Printer Friendly

Opportunistic use of modified and disturbed habitat by the Texas kangaroo rat (Dipodomys elator).

ABSTRACT. -- Distribution of the Texas kangaroo rat, Dipodomys elator, currently is restricted to mesquite grassland of north-central Texas. Populations usually are localized and associated with heavily grazed mesquite rangelands, although association of D. elator with mesquite (Prosopis glandulosa) appears not to be so critical as has been supposed. Detailed study of two populations in Wichita County revealed opportunistic use of manmade terraces and mounds for burrow sites, and a preference for heavily grazed short-grass areas with exposed earth resulting from concentrated traffic of vehicles and livestock. These observations, and the recent demise of some populations from mesquite rangelands in Hardeman County presently overgrown with dense grasses, indicate that D. elator now may rely on certain land use practices to provide suitable habitat. Thus, habitat modification might be employed in the future to insure the viability of local populations, or to permit the existence of the species elsewhere in otherwise suitable habitat. Such changes, in fact, may duplicate past landscaping effects of the extirpated bison and now relictual prairie dog, and perhaps once regularly occurring fires, on the mesquite grasslands of northern Texas. Key words: Texas kangaroo rat; conservation; threatened species; habitat management; Dipodomys elator.


The known historic range of the Texas kangaroo rat, Dipodomys elator, is restricted to southwestern Oklahoma and much of north-central Texas (Carter et al., 1985). The species probably no longer occurs in Oklahoma (Baumgardner, 1987; Moss and Melhop-Cifelli, 1990); in Texas, only Hardeman, Wilbarger, and Wichita counties have been reported to contain substantial populations (Martin and Matocha, 1972; Dalquest and Horner, 1984; Jones and Bogan, 1986; Jones et al., 1988; Stangl and Schafer, 1990).

Land use practices within the limited range of D. elator, and its selection of rather specific habitat--short grasses associated with mesquite (Prosopis glandulosa) on clay or sandy loam soils--are thought to contribute to the vulnerability of the species. The concern for its future is formally expressed in the most recent (1987) listing of threatened species by the Texas Parks and Wildlife Department. Studies have been commissioned on an intermittant basis by state and federal authorities to monitor the status of the Texas kangaroo rat. Comparison of results from the most recent of these (Jones and Bogan, 1986; Stangl and Schafer, 1990) provides evidence of a geographic range that presently seems stable, but severe population decline, and some local extinctions, in Hardeman County indicate that an active conservation program may be warranted.

We present here observations on stable populations of D. elator in Wichita County, Texas. Persistant and long-term success of these populations is attributed to certain land-use practices that can readily be applied to otherwise suitable habitat elsewhere, and may have important management implications for any future attempts to conserve the species. It is speculated that such modifications may represent a recreation of conditions once maintained by bison, prairie dogs, and fire before the advent of current farming and ranching practices.


Observations and experiences forming the basis for this study extend back many years. Most were incidental to other studies of small mammals in northern Texas, and culminated with a solicited status report on D. elator from Hardeman and Wichita counties, Texas (Stangl and Schafer, 1990). Acquaintance with the region and Texas kangaroo rats dates to the late 1960s for one of us (FBS), whereas another (JRG) represents the third generation of a family of farmer-ranchers who first homesteaded the area in the 1880s.

The two primary study sites were located about 10 kilometers apart in north-central Wichita County, Texas. Monoculture (wheat, and to a lesser extent, cotton) is locally important, but most of the area deemed unsuitable for such purposes is set aside for grazing of livestock (mostly cattle), typically on grasslands dominated to varying degrees by mesquite. Each area was thoroughly and methodically searched for D. elator burrows. Sherman live traps were employed to determine relative population densities when permitted.

Study site 1 (Goins property). -- The first study area is mesquite rangeland (Fig. 1a), leased and operated by Terrell Goins of Burkburnett, Texas. This section of land (2.6 square kilometers) is located 8.6 mi. N, and 1 mi. W Iowa Park, Wichita Co., Texas. The pasture has been moderately to heavily grazed continuously since at least 1967. Mesquite trees are small to moderate (1.5-3.0 meters) in height, mostly scattered although some dense stands of mature trees exist. Dirt roads servicing oil pumping units divide the property and are maintained by periodic bulldozing activity. Partly reflecting spot grazing habits of cattle, a mosaic pattern of grasses is evident. Stands of dense and mostly unpalatable species of grasses (for example, Andropogon saccharoides and Aristida sp.) contrast with localized areas of grazed short grasses (mostly buffalograss, Buchloe dactyloides).

Heavy traffic (vehicular and livestock) occurs in the vicinity of an abandoned homestead where an old wooden corral and two metal hay barns are situated (Fig. 1b). Nearby is an earthen stock tank with water. This area of perhaps half a hectare is bounded to the east by a paved farm-to-market road, to the south by a barren and badly eroded salt flat (from contamination of nearby pumping units), and to the west and north by relatively dense stands of grasses and mesquite. The site is devoid of all but a few small and scattered mesquite; grasses are scattered, sparse, and kept closely cropped by the regular and concentrated grazing pressures. Numerous patches of earth are exposed, maintained by the traffic and wind erosion.

Excess surface soil, routinely bladed off to the side of dirt roads during maintenance activities, forms low earthern terraces and mounds in this open area, and along sides of roads elsewhere on the property.

Study site 2 (Goetze property). -- This study area is comprised of two adjacent pastures of mesquite grassland (Fig. 2a, 2b), and is located 4 mi. N Lake Buffalo, Wichita Co., Texas. The northern (about 30 hectares) and southern (about 540 hectares) pastures straddle the site of the original homestead, which was established by the Goetze family in the 1880s and still is represented by the abandoned house, barn, stable, and assorted sheds and structures. Heavy traffic here by livestock and farm vehicles have created a situation comparable to the abandoned homestead on the Goins property, but more extensive.



Composition of vegetation is quite different than that of the Goins property, for this property has a long and documented history of brush control using a variety of methods. As a consequence, the mesquite here have been reduced to small (1-1.5 meters), scattered growth. The south pasture was first sprayed with an herbicide in 1951, and again in 1959, and 1964. Hand-grubbing of individual mesquite was performed in 1975 and again in 1981. Remains of these grubbed mesquite were raked (along with some soil) into piles and burned, leaving low, earthen mounds (Fig. 2b) that are scattered uniformly across the pasture. In 1991, both northern and southern pastures were sprayed with an herbicide.

These pastures serve as year-around holding areas for cows and their calves, and are, therefore, heavily grazed throughout the year. As a consequence, grasses (mostly buffalograss) are short. Pastures are fenced with barbed wire and surrounded by wheat fields. At the bases of fencerows, drifted soil has formed terraces. Like the earthen mounds, grazing and erosion by wind and rain maintain these elevated banks as barren or sparsely vegetated.


Each burrow system of D. elator discovered during this study was dug into elevated, friable soils in the immediate vicinity of open ground and bare earth (Fig. 3). In every case where artificially constructed terraces or mounds, scantily vegetated, were available, one or more burrows was found. Accumulated soils at the bases of buildings and fencerows also provided suitable sites for excavation of burrows. Specific findings for the placement of burrows at each study site are presented below.

Study site 1 (Goins property). -- Evidence of D. elator here was extremely localized. Ten of the 14 burrows of the Texas kangaroo rat were situated in or along the periphery of the cleared half-hectare homestead and were, therefore, in close proximity. Eight of these were in low earthen mounds or terraces produced by past bulldozing activity, one was along a fencerow, and another was found at the earthen base of a collapsed wooden shed. Additionally, four burrows were dug into small bulldozed mounds along a nearby dirt-surfaced oilfield service road that bissected dense mesquite grassland.

A trapping grid of 54 by 54 meters, using Sherman live traps, was employed for a capture-and-release study, and included the entire area where burrows were discovered. From the period of 16-18 November 1990, 4000 trapnights established the presence of at least 13 resident D. elator. None of these was taken in the dense mesquite and grasses surrounding the homestead or away from the dirt road.

Other species taken here (in order of abundance, and respective numbers) were: Sigmodon hispidus (51), Peromyscus maniculatus (42), P. leucopus (29), Neotoma micropus and Chaetodipus hispidus (16), Reithrodontomys montanus (6), Perognathus flavus (3), and Sylvilagus floridanus (2). Each of the cotton rats, woodrats, cottontail rabbits, and most of the white-footed mice were taken only in the denser vegetation avoided by D. elator.

Study site 2 (Goetze property). -- Evidence of the Texas kangaroo rat was abundant (50 burrows) and uniformly widespread over this relatively barren site. Each of the 21 earthen mounds remaining from mesquite brushpiles contained at least one burrow. Seventeen burrows were found along fencelines, and seven were discovered at the earthen bases of buildings on the abandoned homestead. A few other burrow sites included the earthen dam of a stock tank and soils accumulated at the bases of unused farm implements and stands of cactus (Opuntia).



Earlier workers (Dalquest and Horner, 1984; Martin and Matocha, 1972) have suggested that habitat destruction by brush control methods (spraying, chaining, root plowing of mesquite) and conversion of rangelands to agricultural purposes poses a serious threat to the continued existance of D. elator in northern Texas. Monoculture certainly would render an area uninhabitable for this species, but the association of the Texas kangaroo rat with mesquite and the threat of brush control measures to this rodent, may be overstated. Both rodent and mesquite have persisted on virgin sod. The common denominator we have noted for sites occupied by viable populations of this kangaroo rat is the disturbance or modification of grassland habitat through certain land-use practices. Some comments from other workers (association of populations with overgrazed pastures and excavation of burrows in cutbanks--Dalquest and Horner, 1984; location of some burrows along artifically constructed terraces--Roberts and Packard, 1973) strengthen this observation. Additional supportive evidence is the recent local extinction of several earlier-reported concentrations of D. elator in Hardeman County (Stangl and Schafer, 1990), which coincided with the removal of livestock from those rangelands and the subsequent return of dense grasses.

We discuss below findings that pertain to opportunistic use by that species of altered habitat, and the potential value of habitat modification in future management strategies. The historical implications of habitat preference by D. elator also are addressed.


Although the specific habitat requirements of D. elator have yet to be fully detailed, the species always is closely associated with predominantly clay or sandy loam soils supporting mesquite grasslands (Dalquest and Collier, 1964; Dalquest and Horner, 1984; Roberts and Packard, 1973). However, the association of the Texas kangaroo rat and mesquite that we find today seems artifactual (see discussion above). At least in Wichita County, most tillable land has been converted to cropland (mostly wheat). Elsewhere, either the sandstone bedrock is too shallow, the ground is too rocky, or the landscape is dissected by drainage systems that would lead to erosion of a planted field. Only under such circumstances locally is mesquite grassland set aside for pasturage; brush control is practiced to varying degrees. Intensive control efforts, using such methods as spraying herbicides, grubbing or chaining, apparently do not harm the kangaroo rat population, but actually seem to increase or enhance the habitat for D. elator. However, the effects of the more destructive root-plowing method are unknown.

Dalquest and Horner (1984) first remarked on the importance of overgrazing by livestock in producing and maintaining the preferred sparse, short-grass situation for foraging, and patches of bare earth for dust-bathing. They concluded that such land use served to increase suitable habitat for D. elator. Indeed, heavy grazing seems an important component in contributing to suitable habitat, and Chapman (1972) found that several important food items of the species are characteristic of rangelands so used. However, even where grazing pressure is less intense, the increased activity by man and livestock is still capable of combining to create restricted pockets of suitable microhabitat among dense stands of grasses and mesquite as occurs on the Goins property. Bailey's (1905) report of this kangaroo rat occurring around old farm buildings may have been based on similar circumstances.

Burrow Sites

Burrow entrances have been associated with mounds of earth excavated by the kangaroo rat at the bases of mesquite (Dalquest and Collier, 1964; Roberts and Packard, 1973), although cutbanks and terraces also are known to serve as burrow sites when available (Dalquest and Horner, 1984; Roberts and Packard, 1973). The absence of mounds constructed by Dipodomys at our study sites seems related to the availability and preferential use of already existing elevated soils, and indicates to us that the animals resort to construction of a mound only when otherwise suitable burrow sites are not available.

In the vicinity of our study sites, burrows were confined mostly to fencerow drifts or artificial terraces and mounds. These weathered earthen works are characteristically sparsely vegetated, well drained, and comprised of loose, easily excavated soil. The extent to which such suitable sites are utilized on both study areas for burrowing indicates that this may be an important limiting factor in local populations. Any sizeable mound, some separated from burrows in other mounds by as little as three meters, minimally perhaps a meter in diameter and half as high, was found to house at least one burrow.

Historical Implications

The Texas kangaroo rat probably always has been restricted in geographic distribution, and may never have been more than locally common. There is no known fossil record, although Dalquest and Horner (1984) speculated that the species may have evolved in mesquite grasslands of northern Texas and southern Oklahoma. Given the ready exploitation of disturbed or altered habitat by the Texas kangaroo rat, and its recent disappearance from Hardeman County rangelands allowed to return to fallow, it is entirely likely that brush control methods and local grazing practices are now reproducing the ideal habitat for D. elator once created and maintained by naturally occurring factors that have been eliminated or are controlled by man.

Two likely candidates are the bison (Bison bison) and black-tailed prairie dog (Cynomys ludovicianus). Both species are capable of exerting considerable influence on their immediate surroundings, and together may have contributed to conditions closely approximating those that now can be found supporting D. elator in Wichita County. The remains of bison and prairie dog are abundant in late Pleistocene and Holocene sediments throughout north Texas where cattle now are the dominant grazers.

Now extirpated in northern Texas, the bison already was scarce by the mid-1800s (Marcy, 1856). The prairie dog ranged almost continuously and in vast numbers across much of northern Texas until relatively recently (Bailey, 1905), but today is found only in small and scattered local colonies in the region (Cottam and Caroline, 1965). Areas occupied by Cynomys colonies are closely cropped by this rodent to prevent encroachment of lush vegetation and shrubs from the periphery, and the Goetze property (Fig. 2) closely resembles the conditions found in local prairie dog towns today. The vast and numerous towns probably were marked by worn trails and wallows left by the migratory bison that seasonally grazed here. Trails wide enough for covered wagons to follow that sometimes were cut up to two meters into the ground, and wallows for dust bathing, up to five meters across and a half meter deep, were often so numerous as to overlap (Cahalane, 1961).

Naturally occurring and unchecked fires probably were important in brush control prior to settlement of the region. Grass fires started by lightning storms were regular (about every five to six years) and often extensive events in Wichita County during the late 1800s and early 1900s. Firebrakes and backfires were employed regularly to protect the early Goetze homestead.

The effects of grass fires today usually are minimized by roadways, firebreaks, and rapid response by teams of firefighters, but when they occur, the resulting accumulation of dead wood and grass thatch permit fires of an intensity that can destroy mesquite trees. One such example is evident on a small (six hectares) pasture of large mesquite and rank vegetation similar to that pictured in Fig. 1a, and separated from study site 2 by less than a kilometer of wheat field. A grass fire in 1987 consumed half of this property, and the burned section today more closely resembles that depicted in Fig. 2a. Small mesquite plants are just beginning to recover and bud out, and ungrazed grasses are still less than half a meter in height. It seems entirely plausable to us that the contrast between the extremely localized population (restricted to homestead and roadways) on the Goins property and the more uniformly distributed population on the Goetze property closely resemble extreme conditions of a dynamic past with a distributional pattern of repeated expansions and contractions of ideal habitat, in response to the effects of periodic fire.

Given the findings of this study, we can envision a scenario of periodic brush control by naturally occurring fires (and subsequently maintained by the prairie dog), landscape modification by bison (forging trails and constructing wallows) and heavy grazing pressures by both species. Such must certainly have contributed to an ever-changing, but more continuous ideal habitat for D. elator than exists today.


The greatest threat to the survival of D. elator clearly is habitat destruction through conversion of range lands to croplands. Lowered grazing pressure on mesquite-grassland range also seems to have a negative effect, although where dense mesquite grasslands occur, the Texas kangaroo rat still may survive in small, localized populations. Our first study site (Goins property) could prove a useful model in helping to create numerous pockets of locally suitable microhabitat even in lightly or moderately grazed rangelands, with little or no inconvenience to landowners.

We conceive certain minimal steps in helping to establish a more continuous distribution of viable populations of D. elator where the species still exists. We hypothesize that implementation of these measures will reproduce microhabitat as was naturally available before the decline of fires and the demise of the bison and prairie dog.

1) Encourage heavy (and preferably year-around) grazing of some mesquite rangelands, especially where D. elator already is known to exist.

2) Where dense cover is preferred for the maintenance of other wildlife species (quail or rabbit, for example), it would be possible to insure concentrated activity (attracting particularly heavy grazing and traffic) by supplemental feeding of livestock in clearings such as around stock tanks or deserted homesteads. Future efforts could include the clearing of small areas in more heavily vegetated mesquite pasture, to be connected by graded dirt roads serving as routes for dispersal.

3) Construction or maintenance of dirt roads has declined in the past few years, due primarily to inactivity of oil pumping units, and we have noted that erosion and overgrowth of these surfaces occurs rapidly. Regular maintenance of these roads, and deposition of surplus earth in or along the periphery of cleared, heavily grazed areas would provide attractive den sites. Perhaps most importantly, roads could serve as dispersal routes between populations or from established populations to suitable but uninhabited areas elsewhere.

4) Other routine land-modification practices that may favorably affect the landscape for D. elator should be explored. Examples include bulldozed firebreaks and powerline trails that pass through or along the margins of mesquite rangeland; these might afford both habitat and avenues of dispersal.


Local ranchers Ernest and Oscar Goetze of Iowa Park, Texas, and Terrell Goins of Burkburnett, Texas, permitted us access to their property. The recollections and observations of the Goetze brothers were particularly helpful. Parts of this study were supported by cooperative agreement funds made available by the U.S. Fish and Wildlife Service and the Texas Parks and Wildlife Department. For helpful discussions and useful comments on an earlier draft of this manuscript, we thank Peggy Barrett, Troy Best, Walter Dalquest, Clyde Jones, and an anonymous reviewer.


Bailey, V. 1905. Biological survey of Texas. N. Amer. Fauna, 25:1-222.

Baumgardner, G. D. 1987. A recent specimen of the Texas kangaroo rat, Dipodomys elator (Heteromyidae), from Oklahoma. Southwestern Nat., 32:285-286.

Cahalane, V. H. 1961. Mammals of North America. MacMillan Company, New York, x + 682 pp.

Carter, D. C., W. D. Webster, J. K. Jones, Jr., C. Jones, and R. Suttkus. 1985. Dipodomys elator. Mamm. Species, 232:1-3.

Chapman, B. R. 1972. Food habits of Loring's kangaroo rat, Dipodomys elator. J. Mamm., 53:877-880.

Cottam, C., and M. Caroline. 1965. The black-tailed prairie dog in Texas. Texas J. Sci., 17:294-302.

Dalquest, W. W., and G. Collier. 1964. Notes on Dipodomys elator, a rare kangaroo rat. Southwestern Nat., 9:146-150.

Dalquest, W. W., and N. V. Horner. 1984. Mammals of north-central Texas. Midwestern State Univ. Press, Wichita Falls, Texas, 261 pp.

Jones, C., and M. A. Bogan. 1986. Status report, Dipodomys elator Merriam 1894. Rept. U. S. Fish Wildlife Serv., 32 pp.

Jones, C., M. A. Bogan, and L. M. Mount. 1988. Status of the Texas kangaroo rat (Dipodomys elator). Texas J. Sci., 40: 249-258.

Marcy, R. B. 1856. Exploration of the Big Witchita and headwaters of the Brazos rivers in the year 1854. Government Printing Office, Washington, D.C., 48 pp. + map.

Martin, R. E., and K. G. Motocha. 1972. Distributional status of the kangaroo rat, Dipodomys elator. J. Mamm., 53:873-877.

Moss, S. P., and P. Mehlhop-Cifelli. 1990. Status of the kangaroo rat, Dipodomys elator (Heteromyidae), in Oklahoma. Southwestern Nat., 35:356-358.

Roberts, J. D., and R. L. Packard. 1973. Comments on movements, home range and ecology of the Texas kangaroo rat, Dipodomys elator Merriam. J. Mamm., 54:957-962.

Stangl, F. B., Jr., and T. S. Schafer. 1990. Report on the current status of the Texas kangaroo rat, Dipodomys elator, in north-central Texas. Rept. Texas Parks and Wildlife Dept., 17 pp.


Department of Biology, Midwestern State University, Wichita Falls, Texas 76301, The Museum and Department of Biological Sciences, Texas Tech University, Lubbock, Texas 79409 (JRG), and Texas A & M Agricultural Experimental Station, Vernon, Texas 76384 (WP)
COPYRIGHT 1992 Texas Academy of Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Stangl, Frederick B., Jr.; Schafer, Tracy S.; Goetze, Jim R.; Pinchak, William
Publication:The Texas Journal of Science
Geographic Code:1U7TX
Date:Feb 1, 1992
Previous Article:Paleontology of the Glen Rose Formation (Lower Cretaceous), Hood County, Texas.
Next Article:A new species of Eimeria (Apicomplexa: Eimeriidae) from Pseudemys texana (Testudines: Emydidae), from north-central Texas.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters