mat. Selvedge also formed by strands bent 90 degrees, woven over and under two strands, bent 90 degrees again and rewoven parallelin opposite original weave direction. Lay of mat same as weave. Corners formed two ways. One method involved one or two strands which were doubled back and rewoven. Two outside strands bent 90 degrees, woven to other side, bent 90 degrees. Thus warp strand became woof strand, then warp strand. Sharp corner formed. This corner possible only when lay of mat parallel to weave. Second method formed by selvedge edge strands bent 45 
degrees, woven over and under two or three strands, re­woven 45 degrees. Forms rounded corner. Generally,checker woven mats are more loosely woven than twilled matting. Materials used: split and unaltered leaves of sotol and sacahuisti; sotol most common. Strand widths vary from 0.15 cm. to 1.3 cm. The larger the strand, the looser the weave; space between weave as 
great as 0.8 cm. 
Provenience: 4 Zone II; 3 Zone III; 5 Zone IV; 11 Zone V;2 unknown. 
_
Fig. 34. common knots used by pre­historic occupants of Baker Cave. 
Twill Woven (15 specimens; Fig. 36, A, B) 
Description: Tightness of weave and split leaf strands characteristic. First strand taken under one opposingstrand, over two strands, under two strands. Second strand over two opposing strands, over two strands. Repeating process produces "herringbone" effect. Sel­vedge formed by warp and woof strands bent 90 degreesand rewoven. In another method, strand is bent 45 de­grees, passed over two strands, bent 45 degrees and 
rewoven. Corners formed from ten strands divided into units of five elements. Outside strand bent 45 degrees, passed over two strands, under two strands, over two strands until opposite edge is reached, bent 45 degrees and rewoven. Fourth strand treated in same fashion as fifth strand; third strand bent 90 degrees, passed un­der second and first strand, bent 90 degrees and re­woven. Second and first strands bent 180 degrees and rewoven. Opposite five elements woven same way. Corner rounded. Strands vary from 0.2 cm. to 0.5 cm. in width. 
TUMP STRAP 
(2 specimens; Fig. 36, C) 
Description: Both specimens possibly of same object. Made of split sacahuisti, checker woven; six warp and five woof elements. Two fragmentary tump lines attached. The one near end of strap was inserted between woven elements, doubled back. Joined to another strip by 
square knot. Second tump line was inserted between woven parts, doubled back. Provenience: 2 Zone 111. 
PLAITED MATERIAL 
Three-ply (9 specimens; Fig. 36, D-G) 
Description; Split leaves usual method. Second method made from bundles of divided fibers. In third method, leaf is split from thorn end and plaited; no knot 
needed at thorn end. In the fourth method, leaf is 
split into three elements, doubled back and plaited;central strand tucked back into center of split leaf. Neither end needs knotting. Object thus formed is 
too short for known use. 
Provenience: 2 Zone IV; 7 Zone V. 

Four-ply (2 specimens; Fig. 36, h, I) 
Description: Two strands doubled back, flat plaited, form­ing short objects.Provenience; 2 Zone V. 
Six-ply (1 specimen; Fig. 36, J) 
Description: Twilled appearance. Outside strand is over 
-
Fig. 35. Tied lechuguilla, A; twill woven basket fragment, B; coiled basket fragments, C, D; wicker woven basket fragment, E; checker woven matting, F, G. 
one strand, under three strands, over opposite outside strand; bent 45 degrees, and rewoven. Second strand, now outside, is under next strand, over three strands, under fifth strand. Pattern repeated.
Provenience: zone V. 
SPLIT PRICKLY PEAR "LEAVES" 
(4 specimens; not illustrated)
Description: Two perforated near stem base, one midway upand on edge. Fourth too fragmentary to describe. Three ties are of split lechuguilla, one is of either splitsotol or sacahuisti. Two split along sides, inner flesh and thorns removed. On third, only one side split,inner flesh not removed. Fourth split completely about circumference, no flesh removed. 
Provenience: 4 Zone V. 
Comment: Ties possibly used as transportation method. Those with inner flesh and thorns removed possibly were used as small bags. 
FIBER LOOPS 
(11 specimens; Fig. 37, A,B)
Description; Ten of single strand of split lechuguilla,
sacahuisti, or sotol. A circle was made, startingsecond loop, twisted over first loop. Process repeat­ed until strand exhausted. Secured by tucking loose 
end into twisted strands. Eight of three-ply, one of two-ply. Eleventh made by three-ply plaiting, loose ends spliced into other end. 
Provenience: 3 Zone IV; 8 Zone V. 
UNKNOWN SPLIT LEAF OBJECTS 
(Fig. 37, C-D)
Description: Split leaf of sotol or sacahuisti bent double. Second split leaf woven over and under to form zigzagdesign. One instance, of woven leaf tucked back through loops, appears to have three vertical elements. Fifth specimen is more complex. Narrow split leaf (0.25cm.) doubled. Another split leaf is woven over and un­der doubled strand. Woven elements are pulled closelytogether. Thus, it has no zigzag appearance.
Provenience: 4 Zone V. 
Comment; Martin (1933: 59) suggests such objects were used 

Fig. 36.-Twill woven matting. A, B. Tump strap, C. Plaited objects, three-ply, D-G; four-ply, H, I; six-ply, J. 
81 
as plume foundations. 
TIED LOOPS 
(2 specimens; Fig. 37, E)Description; Split lechuguilla tied repeatedly by squareknots to form small loops. Has appearance of chain. Provenience; 2 Zone V. 
TWISTED LECHUGUILLA LEAF 
(1 specimen; Fig. 31, F)
Description: Split lechuguilla leaf doubled and twisted tightly. Fibers separated in twisted part. Ends not twisted are intact. May be method of separating paren­chyma from fiber. 
Provenience; Zone V. 
TIED BUNDLES 
(2 specimens; Fig. 31, G-H)
Description; One is made of 24 split sotol leaves, stalk ends together. Split section of lechuguilla tied with half-hitch 15 cm. from stalk end; lechuguilla strand taken up 5 cm., tied with another half-hitch, extended 4 cm., and tied again. Second specimen is bundle of grass with most of grass stems oriented in one direc­tion. Split lechuguilla tied about one end with over­hand knot. Lechuguilla strand extends over oppositeend of bundle and back to origin of first tie and tied again.
Provenience; 2 Zone 11. 
FORESHAFTS 
(14 specimens; Fig. 37, I)
Description: Complete specimen 18.6 cm. long, 1.2 cm. thick. Notched end cut at sloping angle from surface toward notch. Sides of notch unequal, 1.5 cm. deepone side, 1.6 cm. other side. Specimen contained broken Marshall point. Stem of point 1.1 cm. long;thus foreshaft notch is 0.4 cm. to 0.5 cm. longerthan stem. Notched end transversely roughed for dis­tance of 3.0 cm. by light scoring, possibly to providebetter purchase of binding. Notch tailored to stem. 
Proximal end tapered, beginning 4.0 cm. from end and 
82 
-
Fig. 37. Fiber loops. A, B; split leaf objects, C, D; tied 
loops, E; twisted lechuguilla, F; tied bundles, G, H; fore-shaft with hafted point, I; tenon, J; cupped end of atlatl shaft, K, L; arrow nocks, M, N. 
gradually tapering to blunt point. Tapered end roughed by scoring on long axis, possibly to hold foreshaft in dart shaft. No evidence of gum on tapered end. All specimens are of wood, not flower stalks of sotol, lechu­guilla, or sacahuisti. 
Provenience: 1 Zone II; 2 Zone III; 6 Zone IV; 4 Zone V 1 surface. 
TENON 
(1 specimen; Fig. 37, J)
Description; Tenon was formed by cutting obliquely in direction of tenon. Probably made from small stick and whittled to desired size; 1.0 cm. long, 0.3 cm. thick. No mortises recovered. 
Provenience: Zone V. 
CUPPED END OF ATLATL SHAFTS 
(2 specimens; Fig. 37, K-L)Description; Cup or cone-like depressions from 0.2 cm. to 
0.7 cm. deep; cup diameter 0.5 cm.; shafts diameter 1.0 cm. to 1.1 cm. Shafts gradually tapered distally.Grooved 0.6 cm. to 0.64 cm. from proximal end. One specimen scored transversely near proximal end, possiblyfor fletching. Impression of binding on one specimen.
Provenience: 1 Zone II; 1 Zone IV. 
ARROW NOCKS 
(3 specimens; Fig. 37, M-N)
Description: Cane cut angularly with single thrust of cut­ting tool above node. Probably cut green. No indica­tion of fletching.
Provenience; 1 Zone IV; 1 Zone V; 1 unknown. 
WOODEN SCOOP 
(1 specimen; Fig. 38, A)
Description: Made of pecan or walnut fragment of driftwood. Handle end is smaller and smoothed. Scoop end expandsfrom handle. Charred area centrally located. Length
28.5 cm.; maximum width 7.5 cm. 
Provenience: Surface. 

84 
-
Fig. 38. Wooden scoop. A; pegs, B, C; stake, D; pointed stick Type I, E; wooden arrow point, F; pointed stick Type 11, G; wooden arrow point, H; thorn needle, I; rabbit stick, J; lashed 
sticks, K, L. 
85 
PEGS 
(4 specimens; Fig. 38, B-C)
Description: Larger specimens made with care from splitsticks. Tapered to blunt end; base ends battered and frayed. Length 6.6 cm. Small specimens made from un­split twigs, roughly cut to keener point; base ends battered. Length 3.2 cm. to 10.0 cm. 
Provenience: 1 Zone III; 2 Zone IV; 1 unknown. 
STAKES 
(54 specimens; Fig. 38, D)
Description: Fifty-three specimens constitute part of Feature 4 and are discussed in that section. The fifty-fourth specimen in 11.5 cm. long, oval cross section; widest dimension near butt 2.7 cm. Cut from small limb. 
Provenience: Zone IV. 
POINTED STICKS 
Type 1_ (2 specimens; Fig. 38, E) 
Description: Made from sections of main stalks of monocoty­ledon. Butt broken off, opposite end sharpened to sharp point. Little wear evidenced. Appear to have been used as awls. Length 12.6 cm. to 19.1 cm. 
Provenience: 1 Zone III; 1 Zone IV. 
Type II (3 specimens; Fig. 38, F)Description: Made from small twigs. First specimen sharp­
-
ened on both ends one slanted, the other sharpenedfrom all sides. Second specimen split from pointedend; bark shaved off. Third is better made. Small twig split; semicircular part carefully trimmed and split side smoothed. Point made by gradual thinning.Length 6.2 cm. to 28.8 cm.; diameter 0.03 cm. to 0.6 cm. 
Provenience: 3 Zone V. 
Type 111 (5 specimens; Fig. 38, G-H) 
Description: Possibly wooden arrow points. Base preparedby girding and shaving 0.5 cm. to base. Base diameter reduced 0.1 cm. to 0.2 cm. Tapered distally to point.Length 9.8 cm. to 11.2 cm.;diameter 0.55 cm. to 0.6 cm. 
86 
Provenience: 1 Zone IV; 3 Zone V; 1 surface. 
THORN NEEDLE 
(1 	specimen; Fig. 38, I 
& Fig. 39)

Description: Highly
polished mesquite
thorn. Point is 
slightly blunted, 
probably from use. 
Length 5.5 cm. 

P ovenience: Zone V. 
Comment: Comparison
with unused thorn 
suggests original
length of 8.0 cm. 

RABBIT STICK 
(2 	specimens; Fig. 38, 
J)

Description; One speci­men badly burned. 
Four longitudinal 
grooves each side 
approximately 0.1 
cm. deep, spaced

0.2 cm. apart.
Twenty-six trans­verse grooves are 

Fig. 39. Thorn needle (r.) and spaced 0.15 cm. to unaltered-mesquite thorn enlarged
0.2 cm. apart. The approximately 10 times. other specimen has three sets of four longitudinal grooves approximately 0.05 cm. deep on both sides. Handle end rounded. Two shallow grooves circum­
scribed near handle end. Toward distal end reduced by oblique cutting and broken off at this point. Length
10.2 cm. to 14.6 cm.; width 2.9 cm. 
Provenience: 1 Zone III; 1 Zone V. 

LASHED STICKS 
(3 specimens; Fig. 38, K, L)Description: Two instances, twigs laid parallel, bound bysheer lashing. In third example, stick not joined to another, loosely wound. Lashed with either splitlechuguilla or single fiber strand. 
Provenience: 3 Zone V. 
FIRE DRILLS 
(3 Specimens; Fig. 40, A, B)
Description: Two specimens, broken, made from floweringstalks of monocotyledon. Third specimen, complete,made from large twig. Base ends cut, drill end round­ed and charred from drilling process. Length of com­plete specimen 37.4 cm.; diameters 0.7 cm. to 1.4 cm. 
Provenience: 3 Zone V. 
Comment: The complete fire drill is of particular interest because its material is of some unidentified tree or shrub having a hardness greater than that of the fire hearths. The author has found in making fire-by-friction sets in his Boy Scout work that the best results are ob­tained when the fire hearth and drill are of the same material, and even better results are obtained when the hearth and drill are from the same piece of material. Any of the softer woods such as cottonwood, basswood,willow, or the flower stalks of the various yuccas are satisfactory. The wear, when the fire drill is revolved, produces a fine powder, and friction causes an ember. The fine powder from both the fire drill and the fire hearth aids in producing the ember. The smoldering mass is then carefully introduced to a prepared tinder of shredded cedar bark, grass, or other similar substance and ignition is started by gently blowing or fanning.A hard spindle causes greater wear in the hearth than on the spindle and a live coal is difficult to getsince there is not an equal amount of friction between the spindle and the fire hearth. It is possible that a similar wood was used with the hardwood spindle, but 
it does not seem logical that a people with an abun­
dance of suitable material would do it the difficult 
way. Therefore, it is possible that the hardwood fire drill was used for some other purpose in spite of its characteristic fire drill traits. 
FIRE HEARTHS 
(3 specimens; Fig. 40, C, D)Description: Specimens fragmentary. All are made from flowering stalks of monocotyledon. Two are split sec­tions and are plano-convex; third is made from entire 
-
Fig-40. Firedrills, A, B; fire hearths, C, D; cut stick method A, E; cut stick method B, F; cut stick method C, G; cut stick methods A & B, H; chewed stick, I; unknown wooden artifacts, J, K; fish hook cactus spines, L; shaved stick, M. 
stalk which was smoothed on two sides. Broken throughfire holes in each specimen. Fire hole diameter, 0.9 cm. to 1.2 cm. Provenience: 1 Zone IV; 1 Zone V; 1 unknown. 
CUT STICKS 
Method A (42 specimens; Fig. 40, E) 
Description: Smaller sticks grooved by sawing shallow cut, breaking at groove; result was rough, irregular break. Provenience: 1 Zone III; 23 Zone IV; 15 Zone V; 1 unknown. 
Method B (22 specimens; Fig. 40, F) 
Description: Stick cut or whittled about circumference, then snapped off, resulting in rough, irregular break. Provenience; 12 Zone IV; 8 Zone V; 2 unknown. 
Method (9 specimens; Fig. 40, G) 
Description: Used on larger wooden objects. Hacked byheavy cutting tool by repeated blows. Cut from half to three-quarters through, then broken. Jagged, broken ends characteristic. 
Provenience; 3 Zone IV; 5 Zone V; 1 surface. 
Method A and B (3 specimens; Fig. 40, H) 
Description: Possibly sticks were first cut by whittlingshallow groove; when object did not break readily, groove was sawed until deep enough to snap.
Provenience: 2 Zone IV; 1 surface. 
CHEWED STICK 
(1 specimen; Fig. 40, I) Description; Chewed flower stalk of monocotyledon. Possiblyintended as paint brush. Length 5.5 cm.; diameter 0.7 cm Provenience: Zone V. 
DOUBTFUL WOODEN TOOLS 
(2 specimens; Fig. 40, J, K)Description: Both made of split stalks of monocotyledon.First specimen longitudinally grooved by removal of 
stalk heart. Groove 0.5 cm. deep; groove not full stalk length. Second specimen made from smaller stalk. Unbroken end rounded and smoothed. Both grooves round-bottomed. 
Provenience: 1 Zone IV; 1 Zone V. 
CACTUS SPINES 
(2 specimens; Fig. 40, L)Description: One cut from plant, other pulled from plant as evidenced by cactus flesh on butt end. Provenience: 2 Zone V. 
SHAVED STICK 
(1 specimen; Fig. 40, M)
Description; Probably refuse from manufacture of long,slender pointed stick made by thin, sharp flake. Shal­low longitudinal cuts gradually increase in depth,resulting in sharp point; discarded part has curled shavings attached to stick similar to this specimen.
Provenience: Zone V. 
PRICKLY PEAR SKIN WRAPPED ABOUT STICK 
(1 specimen; Fig. 41, A)Description; Monocotyledon stalk wrapped in skin of prickly pear "leaf". Stalk 8.2 cm. long.Provenience: zone V. 
TOY BOW 
(1 specimen; Fig. 41, B)Description; Twig 37 cm. long bent into arc, tied with single fiber. Bow string 24 cm. long.Provenience; Zone V. 
ARTICLES OF BONE 
AWLS 
Type I (7 specimens; Fig. 41, C, D) 
Description: Made from split long bones, highly polished.Longitudinal striations three-fourths of length. Ar­ticulation generally evident. In one specimen, articula­
tion ground off and decorated by incised cross hatching. Length 8.7 cm. to 19.1 cm. Provenience: 2 Zone II; 2 Zone IV; 2 Zone V; 1 surface. 
Type II (3 specimens; Fig. 41, E) 
Description: Made on distal ends of metatarsals or proximal end of radius. Longitudinal striations and high polish.Entire joint of bone forms base. Distal part more taper­ing and pointed. Length of only complete specimen 12.8 
cm. 
Provenience: 1 Zone IV; 2 Zone V. 

Type 111 (2 specimens; Fig. 41, F) 
Description: Made from entire deer splint bone. Naturallysharp end unmodified except from use. Length 4.7 cm. to 5.5 cm. 
Provenience: 1 Zone V; 1 unknown. 
Type IV (3 specimens; Fig. 41, G) 
Description: Made from sections of ribs or long bone. Complete specimen has blunted end. Highly polishedand striated longitudinally.
Provenience: 1 Zone II; 2 Zone V. 
Type v (7 specimens; Fig. 41, H, I) 
Description; Made from pectoral or dorsal catfish spines.One specimen of operculum of unidentified fish. Bonyprojections of spines worn off distally. End of oper­culum highly polished.
Provenience; 6 Zone V; 1 unknown. 
BEADS 
Type 1_ (5 specimens; Fig. 41, J, K) 
Description; Tubular, made from long bones of rodents or birds. Bones sawed transversely and snapped off. Length 1.6 cm. to 4.85 cm.; diameter 0.3 cm. to 1.0 cm. 
Provenience: 3 Zone IV; 3 Zone V. 
Type II (1 specimen; Fig. 41, L) 
-
Fig. 41. Prickly pear skin wrapped about stick. A; toy bow, B; Bone artifacts: Type I awls, C, D; Type II awl, E; Type 111 awl, F; Type IV awl, G; Type V awls, H, I; Type I beads, J, K; Type II beads, L; Type 111 beads, M, N. Ulnar flaking tool, 0; antler flaking tool, P; wedge-like antler tool, Q; punch-likeantler tool, R; spatulate antler object, S; chopped bone, T; sawed bone, U. 
Description: Tubular bead made like Type I. Decorated by eight grooves about circumference spaced 0.15 cm. to 0.2 cm. apart. Made from small animal bone. Length
2.0 cm.; diameter 0.35 cm. 
Provenience Zone V.

: 
Type 111 (3 specimens; Fig. 41, M, N) 
Description: Small, ring-like. Made from narrow sections of large rodent bones. Probably completely sawed from parent bone. Well finished. Length 0.2 cm.; diameter 
0.7 to 0.9 cm. 
Provenience; 1 Zone IV; 1 Zone V; 1 unknown. 

ULNAR FLAKING TOOLS 
(5 specimens; Fig. 41, 0)Description; Distal ends of ulna blunted and scarred. Provenience; 1 Zone IV; 4 Zone V. 
ANTLER FLAKING TOOLS 
(3 specimens; Fig. 41, P)
Description; Tines broken from main beam of antler. Naturally pointed tine ends battered and scarred Length 6.7 cm. to 10.6 cm. 
Provenience; 1 Zone III; 1 Zone IV; 1 Zone V 
WEDGE-LIKE ANTLER TOOL 
(1 specimen; Fig. 41, Q)
Description: Cut from main beam of antler. Base slightlylarger than distal end. Tapered along two sides to convex wedge-like bit. Cancellous part exposed on one side near distal end. Faint striations along both flattened sides. Length 11.4 cm. 
Provenience; Zone IV. 
PUNCH-LIKE ANTLER TOOL 
(1 specimen; Fig. 41, R) 
Description: Tapering, bluntly pointed object made from tine. Base cut partly through and snapped off. Smoothed and polished until nearly all cut and jaggedparts removed. Distal end slightly battered. Length
7.4 cm. 
94 
Provenience: Zone V. SPATULATE-LIKE ANTLER OBJECT 
(1 specimen; Fig. 41, S)
Description: Sawed either from antler near skull or basal part of tine. Sawed end ground, but splintered area not completely removed. Tine segment sawed longitudi­nally and smoothed. Length 7.0 cm.; width 2.0 cm. 
Provenience: Zone 11. 
CUT BONE 
(49 specimens; Fig. 41, J)
Description: Most often incidental to butchering process. Cut marks often noticeable only under magnification, not at articulations. Disoriented multiple cut marks indicate possible use as "cutting boards." One in­stance of scapula cut deeply as if to detach bone for manufacture of artifact. Scapula not broken cleanlyalong cut. 
Provenience: 4 Zone I; 1 Zone II; 1 Zone III; 18 Zone IV; 20 Zone V; 5 unknown. 
CHOPPED BONE 
(8 specimens; Fig. 41, T)Description; Bones severed, most often by multiple blows. Provenience: 2 Zone II; 1 Zone III; 2 Zone VI; 1 Zone V; 2 unknown. 
FIBER-TIED OPERCULUM 
(1 specimen; Fig. 42, A)
Description: Tied through foramen near articulation bysingle coarse fiber strand; strand then wrapped just below articulation. 
Provenience; Zone V. 
FIBER-TIED RABBIT JAW 
(1 specimen; Fig. 42, B) 
Description: Mandible of cottontail rabbit tied in front of molars by thin, split piece of sotol or sacahuisti. Strand started midway of molars, passed forward of first molar, securing fiber, then wrapped over molars 
95 
-
Fig. 42. Objects of bones fiber-tied operculum. A; fiber-
tied rabbit jaw, B; polished bone, C. Objects of leather: stretch holes, D; thongs, E; pouch, F. Sinew-wrapped dart points, G-J. Objects of shell: shell scraper or spoon, K; perforated shell, L; shell pendant, M. 
96 
toward proximal end. Wrapping continued behind molars Fiber then tucked under second from last and last wrapProvenience: Zone V. 
POLISHED BONE 
(2 specimens; Fig. 42, C)Description: Radii of rabbit highly polished, otherwise unaltered. Provenience: Zone V. 
PIGMENT STAINED BONE 
(4 specimens; not illustrated)Description: Fragmented bones stained by red pigmentacquired after bones broken. Provenience: Zone V. 
ARTICLES OF LEATHER AND SINEW 
LEATHER 
(11 specimens; Fig. 42, D-F)
Description: Three specimens have oval "stretch" holes indicating hides were stretched in frame or pegged to ground in processing. Small size of holes indicates more probable use of frame. Holes 0.4 cm. to 0.7 cm. on long axis. Thongs recovered fairly consistent in width, varying from 0.6 cm. to 1.1 cm. Small triangu­lar pouch. Opening at apex. Possible drawstring inserted through hole in neck of pouch. Damaged area repaired by stitching. Appears bag everted for re­pairs and returned to original side out. Bottom and one side missing.
Provenience: 1 Zone III; 6 Zone IV; 4 Zone V 
SINEW-WRAPPED DART POINTS 
(5 specimens; Fig. 42, G-J) 
Description: Split sinew started on one face, wrapped three to six times. Last wrap tucked under preceding wrap. Split sinews 0.2 cm. to 0.8 cm. wide. 
Provenience: 1 Zone III; 1 Zone IV; 3 Zone V. 
OBJECTS OF SHELL 
SHELL SPOON OR SCRAPER 
(1 specimen; Fig. 42, K)
Description; Smoothly worn over shell edge opposite hinge. If scraper was original purpose, no longer practicalfor this use. Inner part stained with reddish-purplepigment.
Provenience: Zone V. 
PERFORATED SHELL 
(1 specimen; Fig. 42, L)Description; Fragmentary mussel perforated from inner side near umbo. Perforation pecked. 
hole drilled from both sides. Concave surface incised 
Provenience:  Zone IV.  
SHELL PENDANT  
(1 specimen:Description;  Fig. 42, M) Fragmentary specimen, perforated by 1.0 cm.  

by single line from edge of hole to edge of pendant.Lower edge of pendant partly broken, saw-like notches on remaining part.Provenience: Unknown. 
CUT SHELL 
(10 specimens; not illustrated)Description: Shell grooved and snapped off, possibly refuse Provenience: 3 Zone IV; 6 Zone V; 1 surface. 
SUMMARY 
As a consequence of the research at Baker Cave, a chronologycan tentatively be constructed. Since only two Carbon 14 sampleswere run, both from near the provenience of a Plainview point, the dating of the six chronological divisions is extracted from data in a recently published source on the area (Story and Bryant,1966: 9-13, Table 1). The Baker Cave sequence, estimated time intervals, and diagnostic artifactual materials are enumerated below. 
Period 1_ (Zone I) 
98 
-
Time: 7500 6500 B.C. Diagnostic Traits: Lithic Plainview golondrina, choppers, scrapers,
-
utilized flakes, bifacially made knives. 
-
Fiber none recovered. 
-
Bone, shell, and leather no worked items recovered. 
-
Wood none recovered. Comment: Mesic conditions prevailed as indicated by pollen peaks in grass and pine just prior to the end of this period (Story and Bryant, 1966: 154-155). 
Period 2_ (Zone II) 
-
Time: 6500 4000 B.C. Diagnostic Traits: Lithic Bandy, Baker, Early Barbed dart points,
-
utilized flakes, bifacially made knives, spokeshaves,
scratched and unmodified river-worn pebbles. 
-
Fiber possibly knotted fiber, knotted netting, mat­ting, cordage. 
-
Wood foreshafts, dart shafts. 
-
Bone, shell, and leather possibly bone awls. 
Comment: Mesic conditions continue but drying or xeric conditions developing (Story and Bryant, 1966: 154-155). 
Period 3_ (Zone III) 
-
Time: 4000 2500 B.C. 
Diagnostic Traits: 
-
Lithic Almagre, Nolan, Langtry I, Pandale, Tortugas dart points. Pandale knives, large lanceolate knives, utilized flakes, scrapers, flaked burins, choppers,spokeshaves, unmodified river pebbles, manos, hema­
titeFiber . Type D sandals, knotted fiber, knotted netting,
-
cordage, matting. 
-
Bone, shell, and leather awls, leather. 
Wood foreshafts, pegs, wooden awls, rabbit sticks,
-
cut sticks. 
Comment: Marginal mesic conditions change to xeric conditions. Desert plants appear late in the period (Story and Bryant, 1966: 155). 
Period _4 (Zone IV) 
-
Time: 2500 1000 B.C. 
Diagnostic Traits; 
-
Lithic Almagre, Langtry I (diminishing), Langtry 11, Langtry 111 (Val Verde), Langtry IV, Langtry V, straight-based, well made knives, crude oval-based knives, utilized flakes, scrapers, gravers, scratched river pebbles, hematite faceted and unmodified, manos, spokeshaves, angle burins, burins, fist axes, amor­phous cores. 
-
Fiber Types C and D sandals (Type D still dominant), knotted fiber, knotted netting, matting, cordage. 
-
Bone, shell, and leather awls, bone beads, ulna flakers, leather, shell scraps. 
-
Wood foreshafts, dart shafts, arrow nocks, tenons, pegs, stakes, wooden awls, arrow points, fire hearths, many cut sticks. 
Comment; Definitely xeric conditions (Story and Bry­ant, 1966: 155). 
Period _5 (Zone V) 
-
Time; 1000 B.C. A.D. 1000 
Diagnostic Traits; 
-
Lithic Ensor, Catan, Uvalde, Marshall, Montell, 
Langtry II (reducing in number but still common),utilized flakes, scrapers (only period in which well made scrapers found), spokeshaves, gravers, flaked burins, burin gravers, choppers, hammerstones, chisel choppers, knives (Types 11, IV, X, and XIII; Abasolo­Tortugas-like), fist axes, amorphous cores most com­mon in this period. Hematite unmodified, modified.
-
Modified and unmodified river pebbles, manos. Bone, shell, and leather bone awls (including those
-
of fish bones), ulna flakers, leather, shell scrap. 
-
Wood arrow shafts, arrow points, wooden awls, fire hearths, cut sticks. 
Comment; There appears to have been a short mesic interval, possibly making conditions suitable for bison. Returned to xeric conditions in later partof this period (Story and Bryant, 1966; 156). 
-
Period 6 (Surface) Tentative 
-
Time: A.D. 1000 1600 
Diagnostic Traits: 

-
Lithic Catan, Fresno, Perdiz, Toyah, Harrell arrow points, utilized flakes, scrapers. 
-
Fiber knotted fiber rare. 
-
Bone, shell, and leather mussel shell fragments. 
-
Wood none recovered. 
Comment; Xeric conditions prevail. There is an in­
crease in cedar, oak, mesquite, and agave (Story and 
Bryant, 1966: 156). 
Story and Bryant (1966; 9-13, Table 1) developed eighttime periods for the area using Carbon 14 dates from Centipede Cave, Coontail Spin, Bonfire Shelter, Fate Bell Shelter, and Eagle Cave. They would separate Period 5 into two parts, but the associated material found within Baker Cave does not justifythis division. Perhaps a larger sampling and stricter controls would have substantiated Story's and Bryant's chronology. 
The recovery of five fragmentary and one complete Plainview golondrina from Period 1 definitely establishes a Paleo-Indian occupation in Baker Cave. The Carbon 14 dates of 8910lt 140 (6960 8.C.) and 230 (7080 8.C.) (Pearson et. ad., 1965; 305) are considerably later than the radiocarbon dates from Bonfire Shelter (8280 8.C.). Dates from other sites producing Plainview points, however, are not inconsistent with dates from Baker Cave. Campbell (1961: 315) reports a Carbon 14 date derived from fos­sil bison bone of l6O B.C. from the Plainview Site and a date of 5OO B.C. from snail shells from the bone beds of the Plainview Site (ibid.: 315-316). Possibly there was contami­nation from the time the first point was found in Baker Cave and the time the charcoal sample was taken. 
The most significant content of Period 2 was the relative abundance of Bandy, Baker, and Early Barbed points. The Bandypoint appears to be the oldest of this group, with the Baker type and Early Barbed varieties being contemporaneous. Knotted strands of fiber, netting, matting, and cordage demonstrate an early use of the local fibrous plants and a tendency to xeric con­ditions. Of importance in Period 2 is the sudden change of dart point styles from a lanceolate form (Plainview points) to a stemmed, barbed form. It is generally assumed that mesic condi­tions were necessary to provide forage for the large grazinganimals associated with the late Paleo-Indians. Bonfire Shelter (Dibble and Lorrain, 1968) clearly indicates that bison once were sufficiently common in the area to sustain a bison-oriented hunt­ing people. Story and Bryant (1966; 154-155) noted that mesic conditions prevailed during their Periods 1 and 2. With the 
waning of mesic conditions and the waxing of xeric conditions it is doubtful that there was sufficient forage for a consistent supply of bison to sustain a people dependent on this animal. 
area
One possibility is that the Paleo-Indians moved from the and another people with a desert-oriented culture drifted into 
the Trans-Pecos region to fill the vacuum. 
Where did these people come from? Johnson (1965: 105-106) suggests that the area was repopulated by a people adapted to a desert environment. A people new to an area would bring with them their culture and its associated artifacts. Additional re­search in Mexico would greatly assist in contributing knowledge to this question. 
Johnson (1965: 92) suggests the Paleo-Indians took on an Archaic mode of existence based on hunting small game and forag­ing at the end of the Pleistocene. It is further stated (ibid.:93) that the shift from barbless, lanceolate Paleo-Indian projec­tile points to the shouldered and barbed points of the Archaic likely reflects a shift from bison hunting to deer hunting. 
Another possibility is that Paleo-Indians who were barelysubsisting on a diminishing supply of bison were driven out by a people who were adapted to survival in semi-desert conditions. 
Period 3 lithic artifacts such as the Pandale, Langtry I variety, and Tortugas points are consistent with those found in other shelters in the area. The appearance of artifacts of fiber such as Type D sandals, knotted fiber, netting, matting, and cordage suggests greater utilization of local fiber producingplants. 
Period 4 appears to be dominated by makers of variations of the Langtry point and the Val Verde point. It is interesting to note that the Val Verde variation does not appear to be as old in Baker Cave as the Langtry I variation. Bone awls increase in numbers and the first ulna flakers appear. There is a greatervariation in sandal types and an increase in knotted strips of split fibers, cordage, netting, and matting. 
Period 5 produced great masses of fiber intermixed with occupation debris and small roof spalls. There is a marked shift in the lithic assemblage, from Langtry varieties to side-notched Ensor and such corner-notched types as the Marshall, Mon-tell, and Castroville. There is a full range of fibrous arti­facts as well as wooden artifacts. Artifacts of bone are more 
common in this time period. Awls of dorsal and pectoral spine of fish are found in this period. River pebbles and simple grinding 
slabs of native limestone and mullers or manos are more common. Pigments of mineral substances are rare in the other periods but occur with relative frequency in Period 5. Thus, in Period 5 a fuller assemblage of tools is found. 
Campbell (1958: 156-160) suggests the strong possibilityof the mescal bean cult being practiced in Val Verde County sites. The presence of parched and hulled mescal beans in Periods 3 through 5 indicates that these beans were used. In addition, the wound grass ball (Feature 1) enclosing the mescal bean twigs sug­gests the possibility of a more elaborate development of the mes­cal bean cult than was previously known. 
Period 6 is tenuous. A total of only 13 arrow points was found, indicating a transitory occupation of the cave during this time period. Only a few pieces of knotted fiber, occasional flint chips, and a single scrap of shell detritus was noted. Con­sidering the fact that Period 6 represents the maximum surface area excavated, very little material was recovered. The deposits did not have the characteristic fire hearths and other character­istics generally associated with an occupational stratum. 
It is difficult to assign many of the artifacts to definite time periods. While the deposits of Baker Cave were not dis­turbed by pothunting or by any other recent human activity, there were disturbances within the site. Rodent burrows were frequently 
encountered. Artifacts were displaced vertically and continuous­
ly over the centuries by these animals. The activity of the ab­
original occupants resulted in various disturbances, including the construction of hearths and daily activities of a busy people. Consequently, there is no discrete assemblage of artifactual ma­
terial. As a result, only the artifacts represented by several 
examples can be used as typical markers for any time period. It 
is essential that these conditions be understood and that the 
data developed must of necessity deal in generalities and frequen­
cy of occurrence within the site. 
An examination of the artifacts recovered from Baker Cave reveals a gradual change in the mode of existence of the people who occupied the shelter following the Paleo-Indian occupation 
(Period 1). Periods 2 and 3 are dominated by tools of the hunter 
such as dart points and scrapers, with a lesser dependence on 
harvesting of native foodstuffs. The principal source of food 
was the deer. Tools such as awls, to utilize fibrous plants, are 
relatively scarce in Periods 2 and 3. 
Hunting continued to be the prime source of food duringPeriod 4 but there came to be an increasing dependence on local plant products. The products of nearby streams, such as fish and turtle, were utilized to a small degree. Late in the period hunting and food gathering may have been of equal importance. 
In Period 5 the balance was tipped in favor of a gathering economy; hunting became a supplementary element. Late in Period 5 this tendency toward a harvesting economy became even more evi­
dent. Deer were still taken but not as a principal source of diet. In spite of a time of probable lesser population. Periods 2 and 3 produced an abundance of hunting tools (points) and 
scrapers. Period 4 produced a greater number of scrapers and dart points, but not in proportion to the probable increase of aboriginal population. Period 5 shows a decrease in dart points, yet in other facets of the artifactual assemblage there is an increase. The deposits in Period 5 indicate an intensive periodof occupation, which may reflect the population increase. 
The utilization of the shelter by the aboriginal occupantsis of interest. The part of the shelter toward the mouth was utilized primarily for cooking. This is obvious since the concen­tration of fire-cracked rock is greater in Units 1, 7, and 8. The recovery of artifacts from there was less than in other parts of the cave. The central part of the shelter appears to be the area where most of the activities other than cooking took place. Here the frequency of artifacts of all types increased, but there was a lessening of fire-cracked rock. The rear of the shelter probably was intensively utilized for a food processing area. Trimmingsfrom harvested plants and bones are most common in this part of the cave. It appears that trash and worn out or broken artifacts were discarded toward the extreme back of the shelter. Most of the coprolites came from this area. 
The clearly defined stratification of Baker Cave prompts one additional thought. Why would the shelter be unoccupied for periods of time sufficient to deposit spalled roof material and dust in such concentrations that sterile strata were accumulated? In some instances there must have been an occupational hiatus of considerable duration. Epstein (1969) suggests that an altithermal existed in Coahuila and Nuevo Leon between 8000 and 6000 B.C. Story and Bryant (1966: 155) have illustrated by pollen analysisthat during Period I (7000 8.C.) at Bonfire Shelter and Devil's Mouth Site there was an altithermal represented by an increase in 
xeric conditions producing grasses and a decrease of mesic plants such as conifers. The altithermal possibly influenced ecologicalconditions in the Baker Cave area and could have resulted in a hiatus between Baker Periods 1 and 2. Epstein's altithermal the­ory does not account for the occupational hiatuses occurring be­tween later strata in Baker Cave. A possibility is suggested that could account for the time gaps in Periods 2, 3, 4, and 5. If our assumption is correct that beginning in Period 2 there was some dependence on native plants, it may have been that the native food plants in the vicinity were depleted to the degree that exist­ence was not possible. In this event, the occupants would be forced to move to more productive areas. Over an unknown span of time the food plants would recover and as other areas were deplet­ed, the aboriginal population would move back into the area and reoccupy the cave. This situation was probably repeated manytimes, depending on the rate of recovery of the supply of food-producing vegetation. 
BIBLIOGRAPHY 
Butler, C. 	T. 
1948 	A West Texas Rock Shelter. Unpublished M.A. thesis. The University of Texas. 
Campbell, T. N. 
1958 	Origin of the Mescal Bean Cult. American Anthro­pologist, Vol. 60, No. 1, pp. 156-160. 
1961 	A List of Radiocarbon Dates from Archeological Sites in Texas. Texas Archeological Society, Vol. 30 (for 1959), pp. 311-320. 
Davenport, 	J. Walker 
ca. 1938 	Archaeological Exploration of Eagle Cave, Langtry,Texas. Witte Memorial Museum, No. 4. 
Dibble, David S. and D. Lorrain 
1968 	Bonfire Shelter: A Stratified Bison Kill Site, Val Verde County, Texas. Miscellaneous PapersNo. 1, The Texas Memorial Museum. 
Encyclopaedia Britannica 
1953 	Vol. 19. Published by Encyclopaedia Britannica, Inc., p. 546. 
Epstein, Jeremiah F. 
1963 	Centipede and Damp Caves: Excavations in Val Verde County, Texas 1958. Bulletin of the Texas Archeo­logical Society, Vol. No. 33 (for 1962), pp. 1-130. 
r 
1969 	Some Implications of Recent Excavations and Surveysin Nuevo Leon and Coahuila. Paper presented at AAAS meeting; Dallas, Texas. 
Gould, F. W. 
-
1962 	Texas Plants A Checklist and Ecological Summary,Texas Agricultural Experimental Station, The Agricul­
106 

tural and Mechanical College of Texas. 
Graham, John A. and William Davis 
1958 Appraisal of the Archaeological Resources of Diablo 
Reservoir, 	Val Verde County, Texas. Mimeographed 
report prepared by the Archeological Salvage Pro­
gram Field 	Office, Austin. 
Holden, W. 	C. 
1937 	Excavation of Murrah Cave. Bulletin of the Texas Archaeological and Paleontological Society, Vol. 9, pp. 48-73. 
Jelks, Edward B. 
1962 	The Kyle Site, A Stratified Central Texas Aspect Site in Hill County, Texas. Archaeology Series No. 5, Department of Anthropology, University of Texas, Austin. 
Johnson, Leßoy, Jr. 
1964 	The Devil's Mouth Site: A Stratified Campsite at Amistad Reservoir, Val Verde County, Texas. Archae­ology Series No. 6. Department of Anthropology,University of Texas, Austin. 
Kelley, J. 	Charles, T. N. Campbell, and D. J. Lehmer 
1940 	The Association of Archeological Materials with Geological Deposits in the Big Bend of Texas. Sul Ross State Teachers College Bulletin, Vol. 21, No. 3. 
Kelly, Thomas C. 
1963 	Archeological Investigations at Roark Cave, Brewster County, Texas. Bulletin of the Texas Archeological Society, Vol. No. 33 (for 1962), pp. 191-227. 
Kelly, Thomas C., and Harvey P. Smith, Jr. 
1963 	An Investigation of Archeological Sites in Reagan Canyon, Brewster County, Texas. Bulletin of the Texas Archeological Society, Vol. No. 33 (for 1962), 
pp. 167-190. 
Martin, George C. 1933 Archaeological Exploration of the Shumla Caves. Witte Memorial Museum Bulletin No. 3. Nunley, John P., Lathel F. Duffield, and Edward B. Jelks 
Project. 
1965  Excavations  at Amistad Reservoir:  1962 Season.  
Miscellaneous Papers No. Salvage Project.  3,  Texas Archeological  
Parsons,  Mark L.  
1965  Test Excavations  at Fate Bell Shelter.  Miscellan­ 
eous Papers No.  4,  Texas Archeological Salvage  
Pearce, J. 	E., and A. T. Jackson 


1933 	A Prehistoric Rock Shelter in Val Verde County,Texas. Anthropological Papers, Vol. 1, No. 3, University of Texas Publication No. 3327. 
Pearson, F. J., Jr., E. Mott Davis, M. A. Tamers, & Robt. W. John-stone 
1965 	University of Texas Radiocarbon Dates 111, Radio­carbon, Vol. 6, pp. 296-314. 
Ross, Richard E. 
1965 	The Archeology of Eagle Cave. Papers of the Texas Archeological Salvage Project No. 7. 
Sayles, E. 	B. 
1935 	An Archaeological Survey of Texas. Medallion PapersNo. 17. Gila Pueblo. 
Schuetz, Mardith K. 
1956 	An Analysis of Val Verde County Cave Material. Bulletin of the Texas Archeological Society, Vol. No. 27, pp. 129-160. 
Schuetz continued 
1961 	An Analysis of Val Verde County Cave Material: Part 11. Bulletin of the Texas Archeological So­ciety, Vol. No. 31 (for I960), pp. 167-205. 
1963 	An Analysis of Val Verde County Cave Material: Part 111. Bulletin of the Texas Archeological So­ciety, Vol. No. 33 (for 1962), pp. 131-155. 
Semenov, S. A. 
1957 	Prehistoric Technology (Translated by M. W. Thomp­son, 1964; Cory, Adams, and Mackay, London), U.S.S.R. 
Setzler, Frank 
1934 	Cave Burials in Southwestern Texas. Explorations and Fieldwork of the Smithsonian Institution in 1933. 
Story, Dee 	Ann, and Vaughn M. Bryant, Jr. 
1966 	A Preliminary Study of the Paleoecology of the Amistad Reservoir Area. Final Report of Research Under the Auspices of the National Science Founda­tion. 
Suhm, Dee Ann, Alex D. Krieger, and Edward B. Jelks 
1954 	An Introductory Handbook of Texas Archeology.
Bulletin of the Texas Archeological Society, 
Vol. No. 25. 

Taylor, Herbert C. 
1949b 	The Archaeology of the Area about the Mouth of the Pecos, unpublished M.A. thesis. University of Texas. 
Texas Almanac 
1956-57 	The Encyclopedia of Texas. Published by A. H. Belo Corporation. Dallas. 

PART II 
ANALYSIS OF THE FAUNAL REMAINS 
Charles L. Douglas 

TABLE OF CONTENTS 
PART II 
Page 
Introduction 117 
Methods 119 
Data and Discussion Unworked Bone 121 
Distribution of Fauna Numbers of bones 123 Numbers of individuals 125 Relative percentages of bones 127 Relative percentages of individuals 129 
Bone Artifacts Head of bone intact, unmodified 129 Head of bone partly modified 131 Head of bone split, otherwise unmodified 131 Head of bone completely modified 134 Heavy fragments 134 Miscellaneous fragments 137 Bodkins, needles, or awls 137 Worked rabbit bones 137 Other worked fish bone 139 
Antler 139 
Beads 139 
Other Worked Bone 139 
Hair and Fragments of Leather 140 
Summary 143 
Literature Cited 149 

ILLUSTRATIONS AND TABLES 
PART II Figure Page 
1. 	Distribution of fauna: numbers of bones from squares excavated to bedrock 124 
2. Number of bones from partly excavated squares 126 
3. 	Distribution of numbers of individuals from squares excavated to bedrock 128 
4. 	Distribution of numbers of individuals from partly excavated squares 130 
5. Distribution of relative percentages of bones 132 
6. 	Distribution of relative percentages of individuals 133 
7. Tools on mammal bone having head of bone intact 138 
8. Tools on mammal bone having modified head 	139 
9. Awls on mammal bone having split head 	141 
10. Assorted worked bones 	142 
11. Summary of overall percentages and numbers of individuals of major groups 145 
Table 
1. Distribution of bones and individuals 	122 
2. Distribution of animals among artifact types 	135 
3. 	Frequency of species and elements present among worked shells and bones 136 
4. 	Worked and unworked deer bones of majorimportance 136 

INTRODUCTION 

Twenty thousand years ago the Amistad region of southwestern Texas was covered by pinyon pine forest, as is shown by the
a 
pollen record. Climatic changes gradually promoted a change in the forest to a more open pinyon parkland, which was dominant eight to ten thousand years ago when Paleo-Indians began livingin rock shelters of the area (Wendorf, 1961; Bryant, 1968). In Bonfire Shelter, their dart points were found in association with an extinct species of bison, dozens of which were driven over a cliff and killed (Dibble and Lorrain, 1968). The relative lack of evidence makes it difficult to piece together complete informa tion about the lives of the Paleo-Indians. Their artifacts are found infrequently, and their camps are widely scattered. This nomadic culture may have undergone a transition during the times when the last of the large mammals were decreasing in numbers 
(Johnson, 1967: 82). Prehistoric mammals such as the large bi­
son, camel, horse, and elephant were dwindling in numbers 10,000 years ago, and the last disappeared by about 8,000 years ago
(Hester, 1960; Martin, 1963). In Central Texas, bones of white-
tailed deer were found with those of extinct horse in deposits
that were 10,000 years old (Alexander, 1963). Completely modern fauna appears in the archeological record of western Texas and Coahuila by about 8,900 years ago (Taylor, 1956). 
The Paleo-Indian hunting culture was replaced in western Texas, either through immigration or cultural evolution, by Ar­chaic people. The Archaic is one of the most interesting periods in Texas prehistory because occupational sites are abundant, and this period encompassed a time span of about 7,000 years. Since the Archaic cultures prospered for such a long time, many of the habitation sites of these Indians were used, intermittently, for thousands of years. Such a duration of usage promoted develop­ment of deep, stratified deposits of cultural debris that often yield abundant artifacts, and bones of animals that were eaten. 
During the span of Archaic occupancy, the climate of western Texas underwent what appears to be a gradual drying trend. The pinyon parkland that was present 7,000 to 8,000 years ago in Paleo-Indian and Early Archaic times, gradually evolved into a grassland savanna (Bryant, 1968). The occurrence of the grass­land coincided with a period of climatic change known as the Altithermal Period, which occurred from about 7,500 to 4,000 years ago. Severe arroyo erosion and calichification occurred in the Southwest during the Altithermal Period. 
Antevs (1955) thinks that these phenomena were the result of flash flooding and changes in the precipitation-evaporation ratio caused, directly or indirectly, by a more arid climatic regime. In contrast, Martin (1963), on the basis of a pollen analysis of Quaternary sediments, has suggested that climatic conditions in the Southwest during this period were moist and subpluvial, and feels that the severe erosion can be attributable to periods of intense summer rainfall. 
Pollen records from the Amistad region of western Texas in­dicate a reduction in mesophytic species of plants during the 
1
Altithermal, thereby tending to support Antov s hypothesis (Bry­ant, 1968). 
Following the Altithermal, western Texas became more arid, until it became as we know it today. Thus, the grassland savanna was replaced by a xerophytic flora. The last aboriginals to live in southwestern Texas were the Neo-American peoples, who occupiedthe Amistad region, after the Archaic peoples, from about 1,000 to 400 years ago. 
In terms of the presently used cultural chronology, the Alti­thermal period encompassed most of the Early Archaic period. The Middle and Late Archaic (approximately 1,000-4,500 8.P.) followed the Altithermal; these periods were therefore coincident with the xeric trend mentioned above. Such climatic changes undoubtedlyaffected the distribution and abundance of animals living in southwestern Texas, and the Indians who utilized them. 
The temporal distribution of vertebrates is of major interest to many persons. Vertebrates, especially small mammals, are restricted by their physiology to habitats having proper condi­tions for their survival. Larger vertebrates are less subject to the vicissitudes of their environment, and therefore can live under a variety of climatic regimes. The physiological limits for most species are not known, but considerable work has been done in determining present distributions of vertebrates in North America. Thus, after the distribution of a species has been analyzed, we can tell, in a general way, what climatic re­gimes it lives under. Remains of small vertebrates from archeo­logical sites can thus give some clues about the climate that prevailed when Indians occupied the site. Interpretation of climatic conditions on the basis of faunal remains must be ap­proached with caution. As in all scientific endeavors, all parameters must be considered (see also Hole and Heizer, 1965). 
Prehistoric man functioned in the role of a predator on other animals. Food habits of predators are studied by direct observa­tion, by analysis of stomach contents, analysis of fecal drop­pings or pellets (in raptorial birds), or by analyzing remains of food items found at the animal's den. Insofar as primitive man the predator is concerned, occupational strata in his former dwellings usually contain remnants of food items, either alone or incorporated in human droppings. 
About 10,000 bones of vertebrates were recovered from Baker Cave during excavations by Mr. James H. Word, from 1962 through 1966. I was especially interested in analyzing these faunal re­mains because of the long duration of occupation of the cave, and because I anticipated that the fauna might indicate the progres­sive change in climatic conditions associated with the Altither­mal Period. 
METHODS 
About two-thirds of the 10,000 bones recovered during excava­tion of Baker Cave were mammalian, and one-third were fish. Other bones included those of birds, turtles, lizards and snakes. In­vertebrate remains included shells of snails and mussels. 
Cranial and postcranial elements of vertebrates were identi­fied by comparing them with skeletons in the Texas Natural His­tory Collection and in the collection of the Laboratory of Verte­brate Paleontology at The University of Texas. The kinds of mammals represented by postcranial elements were identified byexamining osteological characters such as the shape and position of crests, trochanters, condyles, tuberosities, foramina, etc., which give bones their characteristic appearances. Individual variation in size of bones, and variation owing to increase in age, were taken into account. 
Each bone was identified with regard to what element of the skeleton it was, and to whether it was from the left or right side. By noting the relative development of the bone, and the degree of fusion of epiphyses, the left and right elements could be paired, or separated, for individuals of each taxon. Thus, if there was one left and one right humerus of a specific ani­mal, the decision as to whether one or two individuals was repre­sented was made on the basis of relative age. Otherwise, the most abundant bones from either the left or right sides of 
animals within a taxon were considered to represent the minimum number of individuals present. 
About half of the mammalian bones were small fragments that could not be identified to genus. Because of the large numbers of these fragments, it was felt that an attempt to follow their temporal distribution might result in additional information. Therefore, fragments were placed in one of two generalized cate­gories depending upon the thickness of their walls. Thin-walled bones that were definitely mammalian, were placed under Rodentia-Lagomorpha. Unidentifiable fragments of thick-walled bones were grouped under Artiodactyla. Fragments were assigned to these two categories irrespective of the kinds of intact bones they were found with. I believe there is little to be gained from forcing uncertain identifications into a generic taxon, for this would decrease the value of the precise determinations. Heavy-walled fragments can be placed with more precision under Artio­dactyla, thereby permitting these fragments to be analyzed sepa­rately. 
Because of the small size of some fragments, a few scraps of bird bones might inadvertently have been placed under Rodentia-Lagomorpha. Likewise, fragments of limb bones of large carni­vores might have been classed under Artiodactyla. The incidence of such errors probably was small. Because of the large numbers of fragments involved, the two categories appear to be useful in that they allow analyses of the distribution of fragments that are otherwise unidentifiable to taxa. 
Numerous pieces of leather were found in the shelter; these were identified to taxon by examining fragments of hairs remain­ing on them. Hair was identified to genus or species of mammal represented by comparing the diameter, cuticular pattern, typeof medulla, and location of pigment granules with reference slides of hair from known species (Douglas, 1965). 
All bones and other vertebrate remains from each sample were tabulated under the appropriate taxa on cards. Any unusual fea­tures, such as working of the bones, butchering marks, or charring, were noted. Bones from squares excavated to bedrock were 
con­sidered separately from those from squares that were partly ex­cavated. 
DATA AND DISCUSSION 
Unworked Bone 
On the basis of lithic artifacts and radiocarbon dating, the 16 occupational levels were grouped, by the archeologist, into six cultural zones. Zone 1 represents Late Plainview occupation, and was dated by radiocarbon determinations at about 9,000 B.P. Zones 2 and 3 are Early Archaic (4,500-8,500 8.P.); Zone 4 is Middle Archaic (3,000-4,500 8.P.); Zone 5 is Late Archaic (1,000­3,000 8.P.); and the surface contains evidence of Neo-American occupation (400-1,000 B.P). 
The temporal distribution and the numbers of bones and indi­viduals of each taxon in the faunal assemblage is shown in Table 
1. Most of these species could be found in the area today. 
Unidentifiable fragments of thin and thick-walled bones ac­counted for about half of the recovered bone (Table 1). It is interesting to note that there are more bones and individuals of rodents and rabbits than there are of deer; likewise, there are more small fragments (Rodentia-Lagomorpha) than there are large fragments (Artiodactyla). 
The occurrence of three genera of pocket gophers in the de­posits is unusual, because different genera of gophers usually do not occur sympatrically. Thomomys and Geomys appeared to­gether in Zones 3 and 5; and Thomomys and Cratogeomys appeared together in Zone 4. Remains of burrowing animals cannot be re­garded as indicating that the animals were, in all cases, con­temporaneous with the deposits in which they were found. Never­theless, Thomomys apparently had a wide temporal distribution in Baker Cave, whereas Geomys and Cratogeomys did not (Table 1). 
Thomomys is a small gopher; Geomys is considerably larger than Thomomys, and Cratogeomys is the largest of the three. Thus, one small and one large gopher may have been present in the area at the same times. Ecological differences in habitat selection could have permitted this sympatry. Thomomys customarily inhabits thin, rocky soils where Geomys and Cratogeomys could not exist. Conversely, Thomomys occasionally is found in deep soils; in northern Coahuila this species occurs in deep soils along arroyos (Baker, 1956: 217), and in southwestern Colorado it occurs in deep sandy soils of drainages, as well as in areas having shallow soils (Douglas, 1969). Baker Cave lies within the present ranges of Thomomys bottae, Geomys personatus, and Cratogeomys 
S V IVIIIII I
TABLE 
I.1. Distribution 
of 
bones 
and 
individuals 

MAMMALIA 
Chiroptera
Tadarida brasiliensis 1 X 

Carnivora 8 X X Canis latrans 7 (3) 
XX 
Urocvon cinereoarqenteus 4 (3) X X X Vulpes velox 1 X 
X
Procyon lotor 1 
Bassariscus astutus 2 (2) X Conepatus mesoleucus 5 (2) X X Spiloqale putorius 5 (4) X X X 
Rodentia 133 (43) 
Ammospermophilus interpres 3 (2) X 

Spermophilus sp. 8 (6) X X X Spermophilus mexicanus 9 (5) X X X Spermophilus spilosoma 6 (5) X X X X 
Spermophilus variegatus 50 (27) X X X X X Thomomys sp. 28 (18) X X X X X 
Geomys sp. 4 (4) X X Cratoqeomys castanops 2 (1) X 
Peroqnathus sp. 2 (2) X Peroqnathus hispidus 1 X 
Castor canadensis 3 (3) X 
Peromyscus sp. 11 (7) 

XX 
Siqmodon hispidus 61 (27) X X X X X 
Neotomasp. 53 (26) X X X X X Neotoma micropus 2 (2) X 
Neotoma albiqula 105 (40) X X X X X Neotoma mexicana 16 (9) X X Ondatra zibethicus 2 (2) X X Erethizon dorsatum 3 (3) X X X 
Lagomorpha Lepussp. 169(56) X X X X X X 
Sylvilaqus sp. 1,211 (179) X X X X X X 
ArtiodactylaAntilocapra americana 2 (2) X X Odocoileus sp. 385 (88) X X X X X X Odocoileus virqinianus 1 X 
Unidentified Mammalia 19 X X 
FragmentsRodentia-Lagomorpha 2,520 X X X X X X Artiodactyla 2,092 X X X X X X 
REPTILIA 210 (44) 
Testudinata 145 (40) X X X X X 
Squamata (Sauria)
Phrynosoma cornuturn 2 (2) X X 

Phrynosoma douqlassi 1 X 
Squamata (Serpentes) 20 X X X X X X 
AVES220(56) X X X X X X Meleaqris qallopavo 1 X 
PISCES 1,625 (119) X X X X X X 
MOLLUSCA Gastropoda 105 X X X X X Pelecypoda 35 X X X X X 
122 
castanops (Hall and Kelson, 1959). Geomys and Cratogeomys both require deep soils, and at present would be almost completelyrestricted to river terraces within Val Verde County. 
The muskrat, Ondatra zibethicus, occurs in extreme south­eastern Texas, in the eastern half of the Texas Panhandle, and 
in the Rio Grande and Pecos River drainages (Davis, 1960: 201). Muskrats apparently have not been taken recently from the Devils River, although this species apparently was present there duringlate Archaic times (Table 1). 
The occurrence of beaver. Castor canadensis, in late Archaic deposits appears unusual at first impression. Nevertheless, beavers have been taken from the Pecos and Rio Grande rivers in recent times (Hall and Kelson, 1959; 550). Beavers and musk­rats probably were present in the Devils River and its drainageswhen trees and other vegetation along the banks were more abun­dant and provided better conditions for these species than are presently available. Man probably eliminated beavers from the Devils River, but ecological factors may now play a significantrole in inhibiting recolonization of this drainage system. 
Bison remains were not found in the deposits, but this was not totally unexpected. Of the numerous archeological sites that have been excavated in Val Verde County, bison bones have been found only in Bonfire Shelter, Eagle Cave, and Murrah Cave (Dibble and Lorrain, 1968; Raun and Eck, 1967: 141; Holden, 1937: 69). It seems likely that bison were not common during 
occupation of Baker Cave, and may have ranged County only occasionally (Dibble and Lorrain,  into Val Verde 1968: 71-72).  
DISTRIBUTION OF FAUNA  
Numbers  of Bones  

The distribution of vertebrate remains in the squares exca­vated to bedrock is illustrated in Figure 1, which shows the numbers of bones of various groups found in each zone. The dis­tribution of remains indicates a major occupation of the shelter during Middle and Late Archaic periods (Zones 4 and 5). Cotton­tail rabbits (Sylvilagus), jackrabbits (Lepus), and groundsquirrels (Spermophilus) all have major concentrations in Zone 5 
Cottontail rabbits were more abundant than jackrabbitsthroughout all of the deposits. This trend has been noted at 

other sites in the nearby Amistad region of western Texas (Raun and Eck, 1967). At present, jackrabbits are far more abundant 
than cottontails in this area. Cottontails could have found 
favorable habitats in riparian vegetation along the banks of Phillips Creek, which probably contained water during the times of major occupation of the shelter. 
a 
major concentration in the Late plainview zone. The bones of three species of woodrats, of the genus Neotoma. and of cotton-rats, Sigmodon, have similar increases in numbers in this zone. 
Although the Plainview zone contained more rodent bones than anyother zone, it is questionable whether all of the rodents in this zone were brought into the shelter by humans. The most abundant species of woodrat represented was Neotoma albigula, the white-throated woodrat, which has a propensity to build nests under rocks. It seems likely that some of these woodrats were intru­sive in the site either before or between Paleo-Indian occupa­tions. Nevertheless, various degrees of charring on some of the bones indicate that some of the woodrats were eaten by the Paleo Indians. 
The small fragments placed under Rodentia-Lagomorpha have 
The distribution of heavy-walled fragments classed as Artio­dactyla agrees well with the distribution of deer bones (Odocoi­leus). Both reach their maximum numbers in Middle and Late Archaic zones. It was somewhat surprising to find relatively few deer bones in Zone 1, the Late Plainview occupation, since the Plainview materials have heretofore been associated with the re­mains of big game. 
Fish remains have a major concentration in Zone 5, and are almost completely restricted to Zones 4 and 5. This is the onlymajor group having a pronounced restriction to the upper levels. 
Figure 2 illustrates distributions of vertebrates from the squares that were not completely excavated. Distributions of cottontail rabbits in Fig. 2 differ somewhat from those in Fig.1 in that the major concentration is in Zone 4 rather than Zone 
5. Likewise, small fragments (Rodentia-Lagomorpha) have majorconcentrations in Zones 4 and 5. Other distributions are almost identical in the two sets of squares. 
Numbers of Individuals 
Although analysis of temporal distributions of numbers of bones from various taxa yields important information, the numbers 

of individuals represented by these bones may be of more impor­
tance for interpretive purposes. A vertebrate has numerous bones, 
thus it is possible that 50 rabbit bones may represent as few as two, or as many as 50 individuals. It is unlikely that the lat­ter would ever be the case, unless the Indians selected one bone from an animal for a specific purpose. Assuming that most verte­brate remains found in archeological sites are from animals used as food, the difference between two and 50 rabbits represents, ap­proximately, the difference between four and 100 pounds of edible 
protein. The anthropological implications of this are self-evident. 
The distribution of numbers of individuals of major faunal groups (Fig. 3) closely parallels the distributions of bones from the same groups (Fig. 1). Individuals from most groups in the completely excavated squares again have a major concentration in Zone 5. Individuals of woodrats and cottonrats are most numerous in Zone 1, as were the numbers of bones from these genera (Fig. 1). 
Individuals of fish are most numerous in Zones 4 and 5, and are present in insignificant numbers in the lower The
zones. distribution of individuals is in close agreement with distribu­
tion of numbers of fish bones. 
Again, there are slight differences between the squares ex­cavated to bedrock and those not excavated to bedrock. Figure4 shows that rabbits (Sylvilagus and Lepus) are most abundant in Zones 4 and 5. Sylvilagus is also numerous in Zone 3, as are woodrats and cottonrats. It is difficult to know how much impor­tance should be placed on the increase in numbers of these groupsin Zone 3. In most squares. Zone 3 contained numerous occupation­al strata. It seems likely that the increase in numbers in this zone is real, and was not due to sampling bias. 
Relative Percentages of Bones 
The distribution of numbers of bones of each taxon is shown,by relative percentages, in Figure 5. Again, the distribution of fragments is almost identical to the distributions of the 
respective taxa from which they came. As would be expected,there are larger percentages of cottontail rabbits than of jack­rabbits throughout the various levels. Deer bones and fragmentsare present in about the same percentages through all but the lowest zone, where they decrease in importance as rodents become more important. 
127 

The relative percentages of groups from partly excavated squares are, in general, similar to those in the completely ex­cavated squares. There is a noticeable decrease in percentagesof large fragments, as well as an increase of Svlvilagus in Zone 
3. The percentages of rodents increase in the lowest zone. 
Relative Percentages of Individuals 
The relative percentages of individuals of various taxa are shown in Figure 6. Distributional patterns of the taxa are simi­lar to those shown in Figures 1-5. Fish are more abundant in the upper levels, and rodents are more abundant in lower levels. Of possible significance is the decrease in percentages of deer, and 
a 
concomitant increase in percentages of rabbits in Zones 2 to 4 of the partly excavated squares. 
BONE ARTIFACTS 
The seventy bone artifacts recovered from Baker Cave were constructed from bones of deer, antelope, cottontail rabbits, and fish. Brief morphological descriptions are given for each kind of artifact, and other data are tabulated. Although the tips of some pointed bone tools are missing, they are considered to be awls, because they have the same general shape, and are made from the same bones as some of the completely intact awls. 
The kinds of animals represented among the various types of artifacts are shown in Table 2. The skeletal elements represent­ed among the worked bones and shells are summarized, by species,in Table 3. 
-
Head 
of 
bone 
intact, 
unmodified 
(5). 
Three flakers and 
one awl were constructed on ulnae of Odocoileus sp. These tools 
are from the proximal ends of ulnae, and are from 8.3 to 11 

cm. in overall length (Fig. 7, B-D). The shaft of each ulna was broken 2.5 to 4 cm. distal to the radial notch, and the broken end was worked into a point. One of these has been worked into a pointed, awl-like tool (Fig. 7, B). Two appear to have been broken so that the distal end is pointed; the pointed surfaces show slight wear. The third has a blunter distal end, the edges of which show wear. These tools came from Zones 4 and 5. Word considers these to be flakers; I concur that three probably are, but prefer to consider the specimen with a pointed end a-s an awl. Ulnae of deer were widely used in making both flakers and awls throughout the Archaic periods of the state. The head of the ulna forms a natural handle, and the slender shaft of this bone could be 
129 

broken and worked with ease. 
One awl was manufactured from a radius of Odocoileus sp.The intact, proximal end of the radius forms a handle for this awl, which is 12.8 cm. long (Fig. 7, A). Deep cuts near the articulating head indicate that the awl was formed by chopping away unwanted parts of the bone. Longitudinal striations are present on worked surfaces. Word classifies this as a Type II awl; its provenience was Zone 5. 
-
Head 
of 
bone 
partly 
modified 
(3). 
One awl was construct­ed on a metatarsal of Odocoileus sp. This tool was constructed 
from the distal end of a cannon bone. The unwanted bone was cut 
away, beginning immediately proximal to the articulating head 
(Fig. 8, B). The pointed end of the awl has been broken, and 
the tool is now 11.5 cm. long. The tapering parts indicate the 
awl was originally about 15 cm. long. This tool came from Zone 

4. Word classifies this as a Type II awl. 
A flaker or scraper was constructed on an ulna of Odocoileus sp. Similar to the awls on ulnae described above, it differs in being slightly shorter (8 cm.) and in having a flat end rather than a pointed one (Fig. 8, C). In addition, the radial notch and its process have been removed and that part of the head ground smooth. Longitudinal and transverse scratches are present on the distal end. Dried, membranous tissue still adheres to the semilunar notch. This tool was found in Zone 5. 
One awl was formed on a metatarsal of Odocoileus sp. The curvatures, width and length of this bone suggest that it was constructed from a cannon bone. The distal head has been complete­ly eroded, probably by charring (Fig. 8, A). The worked end is blackened on both sides. The awl presently is 15.8 cm. long; the width and curvatures suggest the bone is from a metatarsal, not a metacarpal. The tool is from Zone 4. Numerous longitudinalstriations are present on the working surfaces. 
-
Head 
of 
bone 
split, 
otherwise 
unmodified 
(6). 
Two awls were constructed on metatarsals of Odocoileus sp. Each awl is made from one side of the proximal end of a cannon bone (Fig. 9, B, D). One awl is 17.2 cm. long, the other is short and stubby, and 8.5 cm. overall. Both show considerable polishing; the smaller has numerous longitudinal scratches and is slightly charred on the anterior end. The longer is from Zone 5, the shorter from Zone 
3. Both are Type I awls in Word's classification. 


An awl was made from one side (11.7 cm.) of the proximal end of a cannon bone of Qdocoileus sp. This tool shows no workingother than splitting of the articulating head and longitudinalstriations on the surface. The point is missing, and half of the bone is fire-blackened. This awl was from Zone 5, and is a Type IV awl by Word's classification. 
Two awls were constructed on metatarsals of Antilocapra americana (Fig. 9, A, C). Both awls are well-constructed and highly polished. The longer is from Zone 2, and is 19.2 cm. in overall length; the shorter is 12.5 cm. long, and is from Zone 5. Word considers both as Type I awls. 
One fragmentary awl consists of one side of the proximal end of a metatarsal of Qdocoileus sp. (Fig. 9, E). This fragment is 
5.3 cm. long, and is slightly darkened by charring. The inner, split surfaces have been smoothed, as in other awls of this type.The fragment was found in Zone 5. Word considers this a Type I awl. 
-
Head 
of 
bone 
completely 
modified 
(1). 
This fragmentarytool appears to be from the proximal end of a metatarsal of Odo­coileus sp. The articulating end of the metatarsal was complete­ly removed, and one side of the shaft was used for the tool (Fig. 8, D). The inner and outer curvatures and the width of this bone lead me to believe it is from a metatarsal. Numerous cuts are present near the former articulating head. This artifact, from Zone 4, is discussed by Word under Type I awls. 
-
Heavy 
fragments 
(3). 
One poorly formed awl or punch, 6.4 
cm. long and 18 mm. wide, was made on a fortuitous splinter of 
heavy bone (Fig. 10, F). One end of the splinter was cut or 
chopped into a point. Longitudinal striations present on the 
worked end may have been made by grinding rather than use. This 
tool was from Zone 5. 

One awl was formed on a fortuitous splinter of a metatarsal of Qdocoileus sp. (Fig. 10, E). The fragment of 6.8 cm. shows longitudinal striations at the pointed end, but no other evidence of having been worked. It was found in Zone 5. 
A fragmentary awl of 14.4 cm. overall length, apparently was made on a cannon bone of deer. The articulating end of the bone is missing, and it is impossible to determine whether it is a metatarsal or metacarpal. The awl has numerous, deep longitudi­nal striations and is polished by usage. The surface is irregular 
134 
t-a> .d 1>t* 1tr1g 	pi i-3
o< H-1H CD 1Pipi 3pjt-3 CD CO 1r+ 13 103 H-h 
>	>
CO Hn 1OH-3 1cno 	3H 
% 
on CD pj3 1P.O CO 1<3Pi P) CDcn rt CO 3r+ 10a 1H1OI-1 M


animals Table 
C pjPi rtH-1Qpi CO 1<H CO CO to 3 i-1 CD 1 O0 13 1H-13 2. 
•
H-3 HO 1H-rt 1 1—J3" PiCb CDH OQ 1i—1'C 1PiPi3 3 1CD I—*3 13P^ 1 CDh-' 1-Q 3 0
among• 
3 co CDV 13PJ 13 H-Dft0 H 1CO 3J 1CO Cb 3}H­H-Pit Pi1 pi1 CD CO Hirt3 1 CO H 1CO 3 PiftH-Hi PJ 113 CD113ft H H(DH-0 3 1 CO 1H-rtH­a COr+ fl>11 Hi H-& 
-
PTPiH 1 	1H-Hi3 
rt H-1a 1h' CD 	Pirt
3 Q 1CD 10)a n H­CO Pi1CD 1vrt0



artifact Distribution
31H 1Or13Pi1 1H-rt1 1ft O
3 of
11CO 	Hi

types
11 CD 11 CO 11 11 
on 	1 1X1 I1 Head of bone intact 
11 

CT« to 	11 4^ 11 Head split, unmodified 
I _L 
11
U) 	1U)1 
Head partly modified
11 
11
H* 
(-*
1 I Head fully modified
11 
U) 	1 ro 1 M Heavy fragments
11 
11
to 	to
1 1 Radius, unmodified
111 '1
Ln 	11 
on
1 1 Rib scrapers
LJ 
11 
UJ 1U)1 

1 1 Antler scrapers
.4 
11 
0> 1011 

Antler flakers
11 
G 	1 1t—1
1 1 U) Tips of heavy awls

11 
h-J 	11
H 00
01 	11Beads 
1i 	I 1
L1 
CTi Ito 	1
00 	i 1 Bodkins, eyeless needles 
_i_ J 
to 	11 to 1 H-1 Other worked bones
1 jI 
'¦J 	Ito 1to 
TOTALS
—J
o 	M 00 tO I00 1Ui 
11 

-
Table 3. Worked bones and shells: frequency of species and elements present in Baker Cave 
Odocoileus sp. (deer) Antilocapra americana (antelope) 
q
anuer 
cannon 9 radius 1 Beads 
¦I • 
uina 	D 1
dS Liciycl J_U S 	JL 
splint bones--2 Lepus sp. 1 Total 27 Total 6 
Sylvilaqus sp. (rabbit) 	Ictalurus (catfish) 
L. mandible 	1 dorsal spine 3 
o 
LdUlUo 
Total 3 pectoral spine 3 Total 8 Unio sp. (mussel) 
Total 	12 
-
Table 4. Kinds and numbers of worked and unworked deer bones recovered from Baker Cave 
Isolated teeth 40 Antler !5 Cranial bones 38 Mandible 50 Forelimb 9 Hindiimb H Cannon 30 Other metapodials !51 Vertebrae !7 Scapula 4 Ribs 20 Total 385 
and poorly smoothed in relation to other awls from this site. This awl was found in Zone 3. Word does not classify these artifacts. 
-
Miscellaneous 
fragments 
(13). 
These tip-end remnants of heavy awls were made primarily from heavy limb bones. The frag­ments range from 2 to 11.5 cm. in length. Twelve are well-formed, and highly polished; the other is rather poorly formed from a thin, flat bone that may be part of a deer's ulna. Eight of the tips are on flat fragments and thus are flattened on one side, or both; three are rounded on all sides; two are four-sided and un­like tips on any intact implements. 
-
Bodkins, 
eyeless 
needles, 
or 
small 
awls 
(8). 
Two splint 
bones of Odocoileus sp. were found. These small, pointed bones are unmodified, but appear to be lightly polished (Fig. 10, K). These bones are ideally shaped, without being worked, for use as bodkins or needles. They are thin bones, and could not withstand use on heavy leather or similar material. Word considers these as Type 111 awls. 
Two dorsal spines of catfish, Ictalurus punctatus, may have been used as needles or awls (Fig. 10, L). 
Four pectoral spines of Ictalurus punctatus were recovered. 
These spines are pointed and of the proper shape and size to have been used as needles (Fig. 10, I). Similar spines of cat­fish have been misidentified in the literature as being mandibles of gar (Epstein, 1963: 82). Pectoral and dorsal spines of cat­fish are discussed by Word under Type V awls. 
-
Worked 
rabbit 
bones 
(4). 
Two radii of Sylvilagus sp. are 
highly polished, but are otherwise unmodified (Fig. 10, J). 
One left mandible of Sylvilagus sp. apparently was broken 
in use, then the two halves were wrapped and tied with plant fibers (Fig. 10, G). The incisors are partly broken, and appear to have been worn by human usage. This jaw might have been used 
for scarification, or perhaps for incising mussel shells. The incisors would not have been sufficiently strong to incise bone. 
One bead was made from the tibia of Lepus sp. The bead was made from a section of the proximal end of a tibia, and is 20 mm. in length. 
Fig. 

7. 


Tools 

on

mammal 

bone 
having

-head 

of
bone



intact 

Other 
worked 
fish 
bone 
(2). 
-An operculum from a bony fish (probably catfish) was used in an undetermined way (Fig. 10, H). Plant fiber was threaded through the foramen in the proximal end of the bone, and wrapped several times around the proximal end. This may have been used as an ornament. It seems too large and delicate to have been used as a gorge for fishing. 
One side of an operculum of an unidentifiable fish was formed into a pointed awl-like implement. Word considers this as a Type V awl. 
ANTLER 
Nine artifacts made of antler of Odocoileus sp. were repre­sented by flakers, scrapers, and a wedge-like tool. These arti­facts are discussed by Word, and additional discussion is not warranted. 
BEADS 
Ten beads were found in Baker Cave. One was made from a section of the proximal end of a tibia from Lepus sp. This was the only bead that could be identified to species. Beads are discussed at length in Part I of this report. 
OTHER WORKED BONE (64) 
The distal, articulating head of a cannon bone (metatarsal) of Odocoileus sp. was grooved and severed by deep incisions made just proximal to the head. A median groove was cut along the fusion furrow between the two parts of the head, but the bone was not broken along this groove while the head was intact. 
An astragalus of Odocoileus sp. was chopped in half by single blows from each of two sides. These are the kinds of cuts one might make when severing the foot from the leg by means of an axe. 
Five fragments of ribs from a large mammal were modified. One is the distal end of a rib that has been severed one inch from the end by incising around the bone and breaking it off. An­other is a curved piece of rib 35 mm. long and 26 mm. wide that has been incised and broken on both ends (Fig. 10, D). One end did not break smoothly; deep incisions were made on either side of the resultant point in an attempt to produce a straight edge. The inner surface of the rib has been cut repeatedly, crosswise, along the entire length of this tool, but these cuts are not 
139 
Fig. 8. Tools 
on mammal bone having modified head 
nearly so deep as the incisions. Three other worked fragments of ribs show slight to complete modification by grinding and smooth­
ing; one is charred (Fig. 10, A-C). 
Fifty-seven fragments of bone showing evidence of chopping or incising appear to have been cut at random, and undoubtedly were incised during butchering procedures. 
HAIR AND FRAGMENTS OF LEATHER 
Lagomorpha (2). One sample consisted of loose hair of Sylvi­lagus. The other was matted hair of Lepus packed in a section of the distal end of a tibia from a young rabbit. This type of bone 
140 
head splithavingbone

mammal

on 
-

Awls 

9. 
Fig. 


and fur arrangement is commonly found in coyote scats. 
Odocoileus sp. (11). Various strips and scraps of leather deer hide were found; most of these had some hair remaining. Only two pieces showed working other than cutting. One was sewn, by means of fibers, in two places. Another piece had been cut from the periphery of a hide; holes were present and appear to have been made when the skin was pegged and stretched. These are discussed in detail by Word in Part I of this report. 
The kinds and numbers of worked bones, and the genera of vertebrates represented, are summarized in Table 2. At least half of the bone tools were constructed of deer bone, and it is likely that most of the tip-end fragments of awls also were made of deer bone. 
The dorsal and pectoral spines of catfish are questionableartifacts. These bones were found in strata containing cultural debris; some spines appear to have been polished, but otherwise there is little concrete evidence that they were used as tools. Until spines are found to have been worked, or with fibers at­tached, their status as implements should remain uncertain. 
SUMMARY 
Thirty-one kinds of mammals were identified in the faunal remains from Baker Cave. The contemporary mammalian fauna of Val Verde County consists of 54 species, seven of which are bats. If bats are excluded, the fauna from Baker Cave contained 65 percent of the species present in the contemporary fauna. Some of the larger mammals, such as the mountain lion, bobcat, and badger, probably were present in the area while the cave was oc­cupied, even though their bones were not recovered. Small mammals such as the short-tailed shrew, some species of pocket mice, har­vest mice, white-footed mice, and kangaroo rats, probably lived in the area but may not have been selected for food. Other mam­mals missing from the faunal assemblage, but living in the area today, include the opossum, peccary, and armadillo. These three species probably were not present in the area when the cave was being occupied. Evidence from other sites in the state indicates that the ranges of the armadillo and peccary have been extending northward during the past 50 to 150 years. The range of the opossum has increased remarkably in the past 60 years. Its pres­ent range includes all of Texas except for the extreme western border of the state. 
The overall percentages of identified bones and numbers of individuals from the site are summarized in Fig. 11. Fish were the most numerous insofar as bones were concerned, but there were more individuals of rodents than any other group. Fish, rabbits, and deer provided the bulk of the identifiable bones, and undoubtedly much of the food for the inhabitants of the shel­
ter. 
One of the striking features of the faunal distributions is the preponderance of cottontail rabbits over jackrabbits throughall zones. There are several possible explanations for this dif­ference. Cottontail rabbits are more easily captured by snaring or netting than jackrabbits, and this might bias the selection even if equal populations of both rabbits were available. Con­versely, jackrabbits are more easily stalked than cottontails. 
It is likely that there was more riparian vegetation along streams and rivers during occupation of the shelter than exists currently. Phillips Creek undoubtedly contained water during most, if not all, of the occupations of the shelter, and pollenanalyses indicate that the uplands were covered by a pinyon park­land or grassland savanna during these times (Bryant, 1968). Such conditions would promote habitats suitable for cottontails. At present, the overgrazed uplands and restricted riparian vege­tation are favorable for jackrabbits, but are barely adequate for cottontail rabbits. As a consequence, jackrabbits far out­number cottontails. 
There is some evidence for minor fluctuations in selection between deer and rabbits. Minor decreases in numbers of deer correspond with increases in numbers of rabbits (Figs. 1-5). These fluctuations are difficult to assess, owing in part to the disparity in size of the animals involved. One deer might con­tain as much usable protein as 50 cottontail rabbits. The fluc­tuations in numbers of individuals through the various proven­iences are not sufficiently large to warrant firm conclusion about relative abundances of the two groups. Deer also migrate from higher to lower elevations, depending upon the season, where­as rabbits do not. Seasonal variations in population densities could account for fluctuations in relative numbers of the two groups. Furthermore, the problem is compounded because there is no way of knowing whether the shelter was occupied equally in all seasons. 
The Middle and Late Archaic peoples that inhabited Baker Cave succeeded in catching numerous fish, many of which were 
144 
-
Fig. 11. Summary of overall percentages and numbers 
of individuals of major groups 
145 
large in size. Their diet of fish, rabbit, and deer was supple­mented by an occasional bird, turtle, jackrabbit, squirrel, or woodrat, and some reptiles other than turtles probably were eaten. Most of the turtles appearing in the deposits were soft-shelled turtles (Tr ionyx sp.). Soft-shelled turtles are difficult to catch, and it seems likely that the Indians relied upon traps. 
Snails and mussels may have been eaten, but were not presentin large enough numbers to have been more than incidental food items. Many mussel shells were incised and broken. Most of the snails probably got into the occupational debris through their own efforts. The few remains of crayfish consisted only of tips
of chelipeds, and probably were from stomachs of fish the Indians brought to the shelter. 
Carnivores were poorly represented in the deposits. All genera of carnivores listed in Table 1 could have provided food as well as hides suitable for use as clothing. Two genera of skunks, Conepatus and Spilogale, were found, and it seems likelythat these species might also have been food items. The Tarahumar Indians of southern Chihuahua are known to eat skunks, as well as gray foxes, otters, and badgers (Pennington, 1963: 92-93). 
Deer played an important part in the economy of each groupthat inhabited the shelter. Maxillary and mandibular tooth-rows and isolated teeth of deer were numerous in the faunal remains (Table 4). Procedures for chronological aging of deer, by exam­ining eruption and wear of the teeth, are well established. Ac­curate aging requires a reference series of jaws and skulls from animals of known ages, and it is preferable that these be taken from the area concerned. Such a series was not available, so teeth were assigned to relative age groupings (juvenile, subadult,adult and old). Teeth in a tooth-row erupt sequentially, and do not wear at an equal rate. Therefore, I did not attempt to as­sign relative ages to isolated teeth other than molars or the last permanent premolar (P4). 
It soon became apparent that the inhabitants of the shelter used deer of all age groups. Each occupational zone contained representatives of all age groups, and fawns and yearlings were selected in equal numbers with mature individuals. This indi­cates that the shelter was occupied throughout the year, rather than seasonally. 
Table 4 lists numbers of identifiable deer bones recovered from the site. It is interesting to note that relatively few 
vertebrae, scapulae, and pelvic girdles are present. This would indicate that deer were butchered away from the shelter, and the meat was transported. If this was the case, it is puzzling that so many mandibles and antlers were brought to the shelter. The access is such that it would be hazardous to carry a carcass into the shelter. The kinds of bones that were found strongly suggest that deer were butchered at some distance from the shelter, and that the skull and lower limb bones were retained for further use. The cannon bones played an important role in the construction of implements, and it can only be assumed that the meat in and on 
the skull warranted the efforts required to bring it back to the 
shelter. 
One of the striking features of the faunal materials was the preponderance of heavy, fragmented bone. Few intact deer bones other than teeth and metapodials were recovered. Only two bones of artiodactyls other than deer were found, thus it was assumed that fragments of heavy-walled bones probably were from deer. The assemblage of fragments indicates that all major bones of deer were broken into small pieces. This breaking was intention­al, and many of the larger fragments are marked by scratches, in­dicating that the bones from which they came were scored by a sharp instrument prior to breaking. Such breaking would have per­mitted the marrow to be extracted, and would have provided raw materials for awls and other tools. Similar breaking of deer bones also was noted at sites in Val Verde County (Johnson, 1964: 80), and the Tehuacan Valley of Mexico (Flannery, 1967: 159). 
Many, but not all, of the bones from each zone were charred. Charring of many, or all, of the bones within a level would in­dicate spontaneous combustion, but this was not the case in Baker Cave. Some bones apparently were charred during cooking, where­as others appear to have been burned as a means of disposing of them. Rabbit bones frequently were darkened by charring, while those of rodents almost always were uncharred. About one-fifth of the fragmented deer bones were charred to a dark brown or black. 
It was surprising to find mandibles of deer, pieces of ant­ler, and large bones of the skull almost completely carbonized by charring. It is not clear why the Indians would have burned these larger bones. Possibly the brain was roasted in the skull prior to eating it. The completeness of charring on numerous mandibles of deer indicates that they were burned after any meat was removed. It is possible that these bones were thrown into the fire after the meal in order to divest them of any meat or odor that could have attracted carnivores. The burning of bones. 
for any purpose, can be expected to introduce a bias into the faunal data. 
The Mohave Indians are known to have cooked rabbits in the ashes of a fire; the meat and bones were then pounded together,and this mixture was roasted until even the bones were edible (Stewart, 1968: 36). 
All of the various taxa of vertebrates and invertebrates are represented throughout the major zones, thus all of the cul­tures relied on some of the same food sources. The data indicate that the Plainview people may have eaten proportionately more rodents than did the other cultures. It is apparent that the Middle and Late Archaic peoples utilized the stream life to a much greater extent than did the Plainview, Early Archaic, or Neo-American peoples. 
LITERATURE CITED 
Alexander, H. L, Jr. 
1963 The Levi Site: a Paleo-Indian campsite in Central Texas. American Antiquity, 28; 510-528. 
Antevs, E. 
1955 	Geologic-Climatic dating in the West. American Antiquity, 20; 317-335. 
Baker, R. 	H. 
1956 	Mammals of Coahuila, Mexico. Univ. Kansas Publ. Museum Nat. Hist., 9; 125-335. 
Bryant, Vaughn 
1968 	Pollen analysis of the Devil's Mouth Site, Val Verde County, Texas, pp. 57-70. jin;The Devil's Mouth Site, the third season 1967, by W. M. Sorrow. Papers of the Texas Archeol. Salvage Proj., No. 14. Austin. 
Davis, W. 	B. 
1960 	The Mammals of Texas. Game and Fish Comm., Austin, Texas, Bull. 41; 1-252. 
Dibble, David S., and Dessamae Lorrain 
1968 	Bonfire Shelter; a stratified bison kill site, Val Verde County, Texas. Misc. Papers, Texas Memorial Museum, No. 1; 1-138. 
Douglas, Charles L. 
1965 	Biological techniques in archaeology. American Antiquity, 31, Part 2: 193-201. 
1969 	Ecology of pocket gophers in Mesa Verde National Park, Colorado. Mus. Nat. Hist., Univ. Kansas. Misc. Publ., 51; 147-175. 
149 
Epstein, Jeremiah F. 
1963 	Centipede and Damp Caves: excavations in Val Verde County, Texas, 1958. Bull. Texas Archeol. Soc., 
33: 1-129 (for 1962). Flannery, K. V. 1967 Vertebrate fauna and hunting patterns, pp. 132-177. In.: The prehistory of the Tehuacan Valley. Vol. I. Environment and subsistence; pp. viii + 1-331, D. S. Byers, ed. Univ. of Texas Press, Austin. 
Hall, 	E. R., and K. Kelson 1959 The Mammals of North America. Ronald Press, New 
++
York, 2 vols. (1: xxx 1-546 79, 2: viii + 547-1083 79).
+ 
Hester, J. J. 1960 Late Pleistocene extinction and radiocarbon dating.American Antiquity, 26: 58-77. Holden, W. C. 1937 Excavations of Murrah Cave. Bull. Texas Archeol. and Paleontological Soc., 9: 48-73. Hole, F., and R. F. Heizer 1965 An introduction to prehistoric archeology. Holt, 
+
Rinehart and Winston, Inc., New York, pp. x 1-306. Johnson, Leßoy, Jr. 1964 The Devil's Mouth Site: a stratified campsite at Amistad Reservoir, Val Verde County, Texas. Archeol. 
+
Ser., No. 6: [B] 1-115. Dept. Anthro., Univ. Texas, Austin. 
1967 	Toward a statistical overview of the Archaic cul­tures of Central and Southwestern Texas. Texas Memorial Museum, Bull. 12: viii + 1-110. Austin. 
Martin, P.  
1963  The last 10,000 years:the American Southwest.  a fossil pollen record of Univ. of Arizona Press,  
pp. viii  + 1-87.  
Pennington,  C. W.  
1963  The Tarahumar  of Mexico,  their environment and  
material culture.  Univ. Utah Press, Salt Lake  
City, Utah, pp.  x  + 1-267.  

Raun, Gerald G., and L. J. Eck 
1967 	Vertebrate remains from four archeological sites in the Amistad Reservoir area, Val Verde County, Texas. Texas Jour. Sci., 19: 138-150. 
Stewart, K. M. 
1968 	Culinary practices of the Mohave Indians, El Palacio, 75: 26-37. 
Taylor, W. W. 
1956 	Some implications of the carbon-14 date from a cave in Coahuila. Bull. Texas Archeol. Soc., 
27: 215-234. 
Wendorf, F., ed. 
1961 	Paleoecology of the Llano Estacado. Ft. Burgwin Research Center, No. 1: 1-144, Museum of New Mexico Press, Santa Fe. 
151