Monday, June 2, 2008

Introduction and History


Preface


       For those interested in prehistoric people of North America the artifacts described in this paper, which are only a few of the excavated remains, provide a view of the material culture of the prehistoric indigenous people of the region of what is now known as the Oklahoma Panhandle.
       Only a few sites exist in Oklahoma where such preservation occurs. Perishable food, seeds, clothing fragments, sandals and other items are represented among the items recovered from the various cave sites in the area.
       No dating of the artifacts directly has been attempted and the stratigraphy of the excavated cave deposits are for the most part undocumented. Thus it is not possible to assign the materials to a specific time frame, absolute or relative, except in a very general way. Keep this in mind when reading this paper

Introduction and History

       Site 34Ci50, is one of seven rock shelters, collectively known as the Kenton Caves located in the four corners region of Northwest Oklahoma, near the town of Kenton. This is a semi-arid region known for its flat topped mesa's including Black Mesa. The term shelter is used in lieu of cave since most of the caves, with the exception of 34Ci50, are no more than shallow recesses in the sandstone cliffs.
        The site, the shelter known as the Basket Maker I cave, was the largest measuring "forty-five feet wide at the front (west side), about sixteen feet high...and forty-five feet from front to rear."(Baker, 1929b,17). This shelter was first excavated by Dr. Etiene B. Renaud from the Colorado Museum of Natural History in July 1929. His excavations resulted in the recovery of perishable artifacts including basket fragments, sandal fragments, cloth, wood and plant remains. The collection at Denver University, Department of Anthropology is a result of this effort (See table xx).
        Later, in 1929, excavations conducted under the direction of Joseph Thoburn of the Oklahoma Historical Society recovered similar materials. These are in storage and on display at the Oklahoma Historical Society, Oklahoma City. Joseph Thoburn returned again in 1930 and continued excavation of the site. The location of the recovered artifacts is unknown.
        The fourth and final excavations conducted by Ele Baker and Forest Clements from the University of Oklahoma, in June 1934, resulted in the final clearing of the site. The artifacts recovered are on display at the Woolaroc Museum, Bartlesvile, Oklahoma and in storage at the Stovall Museum, Norman, Oklahoma.
        Additional materials recovered from the site by private individuals and groups are located at the No Mans Land Historical Museum, Goodwell, Oklahoma. For a detailed historical account see The Kenton caves of Western Oklahoma (Lintz and Zabawa,1984) and The Kenton Caves Rexamined (Zabawa 1981).
        The following descriptions and illustrations were compiled from a number of sources. Where direct examination of the artifacts was not possible, as is the case of items from Denver University, accession records, detailed photographs and museum notes were used to produce the descriptions and illustrations. For items on display and not directly accessible, as at the Wooleroc Museum, in Bartlesville, Oklahoma, on site sketches were made and photographs were taken of the items. Accession records were also used when available. When hands on direct examination were possible, direct measurements and examinations provided detailed information about the nature and construction of the artifact.

Cordage


Cordage

       Forty-nine fragments of cordage are listed in the Denver University and Woolaroc Museum accession records (table 1). Additional fragments are located at the University of Oklahoma, Norman, Oklahoma.
       The majority of the cordage, with only several exceptions, has been described as being made of grass or yucca. This identification, however, has not been verified. The yucca found in the area today is Yucca glauca commonly known as small soapweed yucca. The terms soapweed is used because the crowns were sometimes used as a soap substitute.(1968, Trees, Woody Vines of the Southwest)
       All of the cordage examined was S twist, down to the right. The non wrapped cordage in the D.U. collection ranges, in width, from 0.15 cm to 0.8 cm. The largest amount (22.5%) is approximately 0.3 cm (table 2 and 3).
       The cordage was predominantly two ply (figure 1a) although one example each of three and four ply was recovered. The three ply bast fiber cordage, (Specimen DU 359), is 18 cm long and 4.5 cm thick and made of yucca. This specimen is "brittle, and doubled on itself, and loosely twisted". (D.U. Accession records, n.d.) The four ply is essentially two, two-ply bast fiber cordage, doubled on itself.
       In addition to plain cordage, two examples of wrapped and braided cordage were recovered.
       One example of three ply braiding, (DU 373), is made of leather strands. This specimen is 16/cm long and 0.2/cm in width. The second braided specimen (WM A35) is also three ply and made of bast fiber cordage. This specimen is 8/cm long and has a diameter of 0.6/cm.
       Examples of feather wrapped, rabbit skin wrapped and cedar wrapped cordage were also recovered. One specimen (D.U. 380) is two ply bast yucca fiber wrapped with rabbit fur. It measures 11cm long by 0.8/cm wide (figure 1b). Another specimen, D.U. 384, 15/cm long by 0.4/cm wide, is also two ply bast fiber wrapped with feathers (Figure 1c). The cedar wrapped cordage is a "bast fiber cordage, probably yucca core with cedar wrapping, 9.5/cm in length, 0.3-0.8/cm in width, center core two ply, wrapped with a one strand fiber 0.4-0.8/cm wide" (D.U. Accession records, DU 372, n.d.a.)

Cordage Table One

Knots


Knots

       Many knots, of yucca were recovered from the site. Although a detailed examination of the complete inventory of knots was not possible a representative sample is illustrated in Figures 1 and 2.
       The knots were used in the construction of baskets, trays, mats or sandals. The functions of the knots were twofold. They were used to splice ends of elements together and to form attachment structures.
       Baker, (1929b: 19) reported finding knots of two different types the double half hitch and the square wreath. The square wreath will be referred to as a square knot. The double half hitch was not found in any of the collections. (see figure2a,b,c for basic knot types) The single half hitch was used in construction of D.U. 402, Figure4b.
       The knots examined include the square knot and the overhand.
       Figure 2f is a square knot using a single element to form the knot.
       Figure 2g is a square knot, which uses two single elements to form the knot.
       Figure 2e is a square knot, which incorporates four elements.
       Figure 2d is a double overhand knot with an attached square knot, which is formed from three elements. The double overhand knot is formed from a single element, which extends to and becomes part of the square knot.
       Figure 2h is a structure formed from two elements. These two elements form interlocking square knots. Each element forms half the square knot of the opposing knot. All of the knots, square and overhand, occur on sandals.

Knotted Structures


Knotted Structures

       One knotted structure, figure 3, was recovered. Measurements are 20.5/cm by 7.5/cm. This structure was formed from yucca leaf elements that were knotted to form a "net like" structure. D.U. 402, is a grass bundle, containing shelled corn. Three longitudinal elements are visible on the surface A. The central element forms the main structural unit of the item. A series of overhand knots were tied which passed through, and were secured to the vertical elements (Figure 3c). The central element continues to the base where it is secured to the last vertical element. Surface b has three longitudinal elements. These elements pass around the vertical elements, forming a half hitch knot, and continue to the ends of the structure. The exact method of starting the vertical and longitudinal elements is not visible on the photograph

Plain Weave Structures



Plain Weave Structures

       The technique used in the plain weave structures is exemplified in figures 4a and 4b. In these two examples the horizontal elements interlace the vertical elements in an over one, under one pattern. The vertical elements are continued over the rim or edge, which form a continuous running vertical element. The type of splice, if present, which continue the vertical elements are not visible on these specimens. The square knot located on figure 4a was probably used as a splice in order to continue the rim or edge. Figure 4a clearly shows the end of the element close to the remains of a splice. The width of the element indicates they were reaching the end of that element. Edge selvege is present on Figure 4a. The edge selvege consist of two lengths of yucca over which the vertical elements pass. The vertical element turns 180 degrees over the edge to produce an adjacent parallel vertical element. This pattern continues to either the end of the fiber or the edge of structure.
       The original form of these two fragments is uncertain. The weave is similar to the sandal in Figures 4a and 4b but could also represent fragments of baskets, mats or trays. figure 5a, D.U. 391, is a globular shaped, bottomless basket, which measures 16/cm by 16/cm (D.U. accession records, n.d.a.). The weave in this specimen is plain and open. The warp elements pass over a rim element turn 180 degrees and become the adjacent warp element. The element continues to the opposite end of the basket where it passes over another rim and continues in the same pattern. There are two possible explanations which could account for the bottomless construction. The first has been documented by Mason, 1904, 351. "The Mohave and other tribes have curious granaries for storing mesquite beans, corn, etc., near their houses. A platform is constructed on high poles; upon this is placed a round bottomless basket". These baskets, however, are "bird nest like" structures rather than woven. The second alternative is that the bottom of the basket has become disassociated from the remaining portion of the basket. If this is the case since the basket, with two rims, is an intergraded structure it would have been constructed in two parts, the sides and the bottom.
       D.U. 393 (figure 5b) is the remains of a plain, open weave, bowl shaped basket. The remains of this specimen measures, from rim to rim (circumference) approximately 11/cm. A portion of the rim remains on the right and left hand portion of the specimen. The rim is formed from a single element over which the warp elements pass turn 180 degrees over the edge to produce the adjacent and parallel warp element

Twilled Structures


Twilled Structures

       One example (WM A1) of twilled weaving was recovered. This fragment, is a over 3 under 3, even twill, pattern float weave which forms a divergent diagonal appearance (figure 6). The maximum length is 20.1/cm by 13.2/cm. No edge selvege remains on this specimen. The average width of the elements is 0.8/cm. The material from which this structure was constructed has been described in the WM accession records as being "hemp". This designation, however, has not been verified.

Coiled Structures


Coiled Structures

       The coiled technique of basket making is represented by four fragments. The largest piece, (D.U 389, figure 7), is a base fragment. The type of coiling is rod with lateral bundle and non-interlocking stitch. The method of starting was an overhand knot. The work direction is from left to right" (Adavosio n.d.). The construction of the basket incorporates "A single whole unpeeled rod which is sewn with non-interlocking stitches, some of which are occasionally split accidentally. On one side of the rod a fiber bundle is arranged so as to front one face of the basket. Since the specimen is badly charred there is a possibility that the bundle surrounded the rod completely, though this does not appear to be the case. Splice type is not discernible. This specimen is unpitched and not watertight, nor is there any decoration. Measurements: Coils per cm-2; Stitches per cm-3" (ibid).
       The second fragment (W.M A5,) located at the Woolaroc Museum has been described as coiled basketry with a split stitch foundation. The foundation element is grass. The maximum length is 7/cm; maximum width 1.3/cm. The specimen is tightly woven.
       The third fragment (WM A4-1) has a split stitch with a grass foundation element. The maximum length is 5.4/cm; maximum *** width is 1.7/cm. It is tightly woven and appears to have been blackened by burning. Both two and three have been described as being made of Yucca. (WM accession records)

Twined Structures


Twined Structures

       Three examples of twining were recovered from 34Ci50. One example, D.U.388, is 11 by 7/cm. This specimen has been described as a probable sandal and is made of Yucca. A photograph of this item was obtained, however due to the compacted nature of the weave and encrustation of soil no details of the weave could be observed.
       The second specimen, D.U. 392, is 8.5/cm by 5/cm. and is made of yucca. No other information, concerning this specimen was available.
       The third specimen, figure 8, is a fragment of twined fabric. The elements composing the weave on this item are composed of 2 ply S (down to the right) twist cordage. The warp is 0.15/cm with a half turn twist. It is a single warp and is weft twined. The stitch slant is irregular. On some rows it is down to the right and on others down to the left. There is no discernible decoration either woven into or dyed on this item.
       This specimen is a probable bag with a central orifice. The diameter of the bag is smaller at the orifice and becomes larger toward the central portion of the remaining fragment. A single element of yucca passes through and across across the orifice of the bag. In its present condition this element holds the two remaining sections together. Two cords possibly drawstrings pass through this area and the loose ends are knotted together with Yucca" (W.M. Accession records, n.d.).

Basketry Discussion

       Four types of weaving, plain, coiling, twining, and twilling were used in the construction of the artifacts recovered. Generally, the basketry, except in the broadest terms are non-diagnostic. The exception to this is D.U. 389, Figure 7, a coiled basket fragment. Examples of this type weave can be found in any part of the North America, where there is preservation. The following is a brief summary of the occurrence of different types of basketry.


Twilled Basketry
       In the Anasazi area twilled basketry was first being made by Basket Maker III time. While in the Mogollon area even twilled basketry has been found as early as the Georgetown phase (A.D. 500-700) (Bluhm, 1952, 322) which is roughly equivalent, temporally, to Basket Maker III time (A.D. 450-750). However, it more commonly occurs later, in the three Circle phase in the Mogollon area and Pueblo III time (A.D. 1200) in the Anasazi area (Bluhm, 1952, 178).
       "Patterned float weave basketry...is fairly widespread in occurrence. Such weaves were used by pre-historic and historic southwestern and southeastern groups and by historic eastern Plains Indians." (Scholtz, 1975, 107) "In the Southwestern Anasazi area the Pueblo Indians were making patterned float weave baskets of yucca by Pueblo I (A.D. 750-900) time."(Scholtz,1975,105) However, other than the similarity of the geometrical designs, she sees little similarities between the Southwestern and Eastern specimens, in the overall structures and decorations (Scholtz, 1975, 107).


Coiling
       This technique is a fairly rare one. The only other examples known are from the Hueco Mts and one or two S.W. sites. In all cases the technique is late (post 500A.D)"


Twinning
        "The use of the twining technique in the manufacture of fabric and bags, and to a large extent basketry, by the prehistoric Indians of the United States was a wide spread practice. The earliest known twining in the Southwest is that of the Basket Maker II (250 B.C-A.D. 450) Anasazi groups, who manufactured bags, fabrics, and some baskets by the use of the twining technique." (Scholtz, 1975, 132)
       Twined basketry fragments were found in Cordova Cave of the Mogollon area, but their occurrence there is late in the stratigraphic sequence (Bluhm 1952,315-316).
       Twined basketry was more common prehistorically in the Trans-Pecos area of western Texas. There are regular and alternate-pair twined baskets in the material from the Big Bend aspect (Smith, 1941, 137). This aspect probably dates between A.D. 600 or 700 and A.D. 1300 (Lehmer 1960: 121-122, fig 2).(Scholtz,1975,135)

Sandals and Sandal One

Sandal and Basketry terminology

Sandal types: attributes and terminology. (Note: March 2009)

        The following is a explanation of the types of sandals and terms used to describe the Kenton Cave sandals.

Sandal types

        The sandals described below are divided into four main types based on the construction technique and type of weave attributes. These are plain weave, twilled weave, twined and edge frame sandals. The first three types are based on standard weave patterns found in basketry and weaving.
        The fourth type of sandal, the edge frame sandal, is not considered a plain weave sandal in this paper. This type of construction does not form the weave pattern typical of a woven mat or basket.
        The only types of sandals reported and seen in the collections are plain weave and edge frame.

Sandal subtypes

        Other attributes noted on the sandals and included as sub types of plain weave sandals are plain weave diagonally woven sandals and plain weave vertically woven sandals. These subtypes are based on the direction of the of the main construction elements in relationship to the toe or heel end of the sandal.

Plain weave vertically woven sandals (Figures 9, 11, 12, 13, 14)

        The plain weave vertically woven sandals are constructed with the elements running parallel to the toe or heel end of the sandal. The vertical and horizontal elements are perpendicular.

Plain weave diagonally woven sandals. (figures 16 and 17)

       In plain weave diagonally woven sandals, the elements of the sandal run diagonally from the heel or the toe across the sandal. The vertical and horizontal elements are perpendicular.
        This technique gives the sandal a “fish tail” effect and in the literature are sometimes called fish tail sandals. The “fish tail” occurs because of the convergence, due to the diagonal weave, of the elements at the heel end of the sandal. The heel end elements also become shredded during wear which adds to the “fishtail” effect. Other types of sandals can have this fishtail effect such as the edge frame sandals.

Edge frame sandals (figures 18, 19)

        Edge frame sandals are constructed by folding the horizontal elements over a element which runs vertically along the outside of the sandal. This forms the edge of the sandal and gives this type of sandal its cohesiveness.

Frame of Reference

       Left lateral and right lateral edge.

       The use of these terms are arbitrary and used to describe the position of elements and structures on the sandal in relationship to the edge of the sandals. The toe, is used a the reference point if identifiable. For example with the toe pointing north the left lateral edge is the left side of the sandal as you are looking at it. If the toe or heel is not identifiable then it is simply the left edge of the sandal as you are looking at it in the illustration with the top of the page pointed north or up. In most cases the sandals are arranged with the toe facing left or east so that the photo and illustration will fit on a single page. In this case the left lateral will be at the bottom of the screen or page.

        Vertical is used to describe elements which run parallel to the edge of the sandal from the toe to the heel of the sandal if heel or toe is identifiable.
        Horizontal is used to describe the elements which run across the width of the sandal perpendicular to the vertical elements.
        Diagonal is used to describe the elements of the sandal which run diagonally from the heel or the toe across the sandal.

Other attributes


       Double layered sandals
       Single layered sandals

The terms weft and warp are terms used to describe fabrics or cloth woven on a loom. The terms are also used to describe basketry.

Definitions


       Rim:Outer edge of something circular: an outer edge, often slightly raised, that runs along the outside of something curved or circular.

       Weft or woof is the yarn which is drawn under and over parallel warp yarns to create a fabric. (Wikipedia 2009)

       Warp: Elements that run lengthwise.

       Weft: Elements that run horizontally. This element usually forms the selvage edge of the cloth.

       Selvage: In a woven fabric, the selvage is the uncut edge of the fabric which is on the right- and left-hand edges as it comes out of the loom. As such it is 'finished' and will not fray because the weft threads double back on themselves. (Wikipedia 2009)

Sandals

       Eleven complete and fragmentary sandals were recovered from 34Ci50. Renaud reported finding seven specimens, "Two complete, and one fragment were square-toed diagonally woven sandals, with double thickness soles. Four sandals with rounded toes were made of simple square weaving of yucca leaves with cedar bast reinforcement or padding in between the two layers forming the sole" (Renaud,1930,124). Only three of the seven sandals reported recovered by Renaud remain in the Denver University collection. These are two of the "square toed, diagonally woven sandals (Specimens 6 and 7) and one of the rounded toed sandals with padding (Specimen 4)". There are five specimens in the collection of the Oklahoma Historical Society.(Specimens 1,2,3,5,10) These were recovered by William Baker during "salvage excavation" after the departure of Renaud expedition in 1930. Two others, specimens 9 and 10, were recovered during Oklahoma University excavation in 1934. These are located at the Woolaroc museum. One specimen remains in the collection of the No Mans Land Historical Museum, located near Goodwell, Oklahoma and is part of the William Baker collection.
       All but one of the sandals appears to be made of yucca, possible Yucca glacua. This assignment as with the rest of the collection should be verified. These nine available specimens are described individually.





Sandal One
       Specimen one is a double layered, plain woven, square toed sandal fragment which measures 9cm by 16.5cm. (figure 9). The elements are loosely woven. The transverse elements are continuous in the weave. These elements cross over the frame element, turn 180 degrees and become the adjacent element. This pattern is continuous through the structure.
       The twenty-one longitudinal elements, including the edge frame element, are parallel. Due to the mounting technique used by the museum the toe edge could not be closely examined. This made it impossible to determine whether the sandal was made as one unit and folded over or as two separate units.
       One possible method for attachment of the upper and lower layers was through the use of separate ties on the edges. This could be the function of the element located on the right lateral edge 10.5cm from the toe of the sandal.
       There are three structures which were used to secure the sandal to the foot.
       Structure one, the side loops, are simply looped over the lateral edge elements. This element continues along the edge of the sandal until it reaches the corner where another element is used to secure the loop element to the sandal. It then continues across the front of the sandal where it is again fastened to the main body of the sandal near the center and at the left lateral corner. The side loops are not present on the left lateral edge but presumably continued to the heel edge of the sandal.
       Structure two, the sandal attachment structure, consists of three elements, two of which are joined with a square knot. The square knot enables the doubling of elements where extra strength is needed. The junction, where the two elements meet, is located 4cm in from the toe edge and is centered between the lateral edges of the sandal. The elements continue to the end of the sandal where they looped through the frame element. The elements are severed at this point so it is impossible to tell how they were secured to the main body of the sandal. There is a third element which passes between the main knotted elements. This element was probably used for structural reinforcement.
       The third structure is the tie element. This structure consists of two elements which have been joined with a square knot. One element continues from each side of the knot and passes through the side loops. These elements probably continued through the side loops along the lateral edges until they wrapped around the heel and ankle where they were fastened.(See figure 10)

Sandal Two


Sandal Two

       Specimen two, figure 11, (O.H.S ?) is a plain woven, rounded toe single layered sandal. This specimen measures 8cm by 15cm. The weave is identical to specimen one, described above. The transverse elements are continuous in the weave. These elements cross over a two element edge frame, turn 180 degrees and become the adjacent element.
       The fifteen longitudinal elements, in the structure, are not continuous. The original configuration of the longitudinal elements can not be reconstructed from the remaining longitudinal elements. All of the structural elements of the attachment structures are either missing or disassociated.
       This specimen, in its present condition, appears to be to fragile to have functioned as protective footgear. It is possible that this sandal was originally a double layered sandal but has since become separated from the other half.

Sandal Three


Sandal Three

       The third specimen, figure 12, is a tightly woven, plain weave, double layered sandal. This fragment measure 16cm by 9cm. Both layers of the sandal were visible. One layer, Figure 12, was shiny as though with wear. This will be called the top of the specimen. The bottom, is well worn and encrusted with dirt. The heel portion of this sandal is missing.
       The transverse elements are continuous through the sandal. These elements cross over the frame element, turn 180 degrees and become the adjacent element.
       The longitudinal elements are not continuous. Each longitudinal element is a separate unit. The elements run parallel to each other to the toe end of the sandal where they loop over to form the longitudinal elements in the lower layer. The sandal was woven as one unit and then folded over at the toe.

Sandal Three

Sandal Four


Sandal Four

       Specimen four (D.U. 397) has a rounded toe. (figure 13) It is plain woven, closed weave, double layered sandal. This specimen is 26cm by 11.5cm. The transverse elements are continuous. After crossing the frame element they turn back to become the adjacent element. The pattern is identical to specimen one and two. The longitudinal elements are not continuous. Each longitudinal element is a separate unit. The elements run parallel to each other to the toe end of the sandal where they loop over to form the longitudinal elements in the lower layer. Between the two layers there is shredded vegetable fibers which served as padding.

Sandal Five


Sandal Five

       Specimen five, D.U. 387, figure 14, is a possible sandal fragment. This is plain woven. The longitudinal elements are continuous. They continue the length of the sandal turn 180 degrees and become the adjacent element. There are five remaining longitudinal elements.
       The transverse elements are or were also continuous. They continue the width of the sandal, turn 180 degrees and become the adjacent element. Only one intact continuous transverse element is visible on the upper surface of the structure.

Sandal Six


Sandal Six

       The sixth specimen, figure 15, is a plain weave, double layered, closed weave sandal. The remaining dimension are 6.5cm by 8cm. The elements are doubled and woven diagonally. The starting element is located at the upper right corner of the sandal. This element is bent 90 degrees forming an adjacent parallel element. The second element is woven through the first, in a plain weave fashion and then bent 90 degrees to form parallel elements adjacent to the first. This pattern can continue until the desired width of the sandal has been achieved. When this point has been reached, a series of parallel elements form the heel portion of the sandal leaving the final vertical element projecting from the side of the sandal. This element can be woven into the heel portion of the sandal until the end of the structure is reached where it could be then be tied off.
       This method of weaving produces a structure in which the diagonal elements form the lateral, toe, and heel edges of the sandal.
       The method of attachment to the foot is not clear. The remains of the attachment structure is dissociated except for one element. This element extends from the knot to the left lateral edge approximately 4.5cm from the top or toe edge. This element is looped through a element of the left lateral edge and returns to the knot. One element also passes through the lateral edge at the upper left lateral edge. This element, however, is dissociated. It is likely that the opposite elements extending from the knot to the right lateral edge and also looped through the edge elements.

Sandal Seven


Sandal Seven

       D.U. 398, figure 16, is a plain diagonally woven, tightly woven sandals. The structural elements which form the weave unit appear to be doubled on the top and tripled on the sole. The method of weaving is identical to specimen 5,Figure 6. The weave crosses the sandal in a diagonal fashion until they approach the heel portion of the toe. At approximately 12 Cm. from the toe the structural elements turn and continue to the heel parallel to the longitudinal axis of the sandal. Attachment devices could not be seen on this specimen.
       The last diagonal element continues from the left lateral edge to the right lateral edge where it is tied, with a square knot, to a diagonal element which was continued from near the left lateral corner. This is the only visible structure, at the heel portion of the sandal, which holds the heel together. It is likely that there was a similar structure on the left lateral edge. These two structure, when pulled tight would be adequate to hold the heel end of the structure together. This method of construction, convergence of the elements at the heel, gives the appearance of a fishtail heel which is similar to the fishtail heels on sandal 8 and 9.

Sandal Eight


Sandal Eight

       Sandal eight, D.U. 399, figure 17a and b , is a plain diagonally woven sandal. The weave crosses the sandal in a diagonal fashion until they approach the heel portion of the toe. At approximately (C.M.) from the toe the structural elements turn and continue to the heel parallel to the longitudinal axis of the sandal. Attachment devices remain on this specimen.
       The last diagonally element (see figure 17a) continues from the left lateral edge to the right lateral edge where it is tied, with a square knot, to a diagonal element which was continued from near the left lateral corner. This is the only visible structure, at the heel portion of the sandal, which holds the heel together. It is likely that there was a similar structure on the left lateral edge. These two structure, when pulled tight would be adequate to hold the heel end of the structure together. This method of construction, convergence of the elements at the heel, gives the appearance of a fishtail heel.

Sandal Nine


Sandal Nine

       Sandal nine, figure 18, is 25 cm in length, 11 cm wide and 3cm thick and is woven over a three element edge frame. The elements forming the structure are woven over the edge frame in a criss-cross manner. The frame elements converge toward the heel and toe to form a crescent shaped toe and fishtail shaped heel. The fishtail effect of the heel is formed from the convergence of the frame elements which have become shredded. The transverse elements (there are no longitudinal elements except for the frame element.) cross over the right lateral edge turn 180 degrees loop around and under the right frame element. The element continues to the left lateral edge where it is looped under the frame element and across the top. This criss-cross pattern continues the length of the sandal. Possible attachment structures remain at the toe end of the sandal.

Sandal Ten


Sandal Ten

       Sandal ten, figure 19, is 14.1cm long, 7.2cm wide and 2.7cm thick. The technique for constructing this sandal is similar to the method described above. The elements are woven over a frame edge in a criss-cross manner. This specimen was so tightly woven that internal structures were not visible.

Sandal Eleven


Sandal Eleven

       Sandal eleven, figure 20, is made of hide and is not an interworked structure. This specimen is cm by cm. and was attached to the foot by means of a single leather element which passed through perforations that formed gathers when pulled tight. The last perforation, through which the element passed, was located at the heel end of the structure at each lateral edge. The leather element was then pulled back parallel to each edge, approximately cm where it passed under itself and across the foot where it was pulled tight and tied to the opposite end of the element.

Sandal Discussion

Sandal Discussion

       Eleven sandals or fragments which can be identified as sandals remain in the various collections. From these eleven specimens two categories of sandals have been identified. These, based on the types of weaves are, plain weave sandals, and fishtail sandal or scuffer toe sandal. The plain weave can further be divided into categories based upon the gross morphology of the structure. Based upon the direction of the weave they can be divided into two categories, diagonally woven and vertically woven. In diagonally woven sandals the elements, both longitudinal and transverse, are woven diagonally to the longitudinal l axis of the sandal. In vertically woven sandals the longitudinal elements are woven parallel to the longitudinal axis of the sandal. These can further be subdivided into other categories based upon such attributes as weave type (twilled, twined or plain), the number of layers e.g. single layer or doubled layered; padded or non-padded; attachment structure morphology; continuous or non-continuous elements; shape of the sandal E.G. round toe or square toe.
       If the sandal is doubled layered it can be constructed using two methods. The sandal can be woven as one unit and folded over to form the two layers or woven as two structures and then the layers attached together through the use of ties.
       Possible twined sandals have been identified among the D.U. collection (D.U. Accession records #387 & #388). These however, have been, re-identified as "unusual" plain weave structures, (Page ).
       Twilled sandals have not been reported from the site and none were any seen in the collections.


GEOGRAPHICAL DISTRIBUTIONS ETC.


PLAIN WEAVE SANDALS
       Nine specimens of plain weave sandals were found in the collections. Of these nine three are diagonally woven and six vertically woven.


FISHTAIL SANDAL
       Two examples, specimens eight and nine, of fishtail sandals were recovered. These specimens are similar to the type of sandal described as a "two warp scuffer toe".(Cosgrove,1947,83,fig 87 and 88) The basic structure is identical to this type. However, the attachment devices are dissimilar to any described by Cosgrove, fig. 13b, for the "two warp scuffer toe" sandal.
       Twenty-four of the two warp scuffer toe sandals were recovered from Ceremonial Cave. One specimen each, of this type sandal, were recovered from Chavez Cave on the Rio Grande and Picture Cave in the Hueco area. Among the sandals collected in the South, Cosgrove says, there is nothing comparable in form to the sandals found in the San Juan Basket Maker area.(Cosgrove,1947,167)
       This type of sandal has also been found near Silver City, New Mexico, Van Horn, Texas, Lava, New Mexico, Tularosa Cave in western New Mexico, Cave Valley, Northwestern Chihuahua (Montezuma Creek). "A type very suggestive of the El Paso form was found but varying somewhat in the technique of its manufacture was found in northeastern Arizona, Laguna Creek, 1919, by Kidder and Guernsey in their earlier explorations."(Roberts, ).
       Specimen six and seven are similar to Cosgroves types 9a and 9b, fig 11, which he calls full length turned heel sandals. The similarity which can be seen is in the change in the direction of elements at the heel. In Cosgroves specimen 9a and 9b the diagonally woven elements also change direction near the heel end. These elements are subsequently incorporated in a diagonal weave at the heel end of the sandal.
       Cosgroves conclusion was that the "resemblances of weaves, and in some cases actual cross finds of practically identical specimens have served to tie the Hueco and Upper Gila areas together on both horizons and the Huecos to the Big Bend of Texas on the earlier. (Also) the finding of type three sandals (which is a type of two warp scuffer toe sandal similar to the type above) in central New Mexico serves partly to bridge the great territorial gap between the Basket-maker region of southwestern New Mexico and that of the San Juan Basket-makers of northeastern Arizona.(Cosgrove,1947,97-99)


DIAGONALLY WOVEN SANDALS
       Specimen six and seven are similar to Cosgroves types 9a and 9b, fig. 11, which he calls full length turned heel sandals. The similarity which can be seen is in the change in the direction of elements at the heel. In Cosgroves specimen 9a and 9b the diagonally woven elements also change direction near the heel end. These elements are subsequently incorporated in a diagonal weave at the heel end of the sandal.(Cosgrove, 1947,97-98)
       Adovasio, 1970, also saw similarities between the "textiles" recovered from the Gila-Hueco areas of southern New Mexico and west Texas. However certain techniques he felt were "somewhat different from those in the Southwest proper"(Adovasio,1970,79)
       The round or square-toed sandals, including the fish-tail types, of the Big Bend area of Texas would seem to be sufficiently distinct from the coarser sandals of the north. Just what the relationship between them has not yet been determined. See Sayles, 1935, Table 8, and plates 26-27; Roberts, 1929,pp5-9, plate 2, and figs.5-6; for information concerning these southern sandal types. (Baldwin, a 1939,242) (look for this in library) Roberts, plate 35, and fig. 36). These are made of somewhat coarser yucca leaves, in twilled over-two-under-two weave, with the yucca elements turned back at the heel either on the top or bottom to form a pad across the heel (plate I, fig.5" ('Baldwin, 1938, 1) (Is this a turned heel sandal.)
       "The first definite appearance of twilled woven sandals is in the Pueblo III period."
       "Twilled woven basketry and matting, which are made in similar fashion and of the same materials, likewise do not commonly occur until Pueblo III." (Baldwin, 1938, 4)
       "Thus we now place the beginning of the Basket Maker II period as about the opening of the Christian Era and its end about 400A.D, with the Basket Maker III period lasting up to 700 A.D. Definite dates have been established for a great many sites extending as far back as 348 A.D." (Baldwin, 1938, 5)
       The types of woven or inter-worked sandals found prehistorically, parallel the types of basketry found. There are twined, plain woven, and twilled, and those which use combined techniques. The only technique of weaving which is found in basketry and not in inter-worked footgear is the coiling technique of weaving.
       Since the interest in basketry for use as a tool in archaeology there has been a continuing debate over what attributes of woven items are significant for identifying "cultural groups".
       Due to the morphology of the foot the shape of footgear, unlike pottery and basketry, is relatively restricted. That is, it is restricted to the shape of the foot. The same can be said for attachment devices used to attach the sandal to the foot. The type of weaving used to produce the structure is only diagnostic when it is in conjunction with basketry. For example, if the earliest twilling found in the Mogollon area "was about A.D. 500-700" then twilled sandals would not be expected to be found in this area any earlier than this.
       In spite of these limitations, there are attributes of sandals which could be called diagnostic. This is the gross morphology, which includes the shape of the sandal, internal structure, the type weave, the attachment structures, and any decoration on the sandal. Unfortunately, the attachment structures, the most perishable part of the sandal are not always present and Internal structures are not generally visible unless the specimen is dissected.

Biblography

Adovasio, 3.M.
1970 The Origin, Development and
Distribution of Western Archaic
Textiles. Tebiwa, Volume 13, pp.
1-40. Pocatello.

1975 Prehistoric Basketry of the Lower
Pecos and Trans-Pecos Regions of
Texas. Bulletin of the Texas
Archaeological Society. College
Station.

1976 The Evolution of Basketry
Manufacture in Northern Mexico and.
Trans-Pecos Texas. Special
publication of INAH. Mexico City.

1978a Meadowcroft Rockshelter. A 16,000
Year Chronicle. Transactions of
the New York Academy of Sciences.
New York.

1977 Basketry Technology. A Guide to
Identification and Analysis.
Aldine Publishing Co., Chicago,
Illinois.

Baldwin, Gordon C.
1938a Basket Maker and Pueblo Sandals.
Southwestern Lore, Vol. IV, No. 1,
pp. 1-6, Gunnison.

1938b An analysis of Basket Maker III
sandals from Northeastern Arizona.
American Anthropologist, Vol. 40,
No. 3, pp. 465-485, Menasha.

1939c Further Notes on Basket Maker III
Sandals from Northeastern Arizona.
American Anthropologist, Vol. 41,
pp. 223-244, Menashaa.

Bell, Robert E.
1968 Dating the Prehistory of Oklahoma.
Great Plains Journal, 7(2):111-152.

Bell, Robert E. and David A. Baerreis
1968 A Survey of Oklahoma Archaeology.
Texas Archaeological and
Paleontological Society Bulletin,
Val. 22, pp. 7-100.

Bluhm, Elane
1952 Clothing and Textiles. In
Mogolion Cultural Continuity and
Change: The Stratigraphic
Analysis of Tularosa and Cordova
Caves, by P.S. Martin and others,
pp. 159-179 Fieldiana:
Anthropology 40.

1954 Cordage, Sandals and Textiles. In
Caves in the Reserve Area, by P.S.
Rinaldo and E. Bluhm, pp. 159-179.
Fieldiana: Anthropology 42.

Cornelius, Olive Frazier
1938 Basket Maker Sandals(?).
Southwestern Lore, Vol.III, No. 4,
pp. 74-78, Gunnison.

Cosgrove, C.B.
1947 Caves of the Upper Gila and Hueco
Areas in New Mexico and Texas.
Papers of the Peabody Museum of
Archaeology and Ethnology. Volume
XXIV (2), pp. 1-181. Cambridge.

Dick, Herbert W.
1965 Bat Cave. The School of American
Research, Monograph 27. Santa Fe,
New Mexico.

Friston, George C.
1968 Daugherty Cave, Plains
Anthropologist, Vol. 13, No. 42,
Pt. 1. University of Kansas Press.

Haury, Emil
1950 The Stratigraphy and Archaeology
of Ventana Cave, Arizona.
University of Arizona and
University of New Mexico Press,
Tucson and Albuquerque.

Kidder, A. V.
1924 An Introduction to the Study of
Southwestern Archaeology with a
preliminary account of the
excavations at Pecos. Department
of Archaeoloy, Phillips Academy

Kidder, A. V. and Gurnsey, S. J.
1919 Archaeological Explorations in
Northeastern Arizona. Bulletin 65,
Bureau of American Ethnology.

King, Mary Elizabeth
1974a Textiles and Basketry. In Casas
Grandes: A Fallen Trading Center
of the Gran Chichimeca, by
Charles DiPeso, John B. Rinaldo and
Gloria J. Fenner. Volume 8, pp.
76-113. Northland Press, Flagstaff

1974b Escanoles Period Perishable
Artifacts. In Casas Grandes: A
Fallen Trading Center of the Gran
John B. Rinaldo and Gloria J.
Fenner. Volume 8, pp. 76-113.
Northland Press, Flagstaff.

Lamber, Majorie V. and Ambler, Richard J.
1961 A Survey and Excavation of Caves in
Hidalgo County, New Mexico. The
School of American Research,
Monograph No. 25. Santa Fe, New
Mexico.

Lintz, Christopher and Zabawa, Leon George
1984 The Kenton Caves of Western Oklahoma
Prehistory of Oklahoma, Robert Bell
7:161-172

Lehmer, Donald J.
1960 A review of Trans-Pecos Texas
Archaeology. Bulletin of the
Texas Archaeological Society
36:111-152.

MacNeish, Richard S.
1967 The Prehistory of the Tehuacan
Valley, Vol. II, Non-Ceramic
Artifacts. Robert S. Peabody
Foundation, University of Texas
Press. Austin and London.

Martin, Paul S.
1952 Summary and conclusions. In
Mogollon Cultural Continuity and
Change: The Stratigraphic
analysis of Tularosa and Cordove
Caves, by P.S. Martin and others,
pp. 483-507: Fieldiana: Anthropology 40.

Mason Otis T.
1894 Primitive Travel and
Transportation. Report U.S.
National Museum, 1894. Washington D.C.

1904 Aboriginal American Basketry:
Studies in Textile Art Without
Machinery. Report of the U.S.
National Museum, 1902, pp.
171-548. Washington, D.C.

Miner, Horace
1936 The Importance of Textiles in the
Archaeology of the Eastern United
States. American Antiquity, No. 3

O-----, Dorris L
1965 Brawley's Cave, Bosque County,
Texas. Bulletin of the Texas
Archaeological Society
36: 111-152.

Orchard, William C.
1920 Sandals and other Fabrics from
Kentucky Caves. Museum of the American Indian.
Heye Foundation, Miscellaneous 4.

Roberts, Frank H., Jr.
1929 Recent Archaeological Developments
in the Vicinity of El Paso, Texas.
Smithsonian Miscellaneious
Collectios, Vol.81, No. 7,
Washington.

Sayles, E.B.
1935 An Archaeological Survey of Texas.
Medallion Papers, No. 7, No. 17,
Globe.

Schuetz, Mardith K.
1963 An Analysis of Val Verde County
Cave Material: Part III.
Bulletin of the Texas
Archaeological Society
33:131-165.

Smith, Victor J.
1935 The Split Stitch Basket, a
Distinguishing Culture Trait of the
Big Bend in Texas. Bulletin of
the Texas Archaeological and
Paleontological Society 7:100-104.

1941 Some Unusual Basketry and Bags from
the Big Bend Caves. Bulletin of
the Texas Archaeoloical and
Paleontological Society
13: 133-151.

Sunday, June 1, 2008

The Kenton Caves Re-examined



Excavated Kenton Cave
Pictured -- Pam and Christopher Zabawa (1981)

The Kenton Caves Re-examined
George Zabawa
15 September 1981



The Kenton Caves Re-examined

        Located near the town of Kenton, in northwestern Oklahoma, are six caves which are known as the Kenton or Basketmaker caves. The first of the caves, called the Basketmaker 1 cave was discovered by Ele Baker, the son of William "Uncle Bill" Baker, on Saturday, June 16, 1928. William Baker, Cimarron County farm agent, rancher, and a vocational archaeologist, was to play an important role in arousing public and professional interest in these important archaeological sites. Ele Baker, who was also interested in the prehistory of Oklahoma, later directed the final excavation of the Basketmaker 1 cave.
       In an effort to bring the Basketmaker 1 cave to the attention of professional archaeologists, Uncle Bill wrote to an archaeologist at the University of Oklahoma in which he describes the cave as "a wind-blown place in solid sandstone under an overhanging ledge of rock high up the hillside from the valley below. It faces west but is in the east end of a cove. The ledge of rock extends nearly west from near the side of the cave, which gives it very good protection from the north. Two-thirds the distance from the back to front has apparently been a solid rock wall built up, extending across the entire front of the cave. Dim paintings in red cover the east back wall of the cave. Most of them are so dim as to make it almost impossible to follow the outlines."
        In July 1929, William Baker took a small party of Boy Scouts to the cavern to salvage what they could. Later, on August 26, 1929, Dr. Joseph Thoburn of the Oklahoma Historical Society arrived to survey the situation. He initiated excavation. This and subsequent excavations revealed that although a party from Denver had previously recovered a large number of artifacts, the best were still to be found. Initially, the artifacts from the caves were attributed to the Basketmaker cultures of southwest Arizona, Colorado, and parts of New Mexico.
        This was proposed by the first archaeologist to excavate the largest cave. E. B. Renaud, during his excavation in 1929, recovered numerous artifacts from the caves such as sandals, baskets, and cordage. However, more important was what he did not recover—projectile points and pottery. This to Renaud and other archaeologists of the 1930s was indicative of primitive farmers above the stage of nomadic hunting and gathering but below the level of farmers who made pottery. The prehistoric people who lived in this region were on their way to becoming full-time farmers but had not quite made the transition from their hunting and gathering way of life.
        Since that time (1929) until recently there has been a lot of speculation concerning this assessment. Some believed in the Basketmaker origin of the people who occupied the caves. However, others were more skeptical. All of the ideas concerning the origin of the prehistoric people were based on Renaud's observation that pottery and projectile points were absent from the caves. A recent re-examination of the artifact collections, however, revealed that projectile points and pottery were recovered from the caves during subsequent excavations.



        On the basis of these points and pottery, it is now possible to say that the caves have been occupied for several thousand years by more than one group of people. It is uncertain at this time whether the Basketmaker of the Southwest was one of these groups. The types of projectile points which have been identified: (1) unidentified fish-tail point; (2) Williams, 500 B.C. - A.D. 300; (3) Catan, 100 B.C. - A.D. 1450; (4) Scallorn, A.D 270 - A.D. 1150; (5) Washita, A.D. 900 - European contact; (6) Harrell, A.D. 1200 - A.D. 1700; (7) Fresno; (8) Scraper; (9) Knife. These projectile points are made of the locally available quartzite. Only a few examples of other type materials, such as jasper and flint, are represented in the collections. Although quartzite is inferior in terms of its workability when compared to flint or obsidian (volcanic glass), it was made into good quality, razorsharp projectile points, knives, and scrapers by the prehistoric inhabitants of the region.
        The pieces of pottery found in the caves are cord-marked pottery typical of the plains region. As yet no date has been assigned to these.
        The artifacts including the projectile points and pottery were recovered from the Kenton caves between 1929 and 1941 by various groups ranging from professional archaeologists to picnickers. The artifacts are located at various museums across the state: No Man's Land Historical Museum in Goodwell, Oklahoma; the Woolaroc Museum near Bartlesville; the Oklahoma Historical Society in Oklahoma City; and the Stovall Museum at the University of Oklahoma. The artifacts taken by Renaud's party are located at the Department of Anthropology on the University of Denver campus.



        Due to the dry condition of the caves, artifacts which would normally rot (clothing, wood, baskets, food, etc.) when exposed to moisture are preserved. This is a very special circumstance and rarely occurs on the plains.



        These artifacts can be divided into three categories: by materials such as stone, shell, wood, bone, and vegetable materials; by use—hunting implements, garden implements, personal apparel and adornment items, and household items; and many items fall into the third category which is of unknown use.



        Collectively, the artifacts from the caves tell a story of how the prehistoric people of the area took advantage of the natural resources of the area in obtaining food and shelter.
        Items recovered from the caves indicate that the people utilized a wide range of wild plants and animals found in the area in their quest for food. In addition to the wild foods utilized, crops such as corn and squash were grown and stored in the caves.



        Animals which were hunted ranged from the small crayfish found in the local streams to the antelope and bison which roamed the plains. In addition to these animals, deer, elk, jackrabbit, cottontail, coyote, wildcat, badger, various rodents, birds, eagles, and turkeys were also hunted by the prehistoric people of the Kenton (Renaud, 1930 b). The hunting implements used to obtain the game include the spear thrower, the bow and arrow, the snare, and a type of nonreturnable boomerang or throwing stick.
        Wild fruits and berries were also used as a food source. Some were pressed into cakes and dried for later use, perhaps on hunting trips. In a similar manner parched corn might have also been used. These are light, nourishing, and easily transported.



        In addition to the wild food utilized, crops such as corn and squash were planted and harvested which were used to supplement their diet. Complete ears of corn as well as bags of seeds were found in the caves. These were used during the winter for food and the following spring for planting.
        Shelter from the elements must have been of considerable importance to the native inhabitants since temperatures in the region typically drop into the zero-degree range in the winter. In addition to providing shelter from the rain, snow, wind, and cold, they also served as excellent natural storehouses due to their dryness.



        Little is known of the clothing of the prehistoric people of this area. No complete article of clothing was found in any of the shelters or caves. Only one piece of woven cloth, similar in appearance to burlap, was found.
        Cave paintings in the area, however, illustrate human figures with kilts or mantas (Lintz, Zabawa, 1981 n.d.). Although no clothing was recovered, feather-wrapped twine, hide, and hundreds of fragments of cordage and twine indicate that blankets and similar articles were being manufactured. No loom or other such device has been recognized thus far.







        Although little is known of the clothing worn by these people, sandal fragments are commonly found in the various collections. These are woven of yucca leaves and padded with bark and grass. They appear to be durable, comfortable, and well suited to the type of rough terrain found in the Black Mesa area.



        Three burials were reportedly unearthed from the larger Basketmaker 1 cave by picnickers May 1933. Although three were reported, only one has been located.
        It is now on display at the No Man's Land Historical Museum in Goodwell, Oklahoma. The body, which has been called a mummy, is a naturally desiccated (dried up) body of a three-year-old child. Recent examinations of the body by a representative of the State Medical Examiners office failed to reveal any apparent cause of death.
        From X rays it was determined that the body was a shell of skin and bone with no internal organs present. However, the outside was so well preserved that a partial footprint was obtained. The child was bound buried in a semi-flexed position with a shell necklace and corn which were presumably for use on the child's journey into the afterlife.

Conclusion

        The artifacts found in the Kenton caves represent various groups of people who lived in the area over a period of several thousand years. It is impossible at this time to tell what people made what items. Generally, the prehistoric Kenton people, although self-sufficient, were poor in the material aspects of life. They did not make elaborate projectile points or pottery as did the people to the east and west. Whether this was due to lack of time or inclination is not certain.
        The above description reflects only generally the use of the natural resources of the area by several groups of people. This pattern certainly varied through time as different people with different ideas entered the area and the transition was made from a nomadic hunting and gathering existence to a more sedentary life-style of the farmer.

Day in the Life?

        What was it like to live in this area before the arrival of Europeans? The following story is a fictional account of what might have been. “A day in the life of one individual”

        "It had been a good day. The weather was clear and crisp with the hint of fall in the air. The harvest had been bountiful, and the storehouses in the caves on the cliffs were full. The corn and squash should get them through the hard times in the winter. The thought of the caves, where her child was buried, brought back painful memories of the cold winter day when her youngest child died of the high fever in spite of all that they did. She was at least prepared for the journey into the afterlife for she had sent with her some corn. The shell beads around her neck would also help ward off the demons on her perilous journey. Well, there was nothing she could do now but continue. Still, the pain she felt for her loss, she knew, would always be with them. As she approached the cave where her mother was grinding corn in the prepared mealing bin in the rocks, the baby in the cradleboard on her back began to cry. She saw her husband skinning the rabbit which would be roasted for dinner.
        "The boy children were chasing a small lizard with their play bows while the girl children were sorting and shredding the yucca and bark which were to be made into sandals, baskets, and cloth. With winter coming on she thought that it was a lucky time when her husband and the other men had managed to guide a small group of bison into a box canyon for an easy kill. This would mean extra smoked and dried meat for the winter and robes to keep them warm during the cold winter nights.
        "Now that the work of planting and harvesting was over, her thoughts turned to the work which was still to be done for the fall celebration when all the people met for one last time before winter. There were gifts to make for the exchange and food to gather for repayment. The cloth bag she was handweaving for her husband's brother, even though tedious work, was little in repayment for his help when her husband had injured himself in a fall.
        "As she began to talk to her mother, who was mealing corn, about the coming celebration, the baby on the cradleboard began to cry but ceased as it contentedly began to nurse."

Acknowledgments:
        The author is greatly indebted to many individuals and institutions which aided me during my efforts to gather information concerning the Kenton caves.
These include Ele Baker; Helen Pustmueller, Department of Anthropology, University of Denver; Lynn Barnes and Martha Blaine, Oklahoma Historical Museum; K. D. Meek, Woolaroc Museum; and Candice Green and Robert Bell, Stovall Museum. Many thanks are extended to Christopher Lintz who was instrumental in organizing the new information into a usable form.

Christopher Lintz and Leon G. Zabawa
The Kenton Caves of Western Oklahoma
Manuscript on file at the Department of Anthropology, University of Oklahoma