"I was stunned by the beauty of the workmanship. 
                          The hair on the nape of my neck bristled erect. I had 
                          never seen anything like it, and instinctively knew 
                          it was special." 
                        Gene Mear, on finding a 12,000-year-old fluted 
                          Folsom point while screening backdirt at Kincaid Shelter 
                          in the late 1940s.  
                        From "All Things Considered, It was Fun While It 
                          Lasted," a small volume in which Mear recounts, among 
                          other experiences, how his work at the site changed 
                          the direction of his life from studies in chemistry 
                          to a career in geology. 
                       | 
                     
                     
                         
                        Gene Mear looks over a trench dug 
                          through the shelter by the TMM . The young college student's 
                          discovery of three Folsom points in the shelter is what 
                          triggered the chain of events leading to systematic 
                          investigations of the site. Photo by Glen Evans. 
                       | 
                     
                     
                        
                        Map of TMM excavated units (marked 
                          with dots) and location of treasure-hunters' pits (denoted 
                          with horizontal lines) in Kincaid Shelter, as drawn 
                          by Glen Evans. Click to enlarge and read key.  
                       | 
                     
                     
                      |  
                         Only a small area of the surface deposit within 
                          the shelter remained undisturbed, but fortunately there 
                          had been comparatively little disturbance in the deeper 
                          zones and in front of the shelter. 
                       | 
                     
                     
                        
                        A crew member holds a screen over a wheelbarrow in front of the shelter where it will be filled with 
                          another load of dirt. All sediments from the shelter 
                          were screened for artifacts and then hauled to a dump 
                          site below. TARL archives. Click to see full image. 
                       | 
                     
                     
                        
                        Visitors pose in front of the field 
                          camp tents with crew member Powell Goodwin, right, during 
                          the 1948 investigations at the site. TARL archives. 
                       | 
                     
                     
                        
                        Hard day at the shelter. University 
                          of Texas field school students take a break from excavations 
                          during the 1953 project at Kincaid. The woman sporting 
                          the horned-rimmed glasses, in center, is Dee Ann Suhm, 
                          who went on to become a professor at UT-Austin, director 
                          of TARL, and author of numerous landmark publications 
                          on Texas archeology.  
                       | 
                     
                    
                        
                        A page of field notes written by then-student Dee Ann Suhm at the U. T. field school at Kincaid in 1953. The young student's careful attention to detail was evident even then. 
  | 
                     
                   
                 | 
                
                   
                  In December of 1947 while hunting on the Kincaid 
                    Ranch, young college student Gene Mear and several of his 
                    friends visited a rockshelter in a low, limestone bluff on 
                    the Sabinal River. They examined some loosely piled-up dirt 
                    that had been thrown out of a pit dug in the shelter fill 
                    and found several flint artifacts and burned bone fragments. 
                   
                  His curiosity and latent geological instincts 
                    aroused, Mear returned to the site several times, usually 
                    accompanied by Kenneth Rochat, and screened quantities of 
                    the loose, disturbed fill in the shelter. His labors paid 
                    off: among the items he recovered was a complete, exquisitely 
                    made Folsom point. "I was stunned by the beauty of the 
                    workmanship. The hair on the back of my neck bristled erect. 
                    I had never seen anything like it, and instinctively knew 
                    it was something special." 
                  After finding two additional Folsom points, 
                    Mear reported his finds to Ellen Quillin, director of the 
                    Witte Memorial Museum in San Antonio, and she, in turn, reported 
                    the discovery to Glen Evans of the Texas Memorial Museum. 
                  An experienced geologist who had also investigated 
                    several important "early man" sites, Evans was excited 
                    about the site's potential significance. Roughly 20 years 
                    earlier in 1927 at Folsom, New Mexico, the same type of thin, 
                    fluted projectile points had been discovered embedded in the 
                    bones of extinct bison. Prior to that time, it was believed 
                    that the first cultures in America were no more than several 
                    thousand years old. The New Mexico discovery revolutionized 
                    perceptions about the peopling of North America, pushing estimates 
                    of the time of first migrations back into the Late Pleistocene. 
                  In September 1948, Evans accompanied Mear to 
                    Uvalde County to see the shelter, and the two cleaned off 
                    a wall section of fill exposed in one of the old pits. The 
                    shelter deposits were seen to be stratified (layered), with 
                    bones of extinct animal species in place in the lower deposits 
                    and cultural materials associated with bones of modern animals 
                    in the upper layers. The possibility of gaining important 
                    new information was obvious, and plans for excavation immediately 
                    were initiated.  
                  Controlled Excavations Begin
                  Excavation at the Kincaid site was done in two 
                    stages. The first stage, in the fall of 1948, was sponsored 
                    by the Texas Memorial Museum (TMM) and was directed by geologists 
                    Sellards and Evans. During this first stage of excavation 
                    most of the shelter deposits were removed down to the culturally 
                    sterile fill near the bottom of the shelter, and deep trenches 
                    dug in the terrace deposits in front of the shelter. 
                  Before systematic investigations were underway, 
                    Evans and the TMM crew evaluated the extent of disturbances 
                    to the shelter fill. Several large pits had been dug inside 
                    and near the shelter. In addition to the deepest of these 
                    old pits and trenches, there were numerous smaller areas where 
                    the fill had been dug into and otherwise disturbed. Only a 
                    small area of the surface remained in its original position 
                    but, fortunately, there had been somewhat less disturbance 
                    in the deeper zones and in front of the shelter. 
                   From people living in the area it was learned 
                    that the pits and holes had been dug by "treasure hunters," 
                    men who were down on their luck and hoping for a change in 
                    fortune. Hit hard by the Great Depression of the 1930s, they 
                    hoped to find some of the legendary treasures described in 
                    books by J. Frank Dobie and other authors. 
                  Evaluating the damage, Evans saw that other 
                    forces also had taken a toll on the shelter deposits. Small 
                    rodents had burrowed into the shelter fill. There were also 
                    larger burrows, probably made by badgers or coyotes digging 
                    into the rodent dens. Some of the larger burrows had collapsed, 
                    causing overlying deposits to be displaced. In some of the 
                    open burrows, materials obviously belonging in the uppermost 
                    layer had filtered down into lower layers or deposits. Roots 
                    from brush and small trees growing on the slope in front of 
                    the shelter had worked their way inside, causing considerable 
                    disturbance, particularly in the upper two feet of deposits. 
                  The first step for investigators was to remove 
                    the old waste heaps and other disturbed materials in order 
                    to minimize the danger of mixing artifacts and faunal remains 
                    belonging in different stratigraphic layers, or zones. The 
                    disturbed sediment was hauled out of the shelter in wheelbarrows 
                    and passed through screens to salvage artifacts and bones. 
                    The process of clearing out debris required more than two 
                    weeks. After the debris had been removed from the shelter, 
                    the entire surface of the undisturbed fill was painstakingly 
                    swept clean with whisk brooms. 
                  With a more-level floor in front of them, Evans 
                    and his crew laid out a grid system made up of six-foot squares 
                    across the undisturbed shelter fill. In order to be able to 
                    study the stratigraphic relationships between deposits inside 
                    and outside the shelter, they first dug contiguous rows of 
                    successive squares to form long trenches, extending from the 
                    back wall of the shelter. The two principal trenches thus 
                    developed were between lines B-C and A-C, as shown on the 
                    grid map above, left. 
                  The crew then excavated the remaining squares 
                    and plotted cross-section profiles. When possible, each square 
                    was excavated in six-inch (15 cm) vertical cuts, or levels, 
                    but thicker cuts were made at some places where large boulders, 
                    tree roots, or burrows made the usual six-inch cuts impractical. 
                    Excavated material was screened, typically through one-quarter 
                    inch screens; artifacts were logged by provenience (location); 
                    and sediments were hauled out of the shelter to a spoils dump 
                    area.  
                  The TMM crew succeeded in investigating and documenting 
                    most of the shelter deposits, with the exception of several 
                    units near the back where ancient travertine had hardened 
                    and cemented the fill. The workers completely removed the 
                    ancient stone pavement and excavated the underlying zones 
                    to near bedrock in several areas. A section of fill near the 
                    front of the shelter was left in place as a "witness 
                    column" to preserve a record of the stratigraphy for 
                    future investigators. Artifacts, animal bones, a sample of 
                    paving stones, and field records were packed up and brought 
                    back to Austin for analysis.  
                  U.T. Students Join the Effort
                  Five years later, during the heat of summer, 
                    the second stage of excavation got underway at Kincaid. Led 
                    by anthropology professor T. N. Campbell, the effort was jointly 
                    sponsored by the Texas Memorial Museum and the Department 
                    of Anthropology of the University of Texas. Unlike the previous 
                    work, the project was conducted as a summer field school to 
                    train college students in archeological methods and techniques. 
                   
                  During the field school, the remainder of the 
                    culture-bearing deposits of the shelter were removed, further 
                    excavation was done in the terrace deposits in front of the 
                    shelter, and five test excavations were made in the terrace 
                    deposits under small bluff overhangs immediately east and 
                    west of the main shelter.  
                  Among the field school students was a young woman 
                    destined to become a major figure in Texas archeology. Dee 
                    Ann Suhm (later Story) received her basic training at Kincaid, 
                    went on on to earn a Ph.D in anthropology, teach at the University 
                    of Texas, and become the director of the Texas Archeological 
                    Research Laboratory. Her field notes from Kincaid reflect, 
                    even then, a careful eye for detail that she used effectively 
                    in writing landmark publications such as An Introductory 
                    Handbook to Texas Archeology, published in 1954 with Edward 
                    Jelks and Alex Krieger.  
                   
                    Over the course of roughly four months of excavations during 
                    the two field projects, a total of 55 squares were dug. 
                    All collected artifacts were assigned numbers according 
                    to their provenience, including a generic lot number to designate 
                    materials found out of their original context. Evans estimates 
                    that at least half of the artifacts were from the disturbed 
                    fill, or backdirt scooped out and left in piles by the treasure 
                    seekers.  
                  A Voice from Beyond 
                  While excavating the site in 1948, Evans remembers 
                    having a brush with a furtive individual who, unbeknown to 
                    the crew, was watching over the procedures.  
                  He recalls: "I was digging a profile out 
                    in front of the shelter when I heard a low voice saying, 'You're 
                    diggin' in the wrong place.' It was one of the local people, 
                    hidden in the brushlikely one of the people who had 
                    dug at the site seeking treasure. But it was clear he wasn't 
                    prepared to show me what he thought was the right place." 
                   
                  And of course, Evans notes, the true treasure 
                    at Kincaid was very different than what the strange man was 
                    after.      
                 | 
                
                  
                     
                        
                        
                          Backdirt piles left by treasure hunters 
                          can be seen mounded in the front of the shelter (bottom 
                          right ) after TMM investigators cleared away brush. 
                          TARL archives. Click to see full image.
                           Click images to enlarge 
                         
                       | 
                     
                     
                        
                        Books such as J. Frank Dobie's Apache 
                          Gold and Yaqui Silver, published in 1939, inspired treasure 
                          hunters to dig for legendary riches in area caves and 
                          shelters such as Kincaid. Original Tom Lea cover art 
                          and frontspiece courtesy of Mrs. Tom Lea and the University 
                          of Texas Harry Ransom Center, Dobie Collection. 
                       | 
                     
                     
                        
                        Work log and payroll (click to enlarge) 
                          from Glen Evans' notes lists the small start-up crew 
                          for the TMM project (including an individual named Woodrow 
                          Wilson, presumably not related to the former U.S. president). 
                          Paleontologist Grayson E. Meade was among those added 
                          to the crew, and anthropologists Thomas N. Campbell, 
                          Alex Krieger, and J. Charles Kelley were frequent visitors 
                          to the site during TMM excavations. TARL archives. 
                       | 
                     
                     
                      |  
                         Prior to the 1927 Folsom discovery in New Mexico, 
                          it was believed that the first cultures in America were 
                          no more than several thousand years old. That find revolutionized 
                          perceptions about the peopling of North America, pushing 
                          estimates of the time of first migrations back into 
                          the Late Pleistocene. 
                       | 
                     
                     
                         
                        Geologist Glen Evans stands in one 
                          of the large pits left by treasure hunters inside the 
                          shelter. Evans directed Texas Memorial Museum field 
                          work at the site. TARL archives. Click to see full image. 
                         
                       | 
                     
                     
                        
                        Stones from the pavement, shown lined 
                          up across the front of the shelter, were removed after 
                          documentation so that the investigators could excavate 
                          the layers beneath them. The small tree at left marks 
                          the location of the "witness column" left 
                          in place by TMM crews to preserve a record of the stratigraphy. 
                          TARL archives. 
                       | 
                     
                     
                        
                        The University of Texas Field School 
                          crew, headed by anthropology professor Thomas N. Campbell 
                          (left), poses in front of the shelter before beginning 
                          excavations in 1953. The field school dug the remaining 
                          shelter fill and excavated sections of the terrace in 
                          front of the shelter. TARL archives. Click for more 
                          detail. 
                       | 
                     
                    
                        
                        Glen Evans, shown in trench in front 
                          of the shelter. One day while working, Evans heard a 
                          voice from the underbrush nearby, telling him that he 
                          was digging in the "wrong place."   | 
                     
                   
                 |