INDIAN NOTES 
 AND MONOGRAPHS 
 
 No. 
 
 48 
 
 A SERIES OF PUBLICA- 
 TIONS RELATING TO THE 
 AMERICAN ABORIGINES 
 
 ARCHAEOLOGICAL EXPLORATION 
 
 OF A ROCK SHELTER IN 
 
 BREWSTER COUNTY, TEXAS 
 
 EDWIN F. COFFIN 
 
 NEW YORK 
 
 MUSEUM OF THE AMERICAN INDIAN 
 
 HEYE FOUNDATION 
 
 1932 
 
-"" 
 
 \ 
 
 »©• *«• *W* *M* *^* 
 
 PLEASE HANDLE 
 WITH CARE 
 
 University of 
 Connecticut Libraries 
 
 *v* *©• *w* *m^ *^* 
 
 
 3 T1S3 002b0 
 
 O 
 
 O 
 OO 
 
 ro 
 
 -P- 
 oo 
 
 O 
 00 
 
1-v.^ coots ^«-M« ^avNclo 
 INDIAN NOTES 
 AND MONOGRAPHS 
 
 >pt^ si 
 
 No. HfftftiU 48 n<Ji(/? 
 
 L I v 
 
 A SERIES OF PUBLICA- 
 TIONS RELATING TO THE 
 AMERICAN ABORIGINES 
 
 ARCHAEOLOGICAL EXPLORATION 
 ^ OF A ROCK SHELTER IN 
 BREWSTER COUNTY, TEXAS 
 
 EDWIN F. COFFIN 
 
 NEW YORK 
 
 MUSEUM OF THE AMERICAN INDIAN 
 
 HEYE FOUNDATION 
 
 1932 
 
LANCASTER PRESS, INC. 
 LANCASTER, PA. 
 
ARCHAEOLOGICAL EXPLORATION 
 
 OF A ROCK SHELTER IN 
 
 BREWSTER COUNTY, TEXAS 
 
 BY 
 
 EDWIN F. COFFIN 
 
CONTENTS 
 
 PAGE 
 
 Foreword ix 
 
 Introduction xi 
 
 Archaeological Exploration 1 
 
 Bee Cave Canyon Topography 1 
 
 Rock Shelter 5 
 
 House-sites 6 
 
 Caves 21 
 
 m Fire-pits 22 
 
 Objects Recovered . .- 22 
 
 Stone Artifacts 22 
 
 Paints 27 
 
 Wooden Implements 27 
 
 Reeds 31 
 
 Antler Pestles and Tools 33 
 
 Bone Implements 33 
 
 Gourd Vessels 34 
 
 Matting 36 
 
 Basketry 36 
 
 Netting 38 
 
 String 42 
 
 Knots 43 
 
 Sandals 43 
 
 Tie-strings 47 
 
 Fabrics 50 
 
 Fiber Brushes 53 
 
 Cactus Thorns 53 
 
 Bindings 53 
 
 Grasses and Leaves Otherwise Employed ... 54 
 
 Leather Remnants 56 
 
 Ornaments 56 
 
 Clay Figurines 57 
 
 Vegetable Foods 59 
 
 Animal Bones 60 
 
 Conclusions 60 
 
 Appendix 64 
 
 Index 65 
 
ILLUSTRATIONS 
 
 Map 
 
 Contour map of floor of rock shelter and sites exca- 
 vated, Bee Cave Canyon, Brewster County, Texas 
 
 Frontispiece 
 
 Plates 
 
 PAGE 
 
 I Rock shelter and talus at mouth of Bee Cave 
 
 Canyon 3 
 
 II Chalk Draw from eastern end of rock shelter. . . 4 
 
 III Natural stone reservoir at end of Bee Cave 
 
 Canyon 7 
 
 IV House-site 3, looking east 8 
 
 V Upper bed at one-foot level beneath house-site 3 13 
 
 VI House-site 4, looking east, with western and 
 
 part of southern walls removed 14 
 
 VII Outline of hand and wrist, house-site 4 17 
 
 VIII Grooved rock surface, possibly abraded for shap- 
 ing and sharpening implements 18 
 
 IX Stones decorated with designs in black or red . . 25 
 
 X Sticks of wood with tenon-like ends 26 
 
 XI Fragment of grass mat 35 
 
 XII Network bag 39 
 
 XIII Detail of netting technique of network fragment 40 
 
 XIV Blanket-like fabric of soft fiber 52 
 
VIII ILLUSTRATIONS 
 
 Figures 
 
 1 Stones wrapped with grass 24 
 
 2 Pointed stick with leather straps 29 
 
 3 Detail of wrap-woven edge and warp strand of 
 
 mat shown in pi. xi 34 
 
 4 Pouch containing corn and squash seeds 37 
 
 5 Sandals of yucca, braided, with ends woven in . . 43 
 
 6 Sandals of yucca, reinforced 44 
 
 7 Detail of braiding technique with opposed ele- 
 
 ments, strands kept flat 45 
 
 8 Detail of braiding technique with opposed ele- 
 
 ments, strands turned 46 
 
 9 Detail of braiding technique with opposed ele- 
 
 ments, two flat strands 47 
 
 10 Sandals of yucca, braided with two opposed ele- 
 
 ments 48 
 
 11 Sandals of yucca, braided with two opposed ele- 
 
 ments, a with shredded strands and c with 
 fiber string 49 
 
 12 Technique of tie-strings of sandals 50 
 
 13 Technique of framework and tie-strings of child's 
 
 sandal c, fig. 11 51 
 
 14 Pendants of stone and shell 56 
 
 15 Pendant of stone 57 
 
 16 Seeds and sections of reed strung like beads. ... 57 
 
 17 Figurines of unfired clay 58 
 
FOREWORD 
 
 The rock shelter to which this report refers was 
 discovered on February 16, 1928, by Mr. M. R. 
 Harrington who recorded it under the name "Eagle 
 Canon" rockshelter (Indian Notes, vol. V, no. 3, 
 Museum of the American Indian, Heye Foundation, 
 New York, July, 1928), although given on the map 
 of the United States Geological Survey (Nine Point 
 Mesa Quadrangle, Brewster County, Texas) as Bee 
 Cave Canyon. Mr. Harrington, in a letter dated 
 March 15, 1929, says, "I called the canyon Eagle 
 Canyon in place of Bee Cave, as stated on the map, 
 because Ray Miller of Marathon, an old timer in 
 the district, told me that Eagle Canyon was the 
 original name and that Bee Cave Canyon was given 
 it on the spur of the moment by Government men." 
 As it is now recorded on the Geological Survey maps 
 as Bee Cave Canyon, it is so referred to in this 
 publication. 
 
 After some work in the rock shelter, reference to 
 which is made in the appendix, Mr. Harrington re- 
 turned to New York, and shortly afterwards resigned 
 from the staff of this Museum. The results thus 
 far having been of great interest, it was deemed 
 advisable to continue the archaeological investiga- 
 tions of the rock shelter, and in February, 1929, Mr. 
 
X FOREWORD 
 
 Edwin F. Coffin, a member of the Museum's staff, 
 commenced his work where Mr. Harrington left off, 
 and worked on the site until the middle of June of 
 that year. 
 
 George G. Heye, Director. 
 
INTRODUCTION 
 
 The expedition, during which the research de- 
 scribed in this publication was made, investigated 
 primarily Bee Cave Canyon, and other caves and 
 rock shelters in close proximity to it. Caves in the 
 Hueco Mountains and pueblo sites in the vicinity of 
 El Paso were also examined, as well as a rock shelter 
 in Satan Canyon, a branch of Devil's Canyon, 
 twenty-five to thirty miles northwest of Del Rio, in 
 Valverde County, Texas. The work was carried on 
 under the auspices of the Museum of the American 
 Indian, Heye Foundation, of New York City, and 
 the University Museum of Cambridge, England. 
 
 I want to take this opportunity to express my 
 thanks to Mr. Lee Schuler and family, on whose 
 property Bee Cave Canyon is located, for their great 
 assistance and the many favors they have shown me. 
 I wish to acknowledge my obligations to Mr. and 
 Mrs. R. B. Alves, Mr. and Mrs. Fred VVoodworth, 
 Colonel M. L. Crimmins, and Mr. Victor J. Smith, 
 for the many kindnesses shown, and the help given 
 me while working in the vicinity of El Paso; Mr. 
 Henry T. Fletcher and Mr. E. E. Townsend, while 
 working near Alpine; Mr. C. A. Markward of Del 
 Rio, for giving permission to excavate a cave on his 
 property in Satan Canyon, Valverde County, and 
 
XII INTRODUCTION 
 
 Dr. John K. Small of the New York Botanical Gar- 
 den, who identified many of the samples of vegetable 
 substances found in the shelter and brought to New 
 York from Texas. 
 
 E. F. C. 
 

 
 
 
 
 
 
 
 
 . 
 
 
 
 
CONTOUR MAP OF FLOOR OF ROCK SHELTER AND SITES EXCAVATED, BEE CAVE CANYON, BREWSTER COUNTY, TEXAS 
 
ARCHAEOLOGICAL EXPLORATION 
 
 OF A ROCK SHELTER IN 
 
 BREWSTER COUNTY, TEXAS 
 
 By EDWIN F. COFFIN 
 
 Bee Cave Canyon Topography 
 
 THE rock shelter is located on the ranch of Mr. 
 Lee Schuler, on the north side and at the 
 mouth of Bee Cave Canyon, Brewster County, 
 Texas, about nine miles southeast of Santiago Peak 
 and about one and four tenths miles east, and a little 
 south, of Black Peak. 1 Bee Cave Canyon is a box 
 canyon, about 1500 feet deep and 400 feet wide at 
 its mouth, and derives its name from the colonies 
 of bees that nest in crevices of the wall of the rock 
 shelter. The canyon opens into Chalk Draw, which 
 at this point is very wide and drains toward the 
 southeast. For over ten miles of its course, begin- 
 ning a little more than half a mile northwest of 
 Schuler's Ranch, its southern boundary is an almost 
 continuous cliff of limestone, indented here and there 
 with canyons and caves. Within the distance men- 
 tioned, there is but one horse trail leading from the 
 
 1 See Nine Point Mesa Quadrangle, Brewster County, 
 Texas, published by the United States Geological Survey, 
 Washington, 1918. 
 
 1 
 
2 ROCK SHELTER 
 
 draw to the top land. Wagons and automobiles 
 must detour to reach the top. 
 
 The northern wall of the canyon at its mouth and 
 at the rock shelter rises from 400 to 450 feet above 
 the valley. The talus at its base extends up the 
 wall about one third of this distance, and the outer 
 edge of the sloping roof of the rock shelter is about 
 midway between the upper edge of the talus and the 
 top of the cliff. The northern wall juts from 200 
 to 300 feet beyond the southern one. This was ad- 
 vantageous to its inhabitants, as it allowed one 
 standing at the eastern end of the rock shelter a 
 wonderful view for miles to the east and southeast 
 down the draw, and permitted any breeze coming 
 up the draw in summer to enter the rock shelter, 
 while in winter when the winds come down the draw, 
 it prevented quite effectually their deflection into 
 the canyon, a circumstance taken advantage of to- 
 day by herders, who use the old shelter for a winter 
 fold. 
 
 The rock shelter is well protected from rain, for 
 the drainage above is toward the southwest and the 
 water falls into the canyon at its western end, so that 
 very little comes over the cliff. Occasionally, how- 
 ever, a small portion of the rainwater flowing over 
 the edge of the opposite canyon wall is carried into 
 the shelter by strong winds. 
 
 At the end of the canyon is an abundant water 
 supply, stored in a natural stone reservoir, about 
 thirty feet in diameter, formed by the fall of over 
 
itiif 1 
 
BEE CAVE CANYON 5 
 
 one hundred feet of the waters of Upper Rotten 
 Draw, which drains the southern slope of Black Peak 
 and a large area of the upland. This reservoir has 
 never been known to go dry, according to the present 
 older inhabitants of the surrounding country. 
 
 Rock Shelter 
 
 The rock shelter, measured by the overhang, is 
 768 feet long, and 106 to 110 feet wide in its central 
 portion from overhang to back wall. A mass of 
 rocks which has fallen from its roof near the center 
 almost divides it into two portions. 
 
 It was occupied mostly at the easterly end, this 
 part being better adapted to the purpose, as it was 
 the widest and most level. Much filling and level- 
 ing had been done during the time of occupancy. 
 Over a small area to the east the floor-covering was 
 found to be on the undisturbed talus, while the fur- 
 ther the work was carried to the west the deeper was 
 the fill of ash, stone, grass and discarded material, 
 especially against the rock shelter wall. Many 
 pointed ends of large stones protruded through the 
 lower layers of the flooring, but at the time of the 
 abandonment almost all of these were covered by 
 succeeding layers of similar floors, which were over 
 four feet deep at some points. On the wall at the 
 back were a number of pictographs in red oxide of 
 iron, one of which probably represents a bird with 
 wings spread, and two others human beings. 
 
6 ROCK SHELTER 
 
 HOUSE-SITES 
 
 These sites have been called house-sites for want 
 of a better name. The walls, judging by the amount 
 of loose stone in close proximity, were probably not 
 much higher than when found. No roof beams or 
 posts were found except in house-site 4, where 
 charred ends of posts were driven into the floor, but 
 there is no evidence that these supported a roof. 
 Perhaps enclosures would have been a better term. 
 
 The ruins of six "houses" were found, the founda- 
 tions of which were all laid on the lower floor levels. 
 These, for convenience of reference, will be denoted 
 by number, in the order of excavation. 
 
 The walls of house 1 were partly demolished, and 
 the site completely covered by the later floor layers. 
 The east wall was 15 inches thick and measured 11 
 feet in inside length; it was 21 inches high at its 
 southern end, and 17 inches at its northern end 
 where it abutted on the rock shelter wall, and it 
 bowed 6 inches in the center, toward the east. The 
 northern end of the wall was 4 feet 3 inches long, 
 and was constructed of small stones quite well laid, 
 and chinked with a mixture of adobe and ash. The 
 southern end was made of heavier stones with the 
 inside corner thickly plastered with the same mate- 
 rial, and the center part, 3 feet 5 inches long, was 
 poorly constructed, one stone 12 inches wide by 2§ 
 inches thick having been placed on end. 
 
 Six feet of the southern wall of this house were 
 traceable, it being 17 inches thick at its western end. 
 
BEE CAVE CANYON V 
 
 The second course of stone was laid in adobe, with 
 the base heavily plastered, and showed the finger- 
 prints of the builder. The western wall was missing. 
 Near the eastern wall and outside the house-site 
 were two small stakes driven into the ground, not 
 far from the rock shelter wall. 
 
 Little remained of house-site 2, except a part of 
 its eastern wall which curved at its southern end and 
 was 2 feet high. The foundation stones were all 
 quite large and were set on end. For a distance of 
 9 feet this wall was parallel with the west wall of 
 house 3, forming a wall 2 feet thick. 
 
 Near this wall, but inside of the site and six inches 
 below the surface, was uncovered a sherd of a cook- 
 ing pot. In a row, about six inches apart and at 
 the same distance from the house wall and on the 
 same level, were three fragments of unbaked pottery 
 figurines and a fragment of coiled basketry. 
 
 House-site 3 was immediately east of that of house 
 2. The west wall of this house was built against 
 the east wall of house 2, as before mentioned. Both 
 the east and the west walls were of large stones set 
 on edge; the southern one of smaller stones, laid up. 
 All the walls stood 2 feet high. The floor plan was 
 irregular. Its northern side along the rock shelter 
 wall was 11 feet 8 inches long. The average dimen- 
 sions of the house were 6 feet 6 inches, by 9 feet 8 
 inches. No definitely defined floor was found, but 
 the level of a bed of ashes found six inches below 
 the surface and in the center of the room probably 
 
10 ROCK SHELTER 
 
 had been one that caught fire and burned downward 
 until it reached the stone fill beneath (pi. iv). 
 
 Outside of the burned area, in the first foot below 
 the surface and mixed with grass, were fragments of 
 matting, skin, skin twisted on strings, wood showing 
 marks of tools, arrow foreshafts, a hearth for fire- 
 making, a quantity of chews or quids and fragments 
 of crowns of sotol {Dasylirion sp.), cactus leaves 
 (Opuntia), pieces of gourd, corn-cobs and husks, cac- 
 tus seed-pods, pifion nuts, animal bones, stone chips, 
 a broken stone arrowpoint, a small grinding-stone, 
 a pitted hammerstone, and a small stone painted 
 with a black design. 
 
 In the second foot level, which exposed the bottom 
 of the west wall, the fragmentary material was simi- 
 lar to that of the first with the addition of two worn 
 sandals and two fragments of the carapace of a turtle. 
 In the northwest corner, 14 inches below the sur- 
 face, was a bundle of loose grass which covered a 
 fragment of a blanket-like object made of fine fiber 
 and strings. The fragment was folded and laid upon 
 a checker-weave mat, which in turn was placed upon 
 a piece of leather of about the same size. The grass 
 floor-coverings, which were laid before this house 
 was built, ended a few inches below the west wall. 
 
 Below the house just described and between the 
 rock shelter wall and a large rock, were three frag- 
 ments of an object that may possibly have been a 
 baby-carrier, made of grass, tie-twined together with 
 yucca leaves. One of the fragments was charred by 
 the fire before mentioned. 
 
BEE CAVE CANYON 11 
 
 Just east of these fragments and along the rock 
 shelter wall was a bed of grass, 5 feet by 5 feet 4 
 inches in size. Below the grass was a framework of 
 buckeye branches tied together at different points 
 with strands of fiber. A bent piece of cedar laid 
 on this framework, with the concave side toward the 
 wall, caused a ridge under the grass and may have 
 acted as a partition to divide off the space next to 
 the wall, to be used as a bed for a very small child 
 (pl. v). 
 
 Under the second cross-bar, formed by a branch, 
 toward the eastern end, was a small mass of soft 
 grass which, upon being removed, disclosed a digging 
 stick lying parallel to and under the cross-bar, and 
 a net bag. 
 
 Eight inches below the bed just described was an- 
 other one made of grass on a framework of branches. 
 In removing the grass, which was well worn, a small 
 child's sandal was found. The bed was in a depres- 
 sion and rested on the undisturbed talus (which at 
 this point was of very fine stones), except in the 
 center where there was a hole 15 inches in diameter 
 and 5 inches deep which had been filled in with stone 
 and ash. It was of the same length as the upper 
 bed but narrower by about one foot. Scattered be- 
 tween the branches under the lower bed were quan- 
 tities of small seeds, nuts, animal bones, sotol and 
 cactus leaves, a burned fragment of antler, an arrow- 
 point, pieces of matting, and a worn sandal. The 
 two beds and the baby-carrier were quite separate 
 from the debris of house-site 3 above them. 
 
12 ROCK SHELTER 
 
 On the wall of the rock shelter, about midway 
 between the remains of the east and west walls of 
 house 3, were five impressions made by hands smeared 
 with red oxide of iron, four of which, placed in pairs, 
 one above the other, faced directly into the room. 
 The thumbs of the upper pair were placed opposite 
 each other, while the thumbs of the lower pair faced 
 outward. The lower left impression was 6§ inches 
 long and was 3 feet above the surface of the room 
 before excavating began. The fifth impression was 
 that of a left hand placed on a fractured surface 
 facing southwest and a few inches to the west of the 
 four above described. The imprint of the fingers 
 was elongated, as though the person had drawn his 
 hand down with the fingers touching the wall for a 
 short distance before pressing the palm upon it. 
 The bottom of this imprint was about in line with 
 the top of the highest of the other four. About six 
 inches below the imprint last described there was a 
 smear on the wall, as though the hand had been 
 wiped on it after making the impression. 
 
 There were traces of smoke blackening on the wall 
 at the eastern end of the house. 
 
 House-site 4 was east of that of 3, and the space 
 between the walls of the two houses was filled with 
 stone, making a wall varying from 3 feet 6 inches 
 to 4 feet in thickness. The floor plan was oval in 
 shape, with one end truncated. 
 
 The east and west walls were built of both large 
 stones on end, and small stones laid up, while the 
 
BEE CAVE CANYON 15 
 
 southern section was composed of large stones on 
 end. The bases of the east and west walls were 32 
 inches below the surface, and the base of the south- 
 ern portion was slightly higher. Two fragments of 
 crude metates were incorporated in the walls, one in 
 the east (pi. vi) and one in the west. The east wall 
 was smoke blackened. 
 
 While cleaning out the house to the level of the 
 low spot next to the rock shelter wall, an almost 
 complete sifting basket was uncovered, set on edge 
 against the west wall. Six inches below this level 
 was a remnant of a grass floor, the center of which 
 was covered with ash and small burned stones. Be- 
 tween the eastern edge of this area and the east wall 
 of the house, and on the floor, was about a bushel 
 of pieces of leaves which had been stripped from the 
 outside of roasted crowns of sotol, and in the north- 
 west corner was about half this amount of the same 
 material. 
 
 Scattered through the debris encircling the room 
 were numerous bones of animals and birds (many of 
 which were burned and split), an arrow foreshaft, 
 two paint-sticks, six reed pipes, a fragment of a shell 
 pendant, an arrowpoint, fragments of matting and 
 leather, wood showing marks of tools, a hearth for 
 fire-making, gourds, seed-pods, seeds, stone flakes, 
 pieces of antler, a grinding-stone, and the sotol leaves 
 already mentioned. Finding such a quantity of 
 roasted sotol fragments and animal bones surround- 
 ing the burned area in the center of the room sug- 
 
16 ROCK SHELTER 
 
 gests that the house, after falling into disuse as a 
 habitation, was employed as a pit for cooking. 
 
 The east and west walls extended about two inches 
 below this floor. Slightly below the floor were the 
 charred ends of three posts which stood in a row 
 about one foot east of the center of the room. Their 
 lower ends had been hacked and broken off, as 
 though a stone tool had been used, and they had 
 been driven into the stones of the talus below. 
 
 The grass floor-coverings extended under and down 
 9 inches below the walls of house 4, and the fill of 
 rock and debris another 15 inches. The grass was 
 badly burned, and with it everything that it might 
 have contained. 
 
 Between the rocks of the fill below were several 
 fragments of sandals and checker-weave matting, a 
 short piece of fiber string with two seed and three 
 reed beads threaded on it, three arrowpoints, animal 
 and bird bones, several grinding-stones and pitted 
 hammerstones, a fragment of a wooden fire-tongs, a 
 painted stone, a notched rhythm stick, a fragment 
 of an unbaked pottery figurine, a quantity of small 
 seeds, several Mexican walnuts, a stick of wood 
 charred at one end and having two strips of leather 
 attached with sinew to the other. 
 
 On the wall of the rock shelter, about midway 
 between the east and west walls of house 4, was 
 limned an outline of a left hand and wrist. Inside 
 the outline, which was dark red, the surface was 
 clean with the exception of a trace of lighter red on 
 
COFFIN — ROCK SHELTER 
 
 ; ■ - A y 
 
 OUTLINE OF HAND AND WRIST, HOUSE-SITE 4 
 Length of hand, 8.5 inches 
 
n 
 
 02 o 
 
BEE CAVE CANYON 19 
 
 the two center fingers and across the knuckles. Out- 
 side there were smudges of smoke blackening on the 
 left side of the wrist and between the thumb, first, 
 second and third fingers. The left outline of the 
 third finger was missing and the tips of thumb and 
 fingers were not completed. The total length of 
 this outline was 13j inches (pi. vn). 
 
 To the south, adjacent to the combined east and 
 west wall of house-sites 3 and 4, and on the surface, 
 was a fireplace or pit 26 inches in diameter by 7 
 inches deep, which was filled with ash. It was well 
 defined, having no doubt been cleaned out many 
 times. The grass floorings around and below the pit 
 were charred, but remained firm. 
 
 House-site 5 was excavated by Mr. M. R. Har- 
 rington in 1928. Its inside northwest corner was 7 
 feet 9 inches from the inside northeast corner of 
 house-site 4. The west wall stood on a grass floor- 
 covering 15 inches thick, in which a pendant of 
 steatite was found. Between this flooring and the 
 stone fill, 12 to 15 inches below and under the west- 
 ern part of the house, three wooden plugs and a piece 
 of coiled basketry were recovered. The removal of 
 a large stone from the center of the south wall ex- 
 posed a pouch made from a small checker-weave 
 mat, containing a quantity of corn and squash seeds. 
 
 When house 6 was built, three large stones which 
 lay on the surface were incorporated with other 
 stones on edge to form the wall. The floor, which 
 was immediately below the surface, was of well worn 
 
20 ROCK' SHELTER 
 
 grass. A fireplace, in the shape of a rounded pit, 
 was midway of the room, not far from the southern 
 wall. It was filled with ashes to the depth of 7j 
 inches, and the grass adjoining was charred. The 
 layers of the grass floor-coverings under the house 
 were 9 inches thick and lay on the talus. There was 
 no rock or other fill below. 
 
 Between the grass floor-coverings and next to the 
 large stone at the eastern side of the house were 
 found the left half of the lower jaw and part of the 
 frontal bone of a human skeleton, two fragments of 
 metates colored red on one side and black on the 
 other, and a rubbing stone. Near by was a fragment 
 of a sandal and a short length of reed inserted into 
 what appears to be a piece of dried vegetable sub- 
 stance. 
 
 Along the rock shelter wall, to the west of the 
 house-sites above described, were three groups of 
 stones which were flat on the surface. Their posi- 
 tion suggests that at one time they formed low walls 
 which have been knocked down by the sheep and 
 goats that today occupy the rock shelter during the 
 winter months. While it was impossible to measure 
 these enclosures accurately, they are in such a posi- 
 tion as to indicate rooms about 8 by 9 feet in size. 
 If assembled, the stones would have made a wall two 
 tiers, or about 9 inches high. 
 
 Remnants of seven similar structures were located 
 along the wall in the western part of the rock shelter. 
 They were built of loose stones on the talus and the 
 
BEE CAVE CANYON 21 
 
 floors covered with grass mixed with which strings 
 and fibers were found. There were also eight spots 
 along this section of wall which had been lightly 
 covered with grass and had probably been used for 
 camping. One of these, in a depression, was cov- 
 ered a little more thickly than the others, and was 
 littered with twigs, leaves, animal bones, fiber strings 
 and corn-cobs. 
 
 CAVES 
 
 The mass of stones previously mentioned, which 
 had fallen from the middle of the arch of the rock 
 shelter roof, formed several small caves, two of which 
 showed signs of human habitation. The larger, cave 
 1, was 37 feet 9 inches in extreme length, and about 
 9 feet wide. The smaller was approximately 9 feet 
 square. Cave 1 had been occupied at its northeast- 
 ern end. Back of a retaining wall four to five inches 
 high, formed of three stones placed end to end be- 
 tween a large rock and the southerly wall, was a bed 
 of grass, leaves, twigs and litter. In front of the 
 retaining wall was a patch of grass flooring one to 
 two inches thick. A fireplace 18 inches deep was 
 close to the edge of the flooring and 6 inches from 
 the northerly wall. It was filled with ashes and the 
 wall was smoke blackened. 
 
 Behind a stone, against the southerly wall and 
 at the edge of the grass flooring, were a pair of worn 
 fiber sandals, and scattered through the grass of the 
 bed were a corn-cob, fragments of checker-weave 
 
22 ROCK SHELTER 
 
 matting, coiled-weave basketry, reeds, pieces of wood 
 bearing marks of tools, fiber, strings, gourds, a hearth 
 for fire-making, a notched end of an arrowshaft, a 
 bowed twig, and a fragment of a curved grooved 
 stick, 8 inches long, such as is used for rabbit hunting 
 by some of the living tribes of the southwestern 
 United States today. Upon removal of the retaining 
 wall before mentioned, the grass flooring was found 
 to run under the wall and bed. 
 
 On the northerly face of a large rock at 9-G, 10-G, 
 and on the upper face and edge of a large split rock, 
 18-J, 19-J (see map), were numerous grooves which 
 may have been used to help shape and sharpen bone 
 and other implements (pi. vin). 
 
 FIRE-PITS 
 
 To the east of the rock pile at the center of the 
 shelter were two depressions, the larger without, and 
 the smaller within the area covered with grass. 
 Both these depressions proved to have been fire-pits. 
 The greater one was bare, but the smaller was sur- 
 rounded with pieces of roasted sotol crowns and 
 other refuse. 
 
 Objects Recovered from the Excavations 
 
 stone artifacts 
 
 Many grinding-stones, few of which are over 6 
 inches in length, a size convenient to operate with 
 
BEE CAVE CANYON 23 
 
 one hand, were found. The coarser and harder ones 
 are neatly pecked to shape, but some of the finer 
 grained and softer are ordinary brook stones. The 
 worn surfaces are of interest as they show the motion 
 used while grinding. Some are worn flat by steady 
 pressure, some are convex from rocking, and others, 
 especially the softer, finer grained type, are worn to 
 a sharp edge on one side. A few are pitted on one 
 or both sides, showing a secondary use as hammers. 
 
 Beside the two crude metates used in construction 
 of house 4, and one found under the surface, only a 
 few fragments of others were recovered. 
 
 Pitted hammerstones were abundant, and about a 
 dozen roughly chipped spherical stones, which prob- 
 ably were used as hammerstones, were also found. 
 
 Many chipped or flaked stone implements oc- 
 curred, such as arrowpoints, scrapers, saws, knives 
 and spear-points. The largest of the scrapers is 5§ 
 inches long. 
 
 A fragment of a grooved axe or maul, and a frag- 
 ment of what appears to have been a bowl made 
 from a nodule with the rim rounded by grinding and 
 the outside pecked were found in the fill under the 
 floor-covering, near the wall about in the center of 
 the shelter. 
 
 Other finds within the shelter were: 
 
 Several fragments of flat, fire-blackened stones, 
 which may have been used as griddles or baking 
 stones. 
 
 Two rough stones with fiber cords attached. 
 
24 ROCK SHELTER 
 
 Two fragments of tubular smoking pipes, or " cloud 
 blowers," made of limestone. 
 
 Four small unworked stones wrapped with grass. 
 The ends of the wrapping on one of these are 
 twisted together, while the ends of the grass on two 
 of the others are held in place by cactus spines. 
 Protruding from under the wrappings are what ap- 
 pear to be thin pieces of some vegetable substance. 
 It is possible that these objects were used as charms 
 (fig- 1). 
 
 Fig. 1. — Stones wrapped with grass. Maximum length, 2.6 in. 
 (16/8706-08) 
 
 A number of brook stones and flakes from the 
 shelter wall were decorated with painted designs, 
 two with red, the others with black. The designs, 
 for the most part, although quite clearly defined, are 
 not suggestive of interpretation (pi. ix). 
 
COFFIN ROCK SHELTER 
 
 STONES DECORATED WITH DESIGNS IN BLACK OR RED 
 Maximum length, 3.5 in. (16/8589, 8594, 8596, 8599, 8600) 
 
COFFIN — ROCK SHELTER 
 
 STICKS OF WOOD WITH TENON-LIKE ENDS 
 Length of b, 7.4 in. (16/8619) 
 
BEE CAVE CANYON 27 
 
 PAINTS 
 
 Black, red, and yellow paints were employed by 
 the dwellers in the rock shelter. On some of the 
 painted stones charcoal appears to have been the 
 basis of black paint, on others vegetable juices were 
 the chief ingredient. Oxide of iron was used for red, 
 and ochre for yellow. The beveled ends of thin 
 pieces of stone served to apply the colors, and a num- 
 ber of these were found with the paint adhering to 
 them. 
 
 WOODEN IMPLEMENTS 
 
 A number of round sticks cut at each end, pos- 
 sibly gaming sticks, and two short triangular ones 
 which may have been dice were found. 
 
 Among objects of unknown use are five lengths of 
 wood, round, and with one end cut like a tenon. 
 The opposite ends of two are tool marked (pi. x, 
 c, d). The one marked e is splintered, and the re- 
 maining two are charred. 
 
 A similar object, 9| inches long and yq of an inch 
 in diameter, was found back of a cave at the mouth 
 of Rotten Draw, about two and three quarters of a 
 mile south by east of the shelter. This specimen 
 had a tenon at each end, one at an angle of about 
 forty-five degrees in relation to the other. 
 
 An entire rhythm stick, llj inches long, with 
 twenty-nine notches, and several fragments of others, 
 some of which are very crude, were recovered. 
 
 Many fragments of fire-making drills and hearths 
 
28 ROCK SHELTER 
 
 remained. Only one of the drills was complete. It 
 measured 25f inches in length, and had a rough 
 point at the upper end, proving that it was of the 
 type manipulated by twirling between the hands. 
 
 Small bows, about a foot in length, made from a 
 branch or twig, and bent with a string of fiber, were 
 the only bows recovered. These were probably 
 toys, for although they are strong enough to rotate 
 a fire-drill, it is doubtful if they were used as such, 
 for no parts of the fire-drills found show evidence of 
 having been used in that manner. 
 
 Four fragments of atlatls, or throwing-sticks, all 
 from the end in which the nock of the arrow or spear 
 rests, lay in the same levels as notched arrows. The 
 upper sides of three of them are flat, while the upper 
 surface of the fourth slopes slightly toward the cen- 
 ter. The grooves are round. The under sides are 
 all convex. Two of these specimens had been sev- 
 ered from the rest of the implement, probably after 
 accidental damage to the prongs, by cutting or saw- 
 ing part way through from either side and breaking 
 the remaining fibers. 
 
 Three fragments of curved sticks (rabbit sticks), 
 grooved along the sides, were recovered, but no en- 
 tire specimen. 
 
 Tapered plugs, saw-grooved, and broken off at one 
 end and battered at the other, may have been used 
 as stoppers for gourd bottles. One of these has been 
 cut off and made smaller at one end, the tool marks 
 showing where this operation ceased. In both ends 
 
BEE CAVE CANYON 
 
 29 
 
 of this specimen are what appear to be wedges (al- 
 though these may be the projections of one piece 
 running all the way through, as they are in line) of 
 a material, in its 
 present state, as soft 
 as, and resembling, 
 the edge of a piece 
 of gourd. 
 
 A number of bat- 
 tered and hacked 
 pieces of wood which 
 may have been used 
 as stakes to drive in 
 the ground, or as 
 wedges, were found. 
 
 Many broken and 
 cut fragments are 
 probably remains of 
 tool handles and im- 
 plements. One of 
 these is probably the 
 partly sawed and 
 broken off handle of 
 a rabbit stick, rough- 
 ened to afford a grip. 
 
 A thin stick 4| 
 inches long, flattened 
 
 on tWO sides, is FlG - 2.— Pointed 
 
 • , 1 11 1 stick with leather 
 
 pointed and charred straps Length of 
 at one end. On the stick, 4.1 in. (16/8685) 
 
30 ROCK SHELTER 
 
 other end are bound two pieces of leather which ap- 
 parently originally formed loops, one on each flat 
 side. One of the loops is broken, and part of the 
 other is missing (fig. 2). The object suggests an im- 
 plement held between two fingers passed through 
 the loops, possibly a corn sheller. 
 
 An object Ye of an inch in diameter and 3 inches 
 long is wrapped on one end with sinew. A hole yq 
 of an inch in diameter and 1 inch deep is drilled in 
 one end. The other end is cut off, and judging by 
 the surface this object was used after the cutting 
 took place, as it is more or less smoothed and rounded. 
 Other similar sticks ranging up to 8 inches in length 
 were also found, with drillings in one or both ends, 
 though some do not show usage on the cut end. 
 
 Other sticks have depressions in the upper ends 
 from Y2 to yq of an inch deep. The lower ends are 
 partly cut, then broken off. These objects fit well 
 when laid in the groove of the fragment of the throw- 
 ing-stick before described, with the depressions en- 
 gaged with the spur of it. They may have been ends 
 of light projectiles. 
 
 Other wooden implements were: 
 
 Several pointed digging sticks, varying from 20 to 
 42f inches long, and ranging from f to 1 inch in 
 diameter. 
 
 A number of sides of wooden tongs used for trim- 
 ming the fire, and for other purposes. These had 
 been flattened on one side and grooved at one end, 
 so that when a pair of them is bound together, the 
 
BEE CAVE CANYON 31 
 
 flat surfaces face to face, a very efficient tool for 
 handling embers is obtained. They had all been 
 used, as shown by the charring of the unbound ends. 
 
 Two finely finished wooden implements, pointed 
 at one end and grooved at the other, probably used 
 as paint-sticks. 
 
 Two twigs bound at both ends with fiber strings, 
 probably parts of snares. 
 
 A flat piece of wood f of an inch wide by 2\ inches 
 long, decorated with three black painted designs, 
 one of which appears to represent a dragon-fly, an- 
 other possibly a butterfly, while the last is a rectan- 
 gular oblong outline. 
 
 Notched foreshafts of wood, for projectiles used 
 with throwing-sticks. 
 
 Some complete, and many fragments of wooden 
 foreshafts for arrows. A few of these were pointed 
 at one end and notched at the other, for the purpose 
 of mounting points therein. The majority of these 
 foreshafts, however, were pointed at each end. One 
 of the latter type was found inserted in and attached 
 to a fragment of reed arrowshaft. Two fragments of 
 notched foreshafts were found with fragments of 
 arrowpoints in place, held there by sinew wrapping. 
 
 REEDS 
 
 Of this material there were: 
 
 Fragments of the tapered head ends of arrow- 
 shafts. 
 
 A cut and tapered piece of reed inserted into 
 another section of reed cut for the purpose. 
 
32 ROCK SHELTER 
 
 Fragments of the nock end of reed arrowshafts 
 with wrappings and shreds of feathers in place; two 
 specimens showing that they had been cut from the 
 shaft. All of the nock ends when wrapped had been 
 reinforced by driving pieces of reed or wood into 
 them. 
 
 Small reeds which had had feathers attached to 
 one end, parts of the bindings and feathers being still 
 in place. 
 
 A short piece of reed inserted into what appears to 
 be a dried vegetable substance. 
 
 Pipes were found made from two sections of the 
 stem of a reed. One end is cut off close to, and the 
 other at a distance from the joint; the dividing par- 
 tition between the sections is punctured. The short 
 ends served as mouth-pieces while the longer ends 
 served as bowls, these being in many cases badly 
 charred and shortened through use. A few of the 
 bowls still contained the dottle, one of which was of 
 cedar sprays. 
 
 Tubular containers were made of reed; three of 
 these were found between two clumps of grass, under 
 the edge of a large stone, near the bottom of the 
 trench and below the grass floor-covering. They 
 were filled with what appear to be seeds of Amaran- 
 thus paniculatus, 2 and were stoppered with grass 
 plugs. The tubes were cut similar to the reed pipes, 
 but were longer, and the partitions between the sec- 
 tions were not punctured. 
 
 2 Identified by the Bureau of Plant Industry, Washing- 
 ton, D. C. 
 
BEE CAVE CANYON S3 
 
 ANTLER PESTLES AND TOOLS 
 
 The following specimens were found: 
 
 Several pieces of antler, cut and rubbed smooth 
 at the ends, that may have been used as pestles. 
 
 The basal segment of an antler, hollowed out ye 
 of an inch in diameter and f of an inch deep, on one 
 end, the inside of which is charred. 
 
 A fragment of an antler implement, possibly a 
 flaking tool. 
 
 An implement, the inner surface of which is con- 
 cave, and the ends of which are beveled. It was 
 probably used as a skin scraper. 
 
 Pieces of split antler, the cut surface and edges 
 finished smooth as by rubbing. 
 
 BONE IMPLEMENTS 
 
 Bone implements, awl, chisel and gouge shaped, 
 were quite plentiful, and well polished from use. 
 They do not differ in type from those found so 
 abundantly in Arizona and New Mexico. Two frag- 
 ments of bone have as designs scorings of parallel 
 straight lines. One bone fragment is decorated by 
 a series of black paint dots running lengthwise, and 
 another shows traces of having been smeared with 
 red paint and striped with black. 
 
 Three implements made of scapulae, probably 
 those of deer, were broken away so as to retain the 
 spine and one fossa. The edges of the implements 
 are worn quite smooth and thin at the point, and 
 
34 
 
 ROCK SHELTER 
 
 one has what appears to be some vegetable substance 
 adhering to it. For use as scoops these implements 
 are sturdy and fit well in the hand. 
 
 GOURD VESSELS 
 
 Fragments of gourd vessels, many of which have 
 been mended by placing a caulking of soft fiber over 
 the crack before lacing together with fiber strings, 
 were recovered. A few fragments appear to have 
 been perforated with a friction drill, and one has 
 many indentations, apparently made with a hot, 
 pointed instrument. 
 
 Fig. 3. — Detail of wrap-woven edge and warp strand of mat shown in 
 pi. XI 
 
COFFIN ROCK SHELTER 
 
 FRAGMENT OF GRASS MAT 
 Length, 6.3 in. (16/8688) 
 
36 ROCK SHELTER 
 
 MATTING 
 
 Many fragments of matting of yucca leaves were 
 found, both of the checker type, and of the over-two 
 under-two twilled-weave technique, also three small 
 mats, two of which are of checker-weave, and one 
 of twilled, and a small checker-weave object with 
 several strands allowed to extend out from one side. 
 
 There were also fragments of a mat made of un- 
 identified grass, the edge wrap-woven with a two- 
 strand twisted fiber string. Three inches from the 
 edge the matting was held together by piercing and 
 stringing the grass leaves on a two-strand twisted 
 fiber cord (pi. xi and fig. 3). Fragments of a similar 
 mat were found under the remains of a child in cave 
 3, around the corner and a little to the north of the 
 shelter. 
 
 BASKETRY 
 
 A pouch about four and one-half inches square, 
 made by folding a small checker-weave mat in the 
 middle and stitching the ends and one edge together, 
 was found buried a few inches below the surface 
 and next to the outer side of a stone, forming part 
 of the south wall of house 5. It was filled with 
 about an equal quantity of corn and squash seeds. 
 Part of the matting at the opening of the pouch 
 had been broken away. It was closed by stuffing 
 in a small quantity of soft grass and stitching with 
 a narrow yucca leaf; a fiber cord was tied both ways 
 around the pouch (fig. 4). 
 
BEE CAVE CANYON 
 
 37 
 
 A cylindrical checker-weave basket four inches 
 high, containing a piece of hematite, five small pieces 
 of sinew, and a quantity of fine fiber, may be part 
 
 i 
 
 ■ ....iA : '•-■■'■ ,:■,-:■ w-; -',r, ■ « :; 
 
 ; - 
 
 
 Fig. 4. — Pouch containing corn and squash seeds. Length, 5 in. 
 (16/8690) 
 
 of a fire-making outfit. They were found while 
 scraping off the floor of the shelter preparatory to 
 
38 ROCK SHELTER 
 
 excavation. The only fire-making tools found be- 
 low the surface were drills and hearths. 
 
 Two sifting-baskets of the tray type were made 
 of yucca leaves, with the strands separated xg- to 
 | of an inch. Attached to each of these is a frag- 
 ment of a rim or binding, made of several yucca 
 leaves one laid on another and bound to the edge 
 of the basket with an open-coil stitch. Many frag- 
 ments of basketry with similar binding were found. 
 
 Of several fragments of coiled basketry, all in a 
 very poor state of preservation, three are open-coiled, 
 and nine split-coiled. The foundation of most is 
 splints; one of the open-coiled fragments has a foun- 
 dation of the small stems of some plant. A frag- 
 ment of a base of another specimen was repaired by 
 stitching with yucca fiber. 
 
 A piece of coiled basketry, about two inches in 
 diameter, had both the foundation and the coils of 
 yucca leaves. Two narrow leaves on edge form the 
 foundation and split leaves form simple interlocking 
 coils. 
 
 Many small fragments showed a weaving tech- 
 nique similar to the sandals found in the shelter, to 
 be described later (pp. 43-49). 
 
 NETTING 
 
 Two fragments of netting were made of fiber 
 strings, tied with a netting knot, the meshes approxi- 
 mately If and 2| inches respectively. 
 
COFFIN ROCK SHELTER 
 
 NETWORK BAG 
 Length, 12 in. (16/8719) 
 
COFFIN ROCK SHELTER 
 
 DETAIL OF NETTING TECHNIQUE OF NETWORK FRAGMENT 
 
BEE CAVE CANYON 41 
 
 A network bag 12 inches long, of fiber string, has 
 a mesh averaging f of an inch. A draw string runs 
 through the upper or last mesh; a twisted cord 9 
 inches long with a knot at its end is attached to the 
 center of the bottom (pi. xn). 
 
 The meshes of two other fragments of netted bags 
 measure f and 2j inches approximately. 
 
 Other fragments of netting were made of split 
 yucca leaves. The intersections are not tied, and 
 they are attached to sticks of wood. The net ap- 
 parently was made on a frame, the sticks forming 
 part of it, and the strands of yucca were stretched 
 across, first in one direction and then in the other. 
 Where a strand crossed another it was wrapped one 
 turn around the one already in place. 
 
 A fragment of network, in bad condition, made of 
 split yucca leaves, shows an irregular sized mesh 
 consisting of a series of loops, tied with slip knots, 
 interlaced with other like series. Plate xiii depicts 
 a cotton string netted to illustrate this technique. 
 
 Open-coiled work without foundation, made of 
 fiber strings, was exemplified in a number of frag- 
 ments. One specimen, when held in its proper posi- 
 tion, forms what may have been a cap. Near the 
 edge there are ten coils and six loops to the square 
 inch and near the center seven coils and five loops. 
 Another fragment is of interest because of the extra 
 twist used in making coils. 
 
42 ROCK SHELTER 
 
 STRING 
 
 Examination of the large number of pieces of string 
 show that a variety of fibrous vegetable material was 
 used in making them. The majority appear to be 
 made of the fibers of the leaves of different species 
 of yucca which grow in the neighborhood. 
 
 Many pieces of string of a softer fiber were colored 
 red and yellow; a few are made of grass and some 
 of a material as soft as cotton. 
 
 Many times, split yucca leaves and untwisted fiber 
 strands were used as binders or tie strings. Most 
 of the string found was twisted, varying from y^ to 
 f of an inch in diameter, and of one, two, three, four, 
 and five twisted strands. A few four-strand strings 
 were made by twisting together two two-strand 
 strings. 
 
 Sometimes in making string of a leaf, the point 
 was left intact, thus keeping the end from fraying, 
 and forming a lacer-like tip. A few fragments of 
 sandals have tie-strings of this type attached to 
 them. 
 
 Fringe-like fragments made of one- and two-strand 
 cords of twisted fiber, the units of the fringe hanging 
 down, were probably formed by tightly twisting and 
 looping sections of the main cord. One fragment 
 has two complete units 15f inches long. 
 
 Many pieces of twisted cord have tapering ends. 
 In one case two such pieces are tied together with a 
 square knot, forming a cord almost eight feet long, 
 with the tapered ends free. A twisted fiber cord, 
 
BEE CAVE CANYON 
 
 43 
 
 complete in itself, one end tapered and the other end 
 formed in a loop, was found in a small coil. Several 
 tie-twined bits of fiber string were also collected. 
 
 KNOTS 
 
 The square or reef-knot was generally used. Out 
 of over one hundred knots examined, only three were 
 granny, and two square bow-knots. Half-hitches 
 and slip-knots were also used, the former probably 
 in tying to an object, the latter in snaring. 
 
 Fig. 5. 
 
 -Sandals of yucca, braided, with ends woven in. 
 length, 6.5 in. (16/8814, 8817) 
 
 Maximum 
 
 SANDALS 
 
 A great many sandals made of yucca were found, 
 but unfortunately, almost all of them are badly worn 
 and have but few tie-strings still attached. 
 
44 
 
 ROCK SHELTER 
 
 The sandals may be said to have been braided in 
 two ways: First, several strands braided together; 
 second, one or more strands braided with two op- 
 posed elements. The plaiting of the former is irreg- 
 ular, the ends of the strands are woven back through 
 the braid and it is impossible to trace the whole braid 
 without destroying the specimens (fig. 5). Thick- 
 ness and strength were added to the larger sandals 
 
 Fig. 6. — Sandals of yucca, reinforced. Maximum length, 8.5 in. 
 (16/8812, 8817) 
 
 of this type by weaving in extra strands after the 
 braiding had been completed. The two specimens, 
 b and c, fig. 6, are braided, the latter having been 
 
BEE CAVE CANYON 
 
 45 
 
 reinforced with strands woven in lengthwise, while 
 the former appears to have been reinforced at the 
 ends with an overcast stitch that pierces the center, 
 and coils around the sides of the sandal. The coils 
 have been worn off on the under side. 
 
 The braiding with two opposed elements varies 
 but little (figs. 7-9). In fig. 7 the strand is kept flat 
 
 Fig. 7.- 
 
 -Detail of braiding technique with opposed elements, strands 
 kept fiat. 
 
 while braiding. In fig. 8 the technique is the same 
 as in fig. 7, except that the strand is turned over 
 each time it is passed between the opposed elements. 
 Figure 9 and fig. 7 show the same technique except 
 that in the latter two strands are braided instead of 
 
 When necessary, new strands were added to carry 
 on the work. Loose ends were allowed to project 
 
46 
 
 ROCK SHELTER 
 
 from the under side of the sandal. For opposed 
 elements, broad leaves, narrow leaves, shredded fiber, 
 and two-strand twisted fiber strings were used. 
 
 The sandal c, fig. 10, is made of broad leaves and 
 is constructed in reverse of the technique of fig. 9. 
 B of the same plate is made of narrow leaves and is 
 constructed according to the technique of fig. 9, with 
 
 Fig. 8. 
 
 -Detail of braiding technique with opposed elements, strands 
 turned. 
 
 extra strands woven in lengthwise. A, fig. 10, is 
 made according to the technique of fig. 8, one large 
 leaf forming the two opposing elements. 
 
 The three sandals depicted in fig. 11 are woven 
 in the technique of fig. 7. The opposing elements of 
 a are of yucca which has been shredded, except at 
 the ends which are tied together at the top. This 
 sandal appears to have been made from the bottom 
 
BEE CAVE CANYON 
 
 47 
 
 up, and the shredded yucca strand used to form the 
 opposed elements was wrapped several times spirally 
 around a small bundle of fiber at the start, to help 
 keep the elements apart. The sandal b is made of 
 short narrow leaves, three or four of them being used 
 for each opposing element; a single narrow leaf is 
 
 Fig. 9. — Detail of braiding technique with opposed elements, two fiat 
 strands. 
 
 braided for about five rows from the bottom, after 
 which two or more leaves are woven in at a time. 
 
 TIE-STRINGS 
 
 Judging by the tie-strings found attached to other 
 sandals and fragments of sandals, the tie-strings of 
 
48 
 
 ROCK SHELTER 
 
 the sandals shown in figs. 5, 6 and 10 were of yucca 
 leaf or fiber, and were attached as shown in fig. 12. 
 Six sandals similar to the one illustrated in b, fig. 
 11, were found. All of them are small, ranging from 
 4 to 5j inches in length, and were probably worn by 
 children. They appear to have been held to the foot 
 
 Fig. 10. — Sandals of yucca, braided with two opposed elements. 
 Maximum length, 7.8 in. (16/8812-13) 
 
 with a toe loop, and two side strings, which are 
 attached to the edges of the sandal half to three 
 quarters the distance from the toe end to the heel, 
 passed over the instep, and then tied to the toe loop. 
 
BEE CAVE CANYON 
 
 49 
 
 The opposing elements, or framework, and the 
 tie-strings of the child's sandal (fig. 11, c) were two- 
 strand twisted fiber cords. One long cord was used 
 to form the framework and the front tie-strings, as 
 shown in fig. 13. 
 
 Fig. 11. — Sandals of yucca, braided with two opposed elements, a 
 with shredded strands and c with fiber string. Maximum length, 7 in. 
 (16/8813-14) 
 
 A fragment of what may have been a sandal is of 
 checker-weave, with two strings attached to it, and 
 shows wear on one side. 
 
50 
 
 ROCK SHELTER 
 
 FABRICS 
 
 A fragment of woven material of blanket-like ap- 
 pearance, about 14 by 15 inches, is made of coarse 
 twisted strands of fine fiber, held together at inter- 
 vals by twined similar strands, and small tightly 
 twisted two-strand fiber strings (pi. xiv). 
 
 Fig. 12. — Technique of tie- strings of sandals. 
 
 Another specimen is a small corner piece of fabric 
 in which the warp strands, or two-strand twisted 
 strings about f of an inch in diameter, are twined 
 tightly together at lj inch intervals by smaller 
 strings. The edges are turned, and the twining is 
 
BEE CAVE CANYON 
 
 51 
 
 similar to that of the previous specimen described, 
 and in both specimens the technique resembles that 
 used in making rabbit-skin blankets. 
 
 Many twisted strips of skin such as are used for 
 making skin blankets were also found. Most of 
 them are wrapped spirally around fiber cords; a few 
 have bits of fur or hair still in place. 
 
 Fig. 
 
 13. — Technique of framework and tie-strings of child's sandal 
 fig. 11. 
 
 Of materials prepared for use in weaving there 
 were: 
 
 Quite a number of small bundles of shredded fiber; 
 leaves of sotol {Dasylirion sp.), yucca and other 
 plants, sorted or split to a size; grass, etc. Most of 
 the bundles are loosely wrapped near the center with 
 a tie-string of the same material as that in the bundle. 
 
COFFIN ROCK SHELTER 
 
 BLANKET-LIKE FABRIC OF SOFT FIBER 
 
 (16/8729) 
 
BEE CAVE CANYON 53 
 
 FIBER BRUSHES 
 
 A brush was made by folding a number of strands 
 of fiber so that the ends met, and then wrapping 
 them around with other strands of fiber. 
 
 An object made of fiber strands folded over a 
 loosely twisted fiber string and tie-twined below the 
 string was in a roll when found and had the appear- 
 ance of being a brush. ; 
 
 CACTUS THORNS 
 
 Several cactus thorns were bound together with a 
 fiber string. 
 
 BINDINGS 
 
 Numerous rings, from less than two inches to over 
 six inches in diameter, made of yucca leaves and 
 other fibrous plants, were discovered. They prob- 
 ably were used as bindings in transporting grass and 
 other material to the shelter. They were made 
 either by wrapping a single leaf, or leaves tied to- 
 gether, end to end, with a square knot; or by running 
 fiber strands around the bundle of material several 
 times and then repeatedly passing the loose ends 
 under and around the loop thus formed. Several 
 rings made of narrow leaves have the appearance of 
 having been twisted together before binding, but the 
 ends finish in a wrap and not in the knot which 
 would be necessary if the strands had been twisted 
 
54 ROCK SHELTER 
 
 before binding. All these rings are quite flat, prob- 
 ably made so by being trampled upon after having 
 been discarded on the shelter floor. 
 
 Many plain bindings and wrappings were found, 
 and they were almost invariably tied with one or 
 more square knots. 
 
 GRASSES AND LEAVES OTHERWISE EMPLOYED 
 
 Grasses were put to other uses than those already 
 mentioned. Among articles of this material found 
 were: 
 
 Rings of grass wrapped with yucca leaves, or other 
 fibrous vegetable substances. 
 
 Fragments of pads of matted grass with parts of 
 tie-strings attached. 
 
 Fragments of an object made by tie-twining to- 
 gether strands of grass with strips of yucca leaves. 
 The grass stems used in making the strands do not 
 exceed 9 inches in length, so the strands were made 
 continuous by overlapping small bundles of stems 
 as the work proceeded. Some of the strands are 
 loosely wrapped with narrow strips of yucca leaves. 
 The widest fragment has seven strands and is 5f 
 inches wide. Judging by the fragments found, the 
 bottom of the object was a crude network of yucca 
 strips tied to the lower ends of the tie-twined strips. 
 The rim was finished, after the final tie-twining knot 
 of the upper coil, by tying the ends of the tie-twining 
 strips in pairs with a square knot. The work prob- 
 ably was carried on from left to right as all the loose 
 
BEE CAVE CANYON 55 
 
 ends of the knots to the left are free, while the ends 
 to the right are caught under the following knot. 
 The tie-twining is separated by intervals ranging 
 from 1^ to 4 inches. The fragile construction of this 
 object must have made it unfit for heavy duty, and 
 the finding of the fragments just west of the beds 
 under house 3 suggests that they may have been 
 pieces of a cradle or baby-carrier which was discarded 
 when the lower bed was abandoned and the upper 
 bed made. 3 
 
 Fragments of an object made of sotol leaves twined 
 together to form a bag-like container, the bottom 
 part of which is missing. A large fragment of a 
 similar object, with a worn sandal tied across the 
 opening, was found in a cave at the mouth of Lower 
 Rotten Draw. 
 
 Fragments of a hammock-like specimen made of 
 sotol leaves twined together at intervals with split 
 leaves found in a fill 4 feet 2 inches below the surface. 
 This object, when found, was 35 inches long by 18 
 inches wide near the center, and was broken in two 
 by a worked stone which had been thrown in on top 
 of it. One of the ends, a ball of knotted leaves, was 
 broken off and lay a couple of inches away. Scat- 
 tered over the object were some light twigs and 
 leaves, and toward one end was a handful of grass 
 held together with a leaf. 
 
 3 Kidder and Guernsey in Archeological Explorations in 
 Northeastern Arizona, Bulletin 65, Bureau of American 
 Ethnology, Washington, 1919, illustrate, pi. 72, cedar-bark 
 cradles the sides of which are twined together. 
 
56 
 
 ROCK SHELTER 
 
 LEATHER REMNANTS 
 
 Many fragments of different sorts of leather be- 
 sides the strips previously mentioned (p. 30) were 
 found. The majority seem to be scraps, with the 
 
 Fig. 14.- 
 
 -Pendants of stone and shell. Length of mussel shell, 1.6 in. 
 (16/8682, 8683, 8696) 
 
 holes made in stretching still at the edge, and cut- 
 tings left after making garments or other articles. 
 Some of the pieces have been scored with series of 
 lines which form irregular squares. Four fragments 
 appear to have been parts of objects made for some 
 definite use, but there is not enough left fully to 
 determine what purpose they might have served. 
 
 ORNAMENTS 
 
 Few objects of personal adornment were found. 
 These were: 
 
BEE CAVE CANYON 
 
 57 
 
 Two flat pendants of stone, one of which has had 
 two perforations broken out, but in place of which 
 a third one had been drilled. A fragment of a 
 pendant made of the shell of a fresh-water mussel. 
 A pendant made of a land-snail shell, with a piece 
 of a fiber cord attached for suspen- 
 sion (figs. 14-15). 
 
 Two short strands of beads made 
 from seeds, and a short strand of 
 beads of cut sections of reed, 
 threaded on fiber strings (fig. 16). 
 
 Fig. 15. — Pendant 
 of stone. Length, 
 1.2. in. (16/8684) 
 
 CLAY FIGURINES 
 
 Two complete and twenty-four 
 fragments of small grotesque human 
 figures made of untempered and 
 unfired clay seem to have been the only fictile objects 
 possessed by the ancient inhabitants. The figures 
 
 Fig. 
 
 16. — Seeds and sections of reed strung like beads, 
 length, 4.6 in. (16/8704) 
 
 Maximum 
 
 consist of the head and torso; the noses are long and 
 sharp; the mouths are small round punctations; the 
 
58 
 
 ROCK SHELTER 
 
 eyes are part of the black paint decoration which 
 covers the upper parts of the figures. One fragment 
 has sharp pointed breasts, evidently representing a 
 female. Another has traces of red, yellow and black 
 painted decoration. The lower parts of the torsos 
 
 Fig. 17. 
 
 -Figurines of unfired clay. Maximum height, 2.9 in. 
 (16/8675, 8678, 17/8705) 
 
 are round. The backs are incurved and some have 
 the natural indented line down the buttocks. One 
 figure, with head missing, was wrapped around its 
 center with grass (fig. 17). These figurines are prob- 
 ably of the same class as others more or less like 
 them still found among many of the present day 
 tribes, and are a symbolism of fertility. 
 
BEE CAVE CANYON 59 
 
 VEGETABLE FOODS 
 
 Among the vegetable substances, most of which 
 could be used for food, were: 
 
 Quantities of prickly pear (Opuntia sp.) and other 
 small seeds, corn, mesquite beans (Prosopis glandu- 
 losa), acorns (Quercus), Mexican walnuts (Juglans), 
 Mexican buckeye or soap-berries (Sapinus), pifion 
 nuts {Pinas edulis), desert-willow pods (Chilopsis 
 linearis), pumpkin seeds {Cucurbita pepo), small 
 gourds (Cucurbita foetidissima), squash seeds, seeds 
 and pods of yucca, roots, some of which were no 
 doubt used as amoles are today, and strings of dried 
 cactus fruit of which some were tie-twined together. 
 Cacti of different species, many devil's heads (Ho- 
 molacephala texensis) were split in two horizontally 
 to get at the inside. Great care must have been 
 taken to rid them of the upper spiniferous halves, 
 as none were found. Small cacti (Echinocercus sp.) 
 were dried as were star cacti (Astrocarpus fissuratus). 
 The latter is said to be used by Mexicans in the 
 same manner as the peyote button. Opuntia leaves 
 were all of a large size; one leaf was cut and filled 
 with ash in such a manner as to resemble a quarter- 
 cut of pie, 2| inches thick and 4j inches long on the 
 straight sides. A sharpened piece of wood had been 
 thrust through the center. 
 
 Large quantities of corn-cobs were found, and 
 fragments of sotol crowns (Dasylirioji sp.). Innu- 
 merable quids or "cuds" of fiber, which are no 
 
60 ROCK SHELTER 
 
 doubt remnants of the latter discarded after chewing, 
 were scattered through the debris. 
 
 ANIMAL BONES 
 
 While quite a variety of animal and bird bones 
 were found, they were not numerous, considering the 
 size of the shelter and the quantity of vegetable 
 refuse it contained. Whether this points to dry 
 seasons and poor hunting during part of the summer 
 is a question. Most of the bones were broken in 
 order to obtain the marrow. Among the bones that 
 it was possible to identify are those of deer, rabbit, 
 badger, coyote and rat; while the fragments of a 
 large jaw and a tibia are probably those of a bison. 
 Fragments of the carapace of a turtle were also found. 
 
 Conclusions 
 
 The rock shelter in Bee Cave Canyon seems to 
 have been occupied at different intervals and for no 
 long time at any period; for the artificial deposits, 
 which varied from almost nothing at the edge, to 
 about four feet at the rear of the shelter, were sep- 
 arated from one another by layers of grass only a 
 couple of inches apart, as if the occupants had suc- 
 cessively inhabited and abandoned the site many 
 times, leveling and improving the floor with grass 
 each time they or others returned to reoccupy it. 
 In excavating the successive floorings, each layer of 
 deposits was removed, and within them were found 
 
BEE CAVE CANYON 61 
 
 many artifacts and other objects; but there was no 
 indication of any cultural difference in the layers, 
 regardless of their depth. 
 
 Although many fragments of notched arrowshafts 
 were found, there was no trace of a bow. The occur- 
 rence of the atlatl and the notched arrow, in deposits 
 indicating no great range of time, would seem to 
 suggest that the throwing-stick and the bow had 
 been used contemporaneously. 
 
 A notable feature is the scarcity of pottery, the 
 only trace of it being some unbaked figurines, and 
 three small fragments of vessels, one of which (part 
 of a bowl) is of thin brown ware with black painted 
 decoration. These potsherds were found on or near 
 the surface, and are doubtless intrusive. 
 
 The only small fireplaces found in the area covered 
 by grass were the one south of the combined east 
 and west wall of house-sites 3 and 4, and the one in 
 house-site 6. These were undoubtedly made by 
 sheep-herders who use the shelter in the cold months 
 as a fold. The litter of cardboard boxes, pieces of 
 leather and other rubbish on the surface south of 
 house-site 6, and the finding of a small tin box con- 
 taining beans, and burned wood from boxes, on the 
 surface in the house-site, confirm this opinion. 
 
 Leaving these two fireplaces out of consideration, 
 the only fireplaces left are the two large pits east of 
 the rock pile at the center of the shelter, and house- 
 site 4, which had been used as a fire-pit after falling 
 into disuse as a dwelling. 
 
62 ROCK SHELTER 
 
 Bearing in mind what has been said as to the 
 floor-coverings and fire-pits, the evidence points to 
 the supposition that the rock shelter was occupied 
 only as a summer camp. The inflammable material 
 used as floor-coverings prevented building fires for 
 warmth in the houses in the winter time. Cooking 
 or roasting in large pits suggests a communal mode 
 of living. Probably only one pit at a time was used. 
 The larger one outside of the covered area, judging 
 by the fact that it was bare of refuse, appears to 
 have been in use prior to the one within (see p. 22). 
 
 The finding of such large quantities of pieces of 
 sotol crowns and of corn-cobs suggests that corn was 
 raised in the valley, and that sotol was the main 
 food supply while waiting for the corn to ripen. 
 
 Other circumstances pointing to the theory that 
 the camp was used only in the summer time are that, 
 while numbers of hammerstones and manos were 
 transported into the shelter, but few metates, which 
 are heavy, were found. Also, the finding of so many 
 pieces of leather, all of which, with the exception of 
 four, appear to be discarded cuttings, suggests that 
 the people left the lowlands in the spring, traveling 
 light, and that such skins as were secured during the 
 summer were made up into garments or other objects 
 and taken away with them in the fall. 
 
 No whole gourd vessels were found, and most of 
 the fragments had been carefully mended — facts 
 which might indicate either that the gourds also were 
 brought in by the seasonal immigrants, or at least 
 that gourds were scarce. 
 
BEE CAVE CANYON 63 
 
 The only human remains discovered in the shelter 
 in 1929, besides the piece of a frontal and the left 
 half of a lower jaw-bone found in house-site 6, were: 
 three more left halves of lower jaw-bones found eight 
 inches below the surface in a triangle within two 
 feet of one another and about midway and seven feet 
 south of 9 I and 10 I (see map); a few charred frag- 
 ments of what appear to have been a tibia in cave 2; 
 and a tibia and a fragment of a skull in the rear of 
 cave 1. 
 
 It would be interesting to know if there is any 
 significance to be attached to the fact that four left 
 halves of lower jaws were found. 
 
 It would be impossible to say how many people 
 occupied the shelter at one time, but if the houses 
 both in the east and the west halves of the shelter 
 and the camp sites along the wall were all inhabited 
 contemporaneously, sixty souls could easily have 
 been accommodated. 
 
 The heavy grass floor-covering ended somewhere 
 behind the large rock pile, but a grass walk was laid 
 from there, between the loose stones on the surface, 
 to a point beyond, leading past some of the westerly 
 houses, in the direction of others and of the water 
 supply; and this is naturally the first step in building 
 up the floor of the western part of the shelter in a 
 manner similar to that of the east. 
 
APPENDIX 
 
 In a short account of his work done in Bee Cave 
 Canyon rock shelter in 1928, Mr. M. R. Harrington 
 refers to a burial he uncovered. 4 As this was the 
 only one encountered during the work done in this 
 shelter, and as some of the accompaniments are dif- 
 ferent from those found during the later exploration, 
 a quotation from the above mentioned report is per- 
 tinent. Mr. Harrington writes: 
 
 Beginning another trench between the pile of fallen 
 stones and the cliff, and continuing it eastward along the 
 rear of the shelter, we found almost immediately the 
 remains of a flexed skeleton, minus the skull, with which 
 appeared a bowl-shape coiled basket, part of a twined 
 openwork cigar-shape basket, fragments of a bag made 
 of fiber cord, many small beads made of cane, and part 
 of a necklace made of sections of the legs of some large 
 iridescent green beetle neatly strung on a fine fiber cord. 
 Among the vertebrae was a wicked-looking flint spear- 
 head, and near the pelvis lay a deposit consisting of 
 three red paint-stones and two flint knives. 
 
 Notes on the Ethnobotany of Bee Cave Canyon, 
 by Henry T. Fletcher, Bulletin 33, no. 3, West Texas 
 Historical and Scientific Society, Sul Ross State 
 Teachers College, Alpine, Texas, 1930, is also of 
 interest, Mr. Fletcher having devoted much time to 
 research and study in that region. 
 
 - 1 Indian Notes, vol. v, no. 3, Museum of the American 
 Indian, Heye Foundation, New York, July, 1928. 
 
 64 
 
INDEX 
 
 Acorns, 59 
 
 Adobe, walls laid in, 6, 9 
 
 Alpine, Brewster co., exca- 
 vations near, xi 
 
 Amaranthus paniculatus, 
 seeds of, 32 
 
 Amole, roots used like, 59 
 
 Animal bones, 10, 11, 15, 
 60 
 
 Antler: basal segment, 33; 
 flaking tool, 33; frag- 
 ments, 11, 15; pestles, 33; 
 skin scraper, 33; split, 33 
 
 Arizona, bone implements 
 resembling those of, 33 
 
 Arrows: foreshafts, 10, 15, 
 31; points, 10, 11, 15, 16, 
 23, 31; shafts: head ends, 
 31; nock ends, 32; notch- 
 ed end, 22; in same layers 
 as atlatls, 61 
 
 Ashes, chinking of, 6; in 
 fire-pits, 19, 20, 21; in 
 opuntia leaf, 59; levels of, 
 5, 9, 15 
 
 Astrocarpus fissuratus, 59 
 
 Atlatls, 28; in same layers 
 as notched arrows, 61; 
 projectiles for, 30 
 
 Awls, bone, 33 
 
 Axe, grooved, 23 
 
 Baby-carrier, 10, 54-55 
 Badger, bones of, 60 
 
 Bags, network, 11, 39, 41, 
 64; twined, 55 
 
 Baskets: bowl-shaped, 64; 
 cigar-shaped, 64; cylin- 
 drical, 37; pouch-shaped, 
 19, 36, 37; sifting, 15,38; 
 technique of: checker, 19, 
 36, 37; coiled, 9, 19, 22, 
 38, 64; twined, 64 
 
 Beads, of reed, 16, 57, 64; 
 seeds strung as, 16, 57 
 
 Beds of grass, 11, 13, 21-22 
 
 Bee Cave Canyon, Brewster 
 co., named by U. S. Geol. 
 Survey, ix; report of M. 
 R. Harrington on, ix, 64; 
 researches of H. T. Fletch- 
 er in, 64; rock shelter in, 
 ix, xi, 1, 60; topography 
 of, 1-5 
 
 Beetle, necklace of legs of, 64 
 
 Bindings of grass and yucca 
 leaves, 42, 53-54 
 
 Bird, bones of, 15, 60; picto- 
 graph of, 5 
 
 Bison, bones of, 60 
 
 Black paint, bones deco- 
 rated with, 33; figurines 
 decorated with, 58; me- 
 tate coated with, 20; pot- 
 tery decorated with, 61; 
 stones decorated with, 10, 
 24-25; wood decorated 
 with, 31 
 
 65 
 
66 
 
 ROCK SHELTER 
 
 Black Peak, Brewster co., 
 rock shelter near, 1, 5 
 
 Blanket-like fabric, 10, 50, 
 52 
 
 Blankets, skin, 10, 51 
 
 Bone, implements of, 33-34 
 
 Bones: animal, 10, 11, 15, 
 60; human, 20, 36, 63, 64 
 
 Bowl, stone, 23 
 
 Bows, of twigs, 22, 28; and 
 atlatls discussed, 61 
 
 Brewster county, rock shel- 
 ter in, 1 
 
 Brushes, fiber, 53 
 
 Buckeye, bed frame of, 11; 
 berries of, 59 
 
 Bureau of American Eth- 
 nology, Bulletin 65 of, 
 cited, 55 
 
 Bureau of Plant Industry, 
 seeds identified by, 32 
 
 Burial, child's, 36; flexed, 64 
 
 Butterfly, representation of, 
 31 
 
 Cactus: dried, 59; leaves, 
 10, 11, 59; seed-pods, 10; 
 thorns, 24, 53; varieties 
 of: devil's head, 59; 
 prickly pear, 10, 11, 59; 
 star, 59 
 Cap of open coiled work, 41 
 Caves, 21-22, 36; grooved 
 rock near entrance to, 18, 
 22 ; objects in : arrowshaft, 
 22; basketry, 22; bed, 21- 
 22; burial, 36; fiber, 22; 
 fireplace, 21; gourds, 22; 
 grass flooring, 21; hearth, 
 22; human remains, 36, 
 63; matting, 22; rabbit 
 
 stick, 22; reeds, 22; "san- 
 dals, 21; string, 22; tool- 
 marked wood, 22; twig 
 bow, 22 ; occupancy of , 21 ; 
 of Rotten Draw. 27, 55 
 
 Cedar sprays, dottle of , 32 
 
 Chalk Draw, Bee Cave 
 Canyon opening into, 1, 4 
 
 Charcoal, basis of black 
 paint, 27 
 
 Charms, stones as, 24 
 
 Checker- weave technique, 
 10, 16, 19, 21-22, 36, 37 
 
 Child, bed of, 11; burial of, 
 36; sandal of, 11, 49, 51 
 
 Chilopsis linearis, 59 
 
 Chisels, bone, 33 
 
 Coiled technique: 41; bas- 
 ketry, 9, 19, 22, 38, 64 
 
 Containers, reed, 32 
 
 Corn: 19, 36, 59; cobs, 10, 
 21, 59; cultivated in val- 
 ley, 62; in pouch, 19, 36, 
 37;sheller, 16, 29-30 
 
 Coyote, bones of, 60 
 
 Cucurbita foetidissima, 59. 
 See Gourds 
 
 Cucurbita pepo, 59 
 
 Dasylirion sp., 10, 51, 59. 
 
 See Sotol 
 Deer, bones of, 60; scapulae 
 
 of, as scoops, 33-34 
 Del Rio, Valverde co., a 
 
 rock shelter near, xi 
 Desert-willow pods, 59 
 Devil's Canyon, Valverde 
 
 co., a rock shelter in, xi 
 Dice, wooden, 27 
 Digging sticks, 11, 30 
 Dottle of cedar sprays, 32 
 
INDEX 
 
 67 
 
 Dragon-fly, representation 
 
 of, 31 
 Drills, for fire-making, 27- 
 
 28, 38; perforations made 
 
 by, 34 
 
 Eagle Canyon, original name 
 of Bee Cave Canyon, ix 
 
 Echinocercus sp., 59. See 
 Cactus 
 
 El Paso, Texas, excavations 
 near, xi 
 
 Fabrics, woven : blanket- 
 like, 10, 50, 52; blankets, 
 skin, 10, 51; materials for 
 weaving, 51 
 Feathers in arrowshafts, 32 
 Fertility, symbols of, 58 
 Fiber, brushes of, 53; bun- 
 dles of, 51; caulking of, 
 34; in basket, 37; netting 
 of, 38-41; string of, 11, 
 16, 21, 22, 23, 28, 31, 
 34, 36, 42-43, 50, 57, 64; 
 used in fabrics, 10, 35, 
 36, 38-41, 47, 50-52, 64 
 Figurines, clay, 9, 16, 57-58 
 Fire-making implements: 
 drills, 27-28, 38; hearths, 
 10, 15, 22, 27-28, 38; 
 outfit, 37 
 Fire-pits, 19, 20, 21, 22; 
 house-site 4 used as, 16, 
 61; intrusive, 61; sug- 
 gesting communal cook- 
 ing, 62 
 Fire-tongs, 16, 30-31 
 Flakes, stone, 15; painted, 
 
 24-25 
 Flaking tool, 33 
 
 Fletcher, Henry T., re- 
 searches of, xi, 64 
 
 Food, remains of, animal: 
 10, 11, 15, 60; vegetable: 
 berries, 59; cactus, 10, 11, 
 59; corn, 10, 19, 21, 36, 
 37, 59, 62; nuts, 10, 11, 
 16, 59; roots, 59; seeds, 
 10, 11, 15, 19, 36, 37, 
 59; sotol crowns, 10, 11, 
 15, 22, 59, 62 
 
 Foreshafts, 10, 15, 31 
 
 Fringes, string, 42 
 
 Frontal bone, human, 20, 63 
 
 Gaming sticks, 27 
 
 Gouges, bone, 33 
 
 Gourds: 10, 15, 22, 59; 
 vessels, mended, 34, 62 
 
 Grass: baby-carrier, 10, 54- 
 55; beds, 11, 13, 21-22; 
 bindings, 42, 53-54; floor- 
 ings, 5, 9-10, 16, 19, 20, 
 21, 60, 63; matting, 34- 
 36; pads, 54; plugs, 32; 
 rings, 54; sorted, 51; 
 string, 42 ; wrappings, 24, 
 58 
 
 Griddles, stones used as, 23 
 
 Grinding-stones, 10, 15, 16, 
 22-23, 62 
 
 Grooved rock, 18, 22 
 
 Guernsey, S. J., cited, 55 
 
 Hammerstones, 10, 16, 23, 
 62 
 
 Hammock-like object, 55 
 
 Hand, imprints of, 12; out- 
 line of, 16-17, 19 
 
 Handles, wooden, 29 
 
68 
 
 ROCK SHELTER 
 
 Harrington, M. R., excava- 
 tions of, in rock shelter, 
 ix, 19, 64 
 
 Hearths, 10, 15, 22, 27-28, 
 38 
 
 Hematite, 37 
 
 Homolacephala texensis, 59 
 
 House-sites, excavation of, 
 6-21: no. 1, adobe used in 
 walls of, 6, 9; no. 2, 
 basketry in, 9; figurines 
 in, 9, 57-58; pottery in, 
 9; walls of, 9; no. 3, levels 
 of, 9-11; objects: animal 
 bones, 10, 11, 60; antler, 
 11, 33; arrowpoints, • 10, 
 11, 23; baby-carrier, 10, 
 54-55; beds, 11, 13; blan- 
 ket-like fragment, 10, 50, 
 52; cactus leaves and 
 pods, 10, 11, 59; cara- 
 pace of turtle, 10, 60; 
 corn-cobs, 10, 59, 62; 
 digging stick, 11, 30; 
 gourds, 10, 34, 59, 62; 
 grinding-stone, 10, 22-23, 
 62; hammerstone, 10, 23, 
 62; hearth, 10, 27-28, 38; 
 leather, 10, 56, 62; mat- 
 ting, 10, 11, 36; net bag, 
 11, 39, 41; nuts, 10, 11, 
 59; sandals, 10, 11,49, 51; 
 seeds, 11,59; sotol, 10, 11, 
 59-60, 62; stone, chipped, 
 10; stone, painted, 10, 24- 
 25 ; tool-marked wood, 10; 
 walls, 8-9 ; no. 4, levels of, 
 15; objects: animal bones, 
 15, 60; antler, 15, 33; 
 arrowpoints, 15, 16, 23; 
 basket, 15, 38; beads, 16, 
 
 57; corn sheller, 16,29-30; 
 figurine, 16, 57-58; fire- 
 tongs, 16, 30-31; gourds, 
 15, 34, 59, 62; grinding- 
 stones, 15, 16, 22-23, 62; 
 hammerstones, 16, 23, 62; 
 hand outlined, 16-17, 19; 
 hearth, 15, 27-28; leather, 
 15,56, 62; matting, 15, 16, 
 36; metates, 15, 23, 62; 
 paint-sticks, 15, 27, 31; 
 posts, 6, 16; reed pipes, 15, 
 32; rhythm stick, 16, 27; 
 sandals, 16, 43-51; seeds, 
 15, 16, 59; shell pendant, 
 
 15, 56, 57; sotol, 15,59-60, 
 62 ; stone flakes, 15 ; stone, 
 painted, 16, 24-25; tool- 
 marked wood, 15; walnuts, 
 
 16, 59; walls, 12, 14-16; 
 used as fire-pit, 16, 61; no. 
 5, levels of, 19; objects: 
 basketry, 19, 38; corn, 19, 
 
 36, 37, 59; pouch, 19, 36, 
 37; squash seeds, 19, 36, 
 
 37, 59; steatite pendant, 
 
 19, 57; wooden plugs, 19, 
 28-29; work of M.R.Har- 
 rington, 19; no. 6, levels 
 of, 19-20; objects: fire- 
 place, 20, 61; frontal 
 bone, 20, 63; jaw-bone, 
 
 20, 63; metates, 20, 23, 
 62; reed, 20; rubbing 
 stone, 20; sandal, 20; 
 wall, 19 
 
 Hueco Mountains, El Paso 
 co., caves in, examined, 
 
 XI 
 
 Human beings, pictographs 
 of, 5 
 
INDEX 
 
 69 
 
 Indian Notes, M. R. Har- 
 rington's report in, ix, 64 
 
 Jaw-bones, human, 20, 63; 
 
 of bison, 60 
 Juglans, 59. See Walnuts 
 
 Kidder, A. V., cited, 55 
 Knives, stone, 23, 64 
 Knots, 43 
 
 Leather, remnants, 10, 15, 
 
 56, 62; straps, 16, 29-30. 
 
 See Skin 
 Levels in rock shelter, 5, 
 
 9-11, 15-16, 19, 20, 21; 
 
 continuity of culture in, 
 
 60-61 
 Limestone, cliff of, 1; pipes 
 
 of, 24 
 Lower Rotten Draw, bag of 
 
 sotol leaves in, 55 
 
 Manos, see Grinding-stones 
 Matting, 10, 11, 15, 16, 19, 
 22, 36; pouch, 19, 36, 37; 
 technique of, 10, 16, 19, 
 21-22, 34-36, 37 
 Mesquite beans, 59 
 Metates, 15, 20, 23, 62 
 Mexicans, use of star cactus 
 
 by, 59 
 Miller, Ray, on name of 
 
 Bee Cave Canyon, ix 
 Museum of the American 
 Indian, Heye Foundation, 
 excavation under auspices 
 of, xi ; M. R. Harring- 
 ton's report published by, 
 ix, 64 
 
 Netting, 11, 38-41, 54-55; 
 bag of, 11, 39, 41 
 
 New Mexico, bone imple- 
 ments resembling those 
 of, 33 
 
 Nine Point Mesa Quad- 
 rangle, U. S. Geol. Survey 
 map of, ix, 1 
 
 Nodule, bowl fashioned 
 from, 23 
 
 Notes on the Ethnobotany of 
 Bee Cave Canyon, cited, 
 64 
 
 Nuts, 10, 11, 16, 59 
 
 Ochre, used as yellow paint, 
 27 
 
 Opuntia, 10, 59. See Prickly 
 pear 
 
 Ornaments: beads, 16, 57, 
 64; necklace, 64; pend- 
 ants, 15, 19, 56-57 
 
 Oxide of iron, hand out- 
 lined with, 16-17, 19; im- 
 prints made with, 12; pic- 
 tographs traced with, 5; 
 used as red paint, 27 
 
 Paint, 27; bones decorated 
 with, 33; figurines deco- 
 rated with, 58; hand out- 
 lined in, 16-17, 19; im- 
 prints made with, 12; pic- 
 tographs made with, 5; 
 pottery decorated with, 
 61 ; stones decorated with, 
 10, 16, 20, 24-25; wood 
 decorated with, 31 
 Paint-sticks, 15, 27, 31 
 Paint-stone, in burial, 64 
 
70 
 
 ROCK SHELTER 
 
 Pendants: shell, 15, 56, 57; 
 
 stone, 19, 56, 57 
 Pestles of antler, 33 
 Peyote, star cactus used 
 
 like, 59 
 Pictographs, 5, 12, 16-17, 19 
 Pinon nuts, 10, 59 
 Pinus edulis, 59. See Pinon 
 Pipes, reed, 15, 32; stone, 24 
 Pits, see Fire-pits 
 Plugs, of grass, 32 ; of wood, 
 
 19, 28-29 
 Pottery, 9; figurines, 9, 16, 
 
 57-58; scarcity of, 61 
 Posts, 6, 16 
 Pouch of checker- weave, 19, 
 
 36, 37 
 Prickly pear, seeds and 
 
 leaves of, 10, 11, 59 
 Projectiles, fragments of, 
 
 30. See Arrows, Atlatls 
 Prosopis glandulosa, 59 
 Pumpkin seeds, 59 
 
 Quercus, 59 
 
 Rabbit, bones of, 60 
 
 Rabbit sticks, 22, 28 
 
 Rat, bones of, 60 
 
 Red: paint, bones deco- 
 rated with, 33; figurines 
 decorated with, 58; pic- 
 tographs made with, 5, 
 12, 16-17, 19; stones 
 decorated with, 16, 20, 
 24-25; paint-stone, 64; 
 string, 42 
 
 Reed, arrowshafts of, 31, 
 32; beads of, 16, 57, 64; 
 containers of, 32; pipes 
 of. 15, 32; reeds, 20, 22 
 
 Rhythm sticks, 16, 27 
 
 Rings of fiber or grass, 
 53-54 
 
 Rock, fall of, 5, 21-22, 63; 
 grooved, 18, 22 
 
 Rock shelter, burial in, 64; 
 configuration of, 2, 3, 5; 
 cultural unity of, 60-61; 
 excavations of M. R. 
 Harrington in, 19, 64; 
 fire-pits in, 19, 20, 22; 
 house-sites in, 6-21; loca- 
 tion of, ix, xi, 1, 3; pic- 
 tographs in, 5, 12, 16-17, 
 19; seasonal occupancy 
 of, 62; water supply of, 2, 
 5, 7 
 
 Roots, 59 
 
 Rotten Draw, stick with 
 tenoned ends in, 27, 55 
 
 Rubbing stone, 20 
 
 Sandals, 10, 11, 16, 20, 21, 
 55; child's, 11, 49, 51; 
 technique of, 44-49; tie- 
 strings of, 47-49, 50, 51 
 
 Santiago Peak, Brewster 
 co., rock shelter near, 1 
 
 Sapinus, 59. See Buckeye 
 
 Satan Canyon, Valverde 
 co., a rock shelter in, xi 
 
 Saws, stone, 23 
 
 Scapulae, scoops of, 33-34 
 
 Schuler, Lee, rock shelter on 
 ranch of, xi, 1 
 
 Scoops, bone, 33-34 
 
 Scorings, bone decorated 
 with, 33 
 
 Scrapers, stone, 23 
 
 Seeds, as beads, 16, 57; 
 as food, 10, 11, 15, 19,36, 
 
INDEX 
 
 71 
 
 37, 59; in reed containers, 
 
 32 
 Shell pendants, 15, 56, 57 
 Sifting-baskets, 15, 38 
 Sinew, holding arrowpoints, 
 
 31 ; in basket, 37 ; wrapped 
 
 about sticks, 16, 30 
 Skeleton, flexed, 64 
 Skin : fragments, 10; scraper, 
 
 33; strips for blankets, 10, 
 
 51. See Leather 
 Skull, human, 63 
 Small, John K., vegetable 
 
 substances identified by, 
 
 XII 
 
 Snares, 31, 43 
 
 Sotol: crowns, 10, 11, 15, 
 22, 59-60, 62; leaves, bags 
 of, 55; hammock-like ob- 
 ject of, 55; sorted and 
 split, 51 
 
 Spear-points, 23, 64 
 
 Squash seeds, 19, 36, 37 
 
 Stakes, 9, 29 
 
 Steatite pendant, 19, 57 
 
 Sticks, curved and grooved, 
 
 22, 28 ; with leather straps, 
 16, 29-30; with tenon-like 
 ends, 26, 27 
 
 Stone, objects of: arrow- 
 points, 10, 11, 15, 16, 23, 
 31; axe, 23; beveled, 27; 
 bowl, 23; chips, 10; flakes, 
 15, 24-25; grass- wrapped, 
 24; griddles, 23; grinding- 
 stones, 10, 15, 16, 22-23, 
 62; hammerstones, 10, 16, 
 
 23, 62; hematite, 37; 
 knives, 23, 64; metates, 
 15, 20, 23, 62; painted, 
 10, 16, 20, 24-25; pend- 
 
 ants, 19, 56, 57; pipes, 24; 
 rubbing, 20; saws, 23; 
 scrapers, 23 ; spear-points, 
 23, 64; walls, 6, 8-9, 12, 
 14-16, 19, 20; with fiber 
 cords attached, 23 
 String, 16, 21, 22, 23, 36, 
 42-43; beads strung on, 
 16, 57, 64; colored, 42; 
 employed in fabrics, 10, 
 35, 36, 38-41, 50-52, 64; 
 fringes of, 42; knots in, 
 43 ; lacing broken gourds, 
 34; skin twisted on, 10, 
 51; snares of, 31, 43; 
 tapered, 42; ties of, 42, 
 47-49, 50, 51; twigs 
 bound with, 22, 28, 31 
 
 Tibia, human, 63; of bison, 
 
 60 
 Tie-strings, of grass pads, 
 
 54; of sandals, 47-49, 50, 
 
 51; untwisted, 42 
 Toys, bows as, 28 
 Turtle, carapace of, 10, 60 
 Twigs, bows of, 22, 28; 
 
 snares of, 31 
 Twilled technique: mats, 36 
 Twined technique: baskets, 
 
 64; fabrics, 10, 50, 54-55 
 
 U. S. Geological Survey map 
 of Nine Point Mesa Quad- 
 rangle, ix, 1 
 
 University Museum, Cam- 
 bridge, England, excava- 
 tion under auspices of, xi 
 
 Upper Rotten Draw, water 
 supply from, 5 
 
72 
 
 ROCK SHELTER 
 
 Valverde county, a rock 
 shelter in, xi 
 
 Walls of house-sites, 6, 8-9, 
 12, 14-16, 19, 20 
 
 Walnuts, 16, 59 
 
 Water supply: reservoir, 2, 
 5, 7; walk toward, 63 
 
 Weaving, materials for, 51. 
 See Fabrics 
 
 West Texas Historical and 
 Scientific Society, Bul- 
 letin 33 of, cited, 64 
 
 Wood, objects of: arrow- 
 shafts, 10, 15, 22, 31, 32 
 61; atlatls, 28, 30, .61 
 bows, 22, 28; dice, 27 
 digging sticks, 11, 30 
 drills, 27-28, 38; fire- 
 tongs, 16, 30-31; fore- 
 shafts, 10, 15, 31; frames 
 for beds, 11, 13; gaming 
 sticks, 27; handles, 29; 
 
 hearths, 10, 15, 22, 27-28, 
 38; painted, 31; paint- 
 sticks, 15, 31; plugs, 19, 
 28-29; posts, 6, 16; rabbit 
 sticks, 22, 28; rhythm 
 sticks, 16, 27; snares, 31, 
 43; stakes, 9, 29; sticks: 
 with leather straps, 16, 
 29-30; with tenon-like 
 ends, 26, 27; tool-marked, 
 10, 15, 22 
 
 Yellow, ochre used for, 27; 
 figurine decorated with, 
 58; string, 42 
 
 Yucca: leaves, baby-carrier 
 woven with, 10, 54-55; 
 baskets of, 38; matting 
 of, 36; netting of, 41, 54- 
 55; rings of, 53; sandals 
 of, 43-47, 48, 49; sorted, 
 51; stitching of, 36, 38; 
 string of, 42; seeds, 59. 
 
 'B-6555-25 
 5-17T 
 
University of 
 Connecticut 
 
 Libraries 
 
 39153027668591