Jump to ContentJump to Main Navigation
Biodiversity Response to Climate Change in the Middle PleistoceneThe Porcupine Cave Fauna from Colorado$

Anthony Barnosky

Print publication date: 2004

Print ISBN-13: 9780520240827

Published to California Scholarship Online: March 2012

DOI: 10.1525/california/9780520240827.001.0001

Show Summary Details
Page of

PRINTED FROM CALIFORNIA SCHOLARSHIP ONLINE (www.california.universitypressscholarship.com). (c) Copyright University of California Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in CALSO for personal use. Subscriber: null; date: 17 September 2019

Biology of Wood Rats as Cave Dwellers and Collectors

Biology of Wood Rats as Cave Dwellers and Collectors

Chapter:
(p.74) Eight Biology of Wood Rats as Cave Dwellers and Collectors
Source:
Biodiversity Response to Climate Change in the Middle Pleistocene
Author(s):

Robert B. Finley Jr.

Publisher:
University of California Press
DOI:10.1525/california/9780520240827.003.0008

Abstract and Keywords

Wood rats, genus Neotoma, are terrestrial rodents of the family Muridae. This chapter describes the basic ecology of bushytailed wood rats, Neotoma cinerea, and how they leave evidence of their activities in caves. Porcupine Cave exhibits signs of current wood rat activity in the entrance adit, Badger Room, Gypsum Room, Pit, Velvet Room, and Damp Room. In fact, some floors are carpeted with fecal pellets, plant litter, and urine deposits, including, in many places, great quantities of small bones mixed with sediment.

Keywords:   wood rats, Neotoma, Muridae, Neotoma cinerea, Badger Room, Gypsum Room, Pit, Velvet Room, Damp Room

Wood rats, genus Neotoma, also widely known to speleologists as packrats, are commonly found in caves within their range, but they are primarily adapted for life outside caves. They are terrestrial rodents of the family Muridae, subfamily Sigmo-dontinae. The center of abundance and diversity of Neotoma is in the arid southwestern United States and Mexico. Since the diet of Neotoma consists mainly of foliage and fruits, they must forage outside caves for food, much of which is carried inside for use or storage. The abundant fragments of plant materials, predator scats, and owl pellets dropped on the middens and rock shelves provide a good record of ecological communities and of the climate changes that have occurred outside the dens or caves over thousands of years. Indeed a major account of biotic changes in the Southwest is based entirely on remains of plants and animals recovered from middens of species of Neotoma (Betancourt et al., 1990).

Porcupine Cave exhibits signs of current wood rat activity in the entrance adit, Badger Room, Gypsum Room, Pit, Velvet Room, and Damp Room. In fact, fecal pellets, urine deposits, or nests attest to wood rats’ use of most of the several hundred meters of mapped passageway. Fecal pellets were found mixed with small bones in many of the matrix samples from various levels in Porcupine Cave. Cave explorers commonly report seeing wood rats in caves, sometimes far from any entrance (chapter 2; D. L. Rasmussen, pers. comm.).

Colorado’s pioneer mammalogist, E. R. Warren (1910:117), wrote as follows: “I have found them in abandoned mine tunnels, about the timbers, and in drifts. I took one in an abandoned tunnel near Querida, Custer County, at a point about 225 feet in from the entrance; it had a nest there, and there was another nest in a drift close by.” According to Cary (1911:113), “Miners at the Stevens Mill, at timberline on Mt. McClellan, reported a few rats living in the mine several hundred feet from the entrance, and stated that the animals often passed them on the ladders.” The greatest distance from the surface I have seen reported for N. cinerea within a cave is that cited by Turner (1974:103): “I have observed N. cinerea in Wind Cave [South Dakota] at depths of 325 feet below ground surface, and 2300 feet from any known entrance of the cave.”

Bailey (1928) reported that both N. albigula and N. micropus were present at Carlsbad Cavern, N. albigula being abundant in caves and canyon walls and in the great western entrance but not below the brink of the shaft that drops 30 m into the deep interior. N. micropus occupies stick and cactus houses in the open valley for shelter. Bailey (1933) reported that Neotoma pennsylvanica (now known as N. floridana magister) was common in the limestone cliffs and in most of the caves of the Mammoth Cave region of Kentucky, where individuals followed the whole length of extensive passages.

The only species of wood rat living today in Porcupine Cave and vicinity is the bushy-tailed wood rat, Neotoma cinerea (figure 8.1). Within the genus, it is the species most highly specialized for living in cliffs and caves. It is found only at higher elevations in western North America and is superbly adapted to climbing vertical cliffs. The deep rock fissures and caves it occupies serve as heat sinks that stabilize air temperatures within its dens in both winter and summer. In Porcupine Cave some floors are carpeted with pellets and plant litter, including in many places great quantities of small bones mixed with sediment. The wood rats have also left on walls and rock ledges solid dark deposits and urine stains widely known as amberat.

During the Irvingtonian, at least five extant species of wood rats lived around Porcupine Cave at various times: N. cinerea, N. mexicana, N. floridana, N. micropus, and N. stephensi (chapter 18). This chapter describes the basic ecology of bushy-tailed wood rats, N. cinerea, how they differ in ecology from their congeners, and how they leave evidence of their activities in caves.

Definitions

The literature on wood rat dens and middens is often sketchy and ambiguous. For example, words used by hunters, naturalists, and cavers to describe the “sign” at a den or cave entrance have different meanings to different people. I defined my use of such terms, as widely understood by mammalogists, in Finley (1990:28), and I repeat and augment the definitions here.

(p.75)

Biology of Wood Rats as Cave Dwellers and Collectors

Figure 8.1 Bushy-tailed wood rat, Neotoma cinerea, the species of wood rat living in Porcupine Cave today.

In this chapter, I follow the convention of using “wood rat” as the common name for Neotoma, in keeping with standardized use in the mammalogy literature. “Wood rat” as used here is synonymous with the more vernacular “packrat”.

CACHE A quantity of food materials piled or stuffed into a sheltered space, such as a crevice, log, or burrow, for later use.

DEN Any natural or constructed shelter used regularly for protected living space, rest, or the rearing of young, such as a rock crevice.

FECAL DEPOSIT A layer consisting mainly of fecal droppings that may or may not be indurated by dried urine.

FOOD LITTER Food materials brought to a place for feeding and dropped as unconsumed debris.

HOUSE A den constructed of a pile of various materials and containing one or more passages or chambers. It may be located in the open or at a site under some natural shelter, such as a tree.

MIDDEN A layer or accumulation of materials brought to or dropped at any location by an animal, usually but not necessarily the den occupant and including any other materials, such as leaves, pollen, and silt, deposited there by other means.

NEST A cup or ball of soft material, such a shredded juniper bark, within a house or other den, used as a bed.

PALEOMIDDEN A “fossil” midden, which has persisted long enough to survive environmental changes but has not necessarily been altered over time.

POST A site used by an animal repeatedly for some function, such as feeding, urination, or scent deposition. Posts may give rise to deposits, such as food litter found either within or away from dens. Urination posts on rock ledges or cave walls often give rise to conspicuous dry stains (amberat) on walls or under ledges sheltered from the weather.

SIGN Any direct physical evidence of the animal activity that produced it.

URINARY DEPOSIT A layer of dried urine, usually including some fecal pellets or other materials.

The term “amberat” has been used in many different ways. I have not customarily used the term because of its ambiguity. The oldest reference I have found for “amberat” is by Orr (in litt., 1957, cited in Jackson, 1965). He called it “amberat” because “It has no proper name, but smells of rats” but he was uncertain of its origin. Jackson (1965) showed a photograph of two chunks of material from Defense Cave. One was a rough chunk of indurated rat pellets and plant litter that he called “ancient cave rat guano.” The other was a glossy black, solid deposit that he called “amberat”.

Paleoecologists Spaulding et al. (1990:60–63) called amberat the “crystallized packrat urine” that saturates and preserves solid plant and other materials. In their lengthy discussion of the “Physical Properties of Middens” they seem to regard the term as synonymous with urine. “Amberat” is also sometimes used to refer to the glossy urine deposits that resemble a shiny black curtain on a cliff face.

Ecological Attributes of Neotoma Species

Six species of Neotoma are extant in Colorado: N. cinerea, N. mexicana, N. floridana, N. albigula, N. micropus, and N. lepida (Finley, 1958). They live in different parts of the state such that rarely more than three species are sympatric. The most widespread is N. cinerea, which lives throughout the higher mountains. N. mexicana is found in the foothill zone on both the east and west slopes and occurs only about 700 m lower in elevation than Porcupine Cave. N. floridana lives on the Great Plains north of the Arkansas River, and N. micropus lives on the plains primarily south of the river. N. albigula lives at low elevations in southwestern and southeastern Colorado. N. lepida, having a range west of the Colorado River throughout the Southwest, extends into Colorado only in the lower drainages of northwest Colorado (Finley, 1958). Here I accept the analysis of Hoffmeister (1986), who showed that devia is a subspecies of N. lepida, not a separate species of Neotoma. Of the species currently known from Colorado, only N. albigula and N. lepida have not been found in Porcupine Cave (chapter 18).

Houses, rock dens, and caves provide protection for wood rats against adverse weather (Brown, 1968) as well as most predators. Nevertheless, wood rats do fall prey to their enemies when they emerge from their dens to forage for food or gather den materials. No doubt this danger accounts for the caution and speed with which wood rats usually move from one bit of outdoor cover to another (Finley, pers. obs.). Wood rats (p.76) generally forage within 30 m of the den entrance (Finley, 1990), probably to minimize danger from predation.

Wood rats of different species live in various kinds of dens. Some use rock crevices and fissures, in which varying amounts of materials may be piled. Others build houses of sticks or other materials in or over a base of shrubs or cactus. Tunnels and chambers within the house usually extend below ground and may be more extensive than those within the house. Rock-dwelling species, such as N. cinerea (Smith, 1997) and N. mexicana (Cornely and Baker, 1986), may pile few or many sticks and other materials at the entrance to a rock fissure, but the stick pile rarely forms a house with passages and chambers (Finley, 1958). The species that form large houses, such as N. fuscipes in California, are indefatigable collectors and continue to add sticks to their houses, sometimes to a height of 2 m (Linsdale and Tevis, 1951; Carraway and Verts, 1991). At the other extreme, N. mexicana has a weak collecting habit, and its small dens under rock ledges may be easily overlooked because of the few collected materials other than long twigs and food litter. N. floridana and N. albigula may live in either kind of den, depending on available cover and the presence of other species (Rainey, 1956; Finley, 1958; Wiley, 1980).

In the recent past some rock shelters have been occupied by a different species of wood rat than the one found in them today (Finley, pers. obs.). In the case of fossil deposits, extinct species of Neotoma may have been present. Nine extinct species have been proposed for Neotoma from the Blan-can or Irvingtonian: N. quadriplicata, N. taylori, N. spelaea, and N. fossilis (Kurtén and Anderson, 1980), and N. amblidonta, N. ozarkensis, N. findleyi, N. pygmaea, and N. magnodonta (Repenning, in litt.).

Climbing ability varies with the habitat and kind of den. N. cinerea is an agile cliff climber and readily scrambles up cliffs (Finley, pers. obs.) and over the walls of caves. N. fuscipes is an excellent tree climber and sometimes builds its bulky houses among tree limbs several feet above the ground (Carraway and Verts, 1991). N. albigula uses rock dens in some areas and can also build well-fortified houses of cactus joints (Voorhies and Taylor, 1940). It readily climbs for escape to the top of cholla cactus. Both N. floridana and N. micropus may use rock dens, build stick houses, or live in burrows under a thicket (Finley, 1958; Braun and Mares, 1989).

Species of wood rats have diversified in food preferences as well as kinds of dens. In general, the species that occur in more boreal or mesic life zones, such as N. cinerea, N. mexicana, and N. floridana, depend primarily on the leaves of woody plants and forbs. N. cinerea eats conifer needles as a substantial part of its diet, whereas N. floridana makes little use of conifers, eating mainly a wide variety of broad-leaved plants, cacti, and spiny shrubs (Rainey, 1956; Finley, 1958). Neotoma stephensi is a dietary specialist depending mostly on juniper (Juniperus monosperma) for food (Vaughn, 1982; Jones and Hil-dreth, 1989; Dial and Czaplewski, 1990). The wood rats of more austral life zones or xeric habitats, such as N. micropus, N. albigula, and N. lepida, depend more on the pulp of many kind of cactus and succulent shrubs, which they consume in large quantities for water as well as food. Some species of cholla and prickly pear (Opuntia spp.) are also heavily used for house building and defense against intruders.

In Colorado three species, Neotoma cinerea, N. mexicana, and N. albigula, display ecological as well as morphological divergence within each species on the east and west sides of the state. All three live in divergent habitats and in some areas sympatrically with the other two species (Finley, 1958). In the Southwest two or three species of Neotoma often occur at the same locality in the same or different habitats. In Grand Canyon National Park there are five species: three are confined to the South Rim (N. mexicana, N. stephensi, and N. al-bigula), one is confined to the North Rim (N. cinerea), and one is found in the canyon on both sides of the Colorado River (N. lepida) (Hoffmeister, 1971). These conditions illustrate the difficulty of trying to ascribe environmental conditions or communities to types of wood rats. Where more than one species of Neotoma occurs at the same locality, the species are usually separated by different shelter types or narrow vegetative types, not by broad community types such as grassland or sagebrush. However, some species are confined to different climatic limits of temperature and aridity, irrespective of vegetative communities. A brief summary of the ecological diversity of wood rat species in the United States was given by Finley (1990:32).

Taphonomy and Interpretation of Packrat Sign

Wood rats collect and accumulate at their dens great quantities of sticks, bones, owl pellets, carnivore scats, and almost anything they can carry. In caves and rock shelters these are protected for hundreds or thousands of years. Neotoma can thus be a valuable taphonomic agent for preservation of cave deposits and their bone remains.

Donald L. Rasmussen recognized the richness and nature of the fossil deposits at Porcupine Cave and the unique taphonomy that depended so much on wood rats (Barnosky and Rasmussen, 1988; Rasmussen and Anderson, 1996). In searching for other cave entrances along the dolomite ridges of southern Park County, he scanned the rock overhangs and fissures, which were sometimes located where the wood rats had access into the karst fissures of cave systems. The Trailside Nest (a misnomer—it is not a true nest) near the adit entrance to Porcupine Cave is one such den; it was active as far back as 2000 years ago (Barnosky and Rasmussen, 1988). Another such den was found near the base of the next ridge, 1 km to the west.

Owls have long been recognized by mammalogists as among the best sources of information on the kinds of small mammals at any locality (Finley, 1954), and packrat middens may be almost as useful because they commonly contain both raptor pellets and carnivore scats that the wood rats collected. Owl pellets contain the undigested fur, bones, and teeth of their prey as regurgitated by the owls, and thus provide a widely used record of owl food habits (Craighead and Craighead, 1969 [1956]). Faunal data from unconsolidated wood rat (p.77) middens are comparable to those from owl pellets and may reveal the presence of rare or extralocal species. For example, in July 1990 I found the skull of a black-footed ferret (Mustela nigripes) at a den of N. cinerea in Saguache County, Colorado, where no ferrets had been recorded in decades. The skull was partially buried in a crevice of a rock den with loose soil and midden materials. The upland above the lava rimrock had soil too shallow to support the burrows of prairie dogs (Cyno-mys), the prey of ferrets. The ferret had probably been captured by a raptor at the nearest prairie dog town in the valley, about 3 km distant, and carried to a nest or roost in a pine tree below the rimrock near the wood rat den, probably a few decades ago.

The middens at modern dens of N. cinerea sometimes include larger bones, up to the size of the cannon bone of a deer, collected and piled with sticks at the entrance to a rock den. Coyotes, bobcats, and other carnivores that have their dens in cave openings also carry in the carcasses of their kills to be eaten. The smaller bones may then be carried farther into the cave by wood rats. Wood rats also collect scats from along the ridge on which they live, as well as regurgitated owl pellets from beneath nests on ledges and trees. The owl pellets and scats of predators in old cave deposits disintegrate, leaving the bones mixed with the sediment. In some cases the bones, or the pellets and scats themselves, may be fluvially redeposited within the cave and left in stratified deposits. However, no such stratified deposits that would indicate substantial reworking have been found thus far in Porcupine Cave.

This complex taphonomic history means that the radius of possible occurrence of a fossil animal may be as much as several kilometers and a few hundred meters of elevation from the cave entrance, depending on local topography. Hadly (1999) estimated that most remains in the deposits of Lamar Cave, Wyoming, came from within a radius of about 5 km. The seasonal hunting ranges of many species of hawks and owls are within 5 km (Craighead and Craighead, 1969 [1956]), and the ranges of golden eagles are somewhat larger (Mc-Gahan, 1967). Communities currently within a few kilometers of Porcupine Cave include montane coniferous forest, montane grassland, rock outcrops and ridges with bristlecone and limber pines, shortgrass prairie (grazed), riparian shrubs and forbs, small streams, peat bog, and stock ponds (see Cooper, chapter 3, for more details). Many mammals and birds are restricted more by fidelity to topographic type than by fidelity to vegetative type. Such considerations restrict the possible ecological implications of the presence of any species found in the bone deposits. Species from diverse habitats found within a radius of a few kilometers are represented in the cave samples because of varied terrain and vegetation. Species that are not topographically limited no doubt moved higher or lower with climate changes, and others may have moved within range. These considerations must be kept in mind when using faunal remains from caves as indicators of climate or community dynamics.

Time averaging, that is, the stratigraphic mixing that associates as fossils animals that actually lived at different times, can further obscure the fine details of community composition as communities change with time. In deposits in Lamar Cave similar in taphonomy and thickness to those of the Pit in Porcupine Cave, time averaging per level ranged from 300 to 1400 years (Hadly, 1999). However, the comparability of the fossil “community” with that of the neoecologist must be taken into account. The community of the latter is often of such small dimension or so patchy as to contain a few characteristic vertebrates and many that range over several wider communities. Many carnivores and raptors are highly opportunistic in hunting and frequently move from one locality and prey species to another (Craighead and Craighead, 1969 [1956]). Better communication among neoecologists and paleontologists would result if unambiguous terms were used to specify the time-averaged fauna of fossil assemblages versus the neoecologists’ concept of a community.

In order to discern how faunal data from modern middens compare with records of the local fauna obtained by trapping and observation at each den at the time the den was studied, I compiled my records of associated vertebrates from 159 dens of four species of Neotoma in Colorado. Data were collected during the years 1948–51. The results are shown in tables 8.18.4 for 57 dens of N. cinerea, 51 dens of N. mexicana, 39 dens of N. albigula, and 12 dens of N. floridana. Traces of associated vertebrates include bones of mammals and reptiles in the un-consolidated midden, small mammals trapped at the den, and small mammals and reptiles seen or captured at the den. Observations of birds were not included, although bones of some larger birds were recorded. All remains appeared to be from recent years.

The data show not only the great variety of vertebrates represented, but also that the wood rats effectively sample a diversity of carnivores and ungulates not easily recorded by brief survey. In wood rat houses, particularly, many bones of small commensal mammals and reptiles were found, that is, bones from animals that were attracted to the houses for shelter and that died there. In California and Oregon, the large houses of N. fuscipes are especially rich in vertebrate and invertebrate commensals (Carraway and Verts, 1991). Bones of such cohabitants are less likely to be found in rock dens, which cannot be dismantled for study, but they may account for many of the small bones found in caves. The bone samples, however, cannot be considered to represent precise numbers of individuals. Plant remains, either as food litter or as den materials from dens of different species, may reveal differences in the life zones and communities where the wood rats lived. Plant parts are usually gathered within 30 m of the den (Finley, 1990:35).

Conclusions

The location and geological conditions at Porcupine Cave have combined to provide a unique, rich fossil record of early and middle Pleistocene faunas that have been preserved and sampled largely thanks to the ecological habits of the bushy-tailed wood rat. The efforts of a great many dedicated people have made possible the recovery and study of the wealth of information contained in the cave deposits. Correct interpretation of the data has depended on understanding the complex biology and history of these deposits, as well as the geology of the cave itself. This review of the major role of wood rats in cave taphonomy puts into perspective the favorable factors, as well as the limitations of interpretation, inherent in the biology of wood rats. The most important among the favorable factors are the collecting and cave-dwelling behaviors of Neotoma. Major limitations are the ecological diversity of wood rat populations, the wide range of environmental conditions in the vicinity of the cave, and the many known changes of climatic conditions in the Pleistocene.

(p.78)

Table 8.1 Vertebrate Associates of Neotoma cinerea in Colorado

Species

Caught at or near Den

Seen at or near Den

Remains in Den

Plecotus townsendii

Homo sapiens

Sylvilagus sp.

Lepus americanus

L. townsendii

Tamias minimus

T. quadrivittatus

Marmota flaviventris

Ammospermophilus leucurus

Spermophilus lateralis

S. richardsoni

Cynomys gunnisoni

C. leucurus

Tamiasciurus hudsonicus

Thomomys talpoides

Castor canadensis

Peromyscus boylii

P. crinitus

P. maniculatus

P. truei

Neotoma cinerea

N. mexicana

Microtus sp.

Erethizon dorsatum

Canis latrans

Ursus americanus

Martes americana

Mustela frenata

Mephitis mephitis

Taxidea taxus

Felis catus

Lynx rufus

Equus caballus

Sus scrofa

Cervus elaphus

Odocoileus hemionus

Ovis aries

Bos taurus

Branta sp.

Falco mexicanus

Dendragapus obscurus

Centrocercus urophasianus

Gallus gallus

Bubo virginianus

Pica pica

Corvus corax

Urosaurus ornatus

Uta stansburiana

Crotalus viridis

NOTE: () , present.

(p.79)

Table 8.2 Vertebrate Associates of Neotoma mexicana in Colorado

Species

Caught at or near Den

Seen at or near Den

Remains in Den

Notiosorex crawfordi

Antrozous pallidus

Homo sapiens

Sylvilagus sp.

Lepus sp.

Tamias minimus

T. quadrivittatus

Ammospermophilus leucurus

Spermophilus variegatus

Cynomys gunnisoni

C. cf. leucurus

C. ludovicianus

Thomomys cf. bottae

Pappogeomys castanops

Perognathus flavus

P. hispidus

Dipodomys ordii

Castor canadensis

Onychomys leucogaster

Peromyscus boylii

P. difficilis

P. leucopus

P. maniculatus

P. truei

Reithrodontomys sp.

Neotoma albigula

Ondatra zibethicus

Rattus norvegicus

Erethizon dorsatum

Canis latrans

Spilogale gracilis

Mephitis mephitis

Lynx rufus

Odocoileus hemionus

Bubo virginianus

Asio flammeus

Sceloporus graciosus

S. undulatus

Uta stansburiana

Cnemidophorus tigris

Crotalus viridis

Unidentified snake

Unidentified fish

NOTE: () , present.

(p.80)

Table 8.3 Vertebrate Associates of Neotoma albigula in Colorado

Species

Caught at or near Den

Seen at or near Den

Remains in Den

Sylvilagus sp.

Tamias quadrivittatus

Spermophilus tridecemlineatus

S. variegatus

Cynomys sp.

Thomomys cf. bottae

Dipodomys ordii

Castor canadensis

Peromyscus boylii

P. crinitus

P. leucopus

P. maniculatus

P. truei

Neotoma mexicana

Ondatra zibethica

Erethizon dorsatum

Spilogale gracilis

Odocoileus hemionus

Ovis aries

Bos taurus

Buteo sp.

Phasianus colchicus

Cnemidophorus tigris

Crotaphytus collaris

Phrynosoma cf. cornutum

Sceloporus undulatus

Uta stansburiana

Elaphe guttata

Masticophis flagellum

Unidentified lizard

Unidentified snake

Scaphiopus multiplicatus

NOTE: () , present.

Table 8.4 Vertebrate Associates of Neotoma floridana in Colorado

Species

Caught at or near Den

Seen at or near Den

Remains in Den

Sylvilagus sp.

Geomys bursarius

Dipodomys ordii

Peromyscus maniculatus

Unidentified carnivore

Bos taurus

Equus caballos

Melagrus gallopavo

Sceloporus undulatus

Cnemidophorus sexlineatus

Masticophis flagellum

Crotalus viridis

NOTE: () , present.

(p.81) Acknowledgments

I thank Don Rasmussen and A. D. Barnosky for the invitation to participate in the Porcupine Cave project. I am grateful to Don and Jerry Rasmussen for the benefit of many conversations about wood rats and the history of caves in South Park, and for their guidance both inside the cave and in the surrounding hills. I thank Donna Shay for help in collecting and preparing mammal specimens, and for improvements made to a draft of this chapter. Reviews of this chapter by A. D. Barnosky, Don Rasmussen, Tom Van Devender, and an anonymous reviewer did much to improve it. E. A. Hadly provided a reprint of her 1999 paper on Lamar Cave.