Vertebrate fossils are known from numerous localities of Cretaceous and Quaternary age in the Balcones fault zone. Cretaceous vertebrate remains range in age from Neocomian to Maastrichtian and represent the following groups,Chondrichthyes, Osteichthyes, and Reptilia (Ichthysaura, Plesiosauria, Squamata, Crocodilia, Pterosauria, Saurischia and Ornithischia). Trackways of both ornithischian and saurischian dinosaurs are known from the Glen Rose Formation.
Quaternary vertebrates are known from cave deposits of the Edwards Plateau and terrace deposits on the Gulf Coastal,Plain. The mammalian assemblages from these deposits provide data on the Pleistocene and Holocene environments of Central Texas.
Figure1. Map showing locations of dinosaur tracks in the Glen Rose Formation of Central Texas (Modified from Langston, 1974). Scale in miles.
Figure 2. Map showing the locations of Quaternary localities mentioned in the text. 1) Kyle site, Hill Country; 2) Laubach Cave, Williamson County; 3) Fyllan Cave, Travis County; 4) Mac's Cave, Travis County, 5) Barton Road site, Travis County; 6) Levi Shelter, Travis County; 7) Longhorn Cavern, Burnet County; 8) Miller's Cave, Llano County; 9) Wunderlich site, Comal County; 10) Friesenhahn Cave, Bexar County; 11) Cave Without a Name, Kendall County; 12) Kincaid Shelter, Uvalde County; 13) Rattlesnake Cave, Kinney County; 14) Felton Cave, Sutton County; 15) Centipede Cave, Val Verde County; 16) Damp Cave, Val Verde County; 17) Cueva Quebrada, Val Verde County; 18) Bonfire Cave, Val Verde County. Scale in miles.
Figure 3. Map showing modern distributions of Synaptomys cooperi (vertical lines) and late Pleistocene occurrences in Texas and Mexico. Scale in miles.
Figure 4. Map showing modern distributions of Cynomys ludovicianus (vertical lines) and Tamias striatus (horizontal lines), and the locations of Schulze Cave and Friesenhahn Cave (filled circles), where both occur in Pleistocene faunas. Scale in miles.
Figure 5. Map showing the modern distributions of Mustela erminea (horizontal lines) and Reithrodontomys fulvescens (vertical lines), and the location of Schulze Cave (filled circle), where both occur in a Pleistocene fauna. Scale in miles.
The Balcones fault trend, as a fault system in Texas, extends southward from Waco through Austin and San Antonio. It then turns westward towards Del Rio and crosses the Rio Grande River into Mexico. This fault system, and its associated fault-line scarp, is the dividing line between the Edwards Plateau and the Gulf Coastal Plain. Over most of this distance, the faulting juxtaposes Lower Cretaceous marine limestones, forming the Edwards Plateau, against Upper Cretaceous marine chalks and shales that underlie the western margin of the Gulf Coastal Plain. The area considered here is approximately two counties (50 miles) wide on each side of the fault zone; data from other parts of the Edwards Plateau and the Gulf Coastal Plain have also been used. The vertebrate fossils found in these sedimentary rocks have added significantly to knowledge of the faunal history of this region. Most of these fossils have been recovered from the Cretaceous limestones, shales and chalks and from the Pleistocene cave deposits of the Edwards Plateau and alluvial deposits of the streams that traverse the area. A few marine vertebrates are known from the Kincaid Formation of Paleocene age.
Vertebrate fossils of Cretaceous age are sparsely distributed throughout the Balcones fault trend. Those on the Edwards Plateau are from Comanchean rocks ranging in age from Neocomian to lower Cenomanian. Cretaceous vertebrates from the Gulf Coastal Plain are from Gulfian rocks ranging in age from upper Cenomanian to Maastrichtian. In general, the specimens are much more fragmentary and less common in central Texas than in north and northeast Texas. This is probably the result of much shallower water and lower depositional rates on the San Marcos platform of central Texas (Young, 1986).
The teeth of sharks occur in most of the Cretaceous formations, but are most abundant in the Upper Cretaceous strata from the Eagle Ford through the Navarro. Taxonomic identification of sharks based on teeth is uncertain, but all those from Cretaceous rocks of this area appear to belong to living groups and they have been assigned to modern genera. Teeth of a ray-like shark of the genus Ptychodus are common in the Eagle Ford. Rostral teeth of sawfishes of the genus Ischyrhiza are known from the Taylor Formation (Campanian) of Travis County (McNulty and Slaughter, 1964).
Remains of bony fishes are uncommon in Cretaceous rocks of this area. Tooth plates with rows of button-like teeth from pycnodont fish are found in lagoonal facies of the Lower Cretaceous Walnut and Glen Rose formations. These deep-bodied fish probably inhabited the quiet water lagoons behind the reefs. Their crushing teeth suggest that they fed on hard-bodied animals, perhaps corals or rudistids. At least one genus (Proscincetes = Microdon) is represented.
No turtle material of any consequence has been reported from the region considered here. Fossil turtles of a number of groups representing both fresh-water and marine forms have been found farther north from both Middle and Upper Cretaceous rocks (Langston, 1974; Thurmond, 1969; Zangerl, 1953).
Two vertebrae of marine fish-like ichthyosaurs have been found in Travis County, one from the Del Rio Formation. No locality or stratigraphic data are known for the other. Isolated vertebral centra have been reported from the Ft. Worth Limestone and the Duck Creek Limestone farther north (McNulty and Slaughter, 1962; Slaughter and Hoover, 1963). With one possible exception all Cretaceous ichthyosaurs can be assigned to a single genus, Platypterygius, with two species occurring in North America (McGowan, 1972). Presumably, the material from Texas is assignable to this genus, but isolated vertebrae are inadequate for positive identification.
Another group of large marine reptiles, the Plesiosauria , occur in the Middle and Upper Cretaceous rocks of the Balcones fault trend. These plesiosaurs have been reviewed by Storrs (1981). Two genera of short-necked plesiosaurs are known from the area considered here, Trinacromerum from the Austin Chalk of Williamson County and the Wolfe City Member of the Taylor Formation of McClennan County, and Brachauchenius from the Eagle Ford of Travis County.
Long-necked plesiosaur remains are known from the Eagle Ford Formation of Bell and McClennan counties, but are not sufficient to allow generic identification. Better material from farther north indicates that at least four genera, Elasmosaurus, ?Thalassomedon, Alzadosaurus, and an indeterminate imoliasaurid, are represented in Cretaceous rocks of the Texas Gulf Coastal Plain, and it is likely that the indeterminate specimens from the Balcones fault trend belong to one or more of these taxa.
A poorly known reptile, Coniasaurus, has recently been found in the Eagle Ford Group of Texas (Bell et al., 1982). Remains of these reptiles, previously known only from England, have been recovered from Travis, Bell and McClennan counties in the Balcones fault zone and from farther north in Dallas County where they occur in the Lower Britten Formation of late Cenomanian age. This was a small (1-2 ft. in length) lizard related to the living monitors of the Old World.
Another group of marine reptiles, the mosasaurs, is also found in Gulfian rocks of central Texas. Isolated vertebrae and other bones are not uncommon in the Austin, Taylor and Navarro groups. According to Thurmond (1969) nine taxa of mosasaurs occur in Texas Gulfian rocks, although not all of them are known from the Balcones fault trend. The oldest mosasaur remains from Texas come from low in the Eagle Ford in Dallas County (Thurmond, 1969). Another early specimen is from the Eagle Ford-Austin contact in McLennan County, identified by Lucas as Clidastes (Hill, 1901). The very large genus Mosasaurus, first found in Europe in 1779, occurs in Taylor rocks in the city of Austin.
Crocodilians are represented by a portion of a pelvis from the Travis Peak Formation and by dermal scutes of a large crocodilian (Dakotasuchus) from the Cow Creek Sandstone of western Travis County (Langston, 1974). Two partial skeletons of small (approximately 2 feet) crocodilians are known from the lower Glen Rose of Kendall and Medina counties (Langston, 1974). The Medina County specimen comes from an algal bed about twenty feet below the Corbula bed. This horizon, which also contains pholad borings, mudcracks, dinosaur tracks, casts of dinosaur vertebrae and scratch marks that may have been made by a pterosaur, has been interpreted by Stricklin and Amsbury (1974) as a surface subjected to brief subaerial exposure. The locality and horizon of the Kendall County specimen is unknown. Other fragmentary crocodilian remains have been recovered from the Navarro Formation of eastern Travis County.
Pterosaurs are known from several fragmentary specimens from central Texas. The light, extremely thin-walled bones of these animals facilitate their assignment to this group. Langston (1974) reports a well-preserved first phalanx of a wing finger from the Buda Formation in Hays County. There is also in the collection of the Texas Memorial Museum of the University of Texas a fragment of a long bone of a pterosaur from the Eagle Ford Formation of Travis County. Some tracks from the Glen Rose Formation in Medina County were possibly made by a pterosaur (Langston, 1974).
Skeletal remains of saurischian dinosaurs from the Balcones fault trend are rare, although a number of forms are known from farther north (Langston, 1974). Only one occurrence of any importance is known: fragments of a skull and cervical vertebrae, a radius and a metacarpal referred by Langston (1974) to the sauropod, Pleurocoelus. The material was found on the lee side of a rudistid reef in the middle part of the Glen Rose Formation in Blanco County.
The most common vertebrate fossils in the Cretaceous of the Balcones fault trend are footprints and trackways, mostly of dinosaurs and mostly from the Glen Rose Formation. They are widely distributed in central Texas (Fig. 1) and occur at a number of levels within the Glen Rose, and younger formations (Pittman, Pers. Com; Perkins and Stewart, 1971). The tracks in the Paluxy River near the town of Glen Rose are close to the contact of the Glen Rose Limestone and the underlying Bluff Dale Sandstone (Rodgers, 1967). Tracks on the south San Gabriel River in Williamson County are on the upper surface of the Glen Rose Limestone (Perkins and Stewart, 1971). Some trackways in Medina County are located below the Corbula bed in the lower Glen Rose Formation (Stricklin and Amsbury, 1974). The first report of dinosaur tracks in Texas was that of Shuler (1917) on tracks near Glen Rose, north of the Balcones fault zone. Little technical work was done on these tracks for many years and most of the reports were descriptive. Bird (1939, 1941, 1944, 1954, 1985) reported and described a number of occurrences in a series of popular articles.
Three distinct types of tracks are represented in the Comanchean rocks of central Texas. One consists of tridactyl tracks with long slender toe marks and a narrow "heel," and may or may not have claw marks. These have been assigned to the ichnogenus Eubrontes but Langston (1974) placed them in Irenesauripus and suggests that they were made by a large carnosaur.
A second type of tridactyl track, usually smaller than Irenesauripus, has short, blunt digits that diverge widely and a broader, rounder "heel." These were assigned by Langston (1974) to Gypsichites, and probably represent an ornithischian dinosaur. The exact affinities of the dinosaur that made these tracks is uncertain. Although the tracks resemble those of iguanodontids, Langston (1974) pointed out that the only definitely known iguanodontid from the Comanchean is Tenontosaurus which had four toes while the Glen Rose tracks only show three.
A third category of tracks show a lack ofclaws on the front feet and the presence of four claws on the hind feet; they were made by a large sauropod. Langston (1974) suggested that the animal that made the tracks was Pleurocoelus which is known from skeletal remains from related deposits.
Some trackways have been the basis for speculation about the behaviour of dinosaurs. Bird (1944) described a large number of tracks from the Davenport Ranch in Bandera County. There were 23 individual trackways of sauropod dinosaurs, with a maximum divergence angle of about 250 (Ostrom, 1972). In addition there were four sets of three-toed tracks which Ostrom (ibid) believed to have been made by theropods. Two of the theropod trackways paralleled the sauropod trackways and two were oriented at right angles to the trackways. Bird (1944) suggested that the common orientation of the sauropod trackways indicated that sauropods were gregarious animals. Ostrom (1972) has more recently given a more critical discussion of the subject of the implications of the trackways for dinosaur behaviour, and arrived at essentially the same conclusion as Bird. Later work by Farlow (1981) on three theropod trackways from Kimble County has concluded that the animals that made these tracks were moving at 30, 40 and 43 kilometers per hour.
An analysis of the tracks provides information on the locomotion of some dinosaurs and on water depth in the area the tracks. Bird (1944) reported a sequence of sauropod tracks from Bandera County consisting only of fore-foot prints in a straight line. Where the trackway turns to the right, there is a partial track of a left hind foot with the claws spread, which was used to change the animal's direction. Bird's interpretation that the animal was afloat and used the hind foot only to change direction is probably correct.
The tracks of a carnosaur on the top of a monopleurid biostrome in Bandera County have been used by Perkins (1974) to infer the depth of water for monopleurids. The size of the tracks (1 ft.) indicates an animal no more than ten feet high in walking position. The tracks show no sign of sliding from which Perkins infers that the water was too shallow to offer any support for the body, being probably not more than three or four feet deep. The surface with the tracks is overlain by more monopleurid reefs. Assuming the tracks were made at low tide and excluding large sea-level changes, Perkins concluded that water depth over the monopleurid beds was unlikely to be more than 10-12 feet. There is no conclusive evidence of subaerial exposure of these reefs at that time which suggests that, like the sauropods, theropods also ventured out into shallow marine water.
The Cretaceous-Tertiary contact is present in the eastern part of the area considered here. Although it has been investigated in connection with the Cretaceous extinction event (Hansen, 1982; Jiang, 1980), the record of fossil vertebrates is too sparse to be useful.
Quaternary faunas are known from caves in the Cretaceous limestones of the Edwards Plateau and from fluvial deposits on terraces of the streams that cross the Balcones fault zone. All of the known faunas, with one exception, are Rancholabrean or Holocene in age.
The locations of the faunas discussed here are shown in Fig. 2. The data on late Pleistocene and Holocene faunas for Central Texas was summarized by Lundelius (1967). This paper draws on that report, which dealt essentially with the mammalian species, with additions and corrections resulting from new information.
The Balcones fault zone is today an ecologically diverse area. The Balcones fault has placed the hard resistant limestones of the Edwards Plateau against the less resistant shales and chalk of the Gulf Coastal Plain. This has resulted in the formation of the Balcones Escarpment and the dissection of the eastern and southern edges of the Edwards Plateau to form deep and humid canyons separated by relatively dry divides. The outer edge of the Gulf Coastal Plain is a low-relief rolling surface with broad stream valleys. It is this diversity in topography and substrate that is the basis for the environmental diversity. The change in substrate, topography and elevation that takes place at the Balcones escarpment makes this an important biological boundary, as well as a physiographic boundary as was recognized by Blair (1950). An examination of the distributions of Pleistocene faunas shows that this was also the case at that time.
Most of the fossil localities are caves that have been opened to the surface by the dissection of the edge of the Edwards Plateau. The few fossil localities that are known from the Gulf Coastal Plain are from terrace deposits. Approximately half of the localities are archaeological sites. The assemblage of animals from each type of site is biased in different ways. With some exceptions (e.g. Friesenhahn Cave, Kincaid Shelter, Bonfire Shelter and Cueva Quebrada) larger animals are poorly represented in cave deposits when compared to open terrace sites, which frequently do not have a good representation of small animals. Archaeological sites usually have assemblages strongly biased in favor of those species used as food by aboriginals.
Middle Pleistocene Faunas
The oldest known Pleistocene fauna is from Fyllan and Kitchen Door caves immediately west of the Balcones fault zone in Austin, Travis County. These completely filled caves are exposed in a limestone quarry. The faunas have been studied by Taylor (1982), who considered them to be of mid-Irvingtonian age, partly on the basis of resemblance to Irvingtonian faunas in other parts of North America. The muskrat represented is assignable either to Ondatra hiatidens or 0. annectens, 0. annectans is known from such Irvingtonian faunas as Cudahy, Kansas; Conard Fissure, Arkansas; Trout Cave, West Virginia; and Cumberland Cave, Maryland. 0. hiatidens is known from Port Kennedy Cave, Pennsylvania. Paleomagnetic analysis of the sediments surrounding the fossils indicates that they originally had reversed polarity. This indicates a minimum age of 0.73 m.y. B.P., which agrees with the faunal evidence.
The Fyllan Cave fauna contains remains of animals indicative of a wide variety of habitats. Some, such as the microtine rodents (Atopomys texensis, Pitymys quildayi and Ondatra c.f. annectens or 0. hiatidens) and a tapir (Tapirus sp.) indicate the presence of habitats more mesic than the present. More open habitats are indicated by horse (Equus), antilocaprid and the large extinct peccary, Platygonus. In general, the fauna indicates a climate with a more equable climate than today and the presence of diverse habitats.
Late Pleistocene Faunas and Environments
All of the other known Pleistocene faunas from the Balcones fault zone are Wisconsinan in age. All of the available radiocarbon dates are younger than 30,000 years B.P. Most of the material is from cave deposits but some material has been recovered from terrace deposits. The material from the terrace deposits is not abundant and has not been studied carefully. As a result, the degree, nature and significance of the faunas from the various terraces of the streams are not understood. The following discussion is based almost entirely on cave faunas.
Faunas of late Pleistocene age from this region are made up of three categories of species. One is made up of extinct species such as Mammuthus jeffersonii (Jeffersonian mammoth), Mammut americanum (American mastodon), Smilodon floridanus (sabertoothed cat), Canis dirus (dire wolf), Glossotherium harlani ( Harlan's ground sloth) and other species, mostly of large size, that are usually considered to be characteristic of the Pleistocene. Most of these species disappeared between 10,000 and 11,000 years BP (Martin, 1984).
A second group is composed of extant species that either no longer inhabit the Balcones fault zone or are confined to small refugia in the canyons along the edge of the Edwards Plateau. Examples are Sorex cinereus (masked shrew), Mustela erminea (ermine), Synaptomys cooperi (southern bog lemming), Microtus pennsylvanicus (meadow vole), Blarina brevicauda (short-tailed shrew), Tamias striatus (eastern chipmunk) and Pitymys pinetorum (pine vole). These species are found living today to the north or east in climates that are cooler and/or wetter than those in central Texas (Fig.3).
The third group of species still lives in Central Texas. Most of the vertebrate species living today in Central Texas are found in Upper Pleistocene deposits in this area. Exceptions are Dasypus novemcinctus (nine-banded armadillo), Tayassu tajacu (collared peccary), Spermophilus variegatus (rock squirrel), and Bassariscus astutus (ringtail). The first two are recent arrivals [D. novemcinctus about 1890-1900 A.D.( Buchanan and Talmadge, 1954); T. tajacu somewhat earlier but probably after 1700 A.D.]. The precise time the latter two species appeared in Central Texas is not known, but it was some time in the Holocene.
The late Pleistocene faunas give information on Pleistocene environments. The use of extinct species for paleoenvironmental reconstructions is difficult because it is obviously impossible to obtain experimental or field data on their tolerances and habitat preferences. Data on close living relatives may not be applicable because of specific differences in habitat requirements.
Many of the extinct species were very widely distributed in North America, which can be interpreted either that these species had wide tolerances for environmental conditions or that a relatively uniform environment prevailed over North America during the Pleistocene. A detailed examination of the distributions of extinct species shows regional differences that almost certainly reflect regional environmental differences. The American mastodon (Mammut americanum) is much more abundant on the Gulf Coast Plain and in eastern North America generally, than on the Edwards Plateau or the High Plains. The mammoth (Mammuthus jeffersonii) on the other hand was common over most of Texas including the Gulf Coastal Plain. This difference in the relative abundance of mammoth and mastodon indicates differences in the environments of the two regions. The teeth of the two animals are quite different; those of the mammoth are complex and high-crowned and better adapted to a diet of harsh vegetation such as grass than are the simple low-crowned teeth of the mastodon. This indicates more forested areas on the Gulf Coastal Plain than on the Edwards Plateau. The mastodon was not completely absent from the Edwards Plateau but was probably restricted to stream valleys.
Other extinct species were either entirely or largely confined to the Gulf Coastal Plain. The long-nosed extinct peccary, Mylohyus nasutus, is confined to the eastern half of North America and is not known west of the Balcones fault zone. Extinct species that are either absent or rare west of the Balcones fault zone in addition to the long-nosed peccary, Mylohyus nasutus, are the ovibovine, Symbos cavifrons, which is common in the northeast and midwest and whose remains are frequently associated with woodland pollen (Semken et al., 1964), the large armadillo, Holmesina septentrionalis, which has one late Pleistocene record at Lubbock (Johnson, 1985), the glyptodont, Glyptotherium floridanus, with one late Pleistocene record from Andrews County (Lundelius, 1967), and a capybara, Neochoerus pinckneyi. All of these animals are associated to a considerable extent with forest or woodland conditions and some, such as the glyptodont and capybara, with permanent bodies of water (Gillette and Ray, 1981). The living relatives of Holmesina, Glyptotherium and Neochoerus are warm-climate animals. Their southerly distribution suggests mild temperatures for the Gulf Coastal Plain during the late Pleistocene. Thus, the distributions of many of the extinct forms provide information on Pleistocene environments.
Extant species of Pleistocene faunas are a more reliable source of paleoenvironmental data. Because there is little direct information on animal tolerances, present distributions are used as approximate indicators of their environmental requirements. Pleistocene faunas of Central Texas have a number of species that do not live there today but are found to the north and/or east in areas with cooler and/or wetter climates.
The following species are found today well to the north of Texas. Sorex cinereus (masked shrew), known from several late Pleistocene faunas of Central Texas including Cave Without A Name, is presently found throughout northern North America. Its southern limit on the Great Plains is southern Nebraska but it is found farther south at higher elevations in New Mexico, Tennessee and North Carolina. The southern bog lemming, Synaptomys cooperi, is known from Cave Without a Name in Kendall County. Its present distribution southwestward ends in central Missouri, except for two outliers associated with artesian springs in southwestern Kansas. Microtus pennsylvanicus (meadow vole), whose southern limits are in central Nebraska and northern Missouri, is known from Pleistocene faunas from Cave without a Name in Kendall County. The eastern chipmunk, Tamias striatus, is known from Friesenhahn Cave and Cave Without a Name (Graham, 1984). Its modern distribution covers most of the eastern United States with the exception of Florida and the Atlantic Coastal Plain.
The ermine, Mustela erminea and a large subspecies of raccoon, Procyon lotor simus, are also found in Pleistocene faunas of central Texas, the former from Cave Without a Name, the latter from Pleistocene deposits in Levi Shelter, Travis County, and Holocene deposits in the Kyle Site, Hill County. Sorex cinereus is presently found in the northern part of North America including the Arctic. It, like Sorex cinereus, extends farther south in the Rocky Mountains into northern New Mexico and on the east coast into Maryland. Procyon lotor simus, which has much more massive mandibles and skulls than modern populations from Central Texas is now confined to the Pacific Northwest (Wright and Lundelius, 1963).
Another group contains extant species that are no longer found in central Texas; these range to the north and/or east but do not extend as far north as those in the previous group. The prairie vole, Microtus ochroqaster, and the pine vole, Pitymys pinetorum, are difficult to separate on the basis of fragmentary material but both appear to have been present in Pleistocene faunas of central Texas. The southern short-tailed shrew, Blarina olinensis, is found in many Pleistocene faunas (Cave Without a Name, Friesenhahn Cave and Laubach Cave). Another such species is the prairie dog, Cynomys ludovicianus, which is also represented in several late Pleistocene faunas (Friesenhahn Cave, Cave Without a Name, Laubach Cave). It is widely distributed today on the Great Plains but does not extend southeastward beyond Mason County. All of these extant extralimital species indicate a late Pleistocene climate that was cooler and probably had more available moisture than the present climate of central Texas.
Another characteristic of the Pleistocene faunas of this region is the association of species that are now allopatric and whose habitat requirements seem to be mutually exclusive. Hibbard (1960) called attention to this and proposed that these associations indicate a climate with less seasonality than the present. This is based on the assumption that the northern species are limited today to the south by the summer temperature maxima and their associated high evaporation rates. A comparison of the distribution maps of these species with the climatic maps of Visher (1945) support this hypothesis. Several of these species (Sorex cinereus, Mustela erminea) extend farther south in the Rocky Mountains and Appalachian Mountains than they do in the central part of the continent and Visher's maps show that both higher rainfall and lower temperatures extend farther south in these areas today than in the central part of the continent.
These associations have been termed "disharmonious" by Semken (1974) and "intermingled" by Graham (1985). All well-known late Pleistocene faunas in Central Texas show these associations. The following two examples are typical: Late Pleistocene faunas from Schultze Cave (Dalquest et al., 1969) and Friesenhahn Cave (Graham, 76a) have both Cynomys ludovicianus (black-tailed prairie dog) and Tamias striatus (eastern chipmunk). These two species are today allopatric (Fig. 4). The Schultze Cave fauna also contains both Mustela erminea (ermine) and Reithrodontomys fulvescens (harvest mouse), also allopatric today (Fig. 5). They are an indication that the late Pleistocene climate of central Texas was different from any climate we see today and that there are no precise modern analogues. The Pleistocene faunas had a higher species density than do the modern ones. This is not only because of the presence of the extinct species, but also the mixing of extant northern and southern species contributed to the species richness (Foley, 1984; Graham, 1978, Guilday et al., 1978).
The major biological event that marks the end of the Pleistocene was the extinction of many species of large mammals and a few species of small mammals. The exact number of species involved depends somewhat on the taxonomic arrangement used. Martin (1984) lists genera of large mammals (body weight <44 kilograms) and genera of small mammals (body weight > 44 kilograms) that became extinct in North America at the end of the Pleistocene. A conservative estimate of the number of species involved is 28 large and 4 small mammalian species. The time of extinction has been investigated by means of radiocarbon dates by Jelinek (1957), Hester (1960), and Martin (1958, 1967, 1984). The lastest comprehensive study (Martin, 1984) involving new dates and re-evaluation of many radiocarbon dates indicates that the majority of the species disappeared 10,000 to 11,000 years ago.
The cause or causes of this extinction have been discussed by many authors for decades. Martin (1967, 1973, 1084) has presented the case for human overpredation. Guilday (1967), Slaughter (1967), Axelrod (1967), Lundelius (1967), and Graham and Lundelius (1984) have argued that the climatic change that took place at the end of the Pleistocene was the primary cause. The case for human overpredation rests heavily on the close timing between the arrival of humans in North America and the disappearance of the large mammals. It assumes that the early human populations were numerous and were technologically capable of exterminating these species. The case for climatic change rests largely on the assumption that niches were eliminated by the climatic change. This hypothesis is supported by the disappearance of the disharmonious assemblages at the same time that the large animals became extinct.
Holocene Faunas and Climates
After the extinction of the large mammals 10,000 to 11,000 years ago, the fauna of Central Texas consisted of the modern fauna plus a few species that are now found north and east of Central Texas in areas of cooler and more mesic climates. A few species, such as the armadillo, arrived later in the Holocene.
Except for Procyon lotor simus, the species that are found today farthest north disappeared earliest from central Texas. Sorex cinereus, Microtus pennsylvanicus and Mustela erminea are last known from Cave Without a Name (10,900 yrs B.P.). They apparently disappeared from Central Texas with the extinct fauna. Synaptomys cooperi is known from Miller's Cave associated with a radiocarbon date on bone of 7,300 yr B.P. This date should be regarded as a limiting date; the actual age is probably older as radiocarbon dates on bone tend to be somewhat young (Tamers and Pearson, 1965).
Another group of species that disappeared later is found today at various distances from the Edwards Plateau. Microtus ochroqaster (prairie vole), which today does not occur south of Oklahoma was present at Miller's Cave in Llano County 3,000 years ago. Blarina carolinensis, now found in East Texas, was at Austin 1,000 years ago (Barton Road site) and possibly as recently as 600 years ago (Mac's Cave). A microtine rodent (,either Microtus ochrogaster or Pitymys pinetorum) is also represented in the Mac's Cave deposits.
There is some evidence that Blarina carolinensis, Microtus ochroqaster and/or Pitymys pinetorum retreated eastward through time across the Edwards Plateau. Their latest occurrence in the western part of the Edwards Plateau is at Felton Cave in Sutton County 7,800 years ago. None are known from Holocene deposits of either Centipede or Damp Caves in Val Verde County. This absence is significant because these caves, which have produced abundant small mammal faunas and are located along the canyon of the Rio Grande, would have been expected to be more mesic than the uplands.
The deep, mesic canyons along the dissected edge of the Edwards Plateau are refugia today for a number of organisms such as the sugar maple (Acer saccharum), a salamander (Plethodon glutinosus), and the pine vole (Pitymys pinetorium), whose primary distributions lie east and/or northeast of Central Texas today (Blair, 1958). This is likely to have been the case for the entire Holocene, during which species formerly widespread on the Edwards Plateau persisted as isolated populations for varying lengths of time in these locally favorable areas as the regional climate changed.
Several species, Geomys bursarius (plains pocket gopher), Scalopus aquaticus (eastern mole), and Cryptotis parva (least shrew), are now absent from most of the Edwards Plateau but occur to the north, south and east. The record of Geomys bursarius is the best of these, and the restriction of its range seems to follow no pattern, geographic or chronologic. This, plus the fact that its modern distribution in Central Texas is spotty and related to locally favorable soil conditions, suggests that soil erosion on the Edwards Plateau is responsible for its disappearance over much of the area. Remains of this pocket gopher have been found in very young deposits in this area, which suggests that restriction of range due to soil erosion has been operative very recently and probably is still going on.
Mus musculus (house mouse), a European animal brought to North America by Europeans, has been found in the black fill unit at Longhorn Cavern in Burnet County, indicating a date younger than 1700 A.D. (Semken, 1961). This unit also contains Geomys bursarius, which does not live today in Burnet County. Its nearest occurrence is in Llano County approximately 40 miles away. The top twelve inches of a black soil unit in Rattlesnake Cave in Kinney County also contains both Mus musculus and Geomys bursarius. The nearest modern occurrence of G. burarius is in Dimmit County 75 miles to the south. In both instances, Geomys remains have been recovered from a black sediment similar to the soil found on the Edwards Plateau today. The record of the burrowing mole Scalopus aguaticus is similar but is less complete.
The species listed earlier as being absent from or sparsely distributed on the Edwards Plateau are fossorial and require a reasonable depth of soil. Their scarcity in this area today seems to be related to the generally thin soil. The stage was set for the removal of soil from the Edwards Plateau by the entrenchment of streams during or before the Pleistocene. The process was probably accelerated by the post-Pleistocene change to drier conditions. The last stage in the removal of the topsoil from this area seems to have taken place since 1800 as the result of extensive overgrazing that destroyed much of the vegetation (Semken, 1961).
There is no evidence in the known Holocene faunas that the climate during the interval of 4,000 to 6,000 years B.P. was any drier or warmer than at present as indicated by the occurrence of Lutra canadensis (otter) in the Wunderlich site in Comal County 5,000 years ago and the fauna from Centipede Cave in Val Verde County 5,000 years ago. This picture may be somewhat biased by the concentration of sites along the eastern and southern edges of the Edwards Plateau, where erosion has opened many caves that are the source of much of the fossil material. Most of these caves are adjacent to canyons that maintained more moist environments than the uplands. Vertebrate fossils from more localities away from the canyons will be needed to settle this question.
I thank Keith Young, Wann Langston, Jr., and Jeff Pittman for helpful comments, David Stephens for assistance with the illustrations, and my wife Judith Lundelius for editorial help. The Geology Foundation of the University of Texas provided financial assistance.
Adkins, W. S., 1923. Geology and mineral resources of McLennan County. University of Texas Bulletin, v. 2340, p. 1-202.
Axelrod, D. I., 1967. Quaternary extinctions of large mammals. California University Publications Geological Sciences, v. 74, p. 1-42.
Bell , B. A., Murry, P. A. and Osten, L. W., 1982. Coniasaurus Owen, 1950 from North America. Journal of Paleontology, v. 56, no. 2, p. 520-524.
Bird, R. T., 1939, Thunder in his footsteps. Natural History, v. 47, p. 254-261.
Bird , R. T., 1941 , A dinosaur walks into the museum. Natural History, v. 47, p. 74-81.
Bird, R. T., 1944, Did Brontosaurus ever walk on land? Natural History, v. 53, p. 61-67.
Bird, R. T., 1954, We captured a "live" brontosaur. National Geographic, v. 105, p. 707-722.
Bird, R. T., 1985, Bones for Barnum Brown: Adventures of a dinosaur hunter. Texas Christian University Press, Ft. Worth, 225 pp.
Blair, W. Frank, 1950, The biotic provinces of Texas. Texas Journal of Science, v. 2, P. 93117.
Blair, W. F., 1958, Distributional patterns of vertebrates in the southern United States in relation to past and present environments, in Hubbs, C. L., ed., Zoogeography: American Association for the Advancement of Science. Publication 51, p. 433-568.
Buchanan, G. D. and Talmadge, R. V., 1954, The geographical distribution of the armadillo in the United States. Texas Journal of Science, v. 6, P. 142-150.
Dalquest, W. W., Roth, E. and Judd, F. 1969. The mammal fauna of Schulze Cave, Edwards County, Texas. Bulletin of the Florida State Museum. v. 13, p. 206-276.
Farlow, J. O., 1981, Estimates of dinosaur speeds from a new trackway site in Texas. Nature, v. 294, p. 747-748.
Foley, R. L., 1984. Late Pleistocene (Woodfordian) vertebrates from the driftless area of southwestern Wisconsin, the Moscow Fissure local fauna. Illinois State Museum Reports of Investigations, no. 39, p. 1-50.
Gillette, D. D. and Ray, C. E., 1981. Glyptodonts of North America. Smithsonian Contributions to Paleobiology, no. 40, p. 1-255.
Gould, C. N., 1929. Comanchean reptiles from Kansas, Oklahoma and Texas. Geological Society of America Bulletin, v. 40, p. 457462.
Graham, R. W., 1976A. Pleistocene and Holocene mammals, taphonomy and paleoecology of the Friesenhahn Cave local fauna, Bexar County, Texas. Ph.D. dissertation, University of Texas, Austin.
Graham, R. W., 1976B. Late Wisconsin mammal faunas and environmental gradients of the Eastern United States. Paleobiology, v., p. 343-350.
Graham, R. W., 1984. Paleoenvironmental implications of the Quaternary distribution of the eastern chipmunk (Tamias striatus) in Central Texas. Quaterna y Research, v.21, no.1, p.111-114.
Graham, R. W., 1985. Response of mammalian communities to environmental changes during the late Quaternary, in Diamond, J. and Case, T. J. eds. Community Ecology: Harper and Row, New York, p. 300-313.
Graham, R. W. and Lundelius, E. L. Jr., 1984. Coevolutionary disequilibrium and Pleistocene extinctions, in P.S. Martin and R. G. Klein eds. Quaternary Extinctions, A Prehistoric Revolution: University Arizona Press, Tucson, Arizona, p. 223-249.
Guilday, J. E., 1967. Differential extinction during late-Pleistocene and Recent times. in Martin, P. S. and Wright, H. E., eds., Pleistocene Extinctions, The Search for a Cause: Yale University Press, New Haven, p. 121-140.
Guilday, J. E., Hamilton, H. W., Anderson, E., and Parmalee, P. W., 1978. The Baker Bluff Cave deposit, Tennessee, and the Late Pleistocene faunal gradient: Bulletin of Carnegie Museum in Natural History, v. 11, p. 1-67.
Hansen, T. A., 1982. Macrofauna of the Cretaceous/Tertiary boundary interval in east-central Texas, in Maddocks, R. F. ed. Texas Ostracoda. Guidebook of Excursions and Related Papers for the Eighth International Symposium on Ostracoda: Department of Geosciences, University of Houston, p. 231-237.
Hester, James J., 1960. Late Pleistocene extinction and radiocarbon dating. American Antiquity, v. 26, p. 58-77.
Hibbard, Claude W., 1951. Animal life in Michigan during the Ice Age, Michigan Alumnus Quarterly Revue, v. 57, p. 200-208.
Hibbard, C. W., 1960. An Interpretation of Pliocene and Pleistocene Climates in North America. Annual Report Michigan Academy of Science, Arts and Letters, 62:5-30.
Hill, R. T., 1901. Geography and geology of the Black and Grand Prairies, Texas. U. S. Geological Survey Report v.21, no.7, 1-666.
Jelinek, A. J., 1957. Pleistocene faunas and early man. Michigan Academy of Science, Arts, and Letters Papers. v.6, p.225-237.
Jiang, Ming-Jung, 1980. Calcareous nannofossils from the uppermost Cretaceous and the lowermost Tertiary of Central Texas. Master's Thesis, Texas A&M University, p. 1-121.
Johnson, E. and Holliday, V. T., 1985. A Clovisage megafaunal processing station at the Lubbock Lake Landmark. Current Research in the Pleistocene. v. 2, p. 17-19.
Langston, W., Jr., 1974. Nonmammalian Comanchean tetrapods. Geoscience and Man, v.8, p.77-102.
Lundelius, E. L., Jr., 1967. Late Pleistocene and Holocene Faunal History of Central Texas, in Martin, P. S. and Wright, H. E. Jr. eds. Pleistocene Extinctions, The Search for a Cause: Yale University Press, New Haven, p. 288-319.
Martin, P. S., 1958. Pleistocene ecology and biogeography of North America in Hubbs, C. L., ed. Zoogeography: Publication 51, American Association for the Advancement of Science, p. 375-420.
Martin, P. S., 1967. Prehistoric overkill, in Martin, P. S. and Wright, H. E., Jr., eds., Pleistocene Extinctions, The Search for a Cause: Yale University Press, New Haven, p. 75-120.
Martin, P. S., 1984. Prehistoric overkill: the global model. in Martin, P. S. and Klein, R. G. eds. Quaternary Extinctions, A Prehistoric Revolution: Univ. Arizona Press. Tucson, Arizona, p. 354-403.
McGowan, C., 1972. The systematics of Cretaceous ichthyosaurs with particular reference to the material from North America. University of Wyoming Contributions to Geology, v. II:-29.
McNulty, C. L., Jr. and Slaughter, B. H., 1962. An ichthyosaurian centrum from the Albian of Texas. Journal of Paleontology, v. 36, p. 346-347.
McNulty, C.L., Jr. and Slaughter, B. H., 1972. The Cretaceous selachian genus, Ptychotrygon Jackel 1594. Eclogae Geologicae elvetiae, v. 65, no. 3, p. 647-655.
Ostrom, J. H., 1972. Were some dinosaurs gregarious? Palaeogeography, Palaeoclimatology, Palaeoecology. v. 11, P. 287-301.
Patton, T. H., 1965. A new genus of fossil microtine from Texas. Journal of Mammalogy, v. 46, no. 3, p. 466-471.
Perkins, B. F., 1974. Paleoecology of a rudist reef complex in the Comanche Cretaceous Glen Rose limestone of central Texas. Geoscience and Man, v. 8, p. 131-173.
Perkins, B. F., and Stewart, C. L. 1971. Stop 7. Dinosaur Valley State Park. in Perkins, B. F., Fry, R. W., Hanor, J. S. et al. Trace fossils, a field guide to selected localities in Pennsylvanian, Permian, Cretaceous and Tertiary rocks of Texas and related papers. B. F. Perkins, ed., Baton Rouge, Louisiana State Univ. Misc. Publ. no. 71-1, p. 56-59.
Sams, R. H., 1982. Newly discovered dinosaur tracks, Comal County, Texas. South Texas Geological Society Bulletin, v. 23, p. 19-23.
Semken, Holmes A., 1974. Micromammal distribution and migration during the Holocene. American Quaternary Association Abstracts. 3rd Biennial Meeting, p. 25.
Semken, H. A., Miller, B. B. and Stevens, J. B., 1964. Late Wisconsin woodland musk oxen in association with pollen and invertebrates from Michigan. Journal of Paleontology. v. 38, no. 5, P. 823-835.
Shuler, E. W., 1917. Dinosaur tracks in the Glen Rose limestone near Glen Rose, Texas. American Journal of Science v. 44, p. 294-298.
Slaughter, B. H., 1967. Animal ranges as a clue to Late Pleistocene extinction. in Martin, P S. and Wright, H. E. Jr. eds. Pleistocene Extinctions, The Search for a Cause: Yale University Press, New Haven, p. 155-168.
Slaughter, B. H. and Hoover, B. R., 1963. Occurrences of ichthyosaurian remains in the Cretaceous of Texas. Texas Journal of Science, v. 15, no. 3, p. 339-343.
Storrs, G. W., 1981. A review of occurrences of the Plesiosauria (Reptilia: Sauropterygia) in Texas with description of new material. M. A. Thesis. University of Texas, Austin, 226 pp.
Stricklin, F. L., Jr. and Amsbury, D. L., 1974. Depositional environments on a low-relief carbonate shelf, middle Glen Rose limestone, Central Texas. Geoscience and Man, v. 8, p. 53-66.
Tamers, M. A. and Pearson, F. J., Jr., 1965. Validity of radiocarbon dates on bone. Nature, v. 208, no. 5015, p. 1053-1055.
Taylor, A. J., 1982. The mammalian fauna from the mid-Irvingtonian Fyllan Cave local fauna, Travis County, Texas. M.A. Thesis, University of Texas, Austin, P. 1-106.
Thurmond, J. T., 1969. Lower vertebrates and paleoecology of the Trinity-Group (Lower Cretaceous) in north central Texas. Dissertation, Dallas, Southern Methodist University.
Visher, S. S., 1945. Climatic maps of geologic interest. Geological Society of America Bulletin, v. 56, p. 713-736.
Wright, Thomas and Lundelius, E., Jr., 1963. Post-Pleistocene raccoons from central Texas and their zoogeographic significance. The Pearce-Sellards Series, Texas Memorial Museum, no. 2, p. 5-21.
Young, K., 1986. Cretaceous, marine inundations of the San Marcos Platform, Texas. Cretaceous Research, v. 7, p. 117-160.
Zangerl, R., 1953. The vertebrate fauna of the Selma formation of Alabama. Part III. The turtles of the family Protostegidae. Fieldiana, Geology Memoir, v. 3, p. 57-132. Part IV. The turtles of the family Toxochelidae. Ibid. p. 137-277.
in Abbott, Patrick L. and Woodruff, C. M., Jr., eds., 1986,The Balcones Escarpment, Central Texas: Geological Society of America, p. 41-50