Cueva Pilote: Ritual bloodletting among the

prehistoric hunters and gatherers of Northern Coahuila, Mexico

Solveig A. Turpin and Herbert H. Eling, Jr. With contributions by: Linda Scott Cummings and Kathryn Puseman Leland C. Bement Cover art: México á Través de los Siglos, D. Alfred0 Chavero (1962)

Instituto Nacional de Antropología y Historia

Oficio No. C.A .401-36/953

ii

Abstract

Cueva Pilote, a small cave well-hidden in the precipitous canyons of the Sierra Encantada in northern Coahuila, produced an artifact assemblage that can best be explained in the context of ritualized auto-sacrifice, or bloodletting, a custom well-documented in the complex societies of Mesoamerica. Over a score of deer scapulae painted with red or rarely with black geometric designs, 3 marine shell beads, 56 mussel shell beads and a plaque, 43 ornamental snail shells, 4 pads of agave fiber cut from the trunk of the desert palm, 192 agave spines, fragments of a domesticated gourd, bear bones, a small assortment of human bones and teeth, projectile points, a cache of preforms, and other miscellaneous stone tools were exhumed from the site. The projectile point styles found in the cave have been dated to the period between 3200 and 4100 B.P. in Texas, but four radiocarbon assays produced an age range between 600 and 1000 B.P. for the ritual complex. Historical, ethnographic, and pictorial accounts of bloodletting and the presence of human protein residues on a sample of agave spines support the proposition that Pilote was a ritual site, utilized by hunting and gathering people who shared their religious asceticism with their urban neighbors to the south.

iii

Dedicated to the memory of

Tom Barksdale June 21, 1921 – January 17, 1999

Tom Barksdale was a true man of the border, at home in two cultures and beloved in both. Like his father before him, Tom was a repository of oral history, about the people, the places, and the past of northern Coahuila. He was our passport to remote places and our ambassador to the people who inhabit them. He was one of the folk on both sides of the Rio Grande; their respect and honor were expressed in the way they extended their hospitality to him and his companions. Tom was with us both season at Pilote where he endured the rigors of the worst winter in modern memory, at San Vicente, Santa Rosa,

San Antonio de los Alamos, Acatita la Grande, Las Animas, the Sierra del Fuste, El Cedrito, and finally at Aguaverde where this last photograph was taken in December

1998. Tom died unexpectedly in January as we were planning our return to Aguaverde. Certainly the world is a lesser place.

iv

Figure 1. Topographic map of northern Coahuila (Smith 1970: 10). Cueva Pilote is adjacent to the C in Sierra del Carmen

1

Introduction

A small cave, well hidden in the remote mountains of northern Coahuila (Fig. 1), was the site of ritual activities with a strong Mesoamerican imprint around A.D. 1300. Cueva Pilote, so named for its location at the foot of the highest peak in the northern Encantada Valley, produced a non-utilitarian artifact assemblage that included over a score of painted deer scapulae, 56 mussel shell beads and a mussel shell plaque, 3 marine shell beads, 43 large and decorative snail shells, a cache of 6 preforms, at least 3 complete and 3 partial projectile points, fragments of a domesticated gourd, 4 palma-fiber pincushions, and almost 200 agave spines. The most parsimonious explanation for this unusual assortment is their use in rituals that included self-mutilation and bloodletting. In this context, the only artifacts that could possibly be considered mundane are some of the stone tools - 8 bifaces and 4 unifaces - and debitage, the latter amounting to 45 flakes, chips, and cores. The mussel and marine shell, some of the lithic material, and the gourd are exotic and probably brought in from the Rio Grande valley, the Gulf of Mexico, and one of the horticultural areas far outside the Encantada Valley.

The artifact assemblage is not the only characteristic that sets Pilote apart from the norm of occupied shelters. The juxtaposition of mountains and caves is a persistent element in Mesoamerican religion and a theme in many of the creation or origin myths (Heyden 1981). Set in the midst of a dramatic landscape, Cueva Pilote is transitional between the bulk of its eponymous mountain looming in the west and the narrow cleft of the canyon descending to the east, into the rising sun. Descent is precipitous and the last few meters negotiable solely via hand and toe holds in the rock; the ledge in front of the cave is only a meter wide above a sheer 20-meter drop to the canyon floor. Thus, the descent and ascent are perilous and contribute to the aura of secrecy, privacy, and isolation.

Size and shape, as well as setting, contribute to the transformation of the cave into sacred space. The entrance is virtually invisible except from a vantage point on the lowest rimrock ledge directly across the narrow, steep canyon (Fig. 2). The opening is less than a meter high and further obscured by trees that grow directly in front of it. Once through the shallow opening, however, the cave opens into an oval room 6 meters long and 4 meters wide, its domed ceiling peaking at a height of 4 meters above a floor pocked with relic hunter pits. A short, steeply inclined gallery extends upward from the upstream wall near the rear of the cave; its exposed bedrock floor is littered with roof fall but any other material deposited there has long since worked its way down slope to the main cavern floor.

2

Figure 2. The low entrance to Cueva Pilote, hidden in a

steep narrow cleft and further obscured by trees growing at its

mouth, is only visible from the other side of

the canyon.

3

Figure 2 (cont.).Above: Close up of entrance to the cave before the start of excavation Below: Excavation begun in cramped quarters; string line marks the edge of the steep dropoff.

4

The process of entering the cave physically expresses the stages through which the initiate proceeds in any rite of passage (Van Gennep 1961). The almost inaccessible low entrance and narrow confining passage lead from the ordinary world, the great outdoors and open sunlight, into the sharply contrastive shadowed interior with its encircling walls and darkened field of vision (Fig. 3).

The symbolism of the dangerous descent; the passage through the low entryway into the interior cavity, the womb of the living rock; and the contrast between light and dark set the stage for ritual events as the participants leave the limitless open air, squeeze through the hole in the supernatural fabric, and enter a conceptually different space. Re-emerging, transformed by the ceremonies performed, is analogous to rebirth, reentering the natural world renewed and transfigured. Thus, Pilote is one of the rarest sites - a place dedicated to the performance of arcane rituals by people who pursued a hunting and gathering lifeway, armed with a limited technology and an ancient belief system that included a concept of the religious asceticism that permeated Mesoamerican culture and paralleled that of the Christianity that was yet to come.

Figure 3. Looking out from the excavated interior of the cave through the narrow entrance. The mounded dirt outside the cave is the relic hunter’s discard pile (Unit I).

5

On a larger scale, Cueva Pilote belongs to a time in prehistory when change rippled through the arid lands of northern Mexico and Texas (Taylor 1966). People were evidently on the move, as were the exotic goods imported into the Encantada Valley and deposited in this small site. In this sense, Cueva Pilote adds another dimension to the already dynamic reconstruction of life in the arid lands prior to the coming of the Europeans.

6

The Physical Setting

The Sierra Encantada is a northwest-southeast trending outlier of the Sierra Madre Oriental that parallels the Serranías del Burro and the Sierra del Carmen (see Fig. 1) in northern Coahuila, just south of the Big Bend in the Rio Grande (Smith 1970: Fig 4). The Arroyo de la Babia, on the eastern flank of the Encantadas, provided a lowland route between the Spanish colonial center Santa Rosa del Sacramento, now Múzquiz, over the Cuesta Malena, to the presidio San Vicente on the Rio Grande and the more westerly Spanish settlements at La Junta. The Spanish explorers undoubtedly adhered to pre-existing trails just as the modern highway follows the colonial road, although the current route is free of the constraints imposed by the need to water man and beast. The sheer escarpments and rugged crags of the Encantadas were a formidable barrier to westerly movements. The topographic extremes are so dramatic that even 19th century Mexican troops, accustomed to chasing hostiles in the Sierra Madres, described the Encantadas as "notable for their harshness and height" (Flores 1881: n.p.)

The highest peak in the northern valley is the Cerro Pilote de la Encantada (Fig. 4) which rises to a height of 1980 meters; the plain at its base slopes from 1800 to 1700 meters. Near the head of a canyon that flows east from the base of the mountain is Cueva Pilote at an elevation of about 1720 meters. From the escarpment, the eastern rim of the valley falls away into the Valle de la Babia, a drop of over 1000 meters from the crest of Cerro Pilote.

Verticality promotes differences in temperatures and precipitation, as well as the biotic communities they support. Both basins are carpeted with grass and scattered desert succulents, most notably agaves such as Yucca carnerosa or palma (Gentry 1982; Mason and Mason 1987), trending into pine and oak forest on the high ridges and slopes (Muller 1947; Smith 1970). Cueva Pilote is in the transitional zone with pines and yuccas on the rim above it, more mesic trees and shrubs in the canyon below it, and open grassland only a few hundred meters away at the base of Cerro Pilote.

No record has been kept of the modern Encantada climate but presumably it resembles that of the Trans-Pecos where precipitation averages 11 to 13 inches per year, the mean temperature is 63 degrees Fahrenheit and historical extremes range from over 100 degrees F in summer to below 0 degrees F in winter (Deal 1976:28). In December 1997, during our fieldwork, temperatures plummeted below zero and a freak snowstorm blanketed the valley. The interior of the cave, however, stayed above freezing although the traverse of the cliff face was made more treacherous by ice and snow.

7

Figure 4. Cerro Pilote as seen from the canyon rim above the cave.

The upper valley is an interior drainage system with significant arroyo formation limited to the immediate vicinity of the slopes. Only a few intermittent springs and tinajas still function. However, as late as 1881, a Mexican military expedition against hostile Apaches reported two permanent springs in the Encantada Valley, but the map scale is such that their exact locations cannot be determined (Flores 1881). In addition, within the living memory of local ranchers, runoff from heavy rains sometimes pooled in the basin of the Encantada Valley, providing an intermittent source as well as recharging the water table.

Although paleoenvironmental research has only just begun in the valley, changes in local conditions were noted at Cueva Encantada, some 15 km south of Pilote. Normally perishable materials are abundant in the upper layers of the cave, which were laid down in the last 4000 years, but the earlier levels had only stone tools. A travertine formation that clearly represents a much wetter episode separates the upper and lower strata, with their differing preservational qualities, and accounts for the destruction of the fiber features and artifacts below it. Unfortunately, Cueva Pilote contributed little to the reconstruction of past environments in the Encantada Valley.

8

The Cultural Context

Cueva Pilote is one of only two sites excavated in the whole of northern Coahuila, so much of its context must be derived from neighboring areas such as Cuatro Ciénagas (Taylor 1964, 1966, 1968) and more generally the Laguna district in southern Coahuila (Aveleyra et al. 1956; González 1992; Heartfield 1975; Martínez del Río 1953, 1954; Weitlaner 1977), the Big Bend (Mallouf 1985), and the Lower Pecos region of Texas (Turpin 1991, 1995) where comparative material has been exhumed from rock shelters and open sites. Cueva Pilote's sister site, Cueva Encantada, a dry rock shelter only 15 straight-line km to the south (Turpin and Carpenter 1994), was excavated in 1996 as part of the Encantada Project and the report is in progress. Its deposits, dated by radiocarbon assay to the range from 650 to 4000 years ago, are largely composed of domestic features and debris, including a large sample of projectile points like those of Cueva Pilote, despite the fact that most of its deposits are considerably older.

The Big Bend and Lower Pecos regions of Texas share a general quadripartite chronological sequence composed of Paleoindian (ca. 12,000-9500 B.P.); Early, Middle and Late Archaic (ca. 9500-1200 B.P.); Late Prehistoric (ca. 1200-350 B.P.); and Historic (post-350 B.P.) periods. In the Lower Pecos, this scheme has been further refined into 10 subperiods and a phase but only one, the Middle Archaic San Felipe subperiod (4100-3200 B.P.), is directly relevant to Cueva Pilote.

Based on his general observations of Coahuilan prehistory, and specifically his work at Frightful Cave and other sites in the vicinity of Cuatro Ciénagas, Taylor (1966) defined four complexes, which he described as congeries of traits. The earliest manifestations of human occupation were assigned to the Ciénagas complex and far predate the material culture of Cueva Pilote. The broad continuum of prehistory was designated as the Coahuila complex and is, in many senses, analogous to the Archaic period in Texas. Most of the artifacts and features in Cueva Encantada are typically Coahuilan although some of the most recent deposits are contemporaneous with Cueva Pilote.

Late in prehistory, the Jora complex can be identified, according to Taylor (1966:81), by a number of chronologically significant traits within the broader cultural continuum of the Coahuila complex. Of importance to Cueva Pilote is Taylor's hypothesis that the Jora complex represents a reconstitution and reintegration of culture in northern Coahuila, stimulated by outside influences emanating from the La Junta region at the confluence of the Río Conchos and the Rio Grande. Although Pilote has few of the traits that Taylor cited as distinguishing characteristics of the Jora period, the radiocarbon assays all fall

9

within its presumed temporal range and the material culture suggests wide-ranging contacts with external sources.

Taylor's (1966:83-84) Mayran complex subsumed the material culture exhumed from mortuary caves in the Laguna district (Aveleyra et al. 1956). The defining attributes are largely elaborate grave goods interred with bundle burials. Although much of the material culture of the Mayran complex is undated, Coahuila complex projectile points in the burials perhaps indicate some antiquity (Taylor 1966:84). A terminal date in the 16th or 17th century is inferred. Although Cueva Pilote is not a burial cave, the mortuary goods of Cuevas Candelaria and Paila (Aveleyra et al. 1956) provide possible explanations for some of the artifact assemblage, specifically gourd vessels, clusters of beads, and marine shells.

Due to the lack of comparative material in the immediate vicinity of Cueva Pilote, its interpretation depends upon the radiocarbon chronology developed in neighboring areas of Texas, the material culture of sites in southern Coahuila, rare ethnographic references to ritual practices in the Bolsón de Mapimí, and historic documentation of religious ceremonies in the Valley of Mexico. Cueva Pilote remains one of the rarest site types in hunter-gatherer research - a cave dedicated to ritual practices - and much of its analysis must be based in extra-regional analogies and inferences.

10

Methods

During a trip to the Encantada Valley in search of a looted rock shelter, local ranchers told us about painted deer bones that had been exhumed from a cave in the 1960s. Some months later, three of the painted scapulae were retrieved from the vaquero who had originally dug up the bones and had traded, given or sold them to various individuals. Eventually, 13 scapulas were traced to five different individuals or institutions (Turpin 1996, this report). In 1997, we made our first field trip to Cueva Pilote to assess the possibility of controlled excavation to specifically fix the painted bones in time and space.

The signs of active relic hunting, which had affected approximately 30% of the site surface, and the sure knowledge that our presence in the valley would be met with heightened curiosity contributed to the decision to fully excavate the cave. Field work was carried out in December 1997 and March 1998 under the authority of a permit issued by the Instituto Nacional de Antropología e Historia (INAH). Preparations included the placement of a generator atop the rim to artificially light the darkest recesses of the cave. Ropes were strung as a safety measure that became essential when a freak snowstorm coated the rocks with ice and turned the slight soil cover into slippery mud.

Initially, plan and profile maps of the cave were drawn using tapes and compass (Fig. 5). The interior of the cave is an oval, 6 meters long by 4 meters wide, so a 1-meter-square grid was superimposed on the surface with its central axis bisecting the entrance and the rear wall. Excavation began in the center of the site where a relic hunter hole afforded a partial preview of the subsurface deposits. Units were alphabetized in the order that they were excavated; in two cases, small extensions were identified as RX and SX. Two units, H/Q and O/R, were assigned a second letter when work resumed during the second field season to mark elevational differences between the two episodes of excavation;for example, artifacts from the first season were bagged as H1-3, ending at 53 cm below datum (cmbd) and from the second as H/Q 4, beginning at 40 cmbd to accommodate fill that had accrued in our absence and overlap with backfill.

Vertical control was maintained in arbitrary 10 cm intervals measured from a permanent datum, a lead plug and screw driven into the rear wall of the shelter at the terminus of the cross section and at approximately the level of the highest point on the surface. Relic hunter holes and bioturbation created an irregular surface so measurements below datum do not always equate to depth below the surface. The bowl-like configuration of the bedrock and the undulating surface contributed to variability in unit depth which ranged from less than 10 cm on the edges abutting the slope of the wall and below some of the relic hunter holes to a maximum of 118 cm in Unit C.

11

Figure 5. Plan and profile maps of the cave.

12

Twenty-one one-meter-squares and two partial units, the equivalent of the entire surface of the cave, were excavated to bedrock; two units in the entry way, W and Z, were terminated at depths of 80 and 35 cm respectively when recovery diminished drastically. A pile of dirt discarded by relic hunters just outside the entrance and above the steep drop off, designated as Unit I, was screened to recover any overlooked items. All the matrix was removed with trowels or brushes and all was screened through 1/2, 1/4, 1/8 and 1/16 inch mesh.

13

Artifacts and Manuports

Unlike the domestic site of Cueva Encantada where fiber-lined pits, hearths, ash pits, and a latrine were clearly identifiable loci of specific activities, Cueva Pilote contained no definable features. Scattered charcoal and burned rocks indicate that fires were lit and maintained, largely near the entrance to the cave, but none of the hearths or fire pits remained intact. None of the artifacts were clustered or clearly associated with each other so caches, work areas, or burials could not be identified on the basis of the artifact distributions. In fact, one of the more puzzling aspects of Cueva Pilote is the mechanisms by which very similar artifacts, presumably once parts of a larger whole, became scattered throughout the deposits. In the absence of stratigraphic clarity, the entire assemblage is treated here as a unitary phenomenon, despite discrepancies between the age of the materials dated by radiocarbon assays (Table 1) and that of the artifacts that elsewhere serve as index markers for specific, much older time periods.

The Age of the Site

Although charcoal was plentiful, chunks and flecks were relatively evenly distributed throughout the site, causing some concern about the relevance of any samples that might be collected for radiocarbon dating. Given the lack of features and the disturbances caused by relic hunters and burrowing rodents, the provenience of the vast majority of the charcoal was suspect. One sample, taken from Unit J-4 beneath one of the painted scapulae, consisted of large lumps of carbonized hardwood and ash, the latter suggesting that the coals had burned in place. This sample was submitted to the University of Texas Radiocarbon Laboratory and rendered a 13-corrected age of 1060+50 (TX-9331, see Table 1).

A second charcoal sample consisted of a thoroughly charred but intact section of a walnut tree limb (Juglans cf. microcarpa, Sobolik 1998) that was recovered from the bottom of Unit C, 118 cm below datum (cmbd). This was the deepest unit and presumably its basal levels were the oldest in the site but the 13-corrected date of 700+35 B.P. (TX-9340) is the same age or younger than materials taken from higher levels.

Table 1. Radiocarbon Assays

Lab ID C14 BP Cal BP Cal AD 13 (ppm) Comments

TX-9311 890+40 655 1295 -11.9 Unit H, 10-20 cmbd, agave fiber

TX-9331 1060+50 950 1000 -23.8 Unit J, 40-50 cmbd, wood charcoal

TX-9340 700+35 660 1295 -23.9 Unit C, 118 cmbd, walnut charcoal

B117276 600+50 555 1395 -24.5 Unit X, 40-bdrk, gourd fragments A third radiocarbon date was obtained from one of the slabs of desiccated

fiber that were used as a form of pin cushion for agave spines. The two

14

specimens recovered during excavation shared their vertical provenience and presumably are approximately the same age. The larger of the two produced a 13-corrected age of 690+40 (TX-9311, 13 -11.9). Calibrated using Stuiver and Reimer's (1993) Method B, there is a 57% probability that the true age is between A.D. 1280 and 1310 (670 to 640 B.P.) and a 43% chance that it is in the range between A.D. 1360 and 1380 (590 to 570 B.P.).

A fourth radiocarbon date was obtained from fragments of domesticated gourd (Lagenaria siceraria, Minnis 1998), recovered from Unit X, beneath one of the relic hunter pits and below 40 cmbd. An AMS 13-corrected date of 600+50 B.P. (Beta-117276) calibrates to the range between A.D. 1305 to 1410 (68% probability; Stuiver and Reimer 1993).

The temporally diagnostic artifacts exhumed by excavation are all Middle Archaic types dated to the range between 3200 and 4100 B.P. in the Lower Pecos River region of Texas (Bement 1991; Turpin 1991). The Langtry type, represented by one specimen, is a hallmark of the San Felipe subperiod, a time of increasing regionalization and societal complexity. Jora dart points are much more common south of the border but their absolute age range is known only from Arenosa Shelter, on the Pecos River, where they were contemporaneous with Langtry and another variant, Val Verde (Bement 1991). It is quite possible that, outside the Lower Pecos region, in areas such as the Encantada Valley, the Jora style was in use for a much longer period of time than is indicated by the Texas radiocarbon dates.

Painted Bones and Skeletal Remains

The Scapulae

The most curious artifacts in the Pilote assemblage, and the ones that first and foremost establish the cave as a sanctum sanctorum, are the painted deer scapulae. Six painted scapulae (Fig. 6a-f), one badly burned fragment, and two unpainted shoulder blades were traced to four different collections in the United States but all originally came from Cueva Pilote (Turpin 1996, 1997). Another two complete and three partial painted scapulae were recovered during excavation (Fig. 6g-k). After the publication of the first report, a private collection in Mexico surface but two of the three specimens acquired from the vaquero had disappeared, leaving only one intact painted specimen (Fig. 7). Fifty-eight fragments, both burned and unburned, were found in the site and in private collections (Table 2); using the sockets to calculate the minimum number, at least 21 scapulae are represented by complete and fragmentary specimens. Of these, the designs can be almost completely reconstructed on only eight. One of the complete but unpainted specimens is a bear bone, the species that contributed most of the faunal remains in the cave, so it may not be conceptually the same as the other scapulae which are all deer.

15

16

17

Table 2. Provenience of Scapulae and Fragments

Unit/Level No. Comments

A-2 (30-40) 1 burned ridge

B-4 (58) 1 burned ridge

C-6 (90-100) 2 blade frags w/parallel & perpendicular red lines

D-6 (90-100) 1 frag w/red paint

E-3 (37) 2 unburned blade, ridge w/black paint

F-3 (44-54) 1 partially burned socket

G-5 (46-56) 2 burned ridges, 1 w/red paint

H-1 (23-33) 2 burned ridges

J-2 (10-20) 1 complete, red rectangles

L-2 (27) 1 burned ridge

M-1 (34-37) 1 complete, red lines

M-5 (65-75) 1 blade/spine w/red paint

P-1 (36-46) 1 small burned spine

P-2 (46-56) 3 burned ridges, 1 w/red paint

P-3 (56-66) 1 tiny unburned blade frag w/red paint

T-3 (30-40) 2 burned ridge, spine

U-1 (0-20) 2 burned socket, unburned spine w/red paint

V-1 (4-20) 4 burned socket, burned ridge, 2 burned frags

X-1 (40-bdrk) 8 5 burned ridges, 3 burned frags

Z-1 (0-bdrk) 2 blade w/paint on socket, burned ridge

Collections 22 2 sockets, 16 burned frags, 1 unburned w/paint

() depth below datum in cm

Only one fragment in the cave inventory (Fig. 6j) and one of the missing specimens from a private collection were painted in black pigment; the rest are red and all are geometric, composed of crosshatched or parallel and perpendicular lines on one or both sides of the blade. None is exactly the same but three consist largely of double lines that enclosed triangular or quadrilateral space (Fig. 6e-g); three are combinations of parallel and intersecting lines (Figs. 6b, 7); and two are simple intersecting lines that crisscross the blade (Fig. 6c-d). None of the fragments contradicts the assumption that all the scapula designs are variations on the same theme, one that appears with such consistency in preliterate art that it begs a universal explanation. Recently, considerable verbiage has been devoted to the theory that geometric designs are a product of neurobiology, that these images are hard-wired into the human brain and come to the surface under certain conditions, such as stress, trance, illness, and drugs. Because they are "seen" with the brain and not the eye, Lewis Williams and Dowson (1988) called them entoptic or inner-eye phenomena. Their predecessors, such as Reichel Dolmatoff (1978), Hedges (1982), and Bednarik (1984), used the terms psychograms, phosphenes, or form constants, following the psychological literature. Putting semantics aside, cross-hatches, zigzags, comb patterns, and parallel lines such as those seen on the painted scapulae are included in all the typological catalogues of this phenomenon.

18

Figure 7. One of three scapula exhumed by vaquero, in private collection in Mexico.

Deer unquestionably played a major role in local mythology and religion just as they do today in neighboring areas still populated by Native Americans. Archeologically, deer, their antlers, and their footprints appear in both petroglyphs and pictographs throughout Coahuila and southwest Texas (Turpin 1993). Antler headdresses, strikingly like those painted in the most ancient rock art styles, have been recovered from at least two sites in Coahuila (Taylor 1966:Fig. 26, Martínez del Río 1954:Fig. 38), demonstrating a continuity in thought that spanned thousands of years.

Looking at the scapulae in private collections, Turpin (1996) suggested that the bones were musical instruments, used as rattles or with rasps to produce rhythmic sounds, one pathway to an altered state of consciousness. Smoothing on the necks of some of the specimens suggests that they were strung or hung from cords, like a rattle made of four or five unpainted scapulae that was exhumed from the Shumla caves on the Rio Grande (Schuetz 1961:196). Kachina dancers in the southwestern United States still use painted scapula rattles (Wright 1973, 1985) that produce a clacking rhythm that could be used to induce a trance state or to accompany chanting or other hypnotic sounds.

19

Only two other sites have produced painted scapulae; both are caves in the mountainous Big Bend region of Texas. One specimen, retrieved from a cave in Jeff Davis County, bears curvilinear and crosshatched lines and the image of a quadruped, perhaps a mountain sheep (Jackson 1938). The paint on the other scapula, which is in the Smithsonian Institute (Prewitt 1970), can no longer be discerned. A third painted scapula is described by Schoolcraft (1854), Mallery (1893), and Jackson (1938) who attribute it to Comanches of the Texas Plains. This artifact, which shows two Europeans and a mounted Native American, is a classic example of the Plains Biographic tradition and clearly post-dates the material culture of Cueva Pilote.

Although the scapulae were not directly dated, their context and their possible uses make them a logical part of the Pilote ritual complex radiocarbon-assayed to the range between 600 and 700 B.P. The number of severely burned fragments (Fig. 8a) exceeds expectations based on the quantity of charcoal and burned rock in the site unless some specimens were deliberately destroyed after use. It is also possible but not demonstrable that the scapulae were being painted in the cave. Two of them, one of deer and the bear bone discussed above, are not painted. One of the uniface scrapers bears a trace of red pigment (see below). A small lump of pure ocher in Unit T may have been residue from the manufacturing process although it is equally possible that it was intended as body paint. Regardless of their specific origin and function, the painted scapulae of Cueva Pilote are undoubtedly ritual paraphernalia that derive supernatural power from their association with sacred designs painted in sacred colors, and taken from a sacred animal.

Human Remains

The human remains in Cueva Pilote are as puzzling as the bulk of the artifact assemblage. Three incisors, one deciduous and two permanent; a third molar; a fragment of a neonate scapula; and five juvenile finger bones came from Units A, B, C, F, G, and X at depths ranging from 24 to 90 cmbd (Table 3). The private collections contain no human remains and none were mentioned by any of the informants. No explanation for the skewed representation of the child - scapula, finger bones, and a tooth - is readily available unless other bones were removed in the 1960s and local superstitions intervened to quell any discussion of its disinterment. Another possibility is that scavengers destroyed the other bones and the remaining assortment is purely coincidental. A third option is secondary burial with the infant's skeletal remains removed for interment elsewhere in the far distant past. Finally, infant finger bones have been found strung into necklaces but there is no precedent for using a baby scapula in such a manner.

The human incisors can be envisioned as sacrificial offerings but it is difficult to imagine why a sound third molar would or could have been extracted. Again, various options for the disposal of an adult cadaver can be considered but it is unlikely that any resolution will be achieved based solely on the data from Cueva Pilote.

20

Figure 8. Distribution of a) scapula fragments; b) bear bones; c) shell beads; and d) the Langtry preform cache.

21

Faunal Remains

Excluding the painted deer scapulae, the largest component of the faunal assemblage, both in size and number, is attributable to a minimum number of two black bears, one adult and one juvenile (Table 3). The former is represented by a right tibia, a left radius, a left rib, an acetabulum, a thoracic vertebra, a left scapula, a front claw sheath (Fig. 9) and skull fragments. The claw indicates that the adult was extremely large for a black bear (McKinney 2005). The juvenile consists of left and right tibiae, left and right humeri, left ulna, ribs, a sacrum fragment, and the left rear second phalanx. Unlike the deer scapulae fragments, which were fairly well distributed (Fig. 8a), the immature bear was largely contained within the backdirt labeled as Unit I with a few elements in neighboring Unit W. Presumably, the bones were dumped there as part of the relic hunter's discards and originally came from within the cave, probably the rear pothole. The remains of the adult animal were spread through Units A, B, G, K, O, and W (Fig. 8b). Of particular interest are the right tibia and the left radius which were jammed under protruding bedrock ledges on either side of the entrance, flanking it. Perhaps more relevant than the bones that are in the cave are those that are missing. The absence of so many of the large dense bones, including the teeth and all but one of the claws, suggests the pair did not die natural deaths and decompose undisturbed in the cave. Bears would find it difficult to get to the cave as it is today but in the past a convenient tree may well have afforded entry to its excellent sheltered interior. Then, the problem faced by human hunters would be attacking a potentially lethal opponent with a short-range weapon from the precarious toehold afforded by the narrow ledge, tree or no tree. If the pair was killed in their den by some superhuman act of bravery and agility, the skeletal remains must have been selectively scavenged by human predators who removed many of the parts.

The only bones in the size range of deer, antelope, or goat, other than the scapulae, are fragments of: a distal tibia from Unit S (10-20 cmbd), a femur shaft from Unit U, a 2nd phalanx from Unit V (4-20 cmbd), and a thoracic vertebra from Unit X (40 cmbd-bedrock). Smaller animals, such as jack rabbits, cottontails, gophers, rabbits, rock squirrels, rats, mice, and a non-specific bird are represented by a few fragmentary bones, suggesting they are the residue of predators other than humans (see Table 3).

Figure 9. The claw sheath of the

unusually large adult bear.

22

Table 3. Human and Animal Skeletal Remains

Species Type Provenience

Phalanx, third (infant) A-5 (60-70)

Phalanx, third (infant) A-6 (70-80)

Incisor, deciduous B-5 (60-70)

Incisor, permanent B-5 (60-70)

Phalanx, third (infant) B-5 (60-70)

Molar, third, some wear C-5 (80-90)

Scapula, neonate fragment C-3 (60-70)

Incisor, shovel-shaped F-1 (24-34)

Phalanx, third (infant) G-3 (26-36)

Human

Phalanx, third (infant) X-1 (40-bdrk)

Skull fragments (3) A-1 (30-40)

Right tibia B-5 (61)

Front claw with sheath G-2 (39)

Left radius K (65)

Left ulna O-3 (37-48)

Left rib O-3 (52)

Acetabulum fragment U-surface

Left scapula Collection

Bear – mature

Thoracic vertebra W-1 (0-20)

Left rear 2nd phalanx W-1 (0-20)

Rib W-1 (0-20)

Right tibia RX-2 (29)

Ribs (3) I-backdirt

Left tibia I-backdirt

Left humerus I-backdirt

Right humerus I-backdirt

Left ulna I-backdirt

Bear – immature

Sacrum fragment I-backdirt

Distal tibia S-1 (10-20)

Femur shaft - chewed U-1 (0-20)

Deer or goat-sized

Thoracic vertebra fragment X-1 (40-bdrk)

Other fragments Provenience Lepus (Jack rabbit) A-3 (40-50); B-4 (50-60), O-3 (43-53); S-4 (40-bdrk); T-2 (20-30); T-4 (40-50);

V-1 (4-20); V-2 (20-30) Sylvilagus (cottontail) A-4 (50-60; G-1 (6-16) Neotoma (wood rat) B-4 (50-60) Sigmodon (cotton rat) G-5 (46-56) Perignathus (pocket mouse) B-4 (50-60) Geomys (gopher) V-1 (4-20) Bird A-4 (50-60) Unid. Rodent O-4 (40-50) rock squirrel?; S-1 (10-20); V-1 (4-20) () depth in cm below datum

23

Flora

Plants contributed three of the most important artifact categories in Cueva Pilote: agave fiber pads or "pin cushions" made of the fleshy trunk of the desert palm, Yucca carnerosa, commonly known as palma (Fig. 9); the spiny tips of agave leaves; and a domesticated gourd, Lagenaria siceraria. The only other plant parts that appeared with any regularity in the cave are acorn hulls and pine cones, probably introduced by the rodents whose occupancy was evidenced by tunnels and grass nests. Wood charcoal was also abundant but only one large fragment from Unit C-8, 112 cmbd, was identified to the species level before it was submitted for radiocarbon assay: this wrist-sized branch segment, 56 mm long and 53.5 mm in diameter, was cut from a walnut tree Juglans cf. microcarpa (Sobolik 1998).

Figure 10. Two agave fiber pincushions used to hold spines for bloodletting.

Agave Fiber Pads

Two slabs of desiccated agave were unearthed and given away by the curious vaquero. One of those was subsequently discarded or lost. The surviving piece is 20 cm long, 7 to 9 cm wide, and 8 mm thick and still holds five agave spines that were stuck into its flesh centuries ago (Fig. 10a). Two other slabs were recovered side by side, one in Unit O-2 and one from Unit H-1, both at a

24

depth of 33 cmbd. The smaller of the two retains the horizontal bands of cortex that indicate it was cut from the top of the yucca trunk (Fig. 10b). It is 20 cm long, 8 cm wide, and 8 to 14 mm thick. The larger Unit H specimen, an irregular rectangle, 38.5 cm long, 8.5 to 12 cm wide, and 12 mm thick, was submitted for radiocarbon assay and yielded a 13-corrected, uncalibrated date of 690+40 years B.P. (TX-9311). Presumably, this date applies to the entire agave assemblage in the vicinity of Unit H and in the cave as a whole.

Agave Leaf Tips or Spines

The scores of agave leaf tips or spines collected from the cave are but a portion of the total number observed (Table 4). Before it became apparent that the spines were more than casual inclusions, only a sample was picked from the fine screens. Therefore, the 192 in our inventory represent but a sample of the total left in the cave. Agave spines were most plentiful in the vicinity of the slabs of agave fiber with 39 coming from Unit H, 36 from Unit RX, and 33 from Unit V. Although none of the spines was still in place, the agave slab epidermis was riddled with small punctures, as many as 50 in the largest piece.

Table 4. Partial Distribution of Agave Spines

Unit/Level N Unit/Level N

A-5 (60-70) 1 M-3 (45-55) 4

B-1 (20-30) 8* M-5 (65-75) 1

B-5 (60-70) 1 M-6 (78-85) 2

D-1/2 (50-60) 1 O/R-4 (40-50) 4

E-3 (46-56) 2 P-1 (36-46) 2

G-3 (26-36) 1 RX-1 (0-20) 23

G-4 (36-46) 1 RX-2 (20-30) 7

H-1 (23-33) 22* RX-3 (30-bdrk) 6

H-3 (43-53) 4 U-2 (20-30) 1

H/Q-4 (40-50) 9 U-4 (40-bdrk) 32

H/W-5 (50-60) 4 V-1 (4-20) 11

I-backdirt 6 V-2 (20-30) 8

J-3 (30-40 1 V-3 (30-40) 14

J-4 (40-50) 2 X-1 (40-bdrk 2

L-2 (20-30) 1 Z 1

M-1 (25-35 6

M-2 (35-45) 4 TOTAL 192

() depth in cm below datum; * submitted for blood serum tests

25

The grave goods in Cueva Candelaria included small wooden tubes with shell or stone bases and agave spine inserts (Fig. 11), called guardapúas or spine cases (Martínez del Río 1954; Aveleyra et al. 1956:147-151). Scarifiers made of cactus spines, agave leaves, and rodent mandibles appear with some frequency in the archeological record in Coahuila (Taylor 1966: Figs. 23, 24); one of the latter was retrieved from Cueva Encantada, only 15 km south of Cueva Pilote. Taylor (1966) illustrates agave scarifiers where the spine is bent perpendicular to the leaf, thus positioning it for penetration or scratching without the need for a handle or haft. Curved and straight cactus spines wrapped in fiber bundles were recovered from Murrah Cave on the Pecos River (Holden 1937).

Samples from two proveniences, Unit B-2 and Unit H-2, were submitted to Paleo Research Laboratories for protein residue analyses to test the hypothesis that the spines were used in bloodletting rites, such as those documented in early historic accounts of Mesoamerican and other ceremonies. Extraction techniques are described in Appendix I, which contains excerpts from a report, by Cummings and Puseman (1998).

Five of eight spines from Unit B-2 and 10 of 22 from H-2 were washed for protein residues. The samples were then tested against human, agave, bear, bison,

bovine, cat, chicken, deer, dog, goat,

guinea pig, human, mouse, pig, rabbit, rat, sheep, turkey, catfish, and white perch antisera (see Appendix). As expected, both samples responded positively to agave antiserum. Sample 2, from Unit H and associated with the fiber pincushions, also reacted positively to human antiserum, indicating that some, if not all, of the 10 spines retained traces of human proteins. The negative response of the spines from Unit B can be attributed to a number of factors, ranging from preservation of sufficient quantities of human protein to distance from the center of ritual activity in Unit H. However, the

Figure 11. Guardapuás or spine cases with

shell bases. Both are about 3 cm tall. (Aveleyra et al 1956: Lám 44)

26

presence of human proteins in the Unit H sample is far more important than their absence in Unit B. In the context of the fiber slabs, the gourd vessel, and the exceptional suite of exotic artifacts, the positive protein results support the proposition that the agave spines are ritual paraphernalia most likely used in bloodletting ceremonies.

Domesticated Gourd

The fragments of domesticated gourd, Lagenaria siceraria (Minnis 1998), constitute the remains of one relatively large vessel deposited in the cave sometime around 600 radiocarbon years ago. Seen against a background of self-mutilation and ritual blood letting, the gourd may have been used as a container, an interpretation invoked by Aveleyra (et al. 1956) to explain similar fragments in Cueva Candelaria. Less pragmatically, gourds were one type of resonator used to amplify the sound of rasps, such as those played with painted deer scapulae (Turpin 1996). Two decorated gourds, one that may have been locally grown and one imported from the south, were found in one of the famous Laguna district burial sites, Cueva Paila (Aveleyra et al. 1956:190-193). One is a dipper that, like some of the fragments in Candelaria, has been repaired; a crack has been sewn together with thread (Aveleyra et al. 1956: Lám. X). The other is a shallow bowl with ornate decoration that was probably imported from Mesoamerica (Aveleyra et al. 1956:192-195). A mummy bundle from another cave in the Cuatro Ciénegas basin contained an incised gourd bottle (Taylor 1968). These various findings led Taylor (1966) to include gourd vessels as a trait of his Mayran burial complex. Smith (1937) and Jackson (1938) reported but did not elaborate upon gourd fragments in Texas caves, the former in the Big Bend near Alpine, and the latter in Culberson County.

The few fragments in Cueva Pilote present no definitive challenge to the traditional view of northern Coahuilan hunting and gathering economics (see Taylor 1968:26) although casual cultivation can not be ruled out on the basis of the current evidence. Despite excellent preservation of plant parts, excavations in the nearby habitation site, Cueva Encantada, produced no domesticated species in deposits that dated to the past 4,000 years. The gourd fragments in Cueva Pilote may well have been brought in from any of the surrounding horticultural areas, along with the mussel shell, marine shell, and exotic chert.

Cordage

Three short pieces of string in a private collection are attributed to the cave by the owner and one knot was recovered from Unit X, beneath a relic hunter's pit (Fig. 12). All are of different sizes and manufacture. The thinnest piece is only 3 cm long and 1.6 mm wide and is a single, very tightly Z-twisted cord. A thicker piece, 3.5 mm wide, of similar length is unraveling, showing it consists of two cords, each made of 12 individual fibers, Z-twisted together. The third piece is unusual in that it consists of three cords Z-twisted together. One of the strands, composed of about 10 individual fibers, is half the size of the other two but all three are wound together to form a patterned cord, 5.4 mm thick and now only

27

7.5 cm long. The fourth piece, the only one found during excavation, is from Unit X, beneath one of the relic hunter holes. It is a very fragile coil of relatively thick (2.7 mm) S-twisted cord.

None of the pieces can be associated with any of the other artifacts although it is quite possible that the beads, both marine and freshwater, were originally strung on cords such as these. If the scapulae served as rattles, they were probably connected by leather or fiber cords. Twisted fiber was an integral part of sandals, clothing, and carrying baskets so these small pieces of string may have been part of any one of a number of larger artifacts.

Shell Artifacts

Marine Shells Beads

Three common Atlantic Marginella (Prunum apicina or Marginella apicina) shells (Andrews 1981:61), probably imported from the Gulf of Mexico, were recovered from A-2, B-4, and C-1 (Table 5). All three are virtually identical (Fig. 13); the A-2 and B-4 specimens are 10 mm long; the C-1 shell is slightly smaller, measuring 9 mm in length. Each has a perforation in the body whorl next to the aperture; the B-4 shell also has a pinhole at its apex. Viewed microscopically, the aperture in the C-1 shell is quite rough and jagged except at the upper end where the lip is smooth and polished. Presumably, this wear results from friction with a cord that ran through the hole in the side of the shell and the natural opening at its base.

Although far from common, marginella shells have been found at several i sites in interior northern Mexico and neighboring Texas. Nearer the coast, MacNeish (1958:96) found 27 marginella beads in Armadillo Cave in the Sierra de Tamaulipas; 26 of them were in a level attributable to the Nogales phase, ca. 950 to 1450 B.P. Excavations in Cueva Ahumada, between Saltillo and Monterrey, produced at least three similar shells in 1997. A single marginella shell from Boca de Potrerillos, also near Monterrey in Nuevo León, had been strung on a cord

Figure 12. Fragments of cordage from private collections (above) and

recovered by excavation (below). The knot is approximately 3 cm wide.

Figure 13. Marine shell beads with punctures.

28

that ran through a hole in one side, like the Pilote specimens (Turpin et al. 1994:348).

Several marginella shells have been found in southwestern Coahuila, usually in mortuary contexts. A bundle burial from Coyote Cave, near Torreon, contained 16 beads, 15 of bone and one shell, strung on cords and gathered into a pendant (Taylor 1968:25). The same bundle produced a marine shell pendant of Busycon perversum L. (lightning whelk), another native of the Gulf of Mexico (Andrews 1981:55). NEMAP excavations in the Laguna Mayran district of Coahuila produced one marginella, five olivella, and four unidentified shell beads (Heartfield 1975:166) but these shells were modified by removing the apex, rather than puncturing the whorls.

Table 5. Provenience of Shell Artifacts

Marine Shell Mussel Shell Plaque A-2 (30-40) 1 F-2 (34-44) 1 C-1/2 (50-60) 1 B-4 (50-60) 1

Disc-Shaped Mussel Shell Beads

A-2 (30-40) 3 H-1 (23-33) 1

A-3 (40-50) 1 H-2 (33-43) 1

A-5 (60-70) 1 H-2 (37) 1

A-6 (70-80) 5 I-backdirt 2

B-2 (36,40) 2 J-6 (60-70) 2

B-3 (40-50) 1 M-3 (45-55) 1

B-4 (50-60) 2 M-4 (55-65) 1

B-5 (60-70) 1 M-5 (65-75) 1

C-1/2 (50-60) 1 M-6 (75-85) 1

D-3 (48) 1 P-3 (56-66) 1

D-5 (66-76) 1 RX-1 (10-20) 3

E-surface (26) 1 RX-2 (20-30) 1

E-3 (46-56) 1 S-3 (30-40) 1

F-3 (44-54) 1 S-4 (40-bdrk) 1

F-3 (48) 1 SX-1 (0-bdrk) 1

F-4 (54-64) 1 T-1 (10-22) 2

F-6 (79) 1 V-1 (4-20) 4

F-6 (76-86) 1 X-1 (40-bdrk) 1

G-3 (26-36) 2 Y-Z (surface) 1

G-4 (36-46) 1 TOTAL 56

() depth in cm below datum Clusters of marginella shells bound with lengths of coiled string formed

elaborate ornaments called flores or flowers that were buried with the dead in Cueva Candelaria, also in southwestern Coahuila (Aveleyra et al. 1956:126). These shells were punctured in much the same manner as the Pilote specimens and illustrate one of several ways (Fig. 14) that they could have been strung

29

(Aveleyra et al. 1956: Lám. 24). Although widely separated in time, the recorded examples of Marginella apicina beads corroborate the assumption that the disposal of rare, exotic material in Pilote is consistent with ritual contexts throughout the region.

Freshwater Mussel and Unidentified Shell Artifacts

The largest of the freshwater shell artifacts is the polished section of a right valve of a relatively large fresh water mussel, Cyrtonaias tampicoensis, commonly called the Tampico pearlymussel (Howells 1998; Howells et al. 1996:48-50) (Fig. 15). Both sides of the shell have been abraded to a high sheen, removing the rough exterior and most of the identifying characteristics except one muscle scar. The plaque was once diamond-shaped but one end has since snapped off. The broken corner may have been punctured, making the shell a pendant much like those found in burial contexts in Coahuila and Texas. The mussel shell may have been obtained from the Rio Grande, 100 km to the north, where this is one of the

Figure 14. Two ways of stringing marine shell beads: top) single string of shells

alternating with bone beads; above) flores made of dozens of shells (the flower part is approx. 3 cm across). (Aveyleyra et al. 1956: Lám. 21, 24)

30

more common species, or from the Río Sabinas, 160 km to the south, where no systematic inventory of its fossil mussel populations has been made.

The same shell may have served as raw material for some of the 56 round flat perforated shell beads that were recovered from stratigraphic excavations (see Table 5) and another that is reportedly in a relic hunter's collection. The beads are distributed throughout almost every excavated unit and every arbitrary level (see Fig. 8c). They range from 4.4 to 7.2 mm in diameter and from 1.3 to 3.4 mm in thickness although the latter measurement is exaggerated by one or two anomalous specimens. The hole, which is often slightly off center, is a consistent 1.75 mm in diameter. Taylor (1968:Figs. 2-4) and Aveleyra et al. (1956: Lám. 23) show how multiple beads were fastened together and interred with late prehistoric bundled burials in both Coyote and Candelaria caves in southwestern Coahuila (Fig. 16), providing a plausible analogue for the original configuration of the Cueva Pilote discs.

Snails

Although none of the 43 snails collected from Cueva Pilote is modified, their presence in the cave is culturally induced, thus making them manuports if not artifacts (Table 6, Fig. 17). All of the shells belong to the genus Humboldtiana and probably the species malenae (Metcalf 1983, 1998), which was named for its type locality, the cuesta where the La Babia-Boquillas road crosses the Encantadas just east of Pilote. Thus, these snails are native to the area (Metcalf and Riskind 1976; Metcalf 1984). However, all but four of the specimens are large adults so the sample is not representative of the natural age profile nor do snails commonly live in dry caves. Taylor (1966:62) listed Humboldtiana among his traits of the Ciénagas complex which would be much older than the material from Cueva Pilote. Large adult Humboldtiana shells were, however, recovered from the

Figure 15. Mussel shell plaque and discoidal

beads.

31

occupation site Cueva Encantada, only 15 km southwest of Pilote, in deposits dating from 600 to 4000 years old.

Despite their lack of modification, the shells were probably collected for ornamental or symbolic reasons. Alternating tan-brown and cream colored bands accentuate their spiral morphology (see Fig. 17), a shape with cosmic or religious implications in many art forms and one commonly found in the rock art of Coahuila.

Figure 16. Ways of joining discoidal shell beads: left) clusters of beads strung on cords into flores; center) beads glued to a spine case base; right) beads stacked on a stick (Cueva

Candelaria, Aveleyra et al. 1956: Lám. 23, 26).

32

Table 6. Distribution of Snail Shells

Unit/Level Depth (cmbd) Height Width Comments

A-5 60-70 * * 1 whorl, 1 column fragment

B-4 50-60 14.4 23.3* lower whorl broken

B-4 50-60 6.6 10.820.8 small, lower whorl broken

B-4 50-60 19.4 18.1 lower whorl broken

B-4 50-60 * * 2 whorl frags, burnt

B-5 60-70 11.1 14.6* bottom whorl broken

F-6 74-84 19.4 22.2 slightly flattened, hole in side

G-3 26-36 13.1 20.5 mouth opening rough

G-5 46-56 11.1 16.6 lower whorl broken

H-2 35-45 11.4 25.3 complete

H-3 43-53 16 16.6 bottom whorl broken

H-3 43-53 * * central fragment

I backdirt 20.4 25* partially broken bottom whorl

J-7 70-80 19/8 23.3* partially broken bottom whorl

L-2 20-30 * * large whorl fragment

L-3 30-4- * 22.4 top whorl broken, bottom cracked

L-4 40-50 18.4 28.1 complete

L-4 40-50 16.4 26.1 almost identical pair

M-3 45-55 20.1 23.9 bottom whorl broken

M-3 45-55 11.1 27.7 slightly browner

M-5 65-75 16.6 24.4* distinct brown bands on gray

RX-1 0-20 21.6 * 2 frags, columella measurement

S-3 30-40 * 25.1 cap missing

S-3 30-40 * 19.8 bottom broken

S-4 40-bdrk 15.2 24.4 distinct bands

S-4 40-bdrk 21.3 21.1 distinct bands

S-4 40-bdrk 10.1 11 one of the smaller specimens

SX 0-bdrk * * frags of large specimen

SX 0-bdrk * * frags of large specimen #2

SX 0-bdrk 22.2 27.6 eroded to white

T-3 30-40 * * frags of 2 snails, 1 lge, 1 small

T-4 40-50 15 14.6* one of the smaller specimens

T-4 40-50 16 22.9 indistinct brown band

T-5 50-60 * * fragment

U-2 20-30 11.1 * one of the smaller specimens

U-4 40-bdrk 22.8 23.3 grey indistinct bands

W 0-10 20.1 20.9 distinct brown bands on grey

Y-1, Z Surface * * frags of large specimens

Z 18.1 23.9 large but flatter

Z * 22.9 broken bottom, fragile

Z 9.4 11.3 smallest intact specimen

# measurements in mm; * broken or missing dimension

33

Lithic Artifacts

Projectile Points

Three complete and three fragmentary projectile points were recovered from controlled excavation (Table 7). Three others in various private collections may also have come from Cueva Pilote but several factors detract from any confidence in their actual provenience. One black chert Jora dart point, very like the specimen described below from Unit T, 29 cmbd, was in the same collection as two of the painted scapulae and the fiber pincushion that still held agave spines so it likely did come from Cueva Pilote. Unfortunately, this point was lost in the U.S. mail and cannot be traced. A photograph of one of the other painted scapula in a private collection also shows a hafted dart point (Langtry?), an agave spine, a phalanx, a human tooth, and a rib bone that would be consistent with the assemblage from Cueva Pilote but two other artifacts in the same picture are not - a bone awl and a piece of woven fiber - so the attribution of this specimen to Cueva Pilote is questionable. Finally, one of two projectile points in yet a third collection is purportedly from Cueva Pilote; the base of the dart point is broken so it is not typeable and the second specimen is a thick, stemmed arrow point which would be more consistent with the radiocarbon dates but anomalous in the lithic assemblage.

One complete projectile point (Fig. 18a) with secure provenience is a classic specimen of the Langtry type (Bement 1991: Fig. 2.3; Schuetz 1956:141; Suhm et al. 1954:438), attributable to the Middle Archaic San Felipe period (4100 to 3200 B.P.) in the neighboring Lower Pecos chronology (Turpin 1991:29-30). Intrasite provenience is O-3, 50 to 53 cmbd. This specimen is 43 mm long, 29.4 mm wide, and 4.8 mm thick, all measurements within the typical range for this type (Suhm et al. 1954:438; Bement 1991:61); the stem is just over 14 mm long or almost one-third of its total length. Raw material is grey-tan Edwards Formation chert.

Figure 17. Examples of large decorative snails found in

abundance at Cueva Pilote.

34

Table 7. Dart Points

Type Provenience Length* Width* Thick* Comments Langtry O-3 (50-53) 43 29.4 4.8 grey Edwards chert, cache

Jora T-2 (29) 56.4 26.8 5.2 local black chert

Jora/Langtry T-3 (30-40) ? 38.8 6.6 local black chert, cache

Langtry W-1 (0-10) 58 22.2 6.4 thick patina obscures material

Dart blade A-2 (30-40) ? ? 4.5 exotic yellow chert

Reworked blade I (backdirt) 27.7 12.1 5.5 dart chert

Jora Private coll. lost in U.S. Mail, photo only

* measurement in mm; () depth in cm below datum

An asymmetrical dart point of black chert (Fig. 18b), recovered from Unit T, 29 cmbd, fits well within the Jora type, a Middle Archaic type common in northern Coahuila, the Big Bend, and the Lower Pecos region (Taylor 1966: Fig. 3; Bement 1991). This specimen is 56.4 mm long, 26.8 mm wide and 5.2 mm thick. One barb is broken. The point was laying perpendicular to the cave entrance, with its stem on the interior.

The third complete specimen is a variation on the Langtry type distinguished by more pronounced barbs, one of which is broken, and a more excurvate blade (Fig. 18c). The point was made on a flake of coarse-grained grey chert that has developed a pinkish crusty patina that is more pronounced on one side. This dart point is 58 mm long, 22.2 mm wide, and a maximum of 6.4 mm thick and was recovered from the top 10 cm of Unit W where it was laying perpendicular to the entrance, stem inward, like the complete specimen from Unit T.

Unit T produced a partial dart point between 30 and 40 cmbd. The specimen snapped in mid-blade apparently during the reduction process, leaving a partially shaped Langtry or Jora-like base. This dart point is made of a locally available grey-black chert that is clearly distinguished by numerous minute white cracks that resemble crazing (Fig. 18d). Retrievable dimensions are 38.8 mm wide and 6.6 mm thick. A large knot on one side frustrated the maker's attempts to thin the blade; this face is lightly patinated but the obverse is not.

Only half of the blade of a dart point made of unusual yellowish chert with grey bands was recovered from Unit A, between 30 and 40 cmbd. A transverse break detached the base and one of the barbs as well as part of the blade. The fragment is well made, only 4.5 mm thick, and may well be part of yet another Langtry or Jora-like specimen (not illustrated).

35

A reworked blade and barb fragment of grey-brown chert, now 27.7 mm long, 12.1 mm wide, and 5.5 mm thick, was retrieved from the relic hunters' backdirt, Unit I (Fig. 18e). The morphology and technology suggest that the original point belonged to either the Langtry or Jora type.

A Matched Set of Biface Preforms

Only 14 bifaces and 4 unifaces were recovered from Cueva Pilote (Table 8), a total equaled by the number of scapulae. Similarly, more shell beads were collected than pieces of lithic debris so the manufacture, retooling, or use of stone tools was not a major activity inside the cave.

Table 8. Bifaces and Unifaces

Unit/Level Depth* Length^ Width^ Thickness^ % Blade

A-5 60-70 67.7 37.7 10.5 77.85

B-1 23 73.3 46.6 9 80.2

C-1 50-60 73.6 47.2 10 76.2

H-3 53 78.3 43.4 8.8 75.9

M-3 52 65 37.2 7.7 68

T-1 10-20 67.5 39 7.6 73.9

Other Bifaces Comments

A-2 30-40 42.4 29 6.8 primary flake

B-5 60-70 49.9 ? 10 fragment, local chert

C-4 71 69 32.5-12.7 9.9 spatulate

J-1 10-20 30.5 24.1 10 tear drop, local chert

M-2 25-35 52.2 30 7.4 reused flake

“T-3 30-40 51.1 28 14.1 small

U-4 40-bdrk 46 18.5 5.6 lozenge shaped

W-1 0-10 40.4 24.4 7.6 grey chert

Unifaces

C-8 120 85.6 61.2 24.6 igneous rock

H-2 35-45 56.8 30.5 7.6 side scraper

S-3 30-40 52.8 40.8 12.2 side scraper

S-3 30-40 44.2 32 10.8 side scraper

* in cm below datum; ^measurements in mm; “not illustrated

Six of the 14 bifaces are preforms that illustrate stages in the reduction process that, when added to the broken dart point in Unit T, culminate in the Langtry dart point from O-3 (Fig. 18f-k, Table 8). Bement (et al. in preparation) has analyzed them in detail and the pertinent characteristics are excerpted from his work in progress. These specimens are much older than the radiocarbon assays and were probably cached elsewhere, found, and placed in the cave as offerings.

36

Early Stage Preform

Unit A-5 (50-60 cmbd) produced a preform that is 67.7 mm long, 37.7 mm wide, and 10.5 mm thick (Fig. 18f). The stem and blade are alternately beveled in opposite directions and the edges are sinuous. The incipient blade is 77.85% (52.7 mm) of the specimen. The raw material is a light tan Edwards chert; a patch of cortex remains on one shoulder.

Late Stage Preforms

The preform from C-1 (50-60 cmbd) is 73.6 mm long, 47.2 mm wide, and 10 mm thick (Fig. 18g). The blade is 56.1 mm long or 76.2% of the specimen. Stem and body are alternately beveled. The raw material is banded brown-grey Edwards chert.

The specimen from B-1 (23 cmbd) is 73.3 mm long, 46.6 mm wide, and 9 mm thick (Fig. 18h). The blade is 58.8 mm long or 80.2% of the entire preform and alternately beveled. The raw material is locally available dark grey chert with the white linear inclusions characteristic of Encantada Valley sources.

The specimen from H-3 (53 cmbd) is 78.3 mm long, 43.4 mm wide, and 8.8 mm thick with a blade that is 59.4 mm long or 75.9% of the total artifact (Fig. 15i). The stem is thinner than the body. The raw material grades from reddish brown to grey tan chert and has minute crystalline inclusions.

The M-3 (45-55 cmbd) preform is well along in the reduction process; its stem has been thinned to within the range of finished Langtry points (Fig. 18j). This specimen is 65 mm long, 37.2 mm wide, and 7.7 mm thick but the stem has been thinned to 5 mm. The blade is 44.4 mm long or 68% of the entire artifact. The raw material is mottled tan Edwards chert.

The final specimen, from T-1 (10-22 cmbd), has a contracting stem with straight edges and a sharp point (Fig. 18k). It is 67.5 mm long, 39 mm wide, 7.6 mm thick and its body is 73.78% (49.8 mm) of its entire length. The raw material is speckled grey Edwards chert.

Other Bifaces and Unifaces

A thin bifacially worked end scraper made on a coarse-grained grey and black chert primary flake is 42.4 mm long, 29 mm wide, and 6.8 mm thick (Fig. 19a). One surface retains the rough cortex; the working edge is less than 20 mm long. Provenience is Unit A-2.

Half of a well-made ovate biface that broke longitudinally came from B-5 (Fig. 19b). The raw material is locally available grey-black chert with white linear inclusions. The remnant is 49.9 mm long and 10 mm thick.

An unusual spatulate biface from Unit C-4 is made of dark brown-grey chert with red inclusions (Fig. 19c). It is 69 mm long, 32.5 mm wide tapering to 12.7 on the stem, and 9.9 mm thick.

37

Figure 18. Projectile points and preforms. a) Langtry dart point; b) Jora dart point; c) Langtry variant dart point; d) broken preform; e) reworked dart point fragment; f-k) Langtry-Jora

preforms.

38

Figure 19. Other bifaces and unifaces. a-g) bifaces; h-k) unifaces.

A thumb-size, almost round, thin biface from Unit J-1 is 30.5 mm long, 24.1 mm wide, and 10 mm thick (Fig. 19d). It is made of locally available grey-black mottled chert.

A reused flake from Unit M-2 has been bifacially trimmed. It is 52.2 mm long, 30 mm wide, and 7.4 mm thick.

A crude thick biface from Unit T-3 is made of local grey-black chert but a white inclusion inhibited the thinning process leaving a large wedge on the upper surface of the tool (Fig. 19e). Slight polish on one lateral edge suggests it was used to scrape soft material. The specimen is 51.1 mm long, 28 mm wide, and 14.1 mm thick.

39

A lozenge biface made of fine-grained grey-black chert, from just above the bedrock in Unit U, is 46 mm long, 18.5 mm wide, and 5.6 mm thick (Fig. 19f). Morphologically, this specimen resembles the Fragua type defined by Taylor (1966:65) in Cuatro Ciénagas or the Lerma type described by Epstein in northeastern Mexico but neither has well-developed chronological or cultural associations that would make its type relevant to the interpretation of Cueva Pilote.

A small but crude biface from the upper level of Unit W is also made of mottled grey-black chert and retains patches of cortex on three corners (Fig. 19g). It is 40.4 mm long, 24.4 mm wide, 7.6 mm thick and has a 15 cm long bit.

Unifaces

A side scraper from H2 (35-45 cmbd), fashioned from a pink-grayish tan chert tertiary flake, is basically a uniface, although one edge has been sharpened by the removal of small flakes from both sides (Fig. 19h). The converse side has been reworked, leaving only one patinated patch to indicate that this flake was detached and exposed to the elements for some time before it was made into its final form. The tool is 56.8 mm long, 30.5 mm wide, and 7.6 mm thick.

Two unifacially worked flake scrapers were retrieved from Unit S, between 30 and 40 cmbd (Fig. 19i-j). One is a side scraper made on a secondary flake of dark grey chert with cortex on the lateral opposite the working edge. It is 52.9 mm long, 40.8 mm wide, and 12.2 mm thick with a working edge that is 44 mm long. A slight trace of red pigment remains on the ridge above a number of step fractures on the upper working edge. The smaller of the two is an end scraper 44.2 mm long, 32 mm wide, and 10.8 mm thick with a working bit 22.5 mm long. The raw material is a light grey chert with darker grey mottles and patches.

A large flat piece of igneous rock from Unit C-8 has been shaped by the removal of six or seven flakes from one edge, giving it the morphology of a scraper that admirably fits the hand (Fig. 19k). There is no indication of use and this coarse material is more amenable to tasks such as grinding and polishing than cutting. This artifact was in the deepest level in the site, and adjacent to a charcoal sample that produced a radiocarbon age of 700+35 B.P. or A.D. 1290 (TX-9340, see Table 1).

Flakes

The miscellaneous debitage consists of only 3 utilized flakes, 2 secondary flakes, 21 tertiary flakes, 7 chips, 3 core fragments, and 8 pieces of burned shatter, chunks, or potlids (Table 9). The latter are probably the result of accidental burning; fire-cracked rock, ash, and charcoal were mixed through the deposits but were more concentrated in the right front, near the entrance. None constituted what might be called an intact feature and none were securely provenienced enough to inspire much confidence in their radiocarbon dating.

40

Table 9. Distribution of Flakes

Unit/Level Sec. Tert. Chip Core Util. Misc.

A-2 (20-30) 1 1 1 burned cobble

A-3 (40-50) 1 chunk

A-4 (50-60)

A-5 (60-70) 1

A-6 (70-80) 3

B-2 (30-40) 1 1 burned chunk

B-4 (50-60) 2 1 1 1 core/chunk

B-5 (60-70) 1 heat-treated

C-2 (50-60) 1 1 burned shatter

C-6 (90-100) 1 2 1

I (backdirt) 1 1

K-3 (30-40) 1

S-1 (10-20) 1

S-3 (30-40) 1 1 1

S-4 (40-bdrk) 1 pot lid

SX-1 (48-74) 1 2 1 1 pot lid, 1 hematite

T-2 (20-30) 1

T-3 (30-40) 1 burned shatter

U-2 (20-30) 2

U-4 (40-bdrk) 1

W-1 (0-10) 1 spokeshave

Z-1 (0-10) 2

Totals 2 21 7 3 3 9

() depth in cm below datum

Ocher-stained Rocks

Three ocher-stained rocks of vastly disparate size were recovered from Units A and T. A very small fragment from T-4, 40-50 cmbd, is pure ocher. It measures 1.5 by 1 by 0.9 mm. A larger rock, from T-1 (10-22 cmbd), is the rounded edge of a piece of limestone that retains one smoothed side that is reddened by an ocher film. The third specimen, from Unit A, Level 6, is an irregular, flat piece of limestone, 8.8 cm long, 5.9 cm wide, and 1.5 cm thick. One side is covered with an orange crust that is easily detached to make pigment. One corner has been reddened by exposure to intense heat.

Smoothed Pebble

A flat, oval pebble of pinkish fine-grained limestone from the surface of Unit Z has one smoothed face. Although it is small, only 40 mm long, 23 mm

41

wide, and 7.7 mm thick, this stone would have worked admirably as a tool used in the final finishing of little objects made of softer material, much like fine sandpaper. Regardless of its use, this specimen was carried into the cave and is not a natural by-product of rock decay.

Discussion of Lithic Artifacts

The six preforms, the broken dart point from Unit T, and the complete Langtry point in Unit O are probably part of a matched set or cache that was manufactured thousands of years before the main utilization of Cueva Pilote for ritual events. Some pieces are made of chert from the Edwards Formation, others are of local coarser material. The overall paucity of debitage, and specifically debris attributable to the Edwards Formation (one flake), indicates the cache was not made in the cave, but rather introduced in its current form. Typologically, all of the projectile points, and by extension the preforms, are roughly the same age and all are much older than the range indicated by the four radiocarbon dates. Although there is some ambiguity about the duration of the Jora type in Mexico, where it may have been in favor much longer than the 1000 years defined by the Texas radiocarbon dates, no such doubts can be applied to the Langtry style. Therefore, either the cave was used by Middle Archaic folks who left behind a cache of biface preforms, some projectile points, and little else, some 3000 to 4000 years ago, or these lithic artifacts were put in to the cave, perhaps as an offering, by the people who used it for other ritual activities. What appears to be purposeful placement of two of the projectile points, parallel to each other and perpendicular to the entrance, and the distribution of the presumed biface cache near the Langtry point (see Fig. 8d) hints at ritualized discard and favors the latter explanation although, as noted throughout, the overall distribution of artifacts is suspect.

42

Conclusions

Cueva Pilote is one of the rarest of finds - a hunter-gatherer ceremonial site where arcane rituals involving self-mutilation and bloodletting were performed. The conformation of the cave is the physical embodiment of the symbolism inherent in rites of passage - the transition from open to closed, from light to dark, from normality to sacred space (Van Gennep 1961). The nature of the rituals, as inferred from the physical setting and certain of the artifacts, links the people of the Encantada Valley to their northern Mexican relatives who had already climbed from the Chichimec sea into the Valley of Mexico.

The artifact assemblage finds coherence within a model of ritual performance, atonement, and sacrifice. Ethnographically, the painted scapulae, the most unusual class of material culture in Cueva Pilote, are percussion instruments that accompany dance performances or serve as a background for trance inducement (Turpin 1996). Although few specimens are known archeologically, modern kachina dancers in the southwestern United States still include painted deer scapulae in their ceremonial regalia (Wright 1973, 1985). Either hand-held or fastened to their clothing, scapula rattles contribute to the rhythm of the dance. Other kachinas play rasps with deer scapula abraders, sometimes using gourds as resonators to amplify the sound. Worn areas around the neck of some of the Pilote specimens suggest they were strung on thongs, perhaps to be used as rattles.

Scapulae are also used in some parts of the world for prognostication. In Pilote, the scapulae may also have been offerings, left in the cave like many of the other artifacts for unknown reasons probably associated with propitiation and sacrifice. Regardless of their specific function, the painted scapulae are rare, labor-intensive artifacts that conjoin the sacred qualities of deer, bone, ocher, and music to serve a corollary role to the ritual activities conducted in Cueva Pilote.

The agave fiber pincushions and the scores of agave spines can best be explained by referring to the ethnographic accounts of self-mutilation and bloodletting to propitiate the forces of nature, in atonement and penitence, and as part of initiation ceremonies for young men (Alegre, I 1841:450-451; Aveleyra et al. 1956:149-150; Durán 1971:82, 263; Griffen 1969:124; Katz 1972; Martínez del Río 1954:87-88; Sahagún 1956, II:76-81). Bloodletting was an integral and unquestionably long, more primitive part of the institutionalized politico-religious system in the complex societies that flourished in the heart of Mexico and the instrument of choice was often the spines of maguey (agave), in itself a revered and holy plant (Nicholson 1971:420). Auto-sacrifice was a right and obligation of kings and priests that was extended to others, such as initiates, penitents, and participants in specific rites where it may have substituted for human or animal sacrifice

43

Durán’s (1971:119-120) description of the priestly rituals practiced in the Temple of Tezcatlipoca (Fig. 20) suggests a possible function for the agave slabs:

Once the midnight ceremony [burning incense] had ended, everyone withdrew to an ample chamber which contained many seats made of wood and grass. They seated themselves, each one taking a maguey thorn to pierce the calves near the shinbone. They squeezed out the blood and wiped it on their temples. After they had smeared their temples, all the thorns used for piercing and for self-sacrifice were smeared with the remaining blood. Then the thorns were stuck into some large balls of straw which were placed between the merlons of the courtyard wall. These balls were always there for that purpose, decorated with branches. The thorns were left there so that everyone could see the penance and suffering which [the ministers] had inflicted upon themselves, being men who endured pain for the people. In certain chambers of the temple a large number of these thorns were stored; the used thorns were removed every day, were put away, and new ones set into the straw balls, since no thorn could be used twice.

Figure 20. This illustration in Duran’s (1971: Plate 11) codex shows one priest piercing his shin while another burns incense to the god Tezcatlipoca. After the bloodletting, the used maguey

spines were placed in the ball of straw in the upper right corner to insure that none was reused.

44

The substitution of cactus for straw balls was noted by Sèjourné (1976:123, Fig. 44) in a fresco at Teotihuacán where there is also a mural representation of a deity (Fig. 21) in which the central figure is flanked by five maguey spines stuck in bundled fiber (Fine Arts Museum of San Francisco 1993:43; Uriarte 1996:Fig. 1).

Figure 21. In this Teotihuacan mural entitled “Blood offering with maguey spines,”

the central deity is framed by bundles holding maguey spines (Fine Arts Museum of San Francisco 1993; Uriarte 1996: 391).

The analogy is strengthened by a scene carved into the interior of a colossal jaguar effigy vessel used to hold human hearts. This monumental piece of Aztec sculpture was discovered in 1901, one block from the cathedral in downtown Mexico City (Nicholson and Quiñones 1983). The relief shows two figures identified as seated deities, Tezcatlipoca and Huitzilopochtli, who face each other while drawing blood from their ears (Nicholson and Quiñones 1983:31; Smith 1996:Fig. 9.6). At their feet are two slabs of maguey leaf holding four maguey spines of the sort used for auto-sacrifice (Fig. 22). Bernal (1969: Plate 21) suggests that this sculpture was created in 1473, within a few hundred years of the radiocarbon dates generated from Cueva Pilote. Thus, it appears that the religious ideas and practices illustrated in the murals of Teotihuacán endured in dilute form in the remote Encantada Valley while, in the valley of Mexico, they were elaborated into one of the most sanguinary episodes in the history of human society.

Closer to Cueva Pilote, a ceremony held in 1607 in the vicinity of Mapimí, in the Laguna district of Coahuila and Durango south of the Encantadas, relied upon bloodletting to avert the dire effects of a comet that appeared in the sky.

...These Indians believe that many people will die whenever a comet appears, and so it is entertaining to see what some of these old sorcerers,

45

who are also doctors, do. On the first night, after the comet begins to appear, to appease it and to tell it not to do them harm, they offer it some small baskets filled with fish, fruit and mesquite beans—a native tree—and other fruits and herbs that they eat; they burn them in a bonfire that they have made so that the smoke, once released, will reach up to where the comet appears; and so that the smoke will go straight up without drifting, four or five old men with some very wide lashes whip the smoke all over as it begins to rise; and if at this point, a strong wind blows it apart and disperses it, they take it as a bad omen and everyone begins to cry as one, and in the meantime they puncture their arms and chests with some thorns until the blood flows and one of the old men gathers and mixes it with a little water, cuts the hair of a ten or eleven year old maiden and from the hair they make a hyssop with which they sprinkle the air with that blood, as they turn, they let out some very strange snorts that are horrifying to hear (Documentos para la Historia de México 1857:86-87).

Three different accounts of this event - the Documentos quoted above, Alegre (1841:450-451), and Pérez de Ribas (1992:694) - vary somewhat in minor details but the general scenario is the same in all three (Griffen 1969:123-124). After the smoke created by incinerating baskets full of food failed to placate and divert the comet, old men (magicians or healers) punctured their arms and chest with spines (Pérez de Ribas says peines or combs, and includes newborn infants among the victims) and collected the blood which they diluted with water. They cut the hair of a young maiden (or six maidens according to Pérez de Ribas), dipped it in blood, and flailed the air, in the cardinal directions, while crying out horrendously. The remaining blood was thrown in the fire. Although this response to astronomical threat was met with skepticism by the good padres who recorded it, in some ways, the theme persists in our modern preoccupation with asteroid collisions and other Doomsday devices threatening the world from outer space.

Figure 22. This relief from the interior of a monumental Aztec

sculpture shows the gods Tezcatlipoca and Huitzilopochtli

letting blood from their ears with bone awls. Note the agave (maguey) spines stuck in the maguey leaf slab between the kneeling deities (Smith

1966: Fig. 9.6)

46

Other references to bloodletting in Coahuila identify a variety of implements that were applied to the task. Alonso de León described a Coahuiltecan ceremony in which both sexes danced around the fire, consumed peyote, and drew large quantities of blood by scratching themselves with the beak of a fish, probably the long-nosed gar. Scratching one’s skin with obsidian or ground up maguey leaves was reportedly a remedy for tiredness (Griffen 1969:124).

The importance granted to agave spines as ritual bloodletting instruments is ethnographically documented in countless references and archeologically evident in a large number of unusual artifacts entombed in Cueva Candelaria. Small wooden cylinders, attached to shell or stone discs that served as a base, held agave spines, giving rise to their name guardapúas or spine cases (Martínez del Río 1953; Aveleyra et al. 1956:147-150). From their positions in some of the burials, Aveleyra reasoned that the guardapúas were ear plugs, designed to hold the spines used in auto-sacrifice and scarification, traits that he thought indicated Mesoamerican influences on the people of the Laguna district of southern Coahuila.

The exotic materials in the Cueva Pilote artifact assemblage confirm that the people of the Sierra Encantada had widespread contacts outside the valley and its immediate environs. Marine shell from the Gulf of Mexico had to be transported over 450-500 straight-line km, a distance greatly lengthened by the difficult terrain. Mussel shells could only be obtained in river valleys 100 km north and 160 km south of the Encantada. The landmark Edwards Formation chert used to fashion some of the stone tools has yet to be found near Cueva Pilote although the identification of local lithic sources is admittedly poor.

The fragments of domesticated gourd in Cueva Pilote find their analogues in whole gourd vessels from Cueva Paila and fragments from Candelaria. One of the gourd bowls from Paila was an exotic species imported from the south. Its cloisonné‚ decoration was cited as further proof of a Mesoamerican connection. Aveleyra (et al. 1956) points out that Lagenaria siceraria was grown as raw material for implements and vessels, not for food. Gourds were also part of the suite of musical instruments, serving as resonators for rasps that were played at various ceremonies. The possible uses of a gourd vessel are manifold but a most logical explanation for the Cueva Pilote specimen is that it served as a container for blood shed during self-mutilation rites.

The presence of domesticated gourd does not necessarily challenge the traditional view that the dominant economic strategy was hunting and gathering. They, like the shells and some of the lithic raw material, may have been obtained from horticulturists occupying the Rio Grande valley, the Laguna district to the south, or the Huasteca to the east.

The note of incongruity introduced by the number of stone tools that are minimally 2500 years older than the radiocarbon dates for the agave-gourd-scapula complex can be dispelled if the lithic assemblage is placed within the ceremonial context implied by the rest of the material culture and the site setting.

47

Offerings of antique specimens, including a cache of Langtry preforms—bifaces made of both local and non-local chert—are consistent with the theme of abnegation that permeates the site. Presumably, the technological skills of the ancient crafters were recognized and valued by their successors who may have endowed them with special powers.

Although little confidence can be placed in the distribution of many of the artifacts, it is unlikely that two of the antique dart points ended up parallel to each other and perpendicular to the cave entrance, just as it is improbable that two long bones of the adult bear gravitated to a position jammed under protruding ledges or rocks on either side of the doorway. Their fixed position, flanking the entry, suggests a ritual placement rather than simple discard or natural death. All of the fauna found in the cave was locally available, but the smaller animals were most likely brought in by predatory animals or raptors, unlike the bear and deer which represent sacred animals.

The human skeletal remains are more of a puzzle. There is no evidence that Cueva Pilote was a burial cave other than the presence of five infant finger bones, a neonate scapula fragment, and four teeth, three of which are incisors. The recovery of these durable parts of the human body is not in itself unusual but, in a small, unoccupied cave like Pilote, a wider range of skeletal material might be expected from primary interments. Teeth and infant finger bones have been found strung on cords, presumably for ornamentation and ritual use, but there is no precedent for the neonate scapula.

Another disconcerting anomaly is the distribution of the 56 discoidal shell beads which were found throughout the deposits, horizontally and vertically. One possible method of introducing such a large numbers of beads was observed at Candelaria where similar beads were tied into clusters and looped together into flores or bead flowers. The mechanism by which they were spread throughout the cave is puzzling but a similar randomness is evinced by the scapula fragments, the snails, and to some degree, the flakes. On the other hand, the large items, such as the whole scapulae, the bifaces and projectile points, the agave pincushions, and some of the bear bones appear to be purposefully positioned and in their place of original discard. Although rodent and relic hunter activity are obviously factors, their effects did not permeate the entire cave so it is unlikely that this distributional puzzle can be that easily explained.

Despite the fact that Cueva Pilote was not primarily a burial site, the similarities between it and Cueva Candelaria, as well as the contemporaneity indicated by the radiocarbon dates, align Pilote with Taylor’s (1966) Mayrán complex, which he derived from the mortuary caves of southern Coahuila. The human remains may have arrived at the site by any one of a number of paths, but preservation of more skeletal parts would be expected of a cemetery or primary interments on a larger scale.

The radiocarbon assays also place Cueva Pilote within the Jora complex, Taylor’s (1966) equivalent of the Late Prehistoric period in Texas, a time when external influences can be detected in the material culture. Like Candelaria,

48

Cueva Pilote has affinities with both southern and northern cultural areas, dispelling the idea that the Encantada Valley, in all its remoteness, was totally isolated from developments taking place on a larger scale. Cueva Pilote illustrates the flow of goods into the valley but the medium of exchange is largely speculative at this time. It is perhaps worthy of note that African food producers, expanding into the traditional homeland of hunters and gatherers, routinely entered into exchange networks in which religious expertise, such as rainmaking, hunting success, and healing, was symbolic labor that was paid for with livestock (Hall 1994). Rain-making abilities were particularly prized, as would be expected in arid lands such as those inhabited by the Bushman and those surrounding the Sierra Encantada, including much of highland Mexico.

Teotihuacán, with its juxtaposition of pyramids and caves (Heyden 1981) and mural art that illustrates the close relationship between blood sacrifice and water (Parsons 1988), epitomizes the complex role that topography and hydrology played in the local belief systems. Rain and the consequent fertility were a dominant religious theme in all of pre-Hispanic Mexico where caves and mountains were the dwelling places of lightning, rain, and wind (Nicholson 1971:414). Caves were the door to the abode of the rain and earth deities, as well as the land of the dead (Heyden 1981:19, 27) and the place of emergence for the sun, the moon, and many of the ethnic groups. The association of caves with water, and fertility is evidenced in their use for ritual purification after childbirth, as tombs of drowning victims, for the sacrifice of children, and as depositories of riches.

Elsewhere in the Americas, mountains are so strongly associated with rain that the rock art symbol for precipitation is a mountain goat in the western United States (Schaafsma 1997; Whitley 1992) and its equivalent, a vicuña, in the Chilean Andes (Berenguer 1989). No one knows how the Sierra Encantada came by its name, but it is possible that the Spanish explorers who first encountered these intimidating heights were influenced by local lore or practice, recognizing and perpetuating their mythic qualities or reputation as a place of enchantment. Regardless of the motivation, Cueva Pilote is a unique site where people with a relatively simple technology, economy, and level of social organization, whether they were penitents, initiates, sorcerers or supplicants, practiced arcane rites that were one of the bases of power and authority among their complex neighbors to the south.

49

References Cited

Alegre, Francisco Javier, S.J. 1841 Historia de la Compañía de Jesús en Nueva-España. México, D.F.

Andrews, Jean 1981 Texas Shells: A Field Guide. University of Texas Press, Austin.

Aveleyra, Luis, M. Maldonado-Kordell, and Pablo Martínez del Río 1956 Cueva de la Candelaria. Instituto Nacional de Antropología e Historia,

México, D.F.

Bednarik, Robert 1984 On the Nature of Psychograms. The Artefact 8:27-32.

Bement, Leland C. 1991 The Statistical Analysis of Langtry Variants from Arenosa Shelter, Val

Verde County, Texas. In Papers on Lower Pecos Prehistory, edited by S.A. Turpin, pp. 51-64. Texas Archeological Research Laboratory Studies in Archeology 8. The University of Texas at Austin.

Bement, Leland C., Solveig A. Turpin, and Herbert H. Eling, Jr. n.d. A Middle Archaic Lithic Reduction Sequence from Northern Coahuila.

Ms. in preparation.

Berenguer, José 1989 A Survey of Chilean Rock Art. Paper presented at the International Rock

Art Conference, San Antonio, Texas.

Bernal, Ignacio 1969 100 Great Masterpieces of the Mexican National Museum of

Anthropology. Harry N. Abrams, New York.

Chavero, D. Alfredo 1962 México Á Través de los Siglos, Tomo Primero. Ballescá y Comp.,

Editores, México.

Cummings, Linda Scott and Kathryn Puseman 1998 Pollen and Protein Residue Analysis for Cueva Pilote, Encantada Valley,

Northeastern Mexico. Paleo Research Labs Technical Report 98-42. Golden, Colorado.

Deal, Dwight 1976 The Geologic Environment of the Solitario, Brewster and Presidio

Counties, Texas. In The Solitario, A Natural Area Survey 9:17-66. Lyndon B. Johnson School of Public Affairs, The University of Texas at Austin.

50

Documentos para la Historia de México 1857 Del anua del año 1607. Documentos para la Historia de México,

Cuarta Serie, Tomo III. Imprenta de J.R. Navarro. México

Durán, Fray Diego 1967 Libro de los Ritos y Ceremonias en Las Fiestas de los Dioses y

Celebración de Ellos. In Historia de las Indias de Nueva España e Islas de la Tierra Firme, Tomo I. Biblioteca-Porrua, México.

1971 Book of the Gods and Rites, and The Ancient Calendar. Trans. by Fernando Horcasitas and Doris Heyden. University of Oklahoma Press, Norman.

Fine Arts Museum of San Francisco 1993 Teotihuacan: City of the Gods. Guidebook for the Teotihuacan

Exhibition. San Francisco.

Flores, Blas M. 1881 Reseña Historia de las Campañas Contra los Salvajes en la Frontera del

Norte en los Años de 1880 y 1881. Archivo General de Saltillo, Coahuila.

Gentry, Howard S. 1982 Agaves of Continental North America. The University of Arizona Press,

Tucson.

González Arratia, Leticia 1992 Ensayo Sobre La Arqueología en Coahuila y el Bolsón de Mapimí.

Archivo Municipal de Saltillo, México.

Griffen, William B. 1969 Culture Change and Shifting Populations in Central Northern Mexico.

Anthropological Papers of the University of Arizona 13, The University of Arizona Press, Tucson.

Hall, Simon 1994 Images of Interaction Rock Art and Sequence in the Eastern Cape. In

Contested Images, edited by Thomas Dowson and J. David Lewis-Williams. Witswatersrand University Press, South Africa.

Heartfield, Lorraine 1975 Archeological Investigations of Four Sites in Southwestern Coahuila,

Mexico. Bulletin of the Texas Archeological Society 46:126-177.

Hedges, Ken 1982 Phosphenes in the Context of Native American Rock Art. American

Indian Rock Art 7,8:1-10.

Heyden, Doris 1981 Caves, Gods, and Myths: World-View and Planning in Teotihuacan.

Mesoamerican Sites and World-Views. Dumbarton Oaks Research Library and Collections, Washington, D.C.

51

Holden, W.C. 1937 Excavation of Murrah Cave. Bulletin of the Texas Archeological and

Paleontological Society 9:48-73.

Howells, Robert G. 1998 Personal communication to Solveig A. Turpin. Heart of the Hills

Research Station, Kerrville, Texas.

Howells, Robert G., Raymond W. Neck and Harold D. Murray 1996 Freshwater Mussels of Texas. Texas Parks and Wildlife Press, Austin.

Jackson, A.T. 1938 Picture Writing of Texas Indians. Bureau of Research in the Social

Sciences II, Anthropological Papers 27. University of Texas, Austin.

Katz, Friedrich 1972 The Ancient American Civilizations. Praeger Publishers, Inc. New York.

Lewis-Williams, David and Thomas A. Dowson 1988 Signs of the Times: Entoptic Phenomenon in Upper Paleolithic Art.

Current Anthropology 29:201-245.

MacNeish, Richard S. 1958 Preliminary Archaeological Investigations in the Sierra de Tamaulipas,

Mexico. Transactions of the American Philosophical Society 48(6), Philadelphia.

Mallery, Garrick 1893 Picture-Writing of American Indians. Tenth Annual Report of the Bureau

of American Ethnology for 1888-1889. Washington, D.C.

Martínez del Río, Pablo 1953 La Cueva Mortuoria de la Candelaria, Coahuila. Cuadernos Americanos

4:177-204.

1954 La Comarca Lagunera a Fines del Siglo XVI y Principios del XVIII según las Fuentes Escritas. Publicaciones del Instituto de Historia 30. México, D.F.

Mason, Charles T., Jr. and Patricia B. Mason 1987 A Handbook of Mexican Roadside Flora. University of Arizona Press,

Tucson

Mallouf, Robert J. 1985 A Synthesis of Eastern Trans-Pecos Prehistory. Master's thesis, The

University of Texas at Austin.

McKinney, Bonnie R. 2005 Personal communication, Wildlife Coordinator, Proyetco El Carmen,

Cemex Maderas del Carmen, Coahuila, Mexico.

52

Metcalf, Artie L. 1983 A New Humboldtiana (Pulmonata: Helminthoglyptidae) from

Northwestern Coahuila, Mexico. The Nautilus 97(2):69-72.

1984 A New Humboldtiana (Pulmonata: Helminthoglyptidae) from Extreme Eastern Chihuahua, Mexico. The Nautilus 98(4):145-148.

1998 Notes and comments on the snails collected from Cueva Pilote. University of Texas at El Paso.

Metcalf, A.L. and D.K. Riskind 1976 A New Humboldtiana (Pulmonata: Helminthoglyptidae) from Coahuila,

Mexico. The Nautilus 90:98-100.

Minnis, Paul E. 1998 Notes on gourd fragments from Cueva Pilote. Department of

Anthropology, University of Oklahoma, Norman.

Muller, C.H. 1947 Vegetation and Climate of Coahuila, Mexico. Madroño 9:33-57.

Nicholson, H.B. 1971 Religion in Pre-Hispanic Central Mexico. Handbook of Middle American

Indians Vol. 10, Archaeology of Northern Mesoamerica, Part 1. University of Texas Press, Austin.

Nicholson, H.B. and Eloise Quiñones Keber 1983 Art of Aztec Mexico, Treasures of Tenochtitlan. National Gallery of Art,

Washington, D.C.

Parsons, Mark L. 1988 The Iconography of Blood and Sacrifice in the Murals of the White Patio,

Atetelco, Teotihuacan. Master's thesis, The University of Texas at Austin.

Pérez de Ribas, Andrés 1992 Historia de los Triumphos de Nuestra Santa Fee. Facsimile of 1st

edition, 1645, Madrid. Siglo Veintiuno Editores, México, D.F.

Prewitt, Elton R. 1970 Notes on Some Trans-Pecos Archeological Materials in the Smithsonian

Institute, Washington, D.C. Texas Historical Survey Commission Archeological Report 18, Austin.

Reichel-Dolmatoff, Gerardo 1978 Drug-induced Optical Sensations and their Relationship to Applied Art

Among Some Colombian Indians. In Art and Society: Studies in Style, Culture and Aesthetics, edited by Michael Greehnaigh and Vincent Megaw, pp. 289-304. Duckworth, London.

Sahagún, Bernardino de 1956 Historia General de las Cosas de la Nueva España. México, D.F.

53

Schaafsma, Polly 1997 Rock Art Sites in Chihuahua, Mexico. Museum of New Mexico

Archaeology Notes 171, Santa Fe.

Schoolcraft, H.R. 1854 Historical and Statistical Information Respecting the History, Conditions,

and Prospects of Indian Tribes of the United States, Vol. IV. Bureau of Indian Affairs, Philadelphia.

Schuetz, Mardith 1956 An Analysis of Val Verde County Cave Material. Bulletin of the Texas

Archeological Society 27:129-160.

1961 An Analysis of Val Verde County Cave Material, Part II. Bulletin of the Texas Archeological Society 31:167-206.

Sèjourné, Laurette 1976 Burning Water: Thought and Religion in Ancient Mexico. Shambala,

Berkeley.

Smith, Charles I. 1970 Lower Cretaceous Stratigraphy, Northern Coahuila, Mexico. Bureau of

Economic Geology Report of Investigations 65, The University of Texas at Austin.

Smith, Michael E. 1996 The Aztecs. Blackwell. Oxford, U.K.

Sobolik, Kristin 1998 Letter identifying wood charcoal as Juglans cf. microcarpa. Department

of Anthropology, University of Maine, Orono.

Stuiver, M. and P.J. Reimer 1993 Radiocarbon Calibration Program 3.0.3. Quaternary Isotope Lab,

University of Washington, Pullman.

Suhm, Dee Ann, Alex D. Krieger and Edward B. Jelks 1954 An Introductory Handbook of Texas Archeology. Bulletin of the Texas

Archeological Society 25:1-562.

Taylor, Walter W. 1964 Tethered Nomadism and Water Territoriality: An Hypothesis. Acta 35th

Congreso Internacional de Americanistas: 197-203.

1966 Archaic Cultures Adjacent to the Northeastern Frontiers of Mesoamerica. In Handbook of Middle American Indians, Vol. 4. Archaeological Frontiers and External Connections edited by Gordon F. Ekholm and Gordon R. Willey, pp. 59-94

1968 A Bundle Burial from Coahuila, Mexico. In Collected Papers in Honor of Lyndon Lane Hargrave, edited by Albert H. Schroeder. Papers of the Archaeological Society of New Mexico 1:23-56.

54

Turpin, Solveig A. 1991 Time Out of Mind: the Radiocarbon Chronology of the Lower Pecos River

Region. In Papers on Lower Pecos Prehistory, edited by S.A. Turpin, pp. 1-51. Texas Archeological Research Laboratory Studies in Archeology 8. The University of Texas at Austin.

1992 More About Mortuary Practices in the Lower Pecos River Region of Texas. Plains Anthropologist 37(138):7-17.

1993 Hunting Camps and Hunting Magic: Petroglyphs of the Eldorado Divide, West Texas. North American Archaeologist 13(4):295-316. Baywood Publishing Co., New York.

1996 Painting-On-Bones and Other Media in the Lower and Trans-Pecos Region of Texas and Coahuila. Plains Anthropologist 41(157):261-272.

1997 An Agave Pincushion from a Dry Rockshelter in Coahuila, Mexico. La Tierra 23(3):21-24.

Turpin, Solveig A. and Stephen M. Carpenter 1994 Prehistoric Sandals from the Sierra de la Encantada, Coahuila, Mexico.

La Tierra 21(2):7-14.

Turpin, Solveig A., Herbert H. Eling, Jr., and Moisés Valadez Moreno 1994 The Archaic Environment of Boca de Potrerillos, Nuevo León, Mexico.

North American Archaeologist 15(4):331-357.

Uriarte, María Teresa 1996 De Teotihuacanos, Mexicas, Sacrificios, y Estrellas. In La Pintura Mural

Prehispánica en México I: Teotihuacan, Tomo II: Estudios, cordinated by Beatríz de la Fuente. Universidad Nacional Autónoma de México, D.F.

Van Gennep, Arnold 1961 The Rites of Passage. University of Chicago Press.

Weitlaner Johnson, Irmgard 1977 Los Textiles de la Cueva de la Candelaria, Coahuila. Departamento de

Monumentos Prehispánicos, México, D.F.

Whitley, David S. 1992 Shamanism and Rock Art in Western North America. Cambridge

Archaeological Journal 2(1):89-113.

Wright, Barton 1973 Kachinas: A Hopi Artist's Documentary. Cliff Bahnimptewa. Northland

Press, Flagstaff, Arizona.

1985 The Kachinas of the Zuni. Northland Press, Flagstaff, Arizona.

55

Appendix: Protein Residue Analysis

Linda Scott Cummings and Kathryn Puseman

Two samples of agave spines from different proveniences were tested against prepared animal antisera obtained from Organon Teknika Corporation and Sigma Chemical Company, and against agave and bison sera raised under the direction of Dr. Richard Marlar at the Thrombosis Research Laboratory in the Denver VA Medical Center and the University of Colorado Health Sciences Center. Samples were tested using a technique referred to as cross over immunoelectrophoresis (CIEP or COE). The method for CIEP is based on forensic work by Culliford (1964, 1971) with changes made by Newman (Newman and Julig 1989) following the procedure used by the Royal Canadian Mounted Police Serology Laboratory in Toronto, Canada. Further changes were made at Paleo Research Labs following the advice of Dr. Richard Marlar.

A group of five spines from Sample 1 (Unit B-2) and a group of ten spines from Sample 2 (Unit H-2) were washed using 2 mL of a 0.02M Tris hydrochloride, 0.5M sodium chloride, and 0.5% Triton X-100 solution. The spines were placed in an ultrasonic bath for 30 minutes, on a rotating mixer for 30 minutes, then back in the ultrasonic bath for an additional 30 minutes. The extracted solutions were stored in a refrigerator using polypropylene microcentrifuge tubes.

CIEP was performed using agarose gel as the medium. The samples were first tested against pre-immune goat serum (serum from a non-immunized animal) to detect non-specific binding of proteins. Non-specific binding is absent if a negative result is obtained. Samples were electrophoresed in Barbital buffer (pH 8.6) for 45 minutes at a voltage of 130v. Samples were then pressed and rinsed in 1M saline solution overnight to remove extraneous proteins.

The next morning, the gel was washed, pressed, dried, stained in a Coomassie Blue solution, then destained. Gels were observed to determine if non-specific binding was occurring. Positive reactions appear as a line of precipitation between the two wells. All samples tested negative against pre-immune serum and were then tested against various antisera. Appropriate positive and negative controls were run for each gel. A positive control consists of the blood of each species tested, and a negative control consists of the blood of the species in which the antiserum was raised. Gels were electrophoresed, pressed, washed, dried, stained and destained as before.

Positive reactions were re-tested with dilute antisera to determine between true and false positives. Antisera were diluted to increase specificity of reactions, usually 1:10 or 1:20. Positive reactions obtained after this step were reported.

56

Table A-1. List of Antisera Used to Test Spines

Antisera Sources

PLANTS

Agave Prepared under the direction of Dr. Richard Marlar, UCHSC

ANIMALS

Bear Organon Teknika Corporation

Bison Sigma Chemical Company

Cat Sigma Chemical Company

Chicken Sigma Chemical Company

Deer Organon Teknika Corporation

Dog Sigma Chemical Company

Goat Sigma Chemical Company

Guinea Pig Sigma Chemical Company

Human Organon Teknika Corporation

Mouse Sigma Chemical Company

Pig Organon Teknika Corporation

Rabbit Sigma Chemical Company

Rat Sigma Chemical Company

Sheep Sigma Chemical Company

Turkey Sigma Chemical Company

FISH

Catfish Sigma Chemical Company

White Perch Prepared under the direction of Dr. Richard Marlar, UCHSC

UCHSC = University of Colorado Health Science Center

Discussion

The two groups of spines were submitted for protein residue analysis. Five of the seven spines from Sample 1 were washed for protein residues, while 10 of the 22 spines in Sample 2 were analyzed. In addition to human antiserum, the antisera listed in Table A-1 were tested to determine if any other types of protein residues were present on the spines. Both samples responded positively to agave antiserum, as would be expected from either yucca or agave spines (Table A-2). Sample 2 also tested positive to human antiserum, indicating that some, if not all, of these spines from Unit H contained human proteins. Sample 1 did not yield a positive response to human antiserum. It is possible that these spines were not used or that insufficient amounts of protein were retained on their surfaces. The spines yielded negative results to all other antisera tested (see Table A-1).

57

Table A-2. Positive Protein Residue Results

Sample Description Positive Result

(Antiserum Type) Possible Plant/Animals Represented B-2 Agave spines Agave Agave/Yucca

H-2 Agave spines Agave, Human Agave/yucca, Human

Conclusion

Both groups of spines submitted for protein residue analysis yielded positive results to agave antiserum, confirming the presence of agave/yucca proteins. Sample 2, from Unit H-2, also yielded a positive result to human antiserum, indicating the presence of human proteins on these spines.

References Cited

Culliford, Brian J. 1964 Precipitation Reactions in Forensic Problems. Nature 201:1092-1094.

1971 The Examination and Typing of Bloodstains in the Crime Laboratory. U.S. Dept. of Justice, U.S. Government Printing Office, Stock 27000-0083. Washington, D.C.

Newman, M. and P. Julig 1989 The Identification of Protein Residues on Lithic Artifacts from a

Stratified Boreal Forest Site. Canadian Journal of Archaeology 13:119-132.

58

Acknowledgements

First mention must go to Harry, Rosa Maria and their son Patrick Kelly of Eagle Pass, Texas who told us the story of the painted scapulae and the Encantada Valley. Their interest and unfailing support are gratefully acknowledged. John Woodhull, Alden Scott McKellar, Mr. and Mrs. Billy Finan, and Emery “Smokey” Lehnert also helped us trace the painted scapulae and other material from Cueva Pilote to various private collections.

The Encantada Project, including the excavation of Cueva Pilote, was authorized by a permit issued to Dr. Herbert H. Eling McIntosh by the Consejo de Arqueología for the Instituto Nacional de Antropología e Historia. The analysis was supported in part by the Texas Advanced Research Program under Grant 003658-158, administered by the Institute of Latin American Studies (ILAS) of the University of Texas at Austin to Dr. Solveig A. Turpin. Lic. Eduardo Enríquez Terrazas, director of INAH-Coahuila provided logistical support from Saltillo, including the famous and indispensable Jeep/Camper. Rancher Jesús Santos Maldonado provided access to the site and housing during the difficult winter of 1997 and the spring of 1998. Tom Barksdale of Del Rio, Texas, eased our path through his good offices and contacts in the Encantada Valley.

The original field crew, Leslie Zubieta Calvert, Cristina Corona Jamaica, Nina Amaya Altamirano, Ricardo Cruz Jiménez, José Luis Cruz Romero, Carlos Herrera Lazarini, Jeff Turpin, and our steady and constant companion, Tom Barksdale, endured one of the most difficult field seasons in memory. Temperatures plummeted to below freezing and snow carpeted the ground, but they carried on with valor and good humor. Leslie Zubieta and Cristina Corona processed the first season’s artifacts and records. The spring session was begun anew with the kind consent of Sr. Jesús Santos Maldonado, and relied again upon the talents of Leslie Zubieta, Jeff Turpin and Tom Barksdale, along with a new team composed of Lisa Middleton, Greg Sundborg and Tim and Darylin Schlie. Leslie Zubieta processed the artifacts and records from this field session. Arq. Ricardo Davila magnanimously loaned us rare books from his extensive collection.

Specialists who contributed their expertise to the analysis of the material culture are: Dr. Leland C. Bement, faunal remains and lithics; S. Christopher Caran, soils; Dr. Linda Scott Cummings and Kathryn Puseman of Paleo Research Laboratories, pollen and blood residue; Dr. Robert Howells, mussels; Dr. Artie Metcalf, snails; Dr. Paul Minnis, ethnobotany; Dr. Kristin Sobolik, wood charcoal; Sam Valastro and Alejandra Varela, radiocarbon. Pam Headrick produced the line drawings and adapted the illustrations of artifacts from Cueva Candelaria in Avelyera et al. (1956) and murals from the catalog of the Fine Arts Museum of San Francisco (1993). Juan Chen used his magic to improve upon the quality of the photographed maps and to convert slides to composite prints. Noemi Galván Eling translated the English text into Spanish and edited both.