occasional papers

3 downloads 0 Views 485KB Size Report
Oct 31, 1999 - Paleontological work that was done in Papago Springs Cave, Arizona, ... able for speleothems and fossil bones of Stockoceros onusrosagris; the dates ...... port small fish. ... Jacobs, B. Jones, C. Jones, S. Kennedy, W. May, J.
OCCASIONAL PAPERS OF THE

OKLAHOMA MUSEUM OF NATURAL HISTORY UNIVERSITY OF OKLAHOMA, NORMAN, OKLAHOMA Number 5, Pages 1–41

31 October 1999

PAPAGO SPRINGS CAVE REVISITED, PART II: VERTEBRATE PALEOFAUNA NICHOLAS J. CZAPLEWSKI, JIM I. MEAD, CHRISTOPHER J. BELL, WILLIAM D. PEACHEY, AND TEH-LUNG KU Sam Noble Oklahoma Museum of Natural History and Department of Zoology, University of Oklahoma, Norman, OK 73019 USA (NJC); Quaternary Studies Program and Department of Geology, Northern Arizona University, Flagstaff, Arizona 86011 USA (JIM); Department of Geological Sciences, University of Texas at Austin, Austin, Texas 78712 USA (CJB); 3347 East Seneca, Tucson AZ 85716 USA (WDP); Geochemistry Laboratory, Department of Earth Sciences, University of Southern California, Los Angeles, CA 90089 USA (T-LK). Corresponding author: Nicholas J. Czaplewski e-mail: [email protected]; phone: (405) 325-4579; fax: 405-325-7699

ABSTRACT Paleontological work that was done in Papago Springs Cave, Arizona, over 56 years ago yielded a diverse fauna of mammals of late Pleistocene age. Renewed collecting in this cave included sampling for microvertebrate fossils and pollen. New radiometric dates from Uranium-series isotopes are now available for speleothems and fossil bones of Stockoceros onusrosagris; the dates range from 246 ± 19 ka to 26.7 ± 0.7 ka. A few poorly preserved pollen grains were recovered in a unit of cave fill that is radiometrically dated to the last (Sangamonian) interglacial; the pollen included Quercus and Agave, which are consistent with an interglacial climate. Vertebrate taxa not previously reported in the cave fauna include one fish, seven amphibians, five reptiles, three birds, and eight mammals. Most deposits in the cave produced few fossils. The Sangamonian unit produced only Hyla sp. and Stockoceros onusrosagris. The greatest number of specimens and species richness of vertebrates came from a unit of cave fill that dates to the middle of the last (Wisconsinan) glacial, 42 ka. Members of the vertebrate fauna in the middle Wisconsinan unit suggest the presence then of extensive open grasslands with at least some Ponderosa pine in the vicinity. The same unit contains a no-analog association of mammalian species that includes, among others, Notiosorex crawfordi , Sorex arizonae, Marmota flaviventris , Sciurus cf. S. aberti, Perognathus or Chaetodipus sp., and Sigmodon cf. S. arizonae. Although data pertinent to questions about faunal or climatic change through time are weak, there seems to be little indication of faunal change in Papago Springs Cave through time, possibly indicating that climatic changes in the Southwest were less drastic than those much farther north that were closer to the ice front. Key words: Arizona, cave, fossils, Pleistocene, Quaternary, Rancholabrean, vertebrate paleontology, Wisconsinan glacial.

 1999 Oklahoma Museum of Natural History

ISSN:1080-7004

2

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

Papago Springs Cave (PSC) in southern Arizona is rather famous as a late Pleistocene site for its richly fossiliferous deposits. In particular, the cave deposits preserve a large sample of the extinct pronghorn antelope Stockoceros onusrosagris (Colbert and Chaffee 1939; Roosevelt and Burden 1934; Skinner 1942). This paper is the second of two parts describing the results of our study of PSC. In the first part (Czaplewski et al. 1999) we discussed the geological background and speleogenesis of PSC and provided new radiometric dates from samples of speleothems and vertebrate fossils collected in the cave. The present paper concentrates on the vertebrate fossils and their value for paleoenvironmental interpretation.

METHODS AND MATERIALS Abbreviations used in the text are: est., estimated measurement; F:AM, Frick collection of the American Museum of Natural History; ka, kilo annum (thousands of years before the present); NAU, Northern Arizona University Quaternary Studies Program; OMNH, Oklahoma Museum of Natural History, Section of Vertebrate Paleontology; PSC, Papago Springs Cave; Sec., section; SVL, snout-vent length; U, uranium. Field work was carried out in PSC periodically between 1987 and 1996. A detailed description of the deposits from which we collected fossils is provided in Part I of this study (Czaplewski et al. 1999), as are the radiometric dates for speleothems and fossils. Samples of the cave fill were collected and screenwashed (Cifelli 1996) to recover fossils of small vertebrates. Larger vertebrate bones were collected as they were encountered. The faunal lists reported in this paper are poor for most units. During our work, we identified only one “unit” of cave fill (Unit 7 in stratigraphic Sec. 2; see Czaplewski et al. 1999) that was preserved in sufficient volume to yield a sizeable species list. At stratigraphic Sec. 1, a test pit produced only a few bones. The identifications of vertebrate taxa are based on various keys, standard references, and especially on direct comparison with reference skeletal material and museum specimens. We have not detailed the criteria for the identification of taxa except in some cases where multiple species are known for a given genus in the vicinity of the cave or where additional substantiation was warranted. In such cases we have attempted to make the most precise identification that the preserved elements or frag-

OCCASIONAL PAPERS

ments will bear based on the comparative method. Samples for pollen analysis were taken from some of the same units sampled for vertebrate fossils. Modern environmental setting Papago Springs Cave is situated at about 1560 m elevation in the Canelo Hills in southeastern Arizona. Within a 5-km radius, elevations vary from 1395 m to 1785 m. The main modern vegetation type in the vicinity of PSC is Madrean evergreen woodland (Brown 1994), a warm temperate woodland dominated by oaks (Quercus spp.) and juniper (Juniperus deppeana) with an occasional cypress (Cupressus arizonica ) and an understory of various grasses. Dry south-facing slopes are covered with grasses and occasional agaves (Agave palmeri) and sacahuista (Nolina microcarpa). Along the washes, scattered nogal (Juglans major) and ashes (Fraxinus spp.) occur; these and other trees are sometimes draped with canyon grape (Vitis arizonica ). Within 35 km are the summits of the Santa Rita (2881 m), Huachuca (2885 m), and Patagonia (2201 m) mountains, which are presently clothed with Petran and Madrean montane conifer forests of pine (especially Pinus ponderosa and P. flexilis), white fir (Abies concolor), Douglas fir (Pseudotsuga menziesii), aspen (Populus tremuloides), and other species (Bowers and McLaughlin 1995; Brown 1994; Felger and Johnson 1995; McLaughlin 1995; Wallmo 1955). Papago Springs Cave is near the headwaters of Cienega Creek, and near the drainage divide between Cienega Creek and Sonoita Creek in the Sonoita Basin. Cienega Creek flows northward, eventually into the Tucson Basin; Sonoita Creek flows southwestward into the Santa Cruz River which flows into Mexico and then back northward also to enter the Tucson Basin. These basins, as well as others in southern Arizona, are characteristic developments of the late Tertiary tectonism of the Basin and Range geomorphologic province of North America. Both Cienega Creek and Sonoita Creek are tributaries of the Santa Cruz River, which is part of the Colorado River system. At present, in the vicinity of PSC, Cienega Creek does not normally flow above ground except after local heavy summer rains.

RESULTS Pollen Pollen grains were present in only one sample from Unit 5 of Sec. 2 (associated with radiometric

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

ages of 107, 127, and 133 ka). The grains were rare and poorly preserved but identifiable (Anderson, R. S., pers. comm.). They included Agave sp., Quercus sp., and Compositae, and are suggestive of an interglacial climate and consistent with the radiometric ages for the unit from which they came. Agave pollen is zoochorous and could only have been brought into the cave by an insect (possibly one within the gut of an insectivorous mammal whose body may have been brought into the cave by a predator) or nectarfeeding bat. Flower-visiting bats such as Choeronycteris mexicana presently utilize PSC but have no fossil record there or elsewhere. Vertebrate paleontology We recovered about 27 taxa of Pleistocene animals, mostly small vertebrates, that were not reported by Skinner (1942). In the lists of materials given below, the number of specimens or elements is given in parentheses if more than one (some numbers pertain to specimens cataloged in lots). Specimen numbers are those of OMNH unless otherwise noted. The distribution of taxa in each cave-fill unit is provided in Table 1. Skinner (1942) listed about 12 taxa of Pleistocene vertebrates, mostly large mammals, for which we did not recover additional specimens because of the much more limited extent of our excavations and our emphasis on small vertebrates. These 12 taxa include Threskiornithidae, Myotis (?)evotis, Tadarida brasiliensis, Canis caneloensis, Canis lupus, Taxidea taxus, Spilogale gracilis, Bassariscus astutus, Equus tau, Platygonus alemanii, Cervus sp., and Camelidae. Except where noted, scientific names follow Minckley (fish; 1973), Collins (amphibians and reptiles; 1990), American Ornithologists’ Union (birds; 1983), Wilson and Reeder (extant mammals; 1993), and Harris (extinct vertebrates; 1993). Osteichthyes Cyprinidae Rhinichthys osculus (Girard, 1857)—speckled dace Material.—pharyngeal arch (lower pharyngeal; 51822). Discussion.—Three teeth were present in the major row (all broken, but tooth bases are present), and one tooth is present in the minor row. This (partial) pharyngeal tooth formula is limited to four small native Arizonan genera of cyprinids: Rhinichthys, Meda, Tiaroga, and Plagopterus (Minckley 1973). The PSC specimen is

3

too large to belong to Meda, Tiaroga, or Plagopterus, and represents a Rhinichthys osculus of 90 mm estimated standard length (Minckley, C. O., pers. comm.). Only one mention of Rhinichthys as a late Pleistocene fossil in western North America exists in the literature. Smith (1981) noted “Rhinichthys sp.” based on unpublished data but gave no location or other details. Fishes are virtually unknown in the late Pleistocene of Arizona (Lindsay and Tessman 1974; Miller and Smith 1984). The fish bones (a vertebral centrum of an unidentified small fish also was found in PSC) likely were brought into the cave by a predator. The presence of a fish of 90 mm estimated standard length suggests that, in the late Pleistocene in the vicinity of PSC, the creek or its nearby tributaries probably were larger and certainly less ephemeral than they are now. Amphibia Caudata Ambystomatidae Ambystoma cf. A. tigrinum (Green, 1825)—tiger salamander Material.—dentary (2; NAU 8133); humerus (5; NAU 8134-8136); vertebra (19; NAU 8131-8132, 8170, 8139-8140); radius and tibia (NAU 8138); maxilla (NAU 8171); femur (NAU 8137). Description.—The vertebrae were identified using characters discussed by Holman (1969), Miller (1992), and Tihen (1958). Vertebrae of the Ambystoma tigrinum group of Ambystoma are shorter and wider than those of the remaining species-groups except for those of the A. mexicanum group (Tihen 1958; occurring today in southernmost plateaus of Mexico). Members of the A. mexicanum group are virtually indistinguishable osteologically from those of the A. tigrinum group, although those from the mexicanum group tend to be larger (Tihen 1958). Tihen (1958) presented two ratio measurements of mid-trunk vertebrae that can be useful for differentiating the mexicanum and tigrinum groups. The first is a comparison of the length of the centrum with its anterior width. The second is that of combined zygapophyseal width relative to zygapophyseal length. The term “combined” zygapophyseal width is used to denote the measurement between the lateral borders of the prezygapophyses plus the measurement between the lateral borders of the postzygapophyses. The zygapophyseal length is the longitudinal measurement between the anterior tips of the prezygapo-

4

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

OCCASIONAL PAPERS

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

5

6

physes and the postzygapophyses.

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

posterior

tips

of

OCCASIONAL PAPERS

the

Only three vertebrae from PSC were complete enough to permit these measurements (Table 2). Although the grouping is not large enough to be statistically valid, one measurement would imply that at least one individual represents the Ambystoma tigrinum group. Holman (1969) indicated that the large size of the vertebrae and the upswept neural arch that extends well beyond the end of the centrum is indicative of A. tigrinum. In other species, the posterior part of the neural arch is straighter and does not extend so far posteriorly. Following Holman (1969) and Miller (1992) and the data presented in Table 2, fossil vertebrae from PSC were assigned to A. cf. A. tigrinum on the basis of a high neural arch with a lateral profile of approximately 45 degrees, extending well beyond the end of the centrum (Fig. 1A). Limb bones of A. tigrinum typically are larger than those of the closely related species, A. maculatum (found today east of the Great Plains). Humeri of A. tigrinum and the fossils from PSC have columnar diaphyses, along with an abrupt distal taper, unlike the more curved bones from A. maculatum (Miller 1992). Most fossil dentaries, humeri, vertebrae, and the femur, tibia, and radius are from an Ambystoma having an estimated SVL from 105 to 110 mm (typical adults range from 77 to 162 mm SVL); one maxilla (NAU 8171) is from an animal with an estimated SVL of 65 mm (Fig. 1B).

Figure 1. Pleistocene amphibian and reptile fossils from Papago Springs Cave, Arizona. A, Ambystoma cf. A. tigrinum, trunk vertebra (NAU 8170) in left-lateral view. B, Ambystoma cf. A. tigrinum , left maxilla (NAU 8171) in lingual view. C, Hyla sp. indet., humerus (NAU 8172) in anterior view. D, cf. Gastrophryne sp., humerus (NAU 8169) in lateral view. E, Rana sp. indet., right ilium fragment (NAU 8164) in lateral view. F, Hyla sp. indet., left ilium (NAU 8173) in lateral view. G, Phrynosoma douglasi, right dentary (NAU 8160) in lingual view.

Discussion.— Ambystoma tigrinum (including three subspecies) is the only species of salamander that today lives in the immediate region of PSC. Ambystoma rosaceum is found to the south just into Mexico and the plethodontid salamander, Aneides hardii, is found in south-central New Mexico east of the Rio Grande. Based on modern distributions of salamanders, it is logical that the PSC specimens be assigned to A. tigrinum. However, salamander remains are extremely rare in the Pleistocene record of Arizona as well as most of the Southwest. It is not known whether a presently northern species (e.g., A. macrodactylum in the maculatum group; today in Idaho, Washington, and Oregon) or a more eastern species (e.g., Aneides hardii, not an ambystomatid), or a slightly more southern, Mexican form (e.g., A. rosaceum, in the A. tigrinum

group) occupied southern Arizona during the Rancholabrean glacial or interglacial episodes. The remains from PSC would imply that the A. tigrinum group was in southern Arizona, but there is as yet no evidence of the other groups there. We do not know if A. rosaceum ever occurred farther north than it does today (specimens of this species were not available for the study). As more cave studies produce ambystomatid fossils, it might be possible to determine if A. rosaceum ever occurred as far north as Arizona and possibly what mechanism caused A. tigrinum to differentiate into distinct and hybridized subspecies, as discussed in Jones et al. (1995). More work on the southern Arizona fossil salamanders might indicate that some form of the A. mexicanum group did extend farther northward during a previous interglacial.

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

7

Table 2. Measurements (in mm) and vertebral ratios in modern Ambystoma groups and in specimens from Papago Springs Cave. Modern data from Tihen (1958). See text for discussion of measurements. CL, centrum length; CW, centrum width at anterior end; Prz, prezygapophyseal measurement; Pstz, postzygapophyseal measurement; ZL, zygapophyseal length. Specimen location Papago Springs Cave: Annex Annex Sec. 2, Unit 5 Summary A. tigrinum group A. mexicanum group

Species of Ambystoma including the wideranging A. tigrinum live in communities from quiet water of ponds, lakes, temporary rain pools, and streams, from arid sagebrush plains and rolling grasslands to mountain meadows and forests, south into subtropical environments of northern Mexico. Although not an indicator of a particular biotic habitat, the recovery of Ambystoma does indicate the presence of local and fairly permanent water nearby PSC. Anura Bufonidae Bufo woodhousii Girard, 1854 or B. punctatus Baird and Girard, 1852—Woodhouse’s toad or red-spotted toad Material.—ilium (2; NAU 8141-8142). Description.—Each of the two ilia has a prominent tuber superior (dorsal prominence) that is wide at the base, tall, and produces a knob over the acetabulum; there is no vexilla; the pars descendens is slightly flanged—all traits that are typical of Bufo. Rugosity of the knob is similar to that found in B. woodhousii and B. punctatus, and unlike the low, and sometimes less prominent tuber superior found in other species in the region. The fossils compare well with specimens having an estimated SVL of 52 mm. Because of the small to medium size of the ilia and the level of preservation, a more precise species assignment was not accomplished. The specimens from PSC did not appear to belong to B. mazatlanensis (Sinaloa toad), based on characters used by Van Devender et al. (1985). Discussion.—The following toads are known today from the region of PSC: B. woodhousii, B. cognatus, B. debilis, B. retiformis, B. alvarius, B. punctatus, and B. microscaphus. Van Devender et al. (1985) identified B. cf. alvarius , B. cf. cognatus, B. cf. kelloggi, B. mazat-

CL/CW = ratio

Prz + Pstz/ZL = ratio

3.7/1.7 = 2.2 3.6/1.7 = 2.1 4.4/1.9 = 2.3

— 3.6 + 3.3/4.9 = 1.4 3.8 + 4.1/5.5 = 1.4

1.8-2.3 1.9-2.2

1.3-1.7 1.3-1.6

lanensis, and B. punctatus/retiformis from Rancho la Brisca, Sonora. We are unclear about the characters used to assign the Rancho la Brisca elements to B. kelloggi, and how they compare with other small toads in the B. punctatus group. It is possible that if more Bufo specimens can be located in lowland, valley settings, more can be learned about these toads. It seems doubtful that cave settings such as PSC will contain these Mexican forms, even though they reportedly occurred only about 100 km to the south. Bufo woodhousii and B. punctatus frequent a wide variety of habitats ranging from desert streams and oases, open grasslands and scrublands, oak woodlands, to rocky canyons and arroyos, but they do not seem to venture much higher into the forest habitats than along the ecotone region of open woodlands/forests with adjacent grasslands. Whereas B. woodhousii can be found in more forested communities, B. punctatus is more adapted to desert, grassland, and woodland habitats. Bufo Laurenti, 1768, sp. indet.—toad Material.—ilium (3; NAU 8143-8144). Description.—The ilia show all the characters indicating the genus Bufo, as discussed above. One ilium from the Annex (NAU 8143) is from a toad with an estimated SVL of 45 mm. The other two ilia give an estimated SVL of 40 mm, indicating that all are from small and apparently adult individuals. They could represent B. kelloggi, B. debilis, B. retiformis, or B. punctatus. Without identification characters for B. kelloggi and better preserved specimens, species assignment was not accomplished. Hylidae Hyla Laurenti, 1768, sp. indet.—treefrog Material.—tibiafibula (18; NAU 8152-8154); ilium (10; NAU 8148, 8151, 8173); humerus

8

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

(NAU 8172); femur (10; NAU 8154). Description.—The ilia have a low tuber superior and no vexillum (Fig. 1F). The tuber superior is positioned partly over the acetabulum and partly onto the ala (shaft), as in Hyla, and not entirely over the acetabulum as in Pseudacris. The pars descendens is flanged onto the ala. None of the ilia from PSC have characters diagnostic for Pternohyla (Van Devender et al. 1985). Most of the ilia from PSC are from individuals estimated to have an estimated SVL from 20 to 40 mm. Based on the available comparative collections, we were not able to distinguish Hyla species to our satisfaction, therefore our specimens are referred only to the genus. The single humerus is from an adult individual with an estimated SVL about 35 mm. The lateral epicondyle is prominent but the diaphysis is delicate, as in Hyla and not as large and stout as in Pternohyla or in the larger subtropical treefrog Smilisca. A length of the humerus of 7.4 mm excludes Pseudacris. The tibiafibulae are long and slender, with a flexed middle diaphysis (versus those belonging to Bufo and Scaphiopus/Spea [Pelobatidae]). The middle diaphysis is straight on a long, slender bone in the leptodactylid Eleutherodactylus (barking frog; we follow Lynch, 1986, in using Eleutherodactylus rather than Hylactophryne). The small size (estimated SVL 40 mm) of the adult elements from PSC preclude these from being a small ranid, and are larger than those of the minute Pseudacris. Discussion.—Hylid frogs living in the region of PSC today include: Hyla eximia and H. arenicolor, with Pternohyla fodiens occurring south and slightly west, and Pseudacris triseriata occurring to the north in central Arizona. Hylids are rare in fossil sites of the Southwest. Van Devender et al. (1985) recorded Hyla arenicolor and Pternohyla fodiens from Rancho la Brisca. cf. Hyla sp.—treefrog Material.—ilium (NAU 8149); humerus (NAU 8150). Description.—Skeletal elements appear most like those in Hyla and differed from those found in Pternohyla and Pseudacris. The ilium appears to be from an individual with an estimated SVL of 22 mm. Preservation did not permit further identification. Hyla sp. or Pseudacris Fitzinger, 1843, sp. indet.—treefrog or chorus frog Material.—humerus (3; NAU 8147).

OCCASIONAL PAPERS

Description.—Two of the three humeri are complete (one is shown in Fig. 1C) and appear to be from adult individuals. Length of the two complete elements are 5.9 and 5.6 mm, similar to that of an adult Pseudacris with a humerus measuring 5.6 mm. Estimated SVL is about 30 mm for the individuals from PSC. The taxonomic assignment is based on the small adult size of the obviously hylid humeri. The elements appeared most similar to Pseudacris, although the size could also fit Hyla eximia. Microhylidae cf. Gastrophryne Fitzinger, 1843—narrowmouthed frog Material.—humerus (NAU 8169). Description.—The humerus is small (6.5 mm) and ossified, indicating an adult age (Fig. 1D). Beginning at midshaft, the crista ventralis rises slowly toward the proximal end, in contrast to the abrupt rise found in the humerus of Eleutherodactylus. The lateral epicondyle is poorly developed as compared to the prominent projection found on Pseudacris and Eleutherodactylus. The posterior surface opposite the ulnar condyle is roundish rather than flattened as in other small anurans (Van Devender et al. 1985). These characters imply that the single humerus is most likely from the narrow-mouthed frog. Discussion.—The distribution of Gastrophryne olivacea today extends northward from Mexico into a small section of southernmost Arizona that includes the region of PSC west to the Gran Desierto west of Ajo. Its main distribution is in Sonora, Sinaloa, and north into Texas, Oklahoma, and Nebraska. This species is widespread in grassland habitats in its northern distribution, grading south into the subtropical lowlands of Mexico. Narrow-mouthed frogs are secretive, hiding in damp burrows and crevices, under rocks, bark, and wood, in vicinities of streams, springs, and rain pools. In southern Arizona they can be found from mesquite grasslands up into the oak woodlands. Pelobatidae Scaphiopus Holbrook, 1842a, sp. indet. or Spea Cope, 1866, sp. indet.—spadefoot toad Material.—ilium (NAU 8145); fused sacrum and urostyle (NAU 8146). Description.—The urostyle, typical of pelobatids, is fused to the sacral vertebrae. Although incompletely preserved, the bone indicates an individual with an estimated SVL of 35 mm. The ilium

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

lacks the tuber superior, as is typical of pelobatids. The ilium appears to be from a juvenile with an estimated SVL of 22 mm. These young and incomplete specimens did not allow differentiation between Scaphiopus and Spea. Discussion.—Pelobatids (Scaphiopus couchi, Spea multiplicata, and Spea bombifrons) are fairly common today in the region of PSC. Over the last decade or so, there has been a trend to split Scaphiopus hammondi into two different species (now in the genus Spea): Spea hammondi, now restricted to California and Baja California, and Spea multiplicata in the Southwest. Previous works on the fossils of Arizona recorded Scaphiopus hammondi, but these should read Spea multiplicata under current usage (Mead et al. 1984; Van Devender et al. 1991). Scaphiopus couchi was reported at Deadman Cave (Mead et al. 1984) and Rancho la Brisca (Van Devender et al. 1985). Ranidae Rana Linneaus, 1758, sp. indet.—true frog Material.—ilium fragment (NAU 8164). Description.—The ilium fragment (Fig. 1E) preserves the remnants of the vexillum (dorsal flange, broken in this specimen) that is typical of all ranids and is not found on hylids, microhylids, leptodactylids, pelobatids, and bufonids. The specimen came from a frog with an estimated SVL of 75 mm, but was not complete enough to determine the species. Discussion.—Today the following ranids live in the vicinity of PSC: Rana catesbeiana (introduced), R. tarahumarae, R. pipiens, R. blairi, R. yavapaiensis, and R. chiricahuensis. Rana sp. was identified from Deadman Cave (Mead et al. 1984). Van Devender et al. (1985) identified skeletal elements of R. “pipiens” complex from Rancho la Brisca. No attempt was made to distinguish the Rancho la Brisca fossils from the Southwestern species in this complex (those leopard frogs mentioned above and R. berlandieri and R. magnaocularis). Reptilia Squamata Crotaphytidae Crotaphytus Holbrook, 1842b, sp. indet.—collared lizard Material.—dentary (2; NAU 8157-8158). Description.—Both dentaries are fragmentary, although preservation of diagnostic elements permits generic identification. The dentary is heavily

9

built and contains wide, tapering, blunt teeth. Anterior teeth are more conical, but not as slender or as recurved as in Gambelia. The dentary and teeth are larger, wider, and more robust than those on adult, robust Sceloporus magister (or other sceloporines). Both dentaries are from individuals with an estimated SVL of 95 to 100 mm (maximum for the species is about 137 mm). Discussion.—Recently McGuire (1996) provided a phylogenetic study of the crotaphytid lizards Crotaphytus and Gambelia. Nine living species were recognized in Crotaphytus in his analysis. Two species (C. dickersonae and C. insularis) are found today on islands in the Gulf of California, and are unlikely to have previously occurred as far north as PSC. Two species occur today in Coahuila, Mexico (C. antiquus), and southern Texas (C. reticulatus); although they are remote from the region of PSC, they could have been present in southeastern Arizona in the past. Crotaphytus vestigium is found from southern California into Baja California. There is a remote possibility it once occurred as far to the east as PSC. Four species (C. bicinctores, C. collaris, C. grismeri, and C. nebrius) are found today in Arizona and surrounding regions and are likely to have lived in the region of PSC in the past. At present, we feel that characters found on the dentary are not suitable for distinguishing between the nine species mentioned above, and that any of these could have been present at PSC in the past. Fossils from the Tucson region were originally described as C. collaris by Van Devender and Mead (1978). These were referred by McGuire (1996:91) to C. nebrius, who stated “Because the Tucson Mountains and Organ Pipe Cactus National Monument are currently inhabited by C. nebrius, this [fossil] material probably should be referred to C. nebrius on distribution grounds.” We disagree with the placement of the Tucson fossils in C. nebrius and believe that they and the remains from PSC are best left identified only to the generic level until more data can be generated. Crotaphytus species usually are rockdwellers that frequent canyons, mountain slopes, and boulder fields of alluvial fans, typically where vegetation is sparse, although they can be found in rocky and brushy riparian areas. cf. Crotaphytus sp.—collared lizard Material.—dentary (2; NAU 8155-8156). Description.—Both specimens are fragmented dentaries that are heavily built and contain teeth too wide and blunt for any sceloporine and do not appear to have been recurved as in Gambelia.

10

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

Phrynosomatidae Phrynosoma douglasi (Bell, 1833)—short-horned lizard Material.-dentary (2; NAU 8160, 8174). Description.—These dentaries are heavily built (compared to most other phrynosomatids) and contain the lateral expansion typical of Phrynosoma. One specimen (Fig. 1G) is from an individual with an estimated SVL of 75 mm. This size would omit the smaller P. modestum (SVL up to 69 mm). The teeth are well exposed on the dental shelf as on P. douglasi (spelling follows Hammerson and Smith 1991) and unlike those on P. solare and P. cornutum. The lateral expansion is simple as on P. douglasi, dissimilar to the rippled and ornate expansions on P. solare and P. mcalli. Although the dentary has the typical shape and construction of Phrynosoma species, it is not tightly curved as in P. platyhrinos. Too little is understood about the extinct P. josecitensis (known only by temporal spines from San Josecito Cave, Nuevo Leon, Mexico; Brattstrom 1955) to be able to consider this species among the remains from PSC. Mead et al. (1999) reviewed the characters found on cranial elements of the various species of Phrynosoma. Discussion.—The following horned lizards live today in the region of PSC: P. douglasi, P. modestum, P. solare, and P. cornutum; P. platyrhinos lives slightly to the west of the PSC region. Various species of Phrynosoma are fairly common in the Pleistocene of the Southwest. No species recovered as fossils in the Southwest have been found outside their modern geographic ranges. Today, P. douglasi typically lives in plant communities ranging from open juniper-piñon woodlands, forests, chaparral, desert grasslands, semiarid plains. Phrynosoma Wiegmann, 1828, sp. indet.— horned lizard Material.—frontal (NAU 8159). Description.—The specimen is the anterior half of a frontal bone. The shape of the articulation area with the nasals and surface texture indicates that this specimen represents Phrynosoma. The size indicates an estimated SVL of 95 mm, which is too large for P. modestumor P. mcalli. It seems to belong to one of the larger forms of horned lizard, although further identification is not possible without a more complete specimen. Sceloporus cf. S. magister Hallowell, 1854— desert spiny lizard Material.—maxilla (NAU 8161).

OCCASIONAL PAPERS

Description.—The maxilla is a heavy bone with short, wide, blunt teeth, similar to Sceloporus magister. The estimated SVL is 92 mm (the maximum SVL for the species is about 137 mm). Although other sceloporines grow as large or even larger than 92 mm SVL, (S. poinsetii, S. jarrovii, S. clarkii, S. orcutti, and S. olivaceus), none seem to have the wide, blunt, robust teeth found in S. magister. The large S. serrifer of southern Texas was not available for comparison. Although not a common fossil in Pleistocene deposits in southern Arizona, the desert spiny lizard is fairly abundant as a fossil in other localities in central and western Arizona, and in the lower Grand Canyon (Mead and Phillips 1981; Van Devender and Mead 1978; Van Devender et al. 1977). Discussion.—The following large forms of Sceloporus live today in the area of PSC: S. magister, S. clarkii , and S. jarrovii, with S. poinsettii just to the east into New Mexico and Mexico. S. serrifer and S. olivaceus live today well south and east of PSC in Mexico and into southern Texas. S. magister is found typically in desertscrub lands, up into mesquite-yucca grasslands, juniper woodlands, and subtropical thornscrub communities. “Large-sized” sceloporine lizard Material.—maxilla (NAU 8163). Description.—About 20% of the maxilla is preserved, precluding species identification. The teeth are conical, but not to the same extent as in Crotaphytus and Gambelia. The ascending process also is unlike that in Crotaphytus and Gambelia. It is most similar to the maxilla of a Sceloporus magister with SVL of about 115 to 120 mm. Teeth of S. olivaceus are typically closely spaced and narrow, unlike those of S. magister and the fossil. The fossil also is similar to the maxilla of S. poinsettii with a SVL of 115 mm. “Medium-sized” sceloporine lizard Material.—dentary (4; NAU 8162-8164, 8166); maxilla (NAU 8167). Description.—The seven dentaries and the single maxilla in this group all represent small to medium-sized lizards. Typically the Meckelian canal is open, with its orientation being ventrad at the anterior end. Dentaries and maxilla are not as delicate as those in Holbrookia, Cophosaurus, Uma, or Callisaurus , nor as minute as in Uta, Urosaurus, or the very small species of Sceloporus. Many of the fossils are the size of a sceloporine with an estimated SVL of 60 to 65 mm. Although they are similar to adult S. occidentalis

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

and S. undulatus, they also are similar to the younger (smaller) forms of S. magister and S. clarkii, in addition to S. jarrovii with a SVL of 65 to 75 mm (see also discussion of small species below). Discussion.—A number of medium-sized lizards were identified from Deadman Cave (Mead et al. 1984), the woodrat middens at Picacho Peak (Van Devender et al. 1991), and from Rancho la Brisca (Van Devender et al. 1985). “Small-sized” sceloporine lizard Material.—dentary (NAU 8165). Description.—The single, small left dentary has the Meckelian canal closed and fused medially at its midpoint (not as in Coleonyx and Xantusia), but is open anteriorly. The dentary wall is fairly tall and expands posteriorly, unlike that in Urosaurus. The dental shelf is not thin (due to Meckelian canal on ventral surface) as in Cophosaurus, Holbrookia, and Callisaurus . The dentary is most similar to that of Sceloporus scalaris and S. variegatus, but not like that of S. graciosus. Estimated SVL is 50 to 55 mm. Teeth are more slender than on a small S. jarrovii with a SVL of 49 mm. Without better material, there can be no specific identification. It seems most probable that the fossil belongs to either S. scalaris or S. variegatus , although it could also represent some other small species found today farther south and east in Mexico and southern Texas, such as S. merriami or S. grammicus. Discussion.—Sceloporus graciosus does not live in the region today, occurring no farther south than the northern half of Arizona. PSC is on the northwestern edge of the present distribution of S. scalaris. S. virgatus has a disjunct distribution today immediately south and east of PSC. The small species of Sceloporus that occurred during the Pleistocene in the Southwest are poorly understood. Serpentes The difficulties in identifying isolated skeletal elements (especially vertebrae) of snakes limits our ability to reliably identify snake fossils from PSC. In the almost complete absence of detailed studies of vertebral variation in snake species, the precise placement of most isolated fossil vertebrae into proper position in the vertebral series remains impossible. The recognition and definition of distinct vertebral regions by Hoffstetter and Gasc (1969) provided a useful first step toward resolution of the problem, but left an undifferentiated “trunk” region. A more detailed

11

scheme was presented by LaDuke (1991a), who recognized three subdivisions (used here) of the “trunk region”: anterior trunk vertebrae, midtrunk vertebrae, and posterior trunk vertebrae. The morphological similarities in vertebral form between many species of snakes (for example, within the natricine genus Thamnophis) present a serious problem for paleoherpetologists seeking to understand species-level faunal dynamics of the past. Without invoking modern geographic distribution of taxa as a criterion for at least limiting the species used for comparative purposes, we consider accurate diagnosis of isolated fossil vertebrae to species (and in many cases to genus) an unobtainable goal at present. Detailed studies of morphological and serial variation within extant snake species is, in our opinion, a necessary prerequisite to interpretation of Pleistocene snakes. LaDuke (1991b) provided a useful summary of the minimal literature on this subject and presented a detailed analysis of vertebral form in Thamnophis sirtalis sirtalis, but this remains the only species studied in such detail. Colubridae Material.— anterior trunk vertebra (4; NAU 9415, 9437, 9483, 9493); mid-trunk vertebra (7; NAU 9420, 9451, 9452, 9506, 9509, 9516, 9517); trunk vertebra (10; NAU 9411, 9418, 9430, 9450, 9487, 9500, 9504, 9505, 9512, 9518); vertebral fragment (5; NAU 9419, 9423, 9429, 9468, 9497); compound bone (2; NAU 9459); pterygoid (NAU 9515). Discussion.—Specimens referred to this family are recognized based on the characters outlined by Auffenberg (1963) and Holman (1981), but cannot reliably be placed within any given subfamily. Mid-trunk vertebrae referred to the Colubridae do not show the well developed ventral hypapophysis characteristic of vertebrae of the colubrid subfamily Natricinae and are not referable to that group. Other specimens are of indeterminate subfamilial status. At least 16 species of non-natricine colubrids live in the region surrounding PSC today. Natricinae Material.—mid-trunk vertebra (21; NAU 9412, 9413, 9416, 9417, 9425, 9426, 9427, 9454, 9455, 9460, 9461, 9462, 9464, 9467, 9470, 9473, 9480, 9484, 9486, 9494, 9498); posterior trunk vertebra (4; NAU 9446, 9469, 9477, 9514); trunk vertebra (4; NAU 9434, 9438, 9466, 9472). Discussion.—Vertebrae from the anterior, mid, and posterior trunk regions of natricine colu-

12

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

brids, elapids, and viperids all show well developed ventral hypapophyses. In the Natricinae and Elapidae these structures are much thinner and less robust than the corresponding structures in the Viperidae. The PSC specimens also have relatively high neural spines and are anteroposteriorly more shortened than elapid vertebrae. Natricines living in the vicinity of PSC today are members of the genus Thamnophis, a widely distributed genus with numerous species throughout the United States and Mexico (Stebbins 1985). Viperidae Material.—mid-trunk vertebra (6; NAU 9428, 9449, 9463, 9478, 9491, 9503); posterior trunk vertebra (NAU 9488); trunk vertebra (2; NAU 9441, 9502); juvenile trunk vertebra (NAU 9458); vertebral fragment (2; NAU 9414, 9422). Discussion.—These specimens have very robust, relatively straight and thick ventral hypapophyses and are relatively shortened anteroposteriorly, with relatively high neural spines. Isolated vertebrae of viperids are difficult to identify to genus. The genus Agkistrodon is not found in the region today, but at least seven species of the rattlesnake genus Crotalus (some restricted at present to the mountains) and one species in the genus Sistrurus are. A small dorsal process or spine immediately anterior to the neural spine in Sistrurus vertebrae was used by Auffenberg (1963) and Holman (1981) to separate Crotalus vertebrae from those of Sistrurus. None of the PSC specimens show the spine said to be characteristic of Sistrurus. It is most likely that these specimens belong to the genus Crotalus, but the possibility that they belong to Agkistrodon cannot be excluded. Aves Numerous fragmentary bones of birds were recovered during the screening of sediments. These include many postcranial elements belonging to small passeriforms, but no attempt was made to identify most of them even to family. However, two species of Passeriformes were identified because one was relatively large and one was represented by a complete bone. Skinner (1942) identified only two bird bones, a furcula and a fragment of a tibiotarsus, both of which were referred to the Threskiornithoidae (sic) genus indet. The only other bird currently recorded from PSC is the extinct turkey, Meleagris crassipes (Rea 1980; Steadman 1980), which was found in the Annex.

OCCASIONAL PAPERS

Galliformes Phasianidae Meleagridinae Meleagris cf. M. crassipes Linnaeus, 1758— extinct turkey Material.—phalanx 1 of digit II of left pes (51942). Discussion.—Measurements of the specimen are greatest length, 28.0 mm; distal breadth, 6.3 mm; proximal breadth (damaged), ca. 8.5 mm; anteroposterior diameter of shaft at midpoint, 5.5 mm. The specimen compares favorably with the same bone in Recent skeletons of Meleagris gallopavo but its proportions are different. Comparable phalanges of the extinct species M. crassipes were unavailable, and an isolated phalanx is insufficient for species determination. Nevertheless, referral of this specimen to M. cf. M. crassipes is based on the previously published record of the species from PSC and “North Papago Cave” (Rea 1980; Steadman 1980). Rea (1980) summarized the late Quaternary records of turkeys in the Southwest. Meleagris crassipes was widespread there in the late Quaternary, whereas M. gallopavo is known during the late Pleistocene only from Arizpe and Rancho la Brisca, Sonora, and Conkling Cavern, New Mexico (Rea 1980; Van Devender et al. 1985). Odontophorinae cf. Cyrtonyx montezumae (Vigors, 1830)— Montezuma quail Material.—distal fragment of right tibiotarsus (51944); proximal (scapuloclavicular) end of left coracoid (51943); distal (articular) end of left mandibular ramus (51945). Discussion.—The PSC specimens were compared with museum specimens representing Lophortyx gambelii, Callipepla squamata, Colinus virginianus, Philortyx fasciatus, and Cyrtonyx montezumae. Although the specimens are fragmentary, they are equal to or larger than most available comparative specimens. In the tibiotarsus fragment, the curved dorsal border of the supratendinal bridge passes straight (not obliquely) across the bone, as in Callipepla and Cyrtonyx . In the coracoid fragment, the apex of the procoracoid process is rounded; the furcular facet is indistinct or absent as in Cyrtonyx and unlike the other genera. These characteristics agree with those of Cyrtonyx (Holman 1961). Additional or more complete skeletal elements are desirable for a precise identification, there-

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

fore we tentatively identify them as cf. C. montezumae. Cyrtonyx montezumae is relatively common in oak-grassland in southeastern Arizona today, and we have observed them in the immediate vicinity of PSC (near Entrance B). The species is known from the Pleistocene deposits at San Josecito Cave, Nuevo León (Miller 1943; Steadman et al. 1994), Deadman Cave, Arizona (Mead et al. 1984), and Room of the Vanishing Floor (Dry Cave), New Mexico (Harris 1993). Passeriformes Corvidae Aphelocoma ultramarina (Bonaparte, 1825)— Gray-breasted Jay Material.—distal right humerus (51947). Discussion.—Distal breadth of the bone is 7.9 mm. The specimen is slightly smaller than humeri of Cyanocitta stelleri and Gymnorhinus cyanocephalus, and it is slightly larger than in Aphelocoma coerulescens. It matches the size of A. ultramarina. This species is common today in the vicinity of PSC as well as in other oak woodlands of southeastern Arizona. Troglodytidae Salpinctes obsoletus (Say, 1823)—Rock Wren Material.—virtually complete right humerus (51946). Discussion.—Greatest length of the bone is 17.4 mm; distal width is 4.0 mm. The specimen is larger than in Catherpes mexicanus and much larger than in other local modern wren species. Rock wrens often enter and nest in caves, so the presence of this species is no surprise. This wren has been recorded as a fossil in the Quaternary deposits in U-Bar Cave (Harris 1993) and Shelter Cave, New Mexico, and Jaguar Cave, Idaho (Brodkorb 1978), Crystal Ball Cave, Utah (Emslie and Heaton 1987), and Little Box Elder Cave, Wyoming (Emslie 1985). This is the first record of the species in the late Pleistocene in Arizona. Mammalia Insectivora Soricidae Sorex arizonae Diersing and Hoffmeister, 1977— Arizona shrew Material.—anterior of skull plus articulated mandible (Fig. 2; 51852); posterior portion of right dentary (51853); left dentary with last unicuspid and m1-m3 (51854); posterior portion of left dentary with m3 (52732). Description.—When the skull was found, the lower jaws were articulated with the cranium and

13

Figure 2. Rostrum and jaws of Sorex arizonae (OMNH 51852) from Pleistocene deposits in Papago Springs Cave, Arizona. A, palatal view. B, right side of rostrum. C, medial view of left dentary. D, lateral view of left dentary. E, medial view of ascending ramus of right dentary, further enlarged to show mandibular foramen and postmandibular foramen. F, medial view of ascending ramus of left dentary, showing mandibular and postmandibular foramina. Hatching indicates breakage, uniform open stipple indicates carbonate encrustation.

encrusted with carbonate. The specimen was cleaned with acetic acid and the jaws were disarticulated in order to examine the dentition and other characteristics in more detail. During cleaning, the left first upper unicuspid and the tip of the right i1 were lost. Measurements of the skull and teeth are given in Table 3. The teeth are weakly to moderately pigmented. On each of the dentaries articulated with the skull, a postmandibular foramen is present and is separate from the mandibular foramen (Fig. 2 E, F); the right postmandibular foramen is large and well developed but the left one is very small. The upper unicuspids have weak ridges but the ridges are not pigmented. The third upper unicuspid is slightly larger than the fourth. The upper incisors are broken away at the bases, so it is not possible to describe or measure the tines or tips. Breadth of the palate measured across P4s is less than breadth across M3s. Breadth across the upper fourth unicuspids is narrow as in S. arizonae and narrower than in S. merriami. The ratio of preparacrista length: postmetacrista length in M2 is greater than in S. merriami.

14

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

Table 3. Measurements (in mm; following Diersing and Hoffmeister 1977; landmark letters also are repeated here), of the fossil skull (OMNH 51852) of Sorex arizonae from Papago Springs Cave, Arizona. Due to breakage on the left upper tooth row, those measurements marked with an asterisk (*) were made by doubling the measurement from the midline palatal suture to the right side of the skull or teeth. Measurement Maxillary breadth (C-C') Breadth across upper tooth rows (D-D') "Tooth row length" (A-E') Complex tooth row length (E-E') Unicuspid tooth row length (A-F) Zygomatic plate breadth Breadth across first unicuspids (G-G') Breadth across second unicuspids (H-H') Breadth across third unicuspids (I-I') Breadth across fourth unicuspids (J-J') Breadth across fifth unicuspids (K-K') Breadth across P4s (L-L') Breadth across M3s (M-M') Lateral breadth of unicuspid 1 (G'-G") Lateral breadth of unicuspid 2 (H'-H") Lateral breadth of unicuspid 3 (I'-I") Lateral breadth of unicuspid 4 (J'-J") Lateral breadth of unicuspid 5 (K'-K") Height of unicuspid 1 (N-N') Height of unicuspid 2 (O-O') Height of unicuspid 3 (P-P') Height of unicuspid 4 (Q-Q') M2 paracone-parastyle length (D-S') M2 metacone-metastyle length (T-T')

5.28* 4.98* 5.65 4.27 1.87 1.18 1.90 2.02 2.17 2.35 2.42 4.24* 4.20 0.61 0.60 0.57 0.50 0.40 0.67 0.70 0.53 0.45 0.64 0.80

The fragment of a right lower jaw bears a postmandibular foramen as a separate small, downward-directed opening that connects interiorly with the internal temporalis fossa. It also has a normal, larger mandibular foramen that opens posteriorly. In the isolated left lower jaw from Unit 5, all teeth are darkly pigmented. The mandibular foramen and postmandibular foramen are completely merged into one opening, oval in shape with a broad connection interiorly into the internal temporalis fossa. The resulting oval foramen opens downward and posteriorly. Teeth in the left lower jaw fragment from Unit 7 are moderately pigmented. The postmandibular foramen and mandibular foramen are connected by a broad depression; the postmandibular foramen is very large, much larger than the mandibular foramen. Discussion.—All four specimens seem to be referable to Sorex arizonae. Diersing and Hoffmeister (1977) diagnosed S. arizona e as having a postmandibular foramen, upper unicuspids

OCCASIONAL PAPERS

without pigmented lingual ridges, third upper unicuspid larger than fourth, tines always present on medial sides of I1, zygomatic plate narrow, and length of skull small (other external characteristics also were given). Where the appropriate skull parts are preserved, the Papago Springs skull matches their diagnosis well. The long-tailed shrews that are the most similar in these morphological features are members of the S. trowbridgii species group (George 1988), S. trowbridgii, S. merriami, and S. arizonae (and possibly S. emarginatus and other southern Mexican species). Diersing and Hoffmeister (1977) nominated S. arizonae and compared and contrasted it with numerous species of Sorex, but they did not include S. trowbridgii. Our comparisons were limited to members of the S. trowbridgii species group because of the close match of the PSC skull with the diagnostic characteristics of S. arizonae. The skull differs from S. trowbridgii , as described by Junge and Hoffmann (1981), in having upper unicuspid 3 larger than unicuspid 4, rather than the reverse. In the left dentary fragment from PSC, the confluent condition of the postmandibular foramen and mandibular foramen was observed in five mandibles of S. trowbridgii (from California and Oregon); however, the postmandibular foramen was completely lacking in seven others (from Oregon). Thus, this feature seems to be variable and may be equivocal for identification. In fact, the variability exists even between some members of the different subgenera it was once thought to differentiate (Carraway 1987; George 1988; Junge and Hoffmann 1981), and in the present case, it varies from side to side within a single skull. Further analysis of this feature is required, possibly using specimens identified by molecular or other independent means. According to Diersing and Hoffmeister (1977:329-330), S. arizonae differs from S. merriami in the following ways: “breadth of palate measured across P4 less than breadth measured across M3, whereas in S. merriami the reverse is true; palate narrower, especially as reflected by breadth across unicuspids 4;...unicuspid 5 slightly larger; unicuspid 3 equal to or slightly larger than unicuspid 4...; upper molar with cusps not wshaped as in merriami, but with metacone-metastyle [=postmetacrista] length noticeably greater than paracone-parastyle [=preparacrista] length.” Other characteristics also were given that pertain to body parts not preserved in the PSC fossils.

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

Relative to the characteristics preserved, the Papago Springs skull more closely matches S. arizonae, although many of the differences between it and S. merriami are slight. Our measurements of the preparacrista length and postmetacrista length are significantly different from all of the same measurements reported for numerous specimens of Sorex spp. by Diersing and Hoffmeister (1977). We cannot explain this apparent difference in absolute measurements, but note that the proportion of postmetacrista length is noticeably greater than the preparacrista length, as they reported for S. arizonae. Among the species they examined, Diersing and Hoffmeister (1977) considered S. arizonae to differ the least from S. emarginatus. Sorex emarginatus is a poorly known species of the southern Sierra Madre Occidental in southwestern Mexico. Because of problems in the identity of Middle American shrews, they contrasted S. arizonae only with the type specimen of S. emarginatus, as follows: “skull slightly shorter; toothrow slightly shorter; palate between unicuspids noticeably narrower; unicuspids higher (longer)” (Diersing and Hoffmeister 1977:330) No specimens of S. emarginatus were available for comparison in the present study, but detailed measurements of the type specimen (or, preferably, a large population sample; not currently available in museums) are needed for a valid comparison. Although more detailed comparisons of the PSC fossils with S. emarginatus and other related shrews could be instructive and interesting with respect to the historical biogeography of Sorex (see George 1988), such studies await more and better fossils and a thorough revision, using eclectic data sources, of the genus throughout North America. For purposes of the present paper, the PSC specimens morphologically best fit S. arizonae. As such, they represent the first fossil record for S. arizonae and are slightly extralimital to the modern range of the species, which includes the nearby Huachuca and Santa Rita Mountains (and other northern outliers of the Sierra Madre Occidental; Caire et al. 1978; Conway and Schmitt 1978; Hoffmeister 1986), but not the Canelo Hills. Although the species exists today in the Animas Mountains of New Mexico (Cook 1986), it has not yet been identified among Quaternary fossils from U-Bar Cave, New Mexico (Harris, A. H., pers. comm.), which is only 8 m higher in elevation than PSC. However, S. merriami is relatively common in U-Bar Cave deposits (Harris 1987).

15

Notiosorex crawfordi (Coues, 1877)— desert shrew Material.—associated left and right rostral halves (51486); left dentary with m2-m3 (51485); left dentary fragment with lower molar (51488); left dentary fragment with m2 (53069); right dentary with partial m2 (51483) edentulous right dentary (51484); right dentary fragment with m1 and fragment of m2 (53070); right dentary fragment with m2-m3 (53220); right i1 (53221); left I1 (53228); left M1 (53427); right m1 or m2 (52743); left femur (51482, 51487). Discussion.—Teeth of these specimens lack pigment, and the dentaries have a small mandibular foramen but they lack a postmandibular foramen. These characteristics, as well as the rostra with only three unicuspids, allow the identification of the specimens as N. crawfordi. Chiroptera Vespertilionidae Myotis velifer (J. A. Allen, 1890)—cave myotis Material.—left maxilla fragment with P4-M2; left premaxilla-maxilla fragment with P3-P4 (52761); right maxilla with P2-M3 (52767); left half-rostrum with P2-M3 (52768); right dentary with m2m3 (2; 52047, 52059); left dentary with m1-m3 (52056); right dentary with m1-m3 (52057); anterior right dentary with i2-p4 (52058); left dentary with c1-m3 (52066); right dentary with p4, m2m3 (52080); right dentary with m2-m3 (2; 52081, 52082); right dentary with i3-p4 (52083); left dentary fragment with p4-m2 (52084); left dentary with c1-p4 (52762); left dentary with i1-i2, c1-m3 (52763); right dentary fragment with c1-p4 (52764); right dentary with p3, m1-m3 (52765); right dentary with c1, p3-m3 (52766); left humerus proximal fragment (52048); right humerus distal fragment (2; 52049, 52055); left humerus distal fragment (52067); right radius (51947). Discussion.—The sizes of the bones and, especially, the robustness of the bones and teeth indicate that these specimens belong to M. velifer. The species is common in the cave today. Several bones and mummies were picked up from the floor of the Main Room and individuals were observed on numerous occasions roosting in the Main Room as well as in the more remote, wetter portions of the cave beyond the “squeeze” passage. Myotis thysanodes Miller, 1897—fringed myotis Material.—anterior portion of skull (52068); rostrum with right P4-M3 (52069); left half of ros-

16

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

Table 4. Measurements (in mm) of left jaw (OMNH 51480) of a subadult Ursus cf. U. americanus from Papago Springs Cave. Measurement Distance from posterior edge of condyle to posterior edge of c1 alveolus Dentary depth at postsymphysis Diastema, c1-p4 Alveolar length of p4-m3 Length of p4 Width of p4

OCCASIONAL PAPERS

Table 5. Measurements (in mm) of Urocyon cinereoargenteus (OMNH 51481) from Papago Springs Cave. Skull measurements following those recorded by Hoffmeister (1986) and Martin (1974). Measurement

139.5 32.0 26.4 70.0 9.8 5.8

trum plus dentary fragment with p4-m1 (52061); rostrum with left P2-M2 (52060); left half of rostrum (52052); right dentary fragment with p4-m1 (52070); right dentary with c1-p2 and m1-m2 (52079); left dentary fragment with m3 (52071); left dentary with m2-m3 (52053); left dentary (52050). Discussion.—The size of these specimens is generally similar to or slightly smaller than in M. velifer, but the teeth are much less robust. In those cranial fragments preserving the forehead profile, the profile rises as in M. thysanodes, more abruptly than in M. velifer. Myotis sp. —small mouse-eared bat Material.—right dentary with i1-i2 and m2-m3 (52064); complete left humerus (52496); distal humerus (52051). Discussion.—In addition to this material, a few other rami and fragments probably also represent this small Myotis. Measurements (in mm) of the distal humerus fragment are distal width, 2.32, anteroposterior shaft diameter, 0.87. Measurements (in mm) of the complete humerus are length 20.47, proximal width 2.47, distal width 2.28, midshaft diameter 0.88. Alveolar length of the mandibular tooth row is 5.85 mm. The species represented by these specimens is likely to be M. californicus, M. ciliolabrum, or M. yumanensis based on size and modern geographic distributions, although an extinct species might be represented. Any of these three species would be new to the Pleistocene fauna of PSC, but accurate identification must await recovery of better material. All three species are known from the Recent fauna of southeastern Arizona. Corynorhinus townsendii (Cooper, 1837)— Townsend’s big-eared bat Material.—left dentary with m1-m3 (52072); left dentary with m2 (52073). Discussion.—We use the genus Corynorhinus

Condylobasal length Greatest width of braincase Greatest zygomatic width Least width of rostrum Least interorbital width Greatest width between sagittal crests (measured from the outside of each) Greatest length from inion to horizontal line between postorbital processes Greatest width across postorbital processes Maxillary tooth row length Length of P4 Width of P4 Alveolar length of upper postcanine teeth Alveolar length of lower postcanine teeth

115.6 46.7 64.9 17.8 23.9 26.0 55.6 33.3 51.4 9.9 6.3 42.5 46.8

instead of Plecotus following recent studies by Menú (1987), Bogdanowicz et al. (1998), Frost and Timm (1992), and Tumlison and Douglas (1992). The specimens can be distinguished from the small species of Myotis based on the alveolus for the p4 that reflects a single-rooted tooth rather than the double-rooted type present in Myotis. Townsend’s big-eared bats still roost in PSC today. Carnivora Ursidae Ursus cf. U. americanus Pallas, 1780—black bear Material.—left dentary (51480); left m2 (53206). Discussion.—The dentary is from a juvenile; its measurements are provided in Table 4. The permanent canine was erupting at the time of death and the only other tooth preserved is the p4. Shallow, empty alveoli are present for the molars. Measurements of the m2 are anteroposterior length, 23.5 mm; anterior transverse width, 13.7 mm; posterior transverse width, 15.1 mm. Based on specimens from PSC, Skinner (1942) originated the subspecies Ursus americanus gentryi. The type specimen was a virtually complete skull. The subspecies was characterized as having a skull the size of that of U. a. americanus but with an exceptionally wide palate and rostrum, a relatively long C1-P4 diastema, and an extremely heavy pterygoid region, among other characteristics. Skinner referred a left mandibular ramus with three teeth to this subspecies, noting the exceptional depth of the dentary below the diastema. The p4 in the dentary is similar in size

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

to that of other fossil black bears, but the m2 is larger than m2s of fossil black bears given by Graham (1991:table 5). Unfortunately, because of their incompleteness our specimens provide little new information about U. a. gentryi. Canidae Urocyon cinereoargenteus (Schreber, 1775)— gray fox Material.—partial skeleton including the skull (complete except for broken left zygomatic arch, part of left maxilla and pterygoid/palatine area; teeth preserved are right C1, P2, P4, and both M1-M2), both dentaries (right with c1-p3; left with p1-p4), left femur and tibia, several foot bones and a few vertebrae (51481); left dentary with p2-m2 (53440). Discussion.—No significant qualitative differences were detected between these fossils and modern U. c. scotti skulls. In most measurements (Table 5) the skull is within the range of Recent Arizona U. c. scotti (and within all the ranges given by Martin 1974), approximating the measurements of larger females or small to average males given by Hoffmeister (1986). The only appreciably different measurement is that of P4 width, which is much greater in the PSC fossil than in any modern skull examined (females, x = 5.01 mm, range 4.2-5.3 mm; males, x = 5.11 mm, range 4.8-5.4 mm). Upper molars are slightly more robust in the development of the lingual portions than in Recent specimens, but overall dimensions are similar. Skinner (1942) listed three specimens (a partial skull lacking rostrum and most teeth, three partial rami lacking teeth, and miscellaneous isolated limbs) without discussion. There is no reason to differentiate these nor our new material from the living species. Harris (1985b) noted the presence of this fox in middle Wisconsinan deposits in U-Bar Cave in contrast to its absence in fullglacial faunas elsewhere in New Mexico. Mustelidae Mustela cf. M. frenata Lichtenstein, 1831— long-tailed weasel Material.—premaxillae with left I1-I3 (51494). Discussion.—The specimen represents a small carnivore with incisors much smaller than Spilogale gracilis and equivalent in size to M. frenata. A small portion of the palate is preserved that includes the anterior palatine foramina. Only the posterior borders of alveoli of the right I1-I3 are preserved, but the left incisors are all in place. Incisor morphology resembles that of modern M. frenata from Arizona.

17

Long-tailed weasels are uncommon today in Arizona, where they usually occur in cool mountainous areas (Hoffmeister 1986). Rarer occurrences in arid areas are known but these are close to water. The modern geographic range of M. frenata is limited by the availability of water during the summer months (Svendsen 1982). Allen (1895: 255-256) mentioned collecting a longtailed weasel at 9000 feet elevation (2740 m) in the Huachuca Mountains in 1893, but no other modern records exist for these mountains or the Canelo Hills. This is a new species for the fauna of PSC and is apparently the first Pleistocene record for the species in Arizona. Many Pleistocene records exist elsewhere in North America, including some in New Mexico, California, and Texas (Anderson 1984). Mephitis Geoffroy Saint-Hilaire and Cuvier, 1795, sp. indet.—skunk Material.—left P4 fragment and M1, probably associated (53207, 53208). An isolated right I3 (53076) from the same provenience as the maxillary teeth is identifiable only as Mephitinae and may also represent Mephitis. Discussion.—The common skunk in the vicinity of PSC today seems to be the hooded skunk, Mephitis macroura, based on specimens seen in Cienega Creek wash near PSC and in Vaughn Canyon a few km to the east, although M. mephitis might also occur in the area. It is difficult to differentiate these two skunks osteologically (although Hoffmeister [1986] noted slight differences in the relative inflation of the auditory bullae). It is not possible to identify the PSC skunk with the material at hand. Skinner (1942) identified several specimens from PSC, including the posterior portion of a skull (F:AM 42844), as Mephitis occidentalis (= M. mephitis). This cranium should be reexamined to determine whether it might represent M. macroura. The latter species is known as a fossil only in Deadman Cave (Mead et al. 1984). Perissodactyla Equidae Equus conversidens Owen, 1869—extinct horse Material.—fragment of upper cheek tooth (51472); incisor fragment (53209); scapula fragment (2; 51479, 53210); left cuboid (51477); left unciform (51478); right scaphoid (51471); metapodial II or IV (51469); phalanx 1 (3; 51465, 51473, 51474); phalanx 2 (51468); phalanx 3 (2; 51466, 51467).

18

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

Discussion.—Skinner (1942) assigned the majority of horse specimens he collected from PSC to Equus conversidens; he referred a single slender first phalanx to E. tau. Later, Skinner (1972) changed his mind and believed the postcranial elements he had earlier assigned to E. conversidens belonged to another form of Equus for which no teeth were discovered. Harris and Porter (1980) provided detailed measurements of various postcranial elements and teeth of horses from Dry Cave, New Mexico; based on Skinner’s (1942) measurements, they believed that Skinner’s larger horse did match E. conversidens. Although we recovered few new specimens of horse bones, the specimens probably are assignable to E. conversidens. New PSC specimens (Table 6) are within or slightly larger than the range of measurements given for elements of E. conversidens. We did not find any fossils of small horses such as E. tau. Artiodactyla Bovidae cf. Bison (Hamilton-Smith, 1827)—bison Material.—left p2 (51843). Discussion.—The size and shape of this tooth are similar to the same tooth in Bison bison. Skinner (1942) listed Bison taylori (=Bison antiquus) from the cave. Antilocapridae Stockoceros onusrosagris (Roosevelt and Burden, 1934)—Quentin’s pronghorn Material.—braincase with horncores (51464); cranium (52719); horn core bases (51443); horn core tip (52725); left maxilla fragment with P2-P4 and M3 (51439); right dP2; right dp2 (51416); right dp3; left dp3; right P2 (51423); right P2 (51445); left P3 (52726); right P3 (52727); right P4 (3; 51432, 51408); right p2 (51415); left p3 (51419); left M1 (51460); right m3 (51436); cheek tooth fragments (5; 51409-13); axis (51451); thoracic vertebra (2; 51459, 52730); lumbar vertebra (2; 51438, 51422); neural spine fragment (51420); caudal vertebra (51418); right rib (52219); rib head (3; 51405-6, 51457); sternal rib fragment (3; 51447-9); distal humerus (52729); left scaphoid (2; 51440, 52728); left lunar (51441); right lunar (51444); left unciform (51450); distal fragment of metacarpal (51463); proximal fragment of right tibia (51453); proximal fragment of left tibia (51454); distal tibia fragment (52810); malleolus (51414); right metatarsal (51458); left proximal metatarsal (51452); distal fragment of metapodial (2; 51433); sesamoid (3; 51407, 51421, 51446);

OCCASIONAL PAPERS

phalanx 1 (5; 51403, 51422, 52720-2); fragment of phalanx 1 (52723); phalanx 2 (52724); phalanx 2 distal fragment (51402); ungual phalanx (51404); phalanx fragment (2; 51455-6). Numerous bones and fragments of bones of S. onusrosagris can still be found throughout the Main Room of the cave encased in flowstone in the walls and in floor sediments. Specimens taken from stratigraphic context within the measured sections are listed above. Additional specimens from other parts of the cave include the following: right p4 (51431), incisor (51417), incisor or canine fragment (51424), caudal vertebra (51430), cheek tooth fragments (3; 51426-8), and phalanx 2 (2; 51425, 51429) from a small dome in the roof of the passage just beyond Sec. 2; lumbar vertebra (51462), right calcaneus (51461), from the boulder breakdown of Entrance B in the crawlway just outside the Main Room; frontal fragment with base of horn core (2; 51434-5), distal fragment of horn core (51437), left dentary fragment with p4-m1 (52731), surface pickup with no provenience. Discussion.—Harris (1993) recorded this species in the mid-Wisconsinan of U-Bar Cave, New Mexico. Rodentia Sciuridae Tamias Illiger, 1811 sp. indet.—chipmunk Material.—right dentary with i1 (51844). Discussion.—Measurements of the specimen are as follows: lower incisor width, 0.90 mm; length of mandibular alveolar tooth row, 5.62 mm. Chipmunks do not occur in or near the Huachuca Mountains nor Canelo Hills today. The nearest extant populations to PSC are of T. dorsalis, the cliff chipmunk, in the Santa Catalina and Rincon mountains to the north and in the Chiricahua Mountains and Sierra El Tigre, Sonora, to the east. A relict population of T. dorsalis also occurs along coastal Sonora, to the south of the Huachuca Mountains. The specimen we recovered from PSC is not diagnostic to species. Skinner (1942) listed two rami as “E[utamias]. (?) dorsalis”), a species that is found today in a wide variety of habitats over a wide altitudinal range (from 61 m in Sonora [Caire 1978] to 2865 m in Arizona), where large rocks or cliffs exist (Hoffmeister 1986). However, Skinner’s specimens should be reexamined to confirm or deny the species identification. Based on studies of the distributions of extant species and populations of chipmunks in some

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

mountains of the western United States, these rodents have had a dynamic zoogeographical history there (Brown 1978; Findley 1969, 1996; Patterson 1980, 1981, 1984; Sullivan 1985). Although extant populations in southern Arizona have not been studied in this way, radiometrically dated Quaternary fossils have the potential to contribute empirical data toward such studies. However, more and better fossils than are presently available in PSC will be necessary in order to do this. Marmota flaviventris (Audubon and Bachman, 1841)—yellow-bellied marmot Material.—occipital condyle (51499); left maxilla fragment with P4-M2 (53246); left premaxilla with I1 (53457); incisor fragment (2); left P3 (53249); left P4 (53450); right P4 (53250); right dentary with i1, unerupted p4, and m3 (53461); right dentary with i1-m3 (2; 53245, 53455); right i1 (53248); right i1 tip (53451); right m1 or m2 (2; 53252, 53253); left m1 (53251); right scapula (juvenile; 53456); head of right scapula (53460); left scapula (53454); left proximal humerus (53463); proximal radius (53452); right ilium (53462); diaphysis of left femur (53459); proximal diaphysis of right tibia (53247); right distal tibia (53458); left calcaneus (53453). Discussion.—Although Skinner (1942) found remains of this species to be abundant in PSC deposits, only three fragments occurred in our limited excavations in the Main Room. However, the remains of marmots are abundant in undated cave fill in the Annex. Spermophilus cf. S. spilosoma Bennett, 1833— ground squirrel Material.—left P4 (51847); abraded right upper cheek tooth (51848). Discussion.—Although the material is sparse, these elements clearly indicate the presence of a second species of Spermophilus in the PSC fauna. Measurements (in mm) are P4 anteroposterior length, 1.50, transverse width, 1.92; upper cheek tooth anteroposterior length, 1.61, transverse width, 1.94. The teeth are larger than those of the PSC chipmunk and smaller than those of the rock squirrel. This is the first time S. cf. S. spilosoma has been recorded in PSC. The species is known as a fossil in the Isleta Caves, New Mexico (Harris and Findley 1964), Slaton local fauna, Texas (Dalquest 1967), Easly Ranch, Texas (Dalquest and Schultz 1992), Lower Sloth Cave, Texas (Logan 1983), Jimenez Cave, Chihuahua (Messing 1986), and San Josecito Cave, Nuevo León

19

(Jakway 1958). In Arizona, S. spilosoma utilizes a variety of habitats from desert to mountain meadow conditions (Hoffmeister 1986), and in southeastern Arizona, it tends to be associated with mesquite and acacia. Spermophilus variegatus (Erxleben, 1777)— rock squirrel Material.—broken left M1 or M2 (51846); left dentary with i1, right dp4, unerupted crown of right m1, and left P4; right p4 (51845). Discussion.—Measurements (in mm) of the specimens are as follows: i1 width 2.0; depth of dentary at diastema (est.) 6.7; dp4 anteroposterior length 2.30, transverse width 2.17; p4 anteroposterior length 2.37, transverse width 2.55; m1 anteroposterior length 2.45, transverse width 3.20; P4 anteroposterior length 2.27, transverse width 2.90. The depth of the dentary at the diastema is much shallower than in Sciurus (see below). Dental morphology is that of the rock squirrel and not that of a tree squirrel. Rock squirrels occur today on the hill in which PSC occurs and occasionally enter the cave. Sciurus cf. S. aberti Woodhouse, 1853— tree squirrel Material.—right maxilla with M1-M3 (51850); upper incisor (51851); right M3 (53242); right m3 in a fragment of dentary bone (51849). Discussion.—In the maxilla fragment only the molars remain in place, but the alveolus for P4 and a portion of the posterior wall of the alveolus for P3 are clearly preserved (Fig. 3A). The M1 and M2 are low-crowned and have wide protocones. The metaloph is complete on M1 and M2 but absent in M3. Weak mesostyles are present in all three upper molars. M3 is about the same size as M2; it has a smooth-bottomed posterior basin with a rounded posterior border formed by the posterior cingulum. In the lower jaw fragment, the right m3 is lowcrowned with a simple occlusal pattern as seen in tree squirrels in general (Fig. 3B). The posterolophid is low and rounded posteriorly. The entoconid is indistinct. The broad talonid basin is smooth and featureless. For the maxilla, the alveolar length of P4-M3 measures 12.15 mm; the anteroposterior length of M1-M3 is 8.82 mm. Additional measurements are given in Table 7. The presence of two upper premolars is important for the identification of this squirrel. In the portion of southwestern North America centered around PSC, four species of Sciurus occur in the modern fauna. These are Sciurus arizonen-

20

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

OCCASIONAL PAPERS

Table 6. Measurements (in mm) of Equus conversidens from Papago Springs Cave; measurement numbers correspond to the definitions of Harris and Porter (1980). Element Scapular head (OMNH 51479)

Phalanx 1

Phalanx 2 (OMNH 51468)

Phalanx 3 (OMNH 51466)

Measurement no. 2 3 OMNH 51474 1 2 3 4 6 7 8 9 10 11 12 13 1 2 3 4 5 6 1 2 3 4 5 6 7 8 9 10 11 12

sis, S. nayaritensis, S. aberti, and S. colliaei. Of these, S. arizonensis and S. nayaritensis do not normally possess two upper premolars (although Hoffmeister and Goodpaster [1954:93] noted one out of 15 specimens of S. arizonensis huachuca that possessed “a sliver of a tooth in front of each fourth upper premolar”; this may have represented a vestige of the P3). The other two species, S. aberti and S. colliaei, possess two upper premolars as adults, but P3 erupts rather late and it may not appear in subadult animals. Thus, the dental formula of the PSC fossil agrees with that of adult S. aberti and S. colliaei. Measurements of the PSC fossils were compared with those of nearby populations of other southwestern tree squirrels (Table 7). The PSC m3 most closely matches the size of those in a

Measurement (mm) 58.8(est.) 42.5 OMNH 51465 81.6 73.4 82.3 73.9 36.2 19.8 23.9 30.3 39.0 39.1 30.3 47.8 46.0 44.2 36.1 38.6 30.2 27.4 48.9 47.2 39.5 16.7 21.3 18.2 28.4 27.0(est.) 43.0(est.) 45.8(est.) 47.8(est.) 51.4(est.)

OMNH 51473 81.0 78.6 86.6 78.1 34.3 19.9 23.6 27.2 40.8 41.0 31.3 46.2

81.2 74.2 — 74.3 33.8 21.0(est.) 23.1 29.4 — — 29.9 46.4

small sample of S. aberti barberi. The PSC maxillary teeth exceed those of S. colliaei in all dimensions and most closely approximate the sizes of those of a small sample of S. nayaritensis apache rather than those of the species having two upper premolars. However, the samples are very small, and the teeth probably should not be considered distinguishable from the other southwestern squirrels on size alone. We tentatively conclude that the maxilla belongs to S. aberti largely because of the remnant of a robust alveolus for P3. North American tree squirrels ( Sciurus, Tamiasciurus, Glaucomys ) are rare as fossils, probably in part due to their arboreal habits and woodland habitats. This is especially true in the western United States, where the only Pleistocene records

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

Figure 3. Late Pleistocene fossils from Papago Springs Cave, Arizona. A, Sciurus cf. S. aberti, right maxilla with M1-M3, alveoli for P4, and portion of alveolus for P3 (OMNH 51850), palatal view. B, Sciurus cf. S. aberti, right m3 in a fragment of dentary bone (OMNH 51849), in occlusal and labial views. C, Aztlanolagus agilis , occlusal pattern of right p3 (OMNH 52749). D, Geomys sp. or Pappogeomys sp., occlusal pattern of P4 (OMNH 52750).

of tree squirrels are of Tamiasciurus (in California, Colorado, Texas, New Mexico, and Montana; Barnosky and Rasmussen 1988; Faunmap Working Group 1994; Harris 1985a, 1993), Sciurus sp. (Barnosky and Rasmussen 1988), and Sciurus “cf. Arizonensis” from Blackwater Draw, New Mexico (Slaughter 1964). The latter record is probably in error (Harris 1985a:165; Kurtén and Anderson 1980) and likely represents S. carolinensis. The PSC specimens represent the first fossil specimens potentially referable to S. aberti. Geomyidae Thomomys umbrinus (Richardson, 1829) or T. bottae (Eydoux and Gervais, 1836)— pocket gopher Material.—P4 (12; 52127, 52132, 52138, 52140, 52145-7, 53086, 53233-6); left M1 or M2 (52133); M3 (3; 52129, 53239, 53240); ascending ramus of right dentary (52136); left dentary with i1

21

(53241); left dentary with i1 and p4 (51498); left dentary with i1 and p4-m3 (52152); right dentary fragment with i1 and m1-m2 (52130); right dentary fragment with m1 (52134); edentulous dentary fragment (4; 52142-4 , 53091); p4 (13; 52128, 52131,52135, 52148-50, 53087-9, 53237-8, 53442); molar (20; 52141, 52151, 52779, 52780, 53090); distal left tibia (52139). Discussion.—Pocket gopher remains (especially teeth) are moderately abundant in PSC deposits. The morphology of the p4s recovered indicates that T. talpoides is not present in the fauna. All available specimens have the anterior prism of p4 convex or flat anteriorly, never concave. Unfortunately, no skulls were recovered that might indicate which species, T. bottae, T. umbrinus, or both, occurred in the vicinity in the late Pleistocene. Thomomys umbrinus and T. bottae are difficult to differentiate osteologically, even with complete modern skeletons. Today, T. bottae is widespread in southwestern North America, occurring throughout Arizona; T. umbrinus is restricted in Arizona to the shallow rocky soils of the oak woodlands of a few mountain ranges in the extreme southern part of the state, including the Huachuca, Patagonia, and Santa Rita mountains surrounding PSC (as well as other Madrean mountains in Arizona and Mexico). Thomomys bottae also occurs in the general area, but in modern times, as in Sycamore Canyon in the Patagonia Mountains where both species occur, it inhabits lower elevations (up to about 1250 m) and T. umbrinus inhabits elevations above 1250 m (Hoffmeister 1986). Geomys Rafinesque, 1817 or Pappogeomys Merriam, 1895—pocket gopher Material.—P4 (Fig. 3D; 52750). Discussion.—Measurements (in mm) of the tooth are: anteroposterior diameter, 2.32; transverse width of protoloph, 2.30; transverse width of metaloph, 2.26. The protoloph is flattened anteriorly, forming a broad oval virtually as wide as the metaloph. This condition is similar to that in P4s of Geomys and Pappogeomys (including Cratogeomys) but differs strongly from the narrower, rounded or triangular protolophs seen in Thomomys spp. Neither Pappogeomys nor Geomys has been previously reported in the late Pleistocene of Arizona but Pappogeomys (as Cratogeomys castanops) is known in the late Wisconsinan of U-Bar Cave, New Mexico (Harris 1993).

22

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

OCCASIONAL PAPERS

Table 7. Molar measurements (in mm) of the Papago Springs Cave tree squirrel and some southwestern North American Sciurus species. For the PSC fossils, the measurements represent the dimensions of each individual tooth that was available. For the other squirrels, the figures represent the mean measurement of five specimens. Measurement Taxon PSC fossils S. arizonensis huachuca (from Nogales and Magdalena, Sonora, and Huachuca Mountains, Arizona) S. aberti barberi (from Colonia Garcia, Chihuahua) S. nayaritensis apache (from Chihuahua, Sonora, and San Luis Mountains, New Mexico) S. n. chiricahuae (from Chiricahua Mountains, Arizona) S. colliaei truei (from Sonora and Durango)

M1 length

M1 width

M2 length

M2 width

M3 length

M3 width

m3 length

m3 width

2.75

3.33

2.98

3.40

3.06

3.28

3.13

2.79

2.71

3.17

2.72

3.20

3.13

3.06

3.23

3.00

2.53

2.90

2.66

3.02

2.98

3.03

3.15

2.82

2.84

3.28

2.90

3.38

3.09

3.20

3.24

3.00

2.61

3.23

2.92

3.27

2.97

3.09

3.23

3.01

2.37

2.83

2.56

2.95

2.80

2.80

2.93

2.80

Heteromyidae Perognathus Wied-Neuwied, 1839, or Chaetodipus Merriam, 1889, sp. indet.—pocket mouse Material.—left M1 or M2 (53201); left dentary fragment with m2-m3 (51823); right m1 or m2 (3; 53202-4), right m3 (53205). Discussion.—Measurements (in mm) of the teeth in the dentary fragment are anteroposterior length of m2, 0.90; transverse width m2, 1.12; anteroposterior length of m3, 0.77; transverse width of m3, 0.86. The specimens of a pocket mouse are from a species smaller than Chaetodipus hispidus and about the size of C. intermedius. It is not possible to make a species determination. Today, C. intermedius occurs in open grassy areas within the encinal near PSC but it is not common there. The relative rarity of pocket mouse remains in the late Pleistocene cave sediments suggests conditions outside the cave may have been similar to modern conditions with respect to habitat selection by pocket mice. Skinner (1942) reported a single partial ramus from PSC as Perognathus (?)apache. Perognathus apache now is usually considered a subspecies or synonym of Perognathus flavescens (Patton 1993; Williams 1978). However, Hoffmeister (1986) believed the data for considering the two forms conspecific was weak, and he retained full species status for P. apache. Other than a few specimens from Cochise County, near Willcox, Arizona (about 80 km away from PSC), P. apache occurs no nearer than the valley of the Little Colorado River in northeastern Arizona (Hoffmeister 1986). Better specimens are needed before the

specific identity of the PSC heteromyid(s) can be established. Muroidea Sigmodontinae Peromyscus Gloger, 1841, and Reithrodontomys Giglioli, 1874—mice Material.—Unidentified sigmodontines comprise 166 specimens (including 118 isolated molars, 26 dentaries with at least 1 molar, and 16 maxillae with at least 1 molar). Discussion.—Most of these specimens are referable to Peromyscus but a few probably pertain to Reithrodontomys. A superficial examination suggests that none pertain to Baiomys. Several species of Peromyscus probably are present, including at least P. boylii and P. maniculatus, but our samples consist primarily of isolated molars and edentulous jaw bones and are inadequate for certain identifications and for meaningful interpretation or conclusions. Statistical and computerized analyses, not attempted for the present study, might be useful in sorting out some of the taxa. Such analysis should include Skinner’s (1942) specimens. With larger samples of rami at his disposal, Skinner referred specimens to P. maniculatus and to P. (?)boylii or P. (?)truei, noting that his material did not permit a satisfactory separation of these species. At present, seven species of Peromyscus (P. eremicus, P. merriami, P. melanotis, P. maniculatus, P. leucopus, P. boylii, and P. nasutus) have been recorded within a 100km radius of PSC, and another three species (P. polius, P. truei, and P. gratus) occur within 200 km (Carleton 1989). Baiomys taylori occurred in

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

23

Table 8. Measurements (in mm) of lower first molars of Neotoma spp. from PSC. Measurements were defined by Harris (1984a). Measurement OMNH specimen no.

Length m1

Neotoma cf. N. mexicana 51953 3.15 51954 3.2 51955 3.2 Neotoma cf. N. albigula 53199 3.15 53200 3.1 Neotoma pygmaea or N. stephensi 53196 2.7 53197 2.5 53198 2.7

Width m1

Height of dentine tract

Anterior face to F21

F1 to F22

Groove length (FOLD)3

Groove/ width L1 (RATIO)4

Wear stage

1.3 1.4 1.6

1.45 1.55 1.1

2.4 2.5 2.4

0.9 0.9 0.9

0.4 0.4 0.4

0.31 0.36 0.37

2 3 2

1.55 1.4

0.1 0.1

2.3 2.2

0.95 0.9

0.1 0.2

0.52 0.39

3 2

1.3 1.4 1.2

0.75 0.7 0.75

1.8 1.7 2.1

0.7 0.8 0.8

0.1 0.1 0.4

0.46 0.43 0.25

3 2 2

1

F1 = Base of first lingual fold. F2 = Base of second lingual fold. 3 FOLD = Development of the anterointernal reentrant fold of the first loph. width L1 = Width of the first loph. 4 RATIO = Ratio comparing the depth, in occlusal view, of the anterointernal fold to the width of the first loph. 2

formerly dense grasslands north, east, and west of the Canelo Hills in historic times. Three species of Reithrodontomys, R. fulvescens, R. montanus, and R. megalotis, occur in appropriate habitats around the Huachuca Mountains in modern times. Onychomys Baird, 1858, sp. indet.— grasshopper mouse Material.— right M1 (2; 53273, 53280); left M1 (53274); left m1 (2; 53275, 53276); right m1 (4; 53277, 53278, 53279, 52216). Discussion.—In the modern fauna two species, Onychomys leucogaster and O. torridus, occur near to PSC, and a third, O. arenicola, occurs within 120 km in the Peloncillo Mountains in Hidalgo County, New Mexico. No attempt was made to identify the isolated first molars from PSC to species; indeed, the species may not be distinguishable based on these teeth. Neotoma cf. N. albigula Hartley, 1894— white-throated wood rat Material.—right m1 in a small fragment of dentary (51860); right m1 (3; 51863, 53199, 53200). Discussion.—Skinner (1942) tentatively identified Pleistocene woodrats from PSC as “Neotoma (?)mexicana Baird or N. (?)albigula Hartley, ref.” In our analysis of newly collected specimens, both Neotoma cf. N. albigula and N. cf. N. mexicana were found to be present in the cave deposits in one unit of the Main Room, Sec. 1, 2.57-2.77 m

below datum (Unit 6). There, an m1 lacking a labial dentine tract on its anterolophid (N. cf. N. albigula) and a very high-crowned m2 with a high labial dentine tract on the anterolophid (N. cf. N. mexicana) were found. In the Annex, a third species was present (N. pygmaea or N. stephensi). All PSC Neotoma specimens are single cheek teeth, few of which are in fragments of the dentary bone. Isolated teeth of woodrats are difficult to identify, and we made little attempt to identify woodrat teeth other than m1. The m1s were identified based on criteria and measurements in the useful analysis provided by Harris (1984a). Measurements of two teeth are provided in Table 8. Neotoma cf. N. mexicana Baird, 1855— Mexican wood rat Material.— right m1 (3; 51953-5); fragment of left m2 (51862). Discussion.—Measurements of some specimens are provided in Table 8. In Arizona, the occurrence of this woodrat in PSC is associated with ages of 33.5 and 23.1 ka. Neotoma mexicana is known from middle and late Wisconsinan deposits in U-Bar Cave in New Mexico (Harris 1987). The species may have been absent from southern New Mexico east of the Rio Grande until the early Holocene (Harris, A. H., pers. comm.). In parts of their modern geographic ranges where the two species are sympatric, N. mexicana occupies the steep sides of canyons and cliff bases

24

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

near valley bottoms and N. albigula occupies the valley floor such as in Dolores Canyon, Colorado (Finley 1958); a similar situation occurs in the Guadalupe Mountains, Texas (Cornely 1979). Mexican woodrats are principally montane, often with disjunct populations or populations isolated on mountain ranges (Cornely and Baker 1986). Neotoma mexicana inhabits steep cliffs and rocky slopes, whereas N. albigula builds dens on moderately rocky slopes (Macêdo and Mares 1988). According to Hoffmeister (1986), N. mexicana in Arizona usually occupies mountainous areas above the juniper-piñon woodland, where it is usually associated with montane coniferous forests. In the Huachuca Mountains, Mexican woodrats live in the Douglas fir-pine zone. In the White Mts. they are found mostly in the yellow pines, often in association with Microtus mexicanus. Neotoma pygmaea Harris, 1984b or Neotoma stephensi Goldman, 1905—wood rat Material.—right dentary fragment with m1 (53196); m1 (2; 53197, 53198). Discussion.— The measurements (Table 8) and qualitative features of these teeth most closely match those of N. pygmaea given by Harris (1984a, b). They also resemble those of N. stephensi except that the dentine tract is shorter than in N. stephensi measured by Harris. We hesitate to assign PSC specimens to one of the two species until a better sample becomes available. Neither N. pygmaea nor N. stephensi was previously reported in PSC deposits. Neotoma pygmaea is an extinct species known only from interstadial sites within Dry Cave, New Mexico (Harris 1984b). Neotoma stephensi occurred as a fossil in deposits of mid-Wisconsinan age in UBar Cave, New Mexico (Harris 1993). This suggests that the undated, calcite-cemented fill in the PSC Annex could represent interstadial deposits. However, this conclusion remains tentative until better-identified specimens are found there or until radiometric age estimates are available. Sigmodon Say and Ord, 1825—cotton rat Material.—infraorbital plate (51859); right maxilla with M1 (53360); left M1 (2; 51855, 53361); right M1 (2; 53362, 53257); left M2 (51856); right M2 (2; 51857, 53363); left M3 (2; 53364-5); right M3 (3; 53366-7, 51858); left dentary with i1 (53347); right dentary fragment with m2 fragment (53348); left m1 (53349); right m1 (5; 53350-4); left m2 (53355); right m2 (53356); left m3 (3; 53357-9).

OCCASIONAL PAPERS

Discussion.—A relatively small form of Sigmodon seems to be represented in the teeth and fragments from PSC. Most specimens are unidentifiable to species, but the shape of the infraorbital plate recovered in PSC may be useful in identification of at least that specimen. The anterior spine on the infraorbital plate is described as being long and narrow in Sigmodon arizonae, whereas it is broad and blunt in S. hispidus and presumably S. ochrognathus and S. fulviventer (Hoffmeister 1986; Severinghaus and Hoffmeister 1978). The infraorbital plate from PSC has a long pointed anterior spine and thus is identified as Sigmodon cf. S. arizonae. Cotton rats in southwestern North America are undergoing an adaptive radiation and the differences in their habitat requirements (as well as morphology) remain subtle and are poorly understood. Generally, they indicate the presence of relatively warm, semiarid grassland. Three different species (S. arizonae, S. fulviventer, and S. ochrognathus) occur today in Santa Cruz County and a fourth species (S. hispidus) occurs in southeastern and extreme southwestern Arizona (Hoffmeister 1986). One of these species, S. ochrognathus, is thought by Davis and Dunford (1987) to have colonized this area in very recent times. In view of the poor late Pleistocene record for the genus, and given the current richness of Sigmodon species in southeastern Arizona, the reader should be mindful of the possibility that different species might have occurred in the PSC area during different parts of the Pleistocene. Specimens of Sigmodon occurred together with specimens of Microtus (which are considered to be ecologically similar, grass-eating, runway-making rodents) in Sec. 2, Unit 7 and in the cemented fill of the Annex. Fossils of Sigmodon are known in the late Quaternary of Arizona at Murray Springs Arroyo (late Pleistocene; Lindsay and Tessman 1974), Deadman Cave (mixed late Wisconsinan/early Holocene; Mead et al. 1984), a woodrat midden near Wolcott Peak (mixed late Wisconsinan/middle Holocene; Mead et al. 1983), and a woodrat midden near Wellton Hills (early Holocene; Mead et al. 1983). They are also known from UBar Cave and other sites in New Mexico (Harris 1993), and from Rancho la Brisca, Sonora (Sangamonian; Van Devender et al. 1985). Microtus Schrank, 1798 sp. indet.—vole Material.—A total of 257 specimens including 226 isolated molars (46 m1, 35 m2, 44 m3, 23 M1,

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

41 M2, 37 M3), 21 dentaries, and 10 palates or maxillae. Discussion.—Measurements of lower first molars from the Main Room are given in Table 9. Voles do not occur today in the PSC area, in nearby mountain ranges, in Sonora, nor in southeastern Arizona except for M. longicaudus in the Pinaleño Mountains in Graham County. Although Skinner (1942) identified vole specimens he collected as “Microtus (?)mexicanus”, we see no reason to exclude the possibility that the specimens represent other species of voles (e.g., M. longicaudus, M. montanus, M. pennsylvanicus), or that multiple species might have occurred in the PSC deposits. Unfortunately, all of these species have similar m1s (often used as a diagnostic tooth in some kinds of arvicolines) with five closed triangles. At least one m1 (OMNH 51794) from PSC has three closed triangles that may represent a different species from most of the other teeth. It is difficult or impossible to distinguish among the several possible species based on our isolated or fragmentary specimens without detailed study and multivariate analyses similar to those done by Smartt (1977). Therefore, we choose to leave the species unidentified until better specimens are collected and/or comprehensive studies of dental variation can be made for North American Microtus species. Many voles have modern geographic ranges that are fragmented, with different populations probably isolated from one another to varying degrees by existing on different mountain ranges. Approximate distances between PSC and the nearest extant populations of Microtus are 135 km to the northeast in the Pinaleño Mountains (M. longicaudus), 200 km to the north on the Nantanes Plateau, Arizona (M. mexicanus), 210 km to the northeast in the Mogollon Mountains of New Mexico (M. mexicanus), and 225 km to the southeast in the Sierra Madre Occidental in Chihuahua near the Sonora-Chihuahua border (M. mexicanus) (Anderson 1972; Findley et al. 1975; Hoffmeister 1986). Lagomorpha Leporidae Aztlanolagus agilis Russell and Harris, 1987— Aztlán rabbit Material.—right p3 (Fig. 3C; 52749). Discussion.—-Measurements (in mm) of the tooth are anteroposterior length, 2.27; transverse width, 2.11. The measurements are within the range of variation of p3 dimensions given by Rus-

25

sell and Harris (1987) and the occlusal morphology similar to that shown by Russell and Harris (1987), Tomida (1987), and Winkler and Tomida (1988). The Aztlán rabbit was not previously reported from PSC. Lepus Linnaeus, 1758, sp. indet.—jackrabbit Material.—right distal tibia (53219). A left P2 (53085) might also pertain to this genus or to Sylvilagus . Discussion.—Skinner (1942) previously reported Lepus californicus in the PSC deposits; that species occurs in the region in modern times. Sylvilagus cf. S. audubonii (Baird, 1858)— cottontail rabbit Material.—right dentary with p3-m2 (52042); left dentary with i1 and p3-m1 (52043); right p3 (52044); left dentary with p3-m3 (53426). Other specimens of rabbits were recovered from the cave sediments but are identifiable only as Sylvilagus sp. indet. Discussion.—Based on the strongly crenulated enamel pattern of the p3 (Dalquest et al. 1989), all cottontail specimens with that tooth are identified as Sylvilagus cf. S. audubonii. In the Huachuca Mountains and vicinity today, both S. audubonii and S. floridanus occur. Both species occur more frequently in grasslands than in oak woodlands (Hoffmeister 1986: table 4.1).

DISCUSSION Twenty-seven taxa of vertebrates were recovered by us that were not previously reported from PSC (Colbert and Chaffee 1939; Roosevelt and Burden 1934; Skinner 1942). These 27 include Rhinichthys osculus, Ambystoma cf. A. tigrinum, Bufo woodhousii or B. punctatus, Hyla sp., Hyla or Pseudacris sp., cf. Gastrophryne sp., Scaphiopus or Spea sp., Rana sp., Crotaphytus sp., Phrynosoma douglasi, Sceloporus cf. S. magister, medium-sized sceloporine lizard, small-sized sceloporine lizard, Meleagris cf. M. crassipes, cf. Cyrtonyx montezumae, Aphelocoma ultramarina, Salpinctes obsoletus, Sorex arizonae, Notiosorex crawfordi, Myotis small sp., Mustela cf. M. frenata, Spermophilus cf. S. spilosoma, Sciurus cf. S. aberti, Geomys or Pappogeomys sp., Neotoma pygmaea or N. stephensi, Sigmodon cf. S. arizonae, and Aztlanolagus agilis (Table 10). Undated Pleistocene species reported from PSC by Skinner (1942) include an ibis-like bird (Threskiornithidae), Tamias ?dorsalis, Perognathus ?apache, Onychomys ?leucogaster, Microtus ?mexicanus, Peromyscus maniculatus, Peromyscus ?boylii or

26

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

OCCASIONAL PAPERS

Table 9. Summary measurements (in mm) of lower first molars of Microtus sp. from Papago Springs Cave. Measurements follow Martin (1989). Measurement

n

Range

Mean

s.d.

C.V.

Occlusal length Length of the anteroconid complex Width of the anteroconid complex Greatest width of the anterior cap Width of the dentine isthmus connecting T4/T5 and the anterior cap Width of the dentine isthmus connecting T4 and T5

23 23 27 27

2.54-3.24 1.19-1.56 0.75-1.17 0.40-0.84

2.91 1.37 0.95 0.63

0.193 0.099 0.101 0.104

0.066 0.072 0.106 0.165

28

0.01-0.08

0.03

0.017

0.518

28

0.01-0.14

0.06

0.038

0.645

?truei, Lepus californicus, Myotis ?evotis, Antrozous pallidus, Tadarida ?brasiliensis, Canis latrans, Canis lupus, Taxidea taxus, Spilogale gracilis, Bassariscus astutus, Equus conversidens, Equus tau, Platygonus compressus, Bison antiquus, Cervus sp., and Camelidae (taxonomy updated). The PSC faunal list now includes eight or nine extinct species, Meleagris crassipes, Aztlanolagus agilis, possibly Neotoma pygmaea (unless identity is the extant N. stephensi), Equus tau, E. conversidens, Platygonus compressus, Camelidae sp. indet., Bison taylori (as listed by Skinner [1942]; equated with B. antiquus by Harris 1985a) and Stockoceros onusrosagris. Extant species that no longer occur in the Canelo Hills area include a fish, Rhinichthys osculus, and 9 or 10 mammals, Sorex arizonae, Tamias sp., Sciurus cf. S. aberti, Marmota flaviventris, Geomys sp. or Pappogeomys sp., Neotoma stephensi (unless identity is N. pygmaea), N. mexicana, Microtus sp., Ursus americanus, and Cervus sp. Some of these occur not far away today, others are absent in modern times due to extirpation by humans. The rocky, gravelly wash that passes within 5 m of one entrance to PSC today flows above ground for only part of the year. No fish are observed to occur in the headwaters of the wash in recent times. Rhinichthys osculus, the speckled dace, is the only fish identified in the PSC fauna. It is represented by an individual about 90 mm in standard length. The presence of a fish that size indicates that, 42,000 years ago, the wash that runs near the cave (or a nearby stream) discharged more water or was less ephemeral than it is now. Overall, the herpetofauna is not unusual in that it is not thought to contain extralocal species, although this could be in part due to the conservative approach to specific identifications. Nevertheless, some records are unusual by virtue of

their geological age. Deposits containing herpetofaunas in the Southwest are fairly unusual, but those dating to the middle Wisconsinan and earlier are extremely rare. The most abundant amphibian at PSC was Ambystoma cf. A. tigrinum. Until this report, Ambystoma was extremely poorly known for the Pleistocene of the Southwest. The inadequate record of this genus is probably not due to its low abundance or rare occurrence, but is probably due more to the type of fossil localities that have been excavated. Most late Pleistocene localities in the Southwest that were discovered, excavated, and reported in the literature are typically arid sites (e.g., woodrat middens [in which preservation and location are due to aridity], Deadman Cave [an arid location on the Santa Catalina Mountains], Ventana Cave [an arid shelter on the edge of the Gran Desierto]), or else they are open-air sites that can be highly destructive to small skeletons due either to high energy movement by water or to the destructive nature of a fluctuating water table (as at the Lehner Mammoth site and other related localities). The exception is Rancho la Brisca, Mexico, which is an ideal locality from which to obtain fossils of amphibians and reptiles that require riparian and paludal environments. Yet with such a rich anuran and testudine fauna, it is a mystery why caudates were not also recovered there. PSC is also unique in comparison to the “typical” Rancholabrean-age localities reported in the Southwest, because it is a cave location that records a mesic situation instead of a xeric situation. Ambystoma was recovered in PSC cave-fill units dating as follows: approximately 246 ka; sometime between 102 and 246 ka; approximately 102, 42, and 33 ka; and from the presumably middle Wisconsinan Glacial deposits of the Annex. Regardless of whether the species is actually A. tigrinum, A. rosaceum, or some other species, the

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

recovery of an ambystomatid salamander from PSC is indicative of the availability of streams and/or spring water year-round. The local habitat could have been an open plains grassland to a more wooded forest community. The anurans (Bufo woodhousii/punctatus, Bufo sp., Hyla sp., Hyla/Pseudacris, cf. Gastrophryne, Scaphiopus/Spea, and Rana sp.) are found in cave fill dating from 246 ka up to the Holocene. The recovery of Hyla sp., Hyla/Pseudacris, cf. Gastrophryne, and Rana sp. would imply that the local area was fairly moist, at least with some annual water. However, surrounding communities could have ranged from xeric (such as a dry mesquite grassland) to a more moist woodland to forest. At a minimum, the area around PSC was a wet riparian zone. Most of the lizard remains are from the undated Annex deposits, presumably of Wisconsinan Glacial age. Phrynosoma sp. is known from the cave at 42 ka. Sceloporine lizards of a variety of sizes and possibly species are known from the area back to 102-246 ka and 42 to 33 ka. Finding these taxa at PSC at any of the reported ages is not overly surprising. However, depending on the actual species represented (see various accounts), there may have been more forms living farther north than they do today. It would be easy to imply that the present, local herpetofauna and the taxa recovered from various ages at PSC have not changed over time even though there have been climatic changes from a previous interglacial to glacial to the present interglacial. However, this assumption may not be true, depending on which species are actually represented at PSC. Today’s herpetofauna is one of a presumed ‘typical’ interglacial (Holocene) community and climate. Many of the specimens from deposits in Sections 1 and 2 deposits date to the most recent glacial (Wisconsinan), but some are certainly from the previous interglacial (Sangamonian) if not earlier. The earliest units may actually represent a previous glacial (“Illinoian”?). What seem to be missing are deposits representing the full glacial episode (ca. 22 to 18 ka), unless this time period is found in the undated deposits of the Main Room or Annex. None of the recovered PSC herpetofauna seems to be indicative of what we would perceive (preconceive) as a colder-than-present full glacial fauna. Instead the faunas seem to indicate either ‘warm’ pre-full glacial interstadials, or glacial episodes with more equable climates (less extreme in temperature and therefore moisture budget), or

27

interglacials appearing like that of today. Two factors limit the utility of the recovered herpetofauna (and, for that matter, the whole vertebrate fauna) for climatic and community reconstructions: the less than detailed and precise U-series chronology, and the inability to identify many of the specimens to species. Among the mammals, the directly dated bones of Stockoceros onusrosagris span a range of time from 246 ± 19 ka to 31.0 ± 1.0 ka. Occurrences of vertebrate taxa associated with U-seriesdated Stockoceros bones and speleothems are provided in Table 1. The oldest dated unit (Sec. 1, Unit 6; 246 and >172 ka) includes Ambystoma cf. A. tigrinum, Tamias sp., Thomomys umbrinus/ bottae, Neotoma cf. N. mexicana, N. cf. N. albigula, and Sylvilagus cf. S. audubonii. No vertebrate local faunas dating to the last interglacial (Sangamonian) are known in Arizona (Pinsof 1996). Unfortunately, those units in PSC (speleothem canopy and Sec. 2 bottom of Unit 5) that reflect the last interglacial include only two vertebrate taxa, Hyla sp. and Stockoceros onusrosagris. The most species-rich unit (Sec. 2 Unit 7; Table 1) resulting from our excavations is that which includes two concordant age estimates (42.9 ± 2.7 ka and 42.4 ± 2.2 ka) reflecting the middle Wisconsinan glacial. The association of mammalian species within this middle Wisconsinan unit (Sec. 2, Unit 7), if it represents a single fauna, can be considered as a mammalian community without modern analog. It includes among others Notiosorex crawfordi, Sorex arizonae, Marmota flaviventris, Sciurus cf. S. aberti, Perognathus or Chaetodipus sp., and Sigmodon cf. S. arizonae. Today some of these species are allopatric, probably because they responded differently to changing climates in the late Wisconsinan and Holocene, as did other mammals in North America (Faunmap Working Group 1996; Graham 1986; Graham and Semken 1987). The presence of Abert’s squirrels indicates that ponderosa pine forest, or possibly Madrean pine-oak woodland, occurred near PSC (at least within the hunting range of a raptor or mammalian predator) during the middle Wisconsinan. This woodland may have been open with an understory of grasses, as in the northern Sierra Madre Occidental in the present day, but the number of species adapted to or suggestive of grassland conditions (the badger, Quentin’s pronghorn, pocket mouse, vole, and cotton rat) indicates that there may have been open semiarid grasslands, too.

28

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

OCCASIONAL PAPERS

Table 10. Comparison of Rancholabrean vertebrate faunas from selected localities in southern Arizona, southwestern New Mexico, and northern Sonora, Mexico. PSC = Papago Springs Cave. Data from Skinner (1942), as updated by Harris (1985a, 1990b), Rea (1980), Steadman (1980), and this report. DC = Deadman Cave (Mead et al. 1984). RLB = Rancho la Brisca (Van Devender et al. 1985). PP = Picacho Peak woodrat middens (Van Devender et al. 1991). U-Bar = U-Bar Cave (Harris 1987, 1993). Only the mid-Wisconsinan vertebrates from U-Bar Cave are listed here. X = present as fossil at locality at any age. - = absent from locality. Specimens from localities listed below for "Scaphiopus sp." (as originally published) may now include Spea. Specimens originally published as "Scaphiopus hammondi" or "cf. S. multiplicata" are listed here as Spea multiplicata. Locality Taxon OSTEICHTHYES Cypriniformes—minnows, suckers Cyprinidae Rhinichthys osculus Agosia chrysogaster Cyprinodontiformes—pupfish, killifish, topminnows Poeciliidae Poeciliopsis occidentalis/ monacha—occidentalis AMPHIBIA Caudata—salamanders Ambystomatidae Ambystoma cf. A. tigrinum Anura—toads and frogs Bufonidae Bufo alvarius Bufo cf. B. cognatus Bufo cf. B. kelloggi B. mazatlanensis B. retiformis B. punctatus Bufo cf. B. woodhousii B. woodhousii/punctatus Bufo sp. Hylidae Hyla arenicolor cf. Hyla Hyla sp. Hyla/Pseudacris Pternohyla fodiens Leptodactylidae Eleutherodactylus augusti Leptodactylus melanonotus Microhylidae Gastrophryne cf. G. olivacea cf. Gastrophryne Pelobatidae Scaphiopus couchi Scaphiopus sp. Scaphiopus/Spea Spea multiplicata cf. S. multiplicata Ranidae Rana "pipiens" Rana sp. REPTILIA Testudines—tortoises and turtles Emydidae cf. Terrapene

PSC

DC

RLB

PP

U-Bar

X —

— —

— X

— —

— —





X





X







X

— — — — — — — X X

— — — — — X X — —

X X X X X X — — X

— — — — — — — — —

— — — — — — — — —

— X X X —

— — — — —

X — X — X

— — — — —

— — — — —

— —

— —

X X

— —

— —

— X

— —

X —

— —

— —

— — X — —

X — — — X

X X — — —

— — — X —

— X X — —

— X

— X

X —

— —

— —





X





Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

29

Table 10. (continued) Locality Taxon Pseudemys scripta Kinosternidae Kinosternon flavescens K. sonoriense Kinosternon sp. Squamata—lizards and snakes Crotaphytidae Crotaphytus sp. C. collaris Phrynosomatidae Callisaurus cf. C. draconoides C. draconoides Cophosaurus texanus Holbrookia maculata Phrynosoma douglasi P. cornutum P. modestum P. platyrhinos P. solare Phrynosoma sp. Sceloporus cf. S. clarkii S. cf. S. magister S. cf. S. undulatus Sceloporus sp. "small" sceloporine "medium" sceloporine "large" sceloporine Urosaurus ornatus Teiidae Cnemidophorus tigris Cnemidophorus sp. Helodermatidae Heloderma suspectum Gekkonidae Coleonyx variegatus Colubridae Arizona elegans Gyalopion canum Hypsiglena torquata Lampropeltis getulus Masticophis cf. M. mentovarius Masticophis sp. Pituophis melanoleucus ?Pituophis sp. Salvadora sp. Sonora semiannulata Rhinocheilus lecontei Tantilla cf. T. hobartsmithi Thamnophis cf. T. cyrtopsis Trimorphodon biscutatus Leptotyphlopidae Leptotyphlops sp. Viperidae Crotalus atrox

PSC

DC

RLB

PP

U-Bar





X





— — —

— — —

X X X

— — —

— — —

X —

— X

— —

— X

— X

— — — — X — — — — X — X — — X X X —

— X X X X — X — X — X X X — — — — X

X — — — — — — — — — X — — — — — — —

— — — — — — — X — X — — — X — — — X

— — — — X X X — — — — — — — — — X X

— —

— X

— —

X —

— X



X





X







X



— — — — — — — — — — — — — —

X X X X — X X — X — X — — X

— — X — X — — — X — — — X —

X — X X — — X — — X X X — —

— — — — — — — X X — — — — —







X





X

X





30

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

OCCASIONAL PAPERS

Table 10. (continued) Locality Taxon C. scutulatus Crotalus sp. AVES Ciconiiformes—herons, ibises, storks, etc. Threskiornithidae genus and sp. indet. Anseriformes—waterfowl Anatidae cf. Anabernicula Anas platyrhynchos Anas cf. A. crecca A. ?cyanoptera Falconiformes—eagles, hawks, etc. Cathartidae Breagyps clarki Cathartes aura Coragyps atratus/occidentalis Coragyps occidentalis Accipitridae Accipiter cooperi Aquila chrysaetos cf. Spizaetos genus and sp. indet. Falconidae Falco sparverius Galliformes—gallinaceous birds Phasianidae Meleagris crassipes M. gallopavo Colinus sp. Cyrtonyx montezumae cf. C. montezumae Lophortyx gambelii Columbiformes—pigeons and doves Columbidae Zenaida macroura cf. Z. macroura Strigiformes—owls Strigidae Strix brea Strix cf. S. occidentalis Otus sp. Micrathene whitneyi Asio otus Caprimulgiformes—goatsuckers Caprimulgidae genus and sp. indet. Apodiformes—swifts, hummingbirds Apodidae Aeronautes saxatilis Piciformes—woodpeckers, etc. Picidae Colaptes auratus Passeriformes—perching birds

PSC

DC

RLB

PP

U-Bar

— —

X —

— —

— —

— X

X









— — — —

— — — —

— — X —

— — — —

X X — X

— — — —

— — — —

— — — —

— — — —

X X X X

— — — —

— — — —

— — X X

— — — —

X X — —









X

X — — — X —

— — X X — X

— X — — — —

— — — — — —

X — X — — —

— —

— X

— —

— —

X —

— — — — —

— — X X X

X — — — —

— — — X —

— X — — —



X













X





X





X

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

31

Table 10. (continued) Locality Taxon Corvidae Aphelocoma ultramarina ?Cyanocitta Troglodytidae Salpinctes obsoletus Muscicapidae Turdus cf. T. migratorius Catharus guttatus Emberizidae Emberizinae indet. Icterinae indet. Agelaius phoeniceus ?Xanthocephalus xanthocephalus MAMMALIA Insectivora—lipotyphlans Soricidae Sorex arizonae S. merriami S. cf. monticolus S. preblei Notiosorex crawfordi Chiroptera—bats Vespertilionidae Myotis velifer M. lucifugus M. thysanodes M. ?evotis M. ciliolabrum Myotis "small" sp. cf. Myotis Corynorhinus townsendii Eptesicus fuscus Antrozous pallidus Molossidae Tadarida brasiliensis Phyllostomidae Desmodus stocki Xenarthra—sloths, etc. Megatheriidae Nothrotheriops shastensis Carnivora—carnivorans Ursidae Ursus americanus Arctodus simus Canidae Urocyon cinereoargenteus Canis latrans C. lupus Mustelidae Mustela cf. M. frenata Mephitis mephitis Mephitis macroura Mephitis sp. Spilogale gracilis

PSC

DC

RLB

PP

U-Bar

X —

— —

— —

— —

— X

X







X

— —

X X

— —

— —

— —

— — — —

X X — —

X — X —

— — — —

— — — X

X — — — X

— — — — X

— — — — —

— — — — —

— X X X X

X — X X — X — X — X

— — — — — — X — — X

— — — — — — — — — —

— — — — — — — — — —

X X — — X — — X X X

X







X









X



X





X

X —

— —

— —

— —

X X

X X X

— — —

— — —

— — —

X — —

X — — X X

— — X — X

— — — — —

— — — — —

— X — — —

32

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

OCCASIONAL PAPERS

Table 10. (continued) Locality Taxon Spilogale sp. Taxidea taxus Felidae Felis concolor Lynx rufus Procyonidae Bassariscus astutus Proboscidea—mammoths, mastodons, etc. Elephantidae Mammuthus sp. Perissodactyla—odd-toed ungulates Equidae Equus conversidens E. laurentius E. occidentalis E. tau Equus sp. Artiodactyla—even-toed ungulates Tayassuidae Platygonus alemanii Camelidae genus and sp. indet. Camelops sp. Cervidae Cervus sp. Navahoceros fricki Odocoileus hemionus Odocoileus cf. O. hemionus Odocoileus sp. Bovidae Euceratherium collinum cf. E. collinum Bison sp. Oreamnos harringtoni Antilocapridae Capromeryx cf. C. minor Capromeryx sp. Tetrameryx sp. Stockoceros onusrosagris Rodentia—rodents Sciuridae Tamias cf. T. dorsalis Marmota flaviventris Spermophilus variegatus Spermophilus cf. S. spilosoma Spermophilus sp. Ammospermophilus sp. Cynomys gunnisoni Sciurus cf. S. aberti Geomyidae Thomomys umbrinus/bottae T. bottae Geomys/Pappogeomys Heteromyidae

PSC

DC

RLB

PP

U-Bar









X

X







X

— —

X —

— —

— —

— X

X

X

X



X





X





X — — X —

— — — — X

— — — X X

— — — — —

X X X — —

X









X —

— —

— X

— —

— X

X — — — —

— — — — X

— — X — —

— — — — —

— X — X X

— — X —

X — — —

— — X —

— — — —

— X X X

— — — X

— — — —

X — X —

— — — —

— X — X

X X X X — — — X

— — X — — — — —

— — — — — — — —

— — — — — X — —

X X X — X — X —

X — X

X — —

— — —

X — —

— X —

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

33

Table 10. (continued) Locality Taxon Perognathus/Chaetodipus Perognathus cf. P. flavus Dipodomys spectabilis D. merriami/ordii Dipodomys sp. Muridae Reithrodontomys montanus Reithrodontomys cf. R. megalotis Peromyscus maniculatus P. eremicus P. boylii P. boylii/truei P. truei P. nasutus Peromyscus sp. Peromyscus/Reithrodontomys Onychomys cf. O. torridus Onychomys sp. Neotoma albigula Neotoma cf. N. albigula N. cinerea N. findleyi N. mexicana Neotoma cf. N. mexicana N. pygmaea/stephensi N. stephensi Neotoma sp. Sigmodon cf. S. arizonae Sigmodon sp. Microtus mexicanus M. pennsylvanicus Microtus sp. Lemmiscus curtatus Erethizontidae Erethizon dorsatum Lagomorpha—hares and rabbits Leporidae Aztlanolagus agilis Sylvilagus cf. S. audubonii S. audubonii/floridanus S. nuttalli Sylvilagus sp. Lepus sp.

The PSC fossils of Sorex arizonae occur only 25 km from the nearest modern localities of capture of the species in the Santa Rita Mountains. The cave is only 13 m lower in elevation than the lowest modern records of this shrew in the Huachuca Mountains (Simons et al. 1990), some 30 km away. However, the modern records are from the northeastern slope of the Huachuca Mountains, which is cooler and more mesic than the dry

PSC

DC

RLB

PP

U-Bar

X





X

X

— — — —

X X — —

— — — X

— — X —

X X — —

— — X — — X — — X X — X — X — — — X X — — X X — — X —

X — — — — — — — X — — — X — — — — — — — — — X — — X —

— — — — — — — — — — — X — — — — — — — — X — X — — — —

— — — — — — — — X — — — — X — — — — — — — — — — — — —

— X X X X — X X — — X — X — X X X — — X — — X X X — X









X

X X — — X X

— — — — X X

— — — — — —

— X — — — —

X — X X — X

southwestern slope. In fact, no modern records of the Arizona shrew are available from the southwestern slope of the Huachuca Mountains, probably because of the lack, above 1575 m, of permanent streams or springs with sufficient riparian habitats (Simons et al. 1990) to provide a refuge for this shrew. The modern existence of Sorex arizonae high (2591 m) in the main body of the Sierra Madre

34

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

Occidental (Caire et al. 1978) in southwestern Chihuahua suggests that it is broadly distributed in the Sierra Madre Occidental. At the northern end of its range, the Arizona shrew clearly has a relict distribution today on a few isolated montane “islands” that are northern outliers of the Sierra Madre Occidental. The PSC fossils indicate that this shrew once was continuously distributed, in suitable habitats, in the intervening low hills between the Santa Rita and Huachuca mountains. Three of the Sorex arizonae fossils came from the top of Unit 5 in Sec. 2 where they were associated with two discordant age estimates of 33.5 and 23 ka. Only minor remnants of the originally extensive cave deposits were left at this level following excavations done earlier in this century (see Introduction). Nevertheless, our work on these remnants produced fossils of Sorex arizonae, Neotoma cf. N. mexicana, Myotis velifer, Myotis thysanodes, Thomomys umbrinus/bottae, and Stockoceros onusrosagris. One other Sorex arizonae specimen came from Sec. 2 Unit 7, with age estimates of 43 and 42 ka. The desert shrew Notiosorex crawfordi also exists in the modern fauna of the Huachuca Mountains and surrounding area. Its modern distribution is parapatric and possibly mutually exclusive with that of Sorex arizonae (Simons et al. 1990). From 1175 m elevation at the San Pedro River, N. crawfordi occurs up to at least 1583 m in the Huachuca Mountains, both in upland and riparian habitats. In riparian habitats at higher elevations (above 1583 m) N. crawfordi is replaced by S. arizonae. The same habitat separation might have occurred in the late Pleistocene, since only one or the other shrew is recorded in most units in the PSC deposits (they co-occur only in Sec. 2, Unit 7). Thus today, while S. arizonae is isolated in relatively mesic alpine refuges, it is surrounded by drier communities inhabited by the desert shrew. The fossil bats are not very different from those present in PSC in modern times, except that there is no local fossil record of the phyllostomids Choeronycteris mexicana or Leptonycteris curasoae. The extinct phyllostomid Desmodus stocki, known in the mid-Wisconsinan in U-Bar Cave (Harris 1987, 1993), also seems to have been absent in PSC. However, we recovered few specimens of bats except for Myotis velifer, and the bats do not provide strong data for paleoenvironmental interpretation. The absence of phyllostomids during the mid-Wisconsinan is likely due

OCCASIONAL PAPERS

to temperatures in PSC that were too cold for them to roost; even today the nectar-feeding bats utilize PSC only occasionally during the warmer months. Among the carnivores, fossils of felids and the procyonids Procyon lotor and Nasua narica are noticeably absent in the PSC deposits, despite their historic and modern occurrence in the Canelo Hills and Huachuca Mountains (Hoffmeister 1986; Young and Goldman 1946; Czaplewski, N. J., pers. obs.; Mead, J. I., pers. obs.). Some of these mammals and others characteristic of the Mexican Madrean mountains that reach their modern northern limits in southeastern Arizona and southwestern New Mexico may only have entered the area in very recent (Holocene, possibly late Holocene) times (Armstrong 1996; Davis and Dunford 1987; Martin 1961). Other Southwestern species continue to expand their ranges northward today (e.g., Sciurus aberti, Sigmodon ochrognathus, Microtus mexicanus; Brown and Davis 1995; Davis and Callahan 1992; Davis and Dunford 1987; Patterson 1995). Some of the species of tree squirrels of the PSC region today (see account of Sciurus cf. S. aberti, above) are discontinuously distributed on montane “islands,” with populations differentiated sufficiently to be recognized as different subspecies; some such as S. aberti also have a propensity for dispersal across “unsuitable” habitats to reach suitable habitats (Davis and Brown 1989). In montane mammals in general, this dispersal depends to some extent on the nature of the intervening habitats (Lomolino et al. 1989). During the glacial and interglacial fluctuations of the Pleistocene, all these sciurid species and many other nonvolant mammals probably experienced variable and repeated expansions of their habitats and ranges followed by contraction, fragmentation or homogenization of habitats, and total to partial isolation of populations. Sciurus aberti is currently the most widely distributed of the Southwestern tree squirrels; this may reflect its propensity for dispersal. In Arizona, this species was successfully introduced by humans in numerous mountains beyond its historical range (Brown 1984; Hoffmeister 1986), and from these mountains it readily dispersed naturally to still others (Brown 1984), sometimes through intervening “unsuitable” habitats. Its occurrence in PSC deposits suggests that S. cf. S. aberti probably reached the Santa Rita and Huachuca mountains, too, but has since disappeared from them. Its disappearance from the PSC area since 42 ka

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

is probably due to the complete local loss of ponderosa pine forest and associated hypogeous fungi, on which the species usually depends ecologically (Brown 1984; Davis and Brown 1989). However, its disappearance from the nearby mountains—which still have suitable habitat— might be due to the chance disappearance of small “insular” populations (Harris 1990a), as shown for Great Basin mountain ranges by Brown (1971, 1978) and Grayson (1987; Grayson and Livingston 1993), and for those in the American Southwest by Lomolino et al. (1989). The same could be true for other small vertebrates in Pleistocene deposits in PSC. Marmots (Marmota flaviventris ) often utilize rock piles and talus slopes in open meadows as areas of refuge and hibernation (Mares and Lacher 1987). Aside from the small mountain ranges near PSC, rock outcrops that would have been suitable for habitation by yellow-bellied marmots during the Pleistocene are common in the general area of PSC only where the Basin fill has been stripped off. The hill itself in which PSC exists was a large, isolated outcrop of limestone that was ideal for marmots, as evidenced by the relative abundance of their remains in PSC. Harris (1985a, 1987) noted that this species requires an environment with adequate winter and spring precipitation to maintain continuous green fodder after emergence from hibernation. This vegetation, in turn, requires moderate. The diet includes a variety of grasses, flowers, and forbs (Frase and Hoffmann 1980); these authors further noted that plants favored by yellow-bellied marmots may be inhibited by abundant perennial grasses in the absence of large grazing ungulates. Thus, moderate grazing in the vicinity of PSC by pronghorns Stockoceros could have improved food conditions for the marmots. The presence of a pocket gopher tooth referable to Pappogeomys or Geomys rather than the more frequent Thomomys in the PSC deposits suggests the former presence of better developed soils in the general area. Thomomys generally inhabit rocky or montane valley soils while Pappogeomys and Geomys more typically inhabit lower, broader valleys with deeper, better developed soils (although Pappogeomys also occurs in thin soils of the Chihuahuan Desert). Possibly there were better soils developed on the surface of the basin fill during parts of the Pleistocene than at present. Assuming that the Pappogeomys or Geomys tooth was brought to the cave by a raptor or mammalian predator, such soils proba-

35

bly occurred within the hunting radius of such a predator. The occurrence in southeastern Arizona of either Pappogeomys or Geomys during the middle Wisconsinan could have important biogeographic implications, but better specimens must be recovered before meaningful conclusions can be drawn. Harris (1993) gave Pleistocene records for both Geomys and Pappogeomys in New Mexico. These records range from early to late Wisconsinan in the southern part of that state. At U-Bar Cave, which is about 170 km E of PSC, Harris (1987, 1989) found that Pappogeomys castanops was absent during the midWisconsinan, but that its remains appeared in UBar Cave during the late Wisconsinan. Gophers are a common food of badgers, whose fossils in PSC deposits (Skinner 1942) suggest at least some treeless habitat (Long 1973) in the vicinity. In Arizona and New Mexico, extant populations of Microtus species inhabit montane grasslands, often in or adjacent to ponderosa pine or mixed coniferous forests. Occasionally some species are found in lower or higher communities, including grassy areas in the juniper-piñon woodland, stands of sagebrush, or spuce and fir forest. The relative abundance of voles of the genus Microtus in the mid-Wisconsinan at PSC and their current absence there suggests greater effective moisture and possibly cooler summers than today. Cotton rats (Sigmodon spp.) also inhabit grasslands on or between southeastern Arizonan mountain ranges. These grazing rodents are generally intolerant of one another and when species of these two genera co-occur, one or the other may be competitively excluded (Baker 1971; Davis and Dunford 1987; Davis and Ward 1988). Thus it is interesting to find both genera occurring in the same cave fill units in PSC. Specimens of Microtus are far more common in PSC deposits overall than Sigmodon, but precise species identifications, more and better samples of both genera, and more radiometric age estimates would be necessary to learn anything meaningful about their relative distributional dynamics during the Pleistocene in southern Arizona. One of the few kinds of small mammals to have become extinct by the end of the Pleistocene, Aztlanolagus agilis is known from the late Pliocene (2.5 Ma; Blancan land mammal age) in southeastern Arizona, early Pleistocene (> 0.73 Ma; Irvingtonian land mammal age) in central Texas, and late Pleistocene (Rancholabrean land mammal age) in Chihuahua, Mexico, and New Mexico, USA. Harris (1991) thought Aztlanolagus

36

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

agilis to be a marker for mid-Wisconsinan time in southern New Mexico, where it is associated at several sites with radiometric dates between about 36 ka and 25 ka (Russell and Harris 1987). Unfortunately, the deposits from the PSC Annex are not dated, but they may at least partly date to the mid-Wisconsinan based on the presence of A. agilis and the abundance of Marmota flaviventri s. In summary, the faunal (and meager palynological) evidence suggests that the local area near PSC during at least the middle Wisconsinan glacial was wetter than at present, with perennial flowing water at the surface deep enough to support small fish. Vegetation at the time included open ponderosa pine forest and/or Madrean pineoak woodland with an understory of grasses, interspersed with open areas of grassland. Marmots utilized the limestone hill in which the cave occurs and other nearby rocky areas. Fourhorned pronghorns wandered through the grasslands and open woodlands and frequently visited the cave hill.

ACKNOWLEDGMENTS Field work was supported by grants from the National Geographic Society and the National Speleological Society to N. J. Czaplewski. Funds also were generously provided by M. A. Mares, Director of the Oklahoma Museum of Natural History. For help in field and lab work we thank T. Bethard, R. Bridgemon, C. D. Czaplewski, C. Force, P. Hagen, E. Horn, K. Horstman, D. Jacobs, B. Jones, C. Jones, S. Kennedy, W. May, J. McVickar, L. Murray, A. Pittenger, and S. Smith. We thank C. O. Minckley for identifying the fish specimen, R. S. Anderson for processing and examining pollen samples, E. H. Lindsay for access to PSC specimens in the University of Arizona Laboratory of Paleontology, M. A. Mares and J. K. Braun for use of the OMNH Recent mammal collection, and R. Pape for critically reading an earlier version of the manuscript. D. Schmidt of the National Museum of Natural History kindly measured tree squirrel specimens. Permits for working in PSC were provided by D. A. Bennett and helpful information by W. Gillespie, both of the USDA Forest Service.

LITERATURE CITED Allen, J. A. 1895. On a collection of mammals from Arizona and Mexico, made by Mr. W. W. Price, with field notes by the collector. Bulletin of the American Museum of Natural History, 7:193-258. American Ornithologists’ Union. Committee on

OCCASIONAL PAPERS

Classification and Nomenclature. 1983. Check-list of North American birds: The Species of Birds of North America from the Arctic through Panama, including the West Indies and Hawaiian Islands. 6th edition and supplement of July 1985. Washington, D.C.: The Union, 877 pp. Anderson, E. 1984. Review of the small carnivores of North America during the last 3.5 million years. In Contributions to Quaternary Vertebrate Paleontology: A Volume in Memorial to John E. Guilday. Genoways, H. H., and Dawson, M. R. (eds.), pp. 257-266. Pittsburgh: Carnegie Museum of Natural History Special Publication 8. Anderson, S. 1972. Mammals of Chihuahua: taxonomy and distribution. Bulletin of the American Museum of Natural History 148:149410. Armstrong, D. M. 1996. Northern limits of mammals of northern interior Mexico. In Contributions in Mammalogy : A Memorial Volume Honoring Dr. J. Knox Jones, Jr. Genoways, H. H., and Baker, R. J. (eds.), pp. 261-283. Lubbock: Museum of Texas Tech University. Auffenberg, W. 1963. The fossil snakes of Florida. Tulane Studies in Zoology 10:131-216. Baker, R. H. 1971. Nutritional strategies of myomorph rodents in North American grasslands. Journal of Mammalogy 52:800-805. Barnosky, A. D., and Rasmussen, D. L. 1988. Middle Pleistocene arvicoline rodents and environmental change at 2900-meters elevation, Porcupine Cave, South Park, Colorado. Annals of Carnegie Museum 57:267-292. Bogdanowicz, W., Kasper, S., and Owen, R. D. 1998. Phylogeny of plecotine bats: reevaluation of morphological and chromosomal data. Journal of Mammalogy 79:78-90. Bowers, J. E., and McLaughlin, S. P. 1995. Flora of the Huachuca Mountains, Cochise County, Arizona. In Biodiversity and Management of the Madrean Archipelago: The Sky Islands of Southwestern United States and Northwestern Mexico, General Technical Report RM-GTR-264. DeBano, L. F., Ffolliott, P. F., Ortega-Rubio, A., Gottfried, G. J., Hamre, R. H., and Edminster, C. B. (eds.), pp. 135-143. Fort Collins, Colorado, U.S. Department of Agriculture, Rocky Mountain Forest and Range Experiment Station. Brattstrom, B. H. 1955. Pleistocene lizards from San Josecito Cavern, Mexico, with description of a new species. Copeia 1955:133-134. Brodkorb, P. 1978. Catalog of fossil birds: Part 5 (Passeriformes). Bulletin of the Florida State Museum, Biological Sciences 23:139-228. Brown, D. E. (ed.). 1994. Biotic Communities Southwestern United States and Northwestern

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

Mexico. Salt Lake City: University of Utah Press, 342 pp. + map. Brown, D. E. 1984. Arizona’s Tree Squirrels. Phoenix: Arizona Game and Fish Department, 114 pp. Brown, D. E., and Davis, R. 1995. One hundred years of vicissitude: terrestrial bird and mammal distribution changes in the American Southwest, 1890-1990. In Biodiversity and Management of the Madrean Archipelago: The Sky Islands of Southwestern United States and Northwestern Mexico, General Technical Report RM-GTR-264. DeBano, L. F., Ffolliott, P. F., Ortega-Rubio, A., Gottfried, G. J., Hamre, R. H., and Edminster, C. B. (eds.), pp. 231-244. Fort Collins, Colorado, U.S. Department of Agriculture, Rocky Mountain Forest and Range Experiment Station. Brown, J. H. 1971. Mammals on mountaintops: nonequilibrium insular biogeography. The American Naturalist 105:467-478. Brown, J. H. 1978. The theory of insular biogeography and the distribution of boreal birds and mammals. Great Basin Naturalist Memoirs 2:209-227. Caire, W. 1978. Distribution and zoogeography of the mammals of Sonora, Mexico. Unpublished Ph. D. dissertation, University of New Mexico, Albuquerque. 613 pp. Caire, W., Vaughn, J. E., and Diersing, V. E. 1978. First record of Sorex arizonae (Insectivora: Soricidae) from Mexico. The Southwestern Naturalist 23:532-533. Carleton, M. D. 1989. Systematics and evolution. In Advances in the Study of Peromyscus (Rodentia). Kirkland, G. L., Jr., and Layne, J. N. (eds.), pp. 7141. Lubbock: Texas Tech University Press. Carraway, L. N. 1987. Analysis of characters for distinguishing Sorex trowbridgii from sympatric S. vagrans. Murrelet 68:29-30. Cifelli, R. L. 1996. Techniques for recovery and preparation of microvertebrate fossils. Oklahoma Geological Survey, Special Publication 96-4:1-36. Colbert, E. H., and Chaffee, R. G. 1939. A study of Tetrameryx and associated fossils from Papago Spring Cave, Sonoita, Arizona. American Museum Novitates 1034:1-21. Collins, J. T. 1990. Standard common and current scientific names for North American amphibians and reptiles, third edition. Society for the Study of Amphibians and Reptiles Herpetological Circular 19, 41 pp. Conway, M. C., and Schmitt, C. G. 1978. Record of the Arizona shrew (Sorex arizonae) from New Mexico. Journal of Mammalogy 59:631. Cook, J. A. 1986. The mammals of the Animas Mountains and adjacent areas, Hidalgo County,

37

New Mexico. Occasional Papers, Museum of Southwestern Biology 4:1-45. Cornely, J. E. 1979. Ecological distribution of woodrats (genus Neotoma) in Guadalupe Mountains National Park, Texas. In Biological Investigations in the Guadalupe Mountains National Park, Texas. Genoways, H. H., and Baker, R. J. (eds.), pp. 373-394. USDI, National Park Service, Proceedings and Transactions Series 4. Cornely, J. E., and Baker, R. J. 1986. Neotoma mexicana. Mammalian Species 262:1-7. Czaplewski, N. J., W. D. Peachey, J. I. Mead, T.-L. Ku, and C. J. Bell. 1999. Papago Springs Cave revisited, Part I; geologic setting, cave deposits, and radiometric dates. Occasional Papers of the Oklahoma Museum of Natural History, 3:1-25. Dalquest, W. W. 1967. Mammals of the Pleistocene Slaton local fauna of Texas. The Southwestern Naturalist 12:1-30. Dalquest, W. W., and Schultz, G. E. 1992. Ice Age Mammals of Northwestern Texas. Wichita Falls, Texas: Midwestern State University Press, 309 pp. Dalquest, W. W., Stangl, F. B., and Grimes, J. V. 1989. The third lower premolar of the cottontail, genus Sylvilagus , and its value in the discrimination of three species. The American Midland Naturalist 121:293- 301. Davis, R., and Brown, D. E. 1989. Role of postPleistocene dispersal in determining the modern distribution of Abert’s squirrel. Great Basin Naturalist 49:425-434. Davis, R., and Callahan, J. R. 1992. Post-Pleistocene dispersal in the Mexican vole (Microtus mexicanus): an example of an apparent trend in the distribution of Southwestern mammals. Great Basin Naturalist 52:262-268. Davis, R., and Dunford, C. 1987. An example of contemporary colonization of montane islands by small, nonflying mammals in the American Southwest. The American Naturalist 129:398-406. Davis, R., and Ward, O. G. 1988. A vacant “Microtus niche” now occupied by the yellow-nosed cotton rat (Sigmodon ochrognathus) on an isolated mountain in southeastern Arizona. Journal of Mammalogy 69:362-365. Diersing, V. E. and Hoffmeister, D. F. 1977. Revision of the shrews Sorex merriami and a description of a new species of the subgenus Sorex. Journal of Mammalogy 58:321-333. Emslie, S. D. 1985. The late Pleistocene (Rancholabrean) avifauna of Little Box Elder Cave, Wyoming. Contributions to Geology, University of Wyoming 23:63-82. Emslie, S. D., and Heaton, T. H. 1987. The late Pleistocene avifauna of Crystal Ball Cave, Utah. Journal of the Arizona-Nevada Academy of Science 21:53-60.

38

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

Faunmap Working Group. 1994. FAUNMAP: A database documenting late Quaternary distributions of mammal species in the United States. Springfield: Illinois State Museum Scientific Papers, 25:1-690. Faunmap Working Group. 1996. Spatial responses of mammals to late Quaternary environmental fluctuations. Science 272:1601-1606. Felger, R. S., and Johnson, M. B. 1995. Trees of the northern Sierra Madre Occidental and sky islands of southwestern North America. In Biodiversity and Management of the Madrean Archipelago: The Sky Islands of Southwestern United States and Northwestern Mexico, General Technical Report RM-GTR-264. DeBano, L. F., Ffolliott, P. F., Ortega-Rubio, A., Gottfried, G. J., Hamre, R. H., and Edminster, C. B. (eds.), pp. 71-83. Fort Collins, Colorado, U.S. Department of Agriculture, Rocky Mountain Forest and Range Experiment Station. Findley, J. S. 1969. Biogeography of southwestern boreal and desert mammals. University of Kansas Museum of Natural History, Miscellaneous Publication 51:113-128 Findley, J. S. 1996. Mammalian biogeography in the American Southwest. In Contributions in Mammalogy: A Memorial Volume Honoring Dr. J. Knox Jones, Jr. Genoways, H. H., and Baker, R. J., (eds.), pp. 297- 307. Lubbock: Museum of Texas Tech University,. Findley, J. S., Harris, A. H., Wilson, D. E., and Jones, C. 1975. Mammals of New Mexico. Albuquerque: University of New Mexico Press, 360 pp. Finley, R. B., Jr. 1958. The wood rats of Colorado. University of Kansas Publications, Museum of Natural History 10:213-552. Frase, B. A., and Hoffmann, R. S. 1980. Marmota flaviventris . Mammalian Species 135:1-8. Frick, C. 1937. Horned ruminants of North America. Bulletin of the American Museum of Natural History 49:1-669. Frost, D. R., and Timm, R. M. 1992. Phylogeny of plecotine bats (Chiroptera: “Vespertilionidae”): summary of the evidence and proposal of a logistically consistent taxonomy. American Museum Novitates 3034:1-16. George, S. B. 1988. Systematics, historical biogeography, and evolution of the genus Sorex. Journal of Mammalogy 69:443-461. Graham, R. W. 1986. Plant-animal interactions and Pleistocene extinctions. In Dynamics of Extinction. Elliott, D. K. (ed.), pp. 131-154. New York: John Wiley & Sons. Graham, R. W. 1991. Variability in the size of North American Quaternary black bears (Ursus americanus) with the description of a fossil black

OCCASIONAL PAPERS

bear from Bill Neff Cave, Virginia. In Beamers, Bobwhites, and Blue- points: Tributes to the Career of Paul W. Parmalee. Purdue, J. R., Klippel, W. E., and Styles, B. W. (eds.), pp. 237250. Springfield: Illinois State Museum Scientific Papers, vol. 23. Graham, R. W., and Semken, H. A., Jr. 1987. Philosophy and procedures for paleoenvironmental studies of Quaternary mammalian faunas. In Late Quaternary Mammalian Biogeography and Environments of the Great Plains and Prairies. Graham, R. W., Semken, H. A., Jr., and Graham, M. A. (eds.), pp. 1-17. Springfield: Illinois State Museum Scientific Papers, vol. 22. Grayson, D. K. 1987. The biogeographic history of small mammals in the Great Basin: observations on the last 20,000 years. Journal of Mammalogy 68:359-375. Grayson, D. K., and Livingston, S. D. 1993. Missing mammals on Great Basin mountains: Holocene extinctions and inadequate knowledge. Conservation Biology 7:527-532. Hammerson, G. A., and Smith, H. M. 1991. The correct spelling of the name for the short-horned lizard of North America. Bulletin of the Maryland Herpetological Society 27:121-127. Harris, A. H. 1984a. Neotoma in the late Pleistocene of New Mexico and Chihuahua. Special Publication, Carnegie Museum of Natural History 8:164-178. Harris, A. H. 1984b. Two new species of late Pleistocene woodrats (Cricetidae: Neotoma) from New Mexico. Journal of Mammalogy 65:560-566. Harris, A. H. 1985a. Late Pleistocene vertebrate paleoecology of the West. Austin: University of Texas Press, 293 pp. Harris, A. H. 1985b. Preliminary report on the vertebrate fauna of U-Bar Cave, Hidalgo County, New Mexico. New Mexico Geology 7:74-77. Harris, A. H. 1987. Reconstruction of mid-Wisconsin environments in southern New Mexico. National Geographic Research 3:142-151. Harris, A. H. 1989. The New Mexican late Wisconsin—east versus west. National Geographic Research 5:205-217. Harris, A. H. 1990a. Fossil evidence bearing on southwestern mammalian biogeography. Journal of Mammalogy 71:219-229. Harris, A. H. 1990b. Taxonomic status of the Pleistocene ringtail Bassariscus sonoitensis (Carnivora). The Southwestern Naturalist 35:343346. Harris, A. H. 1991. Preliminary report on the fauna of Pendejo Cave, Otero County, New Mexico. Current Research in the Pleistocene 8:92-93. Harris, A. H. 1993. Quaternary vertebrates of New

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

Mexico. New Mexico Museum of Natural History and Science Bulletin 2:179-197. Harris, A. H., and Findley, J. S. 1964. PleistoceneRecent fauna of the Isleta caves, Bernalillo County, New Mexico. American Journal of Science 262:114-120. Harris, A. H., and Porter, L. S. W. 1980. Late Pleistocene horses of Dry Cave, Eddy County, New Mexico. Journal of Mammalogy 61:46-65. Hoffmeister, D. F. 1986. Mammals of Arizona. Tucson: University of Arizona Press and Arizona Game and Fish Department, 602 pp. Hoffmeister, D. F., and Goodpaster, W. W. 1954. The mammals of the Huachuca Mountains, southeastern Arizona. Illinois Biological Monographs 24:1-152. Hoffstetter, R., and Gasc, J.-P. 1969. Vertebrae and ribs of modern reptiles. In Biology of the Reptilia, Volume 1, Morphology A. Gans, C., Bellairs, A. d’A., and Parsons,T. R. (eds.), pp. 201-310. New York: Academic Press. Holman, J. A. 1961. Osteology of living and fossil New World quails (Aves, Galliformes). Bulletin of the Florida State Museum, Biological Sciences 6:131-233. Holman, J. A. 1969. Herpetofauna of the Pleistocene Slaton local fauna of Texas. The Southwestern Naturalist 14:203-212. Holman, J. A. 1981. A review of North American Pleistocene snakes. Publications of the Museum, Michigan State University (Paleontological Series), 1:261-306. Jakway, G. E. 1958. Pleistocene Lagomorpha and Rodentia from the San Josecito Cave, Nuevo León, Mexico. Transactions of the Kansas Academy of Science 61:313-327. Jones, T.R., Routman, E.J., Begun, D. J., and Collins, J.P. 1995. Ancestry of an isolated subspecies of salamander, Ambystoma tigrinum stebbinsi Lowe: the evolutionary significance of hybridization. Molecular Phylogenetics and Evolution 4:194-202. Junge, J. A., and Hoffmann, R. S. 1981. An annotated key to the long-tailed shrews (genus Sorex) of the United States and Canada, with notes on Middle American Sorex. Occasional Papers of the Museum of Natural History, University of Kansas 94:1-48. Kurtén, B., and Anderson, E. 1980. Pleistocene Mammals of North America. New York: Columbia University Press, 442 pp. LaDuke, T. C. 1991a. The fossil snakes of Pit 91, Rancho La Brea, California. Natural History Museum of Los Angeles County Contributions in Science, 424:1-28. LaDuke, T. C. 1991b. Morphometric variability of the precaudal vertebrae of Thamnophis sirtalis sirtalis

39

(Serpentes: Colubridae), and implications for interpretation of the fossil record. Unpublished Ph.D. thesis, City University of New York, 235 pp. Lindsay, E. H., and Tessman, N. T. 1974. Cenozoic vertebrate localities and faunas in Arizona. Journal of the Arizona Academy of Science 9:3-24. Logan, L. E. 1983. Paleoecological implications of the mammalian fauna of Lower Sloth Cave, Guadalupe Mountains, Texas. National Speleological Society Bulletin 45:3-11. Lomolino, M. V., Brown, J. H., and Davis, R. 1989. Island biogeography of montane forest mammals in the American Southwest. Ecology 70:180-194. Long, C. A. 1973. Taxidea taxus. Mammalian Species 26:1-4. Lynch, J. D. 1986. The definition of the Middle American clade of Eleutherodactylus based on jaw musculature (Amphibia: Leptodactylidae). Herpetologica 42:248-258. Macêdo, R. H., and Mares, M. A. 1988. Neotoma albigula. Mammalian Species 310:1-7. Mares, M. A., and Lacher, T. E., Jr. 1987. Ecological, morphological, and behavioral convergence in rock- dwelling mammals. In Current Mammalogy vol. 1. Genoways, H. H. (ed.), pp. 307-348. New York: Plenum Publishing. Martin, P. S. 1961. Southwestern animal communities in the late Pleistocene. In Bioecology of the Arid and Semiarid Lands of the Southwest. Shields, L. M., and Gardner, L. J., (eds.), pp. 56-66. New Mexico Highlands University Bulletin 212. Martin, R. A. 1974. Fossil mammals from the Coleman IIA fauna, Sumter County. In Pleistocene Mammals of Florida. Webb, S. D. (ed.), pp. 35-99. Gainesville: University Presses of Florida. Martin, R. A. 1989. Arvicolid rodents of the early Pleistocene Java local fauna from north-central South Dakota. Journal of Vertebrate Paleontology 9:438-450. McGuire, J. A. 1996. Phylogenetic systematics of crotaphytid lizards (Reptilia: Iguania: Crotaphytidae). Bulletin of Carnegie Museum of Natural History 32:1-143. McLaughlin, S. P. 1995. An overview of the flora of the sky islands, southeastern Arizona: diversity, affinities, and insularity. In Biodiversity and Management of the Madrean Archipelago: The Sky Islands of Southwestern United States and Northwestern Mexico, General Technical Report RM-GTR-264. DeBano, L. F., Ffolliott, P. F., Ortega-Rubio, A., Gottfried, G. J., Hamre, R. H., and Edminster, C. B. (eds.), pp. 60-70. Fort Collins, Colorado, U.S. Department of Agriculture, Rocky Mountain Forest and Range Experiment Station.

40

CZAPLEWSKI, MEAD, BELL, PEACHEY, AND KU

Mead, J. I. and Phillips, A. M. 1981. The late Pleistocene and Holocene fauna and flora of Vulture Cave, Grand Canyon, Arizona. The Southwestern Naturalist 26:257-288. Mead, J. I., Arroyo-Cabrales, J., and Johnson, E. 1999. Pleistocene lizards (Reptilia: Squamata) from San Josecito Cave, Nuevo León, México. Copeia 1999(1):163-173. Mead, J. I., Roth, E. L., Van Devender, T. R., and Steadman, D. W. 1984. The late Wisconsinan vertebrate fauna from Deadman Cave, southern Arizona. Transactions of the San Diego Society of Natural History 20:247-276. Mead, J. I., Van Devender, T. R. and Cole, K. L. 1983. Late Quaternary small mammals from Sonoran Desert packrat middens, Arizona and California. Journal of Mammalogy 64:173-180. Menú, H. 1987. Morphotypes dentaires actuels et fossiles des chiroptères vespertilioninés. 2eme partie: Implications systématiques et phylogéniques. Palaeovertebrata, Montpellier 17:77-150. Messing, H. J. 1986. A late Pleistocene-Holocene fauna from Chihuahua, Mexico. The Southwestern Naturalist 31:277-288. Miller, L. H. 1943. The Pleistocene birds of San Josecito Cavern, Mexico. University of California Publications in Zoology 47:143-168. Miller, M. D., Jr. 1992. Analysis of fossil salamanders from Cheek Bend Cave, Maury County, Tennessee. Unpublished M.S. thesis, Appalachian State University. 99 pp. Miller, R. R., and Smith, G. R. 1984. Fish remains from Stanton’s Cave, Grand Canyon of the Colorado, Arizona, with notes on the taxonomy of Gila cypha. In The Archaeology, Geology, and Paleobiology of Stanton’s Cave, Grand Canyon National Park Arizona. Euler, R. C. (ed.), pp. 5965. Grand Canyon Natural History Association, Monograph No. 6. Minckley, W. L. 1973. Fishes of Arizona. Phoenix: Arizona Game and Fish Department, 293 pp. Patterson, B. D. 1980. Montane mammalian biogeography in New Mexico. The Southwestern Naturalist 25:33-40. Patterson, B. D. 1981. Morphological shifts of some isolated populations of Eutamias (Rodentia: Sciuridae) in different congeneric assemblages. Evolution 35:53-66. Patterson, B. D. 1984. Mammalian extinction and biogeography in the southern Rocky Mountains. In Extinctions. Nitecki, M. H. (ed.), pp. 247-293. Chicago: University of Chicago Press. Patterson, B. D. 1995. Local extinctions and the biogeographic dynamics of boreal mammals in the Southwest. In Storm over a Mountain Island:

OCCASIONAL PAPERS

Conservation Biology and the Mount Graham Affair. Istock, C. A., and Hoffmann, R. S. (eds.), pp. 151-176. Tucson, University of Arizona Press. Patton, J. L. 1993. Family Heteromyidae. In Mammal Species of the World. A Taxonomic and Geographic Reference. Second edition. Wilson, D. E., and Reeder, D. M. (eds.), pp. 477-486. Washington, D. C.: Smithsonian Institution Press. Pinsof, J. D. 1996. Current status of North American Sangamonian local faunas and vertebrate taxa. In Palaeoecology and Palaeoenvironments of Late Cenozoic Mammals: Tributes to the Career of C. S. (Rufus) Churcher. Stewart, K. M., and Seymour, K. L. (eds.), pp. 156-190. Toronto: University of Toronto Press. Rea, A. M. 1980. Late Pleistocene and Holocene turkeys in the Southwest. In Papers in Avian Paleontology Honoring Hildegard Howard. Campbell, K. E., Jr. (ed.), pp.209-224. Los Angeles: Contributions in Science, Natural History Museum of Los Angeles County 330. Roosevelt, Q., and Burden, J. W. 1934. A new species of antilocaprine, Tetrameryx onusrosagris, from a Pleistocene cave deposit in southern Arizona. American Museum Novitates 754:1-4. Russell, B. D., and Harris, A. H. 1987. A new leporine (Lagomorpha: Leporidae) from Wisconsinan deposits of the Chihuahuan Desert. Journal of Mammalogy 67:632-639. Severinghaus, W. D., and Hoffmeister, D. F. 1978. Qualitative cranial characters distinguishing Sigmodon hispidus and Sigmodon arizonae and the distribution of these two species in northern Mexico. Journal of Mammalogy 59:868-870. Simons, L. H., Szaro, R. C., and Belfit, S. C. 1990. Distribution of Notiosorex crawfordi and Sorex arizonae along an elevational gradient. Journal of Mammalogy 71:634-640. Skinner, M. F. 1942. The fauna of Papago Springs Cave, Arizona, and a study of Stockoceros; with three new antilocaprines from Nebraska and Arizona. Bulletin of the American Museum of Natural History 80:143-220. Skinner, M. F. 1942. The fauna of Papago Springs Cave, Arizona, and a study of Stockoceros; with three new antilocaprines from Nebraska and Arizona. Bulletin of the American Museum of Natural History 80:143-220. Skinner, M. F. 1972. Order Perissodactyla. In Early Pleistocene pre-glacial and glacial rocks and faunas of north-central Nebraska. Skinner, M. F., and Hibbard, C. W. (eds.), pp. 117-130. Bulletin of the American Museum of Natural History 148:1-148. Slaughter, B. H. 1964. An ecological interpretation of the Brown Sand Wedge local fauna, Blackwater Draw, New Mexico, and a hypothesis concerning

Number 5

PAPAGO SPRINGS CAVE, VERTEBRATE PALEOFAUNA

late Pleistocene extinction. Paleoecology of the Llano Estacado 2:1-37. Smartt, R. A. 1977. The ecology of late Pleistocene and Recent Microtus from south-central and southwestern New Mexico. The Southwestern Naturalist 22:1-19. Smith, G. R. 1981. Late Cenozoic freshwater fishes of North America. Annual Review of Ecology and Systematics 12:163-193. Steadman, D. W. 1980. A review of the osteology and paleontology of turkeys (Aves: Meleagridinae). In Papers in Avian Paleontology Honoring Hildegard Howard. Campbell, K. E., Jr. (ed.), pp. 131-207. Los Angeles: Contributions in Science, Natural History Museum of Los Angeles County 330. Steadman, D. W., Arroyo-Cabrales, J., Johnson, E., and Fabiola Guzman, A. 1994. New information on the late Pleistocene birds from San Josecito Cave, Nuevo León, Mexico. Condor 96:577-589. Stebbins, R. C. 1985. A field guide to western reptiles and amphibians. Boston: Houghton Mifflin Company, 334 pp. Sullivan, R. M. 1985. Phyletic, biogeographic, and ecologic relationships among montane populations of least chipmunks (Eutamias minimus) in the Southwest. Systematic Zoology 34:419-448. Svendsen, G. E. 1982. Weasels. In Wild Mammals of North America. Chapman, J. A., and Feldhamer, G. W. (eds.), pp. 613-628. Baltimore: Johns Hopkins Press. Tihen, J. A. 1958. Comments on the osteology and phylogeny of ambystomatid salamanders. Bulletin of the Florida State Museum 3:1-50. Tomida, Y. 1987. Small mammal fossils and correlation of continental deposits, Safford and Duncan basins, Arizona, USA. Tokyo: National Science Museum, 141 pp. Tumlison, R., and Douglas, M. E. 1992. Parsimony

41

analysis and the phylogeny of the plecotine bats (Chiroptera: Vespertilionidae). Journal of Mammalogy 73:276-285. Van Devender, T. R., and Mead, J. I. 1978. Early Holocene and late Pleistocene amphibians and reptiles in Sonoran Desert packrat middens. Copeia 1978:464-475. Van Devender, T. R., Mead, J. I., and Rea, A. M. 1991. Late Quaternary plants and vertebrates from Picacho Peak, Arizona. The Southwestern Naturalist 36:302-314. Van Devender, T. R., Phillips, A. M., and Mead, J. I. 1977. Late Pleistocene reptiles and small mammals from the lower Grand Canyon of Arizona. The Southwestern Naturalist 22:49-66. Van Devender, T. R., Rea, A. M., and Smith, M. L. 1985. The Sangamonian interglacial vertebrate fauna from Ranch la Brisca, Sonora, Mexico. Transactions of the San Diego Society of Natural History 21:23-55. Wallmo, O. C. 1955. Vegetation of the Huachuca Mountains, Arizona. American Midland Naturalist 54:466-480. Williams, D. F. 1978. Systematics and ecogeographic variation of the Apache pocket mouse (Rodentia: Heteromyidae). Bulletin of the Carnegie Museum of Natural History 10:1-57. Wilson, D. E., and Reeder, D. M. 1993. Mammal Species of the world. A Taxonomic and Geographic Reference. Second edition. Washington, D.C.: Smithsonian Institution Press, 1206 pp. Winkler, A. J., and Tomida, Y. 1988. New records of the small leporid Aztlanolagus agilis Russell and Harris (Leporidae: Leporinae). The Southwestern Naturalist 33:391-396. Young, S. P., and Goldman, E. A. 1946. The Puma. Mysterious American Cat. American Wildlife Institute (New York: Dover reprint 1964), 358 pp.

OKLAHOMA MUSEUM OF NATURAL HISTORY PUBLICATIONS

The Oklahoma Museum of Natural History publishes books, monographs, and occasional papers based on original research by professionals affiliated with the museum. Authors should contact the editor for a copy of instructions to authors. All prospective papers are reviewed both internally and externally by experts in the appropriate field. Acceptance is based on evaluations by reviewers and the editors: final acceptance is contingent on approval by the Director. This publication is printed on acid-free paper. All formatting is done on Macintosh computers using FrameMaker® and all manuscripts are copyrighted by the Oklahoma Museum of Natural History. Institutional libraries may obtain the Occasional Papers by writing the editor at the Oklahoma Museum of Natural History, 2401 Chautauqua, Norman, Oklahoma 730727029. Individuals may purchase copies of this publication for $2.00 by writing the editor. Reprint requests should be directed to the author(s).

Editor: Janalee P. Caldwell Associate Editor: Janet K. Braun Production Editor: Laurie J. Vitt

PRINTED BY THE OKLAHOMA MUSEUM OF NATURAL HISTORY 2401 CHAUTAUQUA

Norman, Oklahoma 73072-7029 USA