Vegetation classifications of Big Bend National Park, Texas.
The Chihuahuan Desert, one of the three great warm deserts of the American Southwest, represents the easternmost region of arid environments in North America. This subtropical desert is distributed from southern New Mexico through western Texas and extends into the north-central parts of the Mexican states of Coahuila and Chihuahua (Vankat, 1979).
Big Bend National Park is an outstanding area for experiencing and studying the Chihuahuan Desert environment. Big Bend is the sixth largest national park within the contiguous United States, comprising 286,637 hectares (708,281 acres). It is located in the southern extreme of Brewster County in southwestern Texas, where the Rio Grande makes a grand sweep along its eastwardly route toward the Gulf of Mexico (Fig. 1).
Except at coarse scales, it is insufficient to describe the vegetation of Big Bend simply as "Chihuahuan Desert type." The composition of plant species is not homogeneous throughout the region, but changes spatially with differences in characteristics of the physical landscape. The terrain of Big Bend National Park is diverse, consisting of desert plains, badlands, desert mountains, foothills, mesas, high mountains, and riparian areas. Many distinctive plant assemblages occur upon these surfaces. Over the past five decades, several classifications have been developed for describing the varied vegetation cover of the national park. This essay reviews and compares each of these classifications.
VEGETATION CLASSIFICATIONS ENCOMPASSING THE REGION
The vegetation of the Big Bend National Park has been categorized by eight independent investigators since the establishment of the park in 1944. There are several reasons why so many classifications have been developed. Some of the schemes serve different purposes. The areas of coverage may also vary, extending well outside the park or focusing upon a particular environmental subregion within the park. Other approaches represent attempts to improve upon prior classifications.
[FIGURE 1 OMITTED]
A summary of the vegetation classifications of Big Bend is presented in Figure 2. The classifications are arranged in column order. They are listed chronologically, reading from left to right on the diagram. Individual vegetation types may be compared among the various classifications by examining the diagram in horizontal fashion. The diagram was constructed through iterative interpretations of the original publications describing each individual classification. Attention was placed primarily upon plant species composition and lifeform structure, and secondarily upon descriptions of spatial distribution.
Discussion of each individual classification follows, with comparisons made with those previously devised. The scientific names of plants are from Correll and Johnston (1970), except the grasses, which are after Gould (1975). Common names are taken from McDougall and Sperry (1951), Correll and Johnston (1970), and Gould (1975).
Classification of Taylor, McDougall, and Davis
The ecological characteristics of Big Bend National Park were initially surveyed and classified by Taylor et al. (1944). The classification operates at two hierarchical levels of detail. At the coarsest level, plant-animal assemblages were grouped into five biomes--River Floodplain, Desert Scrub, Sotol-Grass, Woodland, and Forest. The term "biome" was used by the authors as a regional descriptor, unlike its modern areal connotation of a continental or global scale.
Each biome was subdivided into finer-detailed categories, termed biotic communities. Although the concept of a biotic community included both flora and fauna, most of the communities were distinguished by the authors solely by the plant assemblages. The faunal component tended to be uniform within each biome; animals were mentioned primarily at the biome level and were described for only a few of the biotic communities.
The River Floodplain biome occurs adjacent to the Rio Grande and along other large, sandy washes and arroyos. Although the plant species found within this biome are variable and interspersed, they were grouped by the authors into the River Floodplain biotic community.
The Desert Scrub biome encompasses the largest areal extent and harbors the greatest number of biotic communities of the five biomes distributed within the park. Seven biotic communities were recognized within the Desert Scrub--Tobosa Grass, Larrea, Larrea-Agave-Flourensia, Larrea-Fouquera-Jatropha, Candelilla, Hechtia, and Larrea-Yucca. The names of these categories are indicative of the dominant taxa comprising each vegetation type. Most of the plant communities are named in this manner within each classification.
The third biome, Sotol-Grass, consists of a single biotic community by the same name. The dominant graminoids of this community are species of Bouteloua and Aristida. They co-exist with the semi-succulent Dasylirion leiophyllum as part of the Sotol-Grass community. Sotol-Grass is distributed between the Desert Scrub and Woodland biomes.
The Woodland biome also consists of a single biotic community. This assemblage, Pinyon Pine-Juniper-Oak, is distributed throughout most of the higher elevations of the Chisos Mountains. The arboreals, Pinus cembroides, Juniperus, and Quercus species, and grasses, especially species of Muhlenbergia, characterize this community.
The Forest biome consists primarily of relict stands of montane coniferous forest, located toward the upper reaches of two canyons in the Chisos Mountains. Two biotic forest communities occur in Big Bend. Arizona cypress-Douglas-fir (Cupressus arizonica-Pseudotsuga menziesii) is found in Boot Canyon, whereas Ponderosa pine-Graves oak (Pinus ponderosa-Quercus gravesii) occupies a portion of Pine Canyon. The authors considered these assemblages as members of the same plant association, but classified them into distinct communities based upon their geographic separation and differing dominant tree species.
This vegetation classification system represents an historically valuable contribution to understanding the ecology and distribution of vegetation in the Big Bend area. No prior classification had been established specifically for this region, making this system a benchmark for subsequent classification studies.
Classification of Denyes
A second vegetation classification was developed by Denyes (1956) to compare plant community types with the occurrence of mammalian species. Three levels of hierarchy are described. Denyes discussed two classes at the coarsest level--the Chisos and the Davis biotic districts. Only the Chisos biotic district is distributed within Big Bend National Park.
The Chisos district is subdivided into three units--the Desert Plains, Foothills, and Encinal life belts. The author defined each life belt by its characteristic climate and general land morphology. Brief mention was made of the biotic components at the life belt level. Figure 2 shows an apparent relationship between the biomes of Taylor et al. (1944) and the life belts of Denyes. They correspond fairly well, although the Desert Scrub biome overlaps with both the Desert Plains and Foothills life belts.
The finest level of hierarchy in Denyes' system is the vegetation association. Seven associations occur within the Desert Plains life belt of the park. Four of these, Tobosa-Mesquite, Creosote-Tasajillo, Creosote-Tarbush, and Creosote-Ocotillo-Mesquite, correspond well with counterparts in the Taylor et al. (1944) classification (Figure 2). No category of Yucca-dominant species is recognized in this scheme, although Yucca is mentioned as a constituent in each of the preceding associations containing Larrea tridentata.
The riparian associations, Mesquite and Baccharis, represent distinct communities within the Rio Grande floodplain. Denyes also described a Riverbank category. It was not defined by any specific group of species, but was characterized by the presence of plants overhanging occasionally flooded bare flats and stream banks. I consider Riverbank to be coarser than the association level of classification, because a portion of any riparian community may fit this definition (Figure 2).
Three vegetation associations were described by Denyes as occurring within the Foothills life belt. Two of these, Sotol-Lechuguilla and Sotol-Nolina, correspond to the Sotol-Grass community of Taylor et al. (1944). The third association, Creosote-Lechuguilla, occurs on the lower elevations and gentler slopes of the foothills. Its sparse cover consists mostly of scattered Larrea and Agave lecheguilla. Small clumps of the grasses, Bouteloua sp., Aristida sp., and Erioneuron pulchellum also occur. It was not described previously, although Taylor et al. (1944) frequently mentioned the occurrence of Agave in the Desert Scrub biome. Conversely, Denyes did not place candelilla or Hechtia in unique categories, although he noted that these species occasionally dominate in the Sotol-Lechuguilla association.
Within the Encinal life belt, two plant assemblages were newly described for the Big Bend region--Oak Chaparral and Feathertop-Grama. The Oak Chaparral association is characterized by low, shrubby vegetation. Quercus grisea and Quercus gravesii are found in their shrubby form accompanied by stunted individuals of Juniperus monosperma and Cercocarpus montanus. Denyes considered this to be a transition between both the Sotol-Nolina and Grama-Bluestem associations and the Pinyon-Oak-Juniper association. The Feathergrass-Grama association is best typified in the Laguna Meadow area of the Chisos Mountains. These lush patches of grass are characterized by the dominance of Stipa and Bouteloua species.
[FIGURE 2 OMITTED]
The remaining Encinal associations corresponded well with the earlier classification. Grama-Bluestem association represents the grassy extents of the Pinyon-Juniper-Oak biotic community. Trees, especially Quercus species, create a woodland savanna around the periphery of this vegetation type. Denyes did not recognize the geographic separation between Pinus ponderosa, Cupressus arizonica, and other large arboreals, and thus combined them into a single category, which he termed the Yellowpine-Fir association.
The classification of Denyes is slightly more detailed than the version of Taylor et al. (1944). Riparian and Chisos Mountains vegetation types were more finely subdivided in the Denyes system. Although the number of Larrea-dominated communities remained unchanged, two of four prior classes were omitted in the latter classification and are replaced by two other vegetation types. Also, Denyes chose to omit from his system the candelilla and Hechtia assemblages.
Classifications of Warnock
The 1970s harbingered an increase in the frequency of classifications developed for the region. A third set of vegetational categories, developed by Warnock (1970), is comprised of plant associations he observed within the national park. A brief description was given of the distribution of each association, but the composition of dominant species was not mentioned. Only the taxa used to name each association can be implied as high-constancy species occurring in that category. A modified version of this vegetation classification was presented in map form by Warnock and Kittams (1970). This is the only known published vegetation map of Big Bend National Park (scale 1:380,000). The vegetation is mapped by community and is hierarchically grouped in the map legend into Desert Shrub, Grassland, and Woodland formations. These formations correspond fairly well with the prior classifications, although some overlap exists between them. For example, two communities assigned to the Grassland formation belong to the Desert Scrub biome of Taylor et al. (1944) (Fig. 2). Rather than either classification being incorrect, it is likely that their species variability is sufficient to occur within both Grassland and Desert Scrub.
The communities differ only slightly from the plant associations described by Warnock. Most of the categories are surely coincident, for example, Larrea-Flourensia as opposed to Creosote-Tarbush. Common species names are used instead of scientific names for labeling the mapped categories.
Some subtle differences are evident between the two classifications (Fig. 2). Two different riparian types were described--Mesquite-Giant reed community and the Tamarix gallica-Prosopis glandulosa association. Apparently the latter was replaced by Arundo as a dominant descriptive floristic component for mapping at a small scale. Three instances occur in the Warnock and Kittams scheme where single categories represented two associations from its sibling classification--Chino Grama-Lechuguilla from Cathestechum-Larrea and Bouteloua-Agave, Giant Dagger-Sotol from Yucca-Dasylirion and Larrea-Yucca, and Mexican Pinyon-Juniper from Pinus cembroides-Juniperus and Pinus-Juniperus-Quercus. Each grouping may have been made either due to reassessment by these authors of their individual importance or the impracticality of mapping their distribution separately, or both.
Two important comparisons need to be made between the Warnock classification and its predecessors. First, tobosa grass was omitted as a category for the first time. It was likely given no significance in a park-wide context due to its scarce distribution within the park (Taylor et al., 1944). Second, additional categories were described within the Grassland formation--Chino Grama-Lechuguilla and Grama-Viguiera. The Chino Grama-Lechuguilla community (Bouteloua breviseta-Agave lecheguilla association) was mapped throughout the hills surrounding the Chisos Mountains. Chino Grama-Black Grama-Skeletonleaf goldeneye community (Bouteloua eriopoda-Viguiera stenoloba association) was depicted east of both Panther Junction and Government Spring.
There are two principal contributions made by the classifications of Warnock. One was the mapping of vegetation. The other was a more detailed description of the grassland assemblages within the park. Conversely the riparian and mountainous vegetation types are less detailed than in the system of Denyes (1956). Larrea-dominated vegetation is also more generalized than the prior classification, with the number of categories reduced from four to two.
Classification of Wauer
A special-purpose vegetation classification was devised by Wauer (1971, 1973) to correspond with avian distributions within Big Bend National Park. He developed his system based primarily upon field analysis of vegetation and distributional selection of breeding birds. The system is divided into five ecological zones or formations, which correspond extremely well with the biomes of Denyes (1956).
Four of the five formations consist of a single vegetation association. The phreatophytic vegetation of the River Floodplain-Arroyo formation was placed in the Arroyo-Mesquite-Acacia association. The low density cover occurring within the Shrub Desert formation was assigned to the Lechuguilla-Creosotebush-Cactus association. The Sotol-Grass association corresponded with the Sotol-Grassland formation. The Moist Chisos formation contained the closed-canopy cover of the Cypress-Pine-Oak association.
Wauer subdivided the Woodland formation into two plant associations, the Deciduous Woodland and Pinyon-Oak-Juniper assemblages. The former had not been described in prior classifications. Wauer described Deciduous Woodland as primarily confined to washes and canyons, with its vegetation composition tending to vary with elevation. Lower and mid-elevation canyons harbor trees such as Quercus gravesii, Juniperus flaccida, and Acer grandidentatum, shrubs including Rhus virens and Ungnadia speciosa, and the semi-succulent Nolina erumpens. The plant assemblages in the washes of the relatively lower elevations intergrade with those of the Arroyo-Mesquite-Acacia association.
The purpose of Wauer's classification was to characterize vegetation assemblages with the breeding habits of birds. He found that a classification system less detailed than those mentioned previously was sufficient to meet this objective. Although this system is much more generalized than the previous classifications, the Deciduous Woodland association was described for the first time.
Classification of Whitson
In his study on vegetation dynamics over desert terrain, Whitson (1970) described four plant communities that occur on the hills and small ranges interrupting the broad and gradually sloping plains of Big Bend. Whitson placed these assemblages into the Succulent Desert formation, which implies a coarseness to the Shrub Desert/Desert Scrub categories described earlier (Fig. 2). It also appears to be representative of less mesic conditions of the Grassland formation of Warnock and Kittams.
Whitson may have been one of the first researchers to employ quantitative techniques for classifying Big Bend vegetation. It is uncertain how the prior classifications were developed as their methodological approaches were not reported. Considering the harsh, rugged, and extensive terrain of Big Bend National Park, it is likely that they were devised qualitatively through field observation, especially the earliest classifications.
Classification of Dick-Peddie and Alberico
As a part of a fire ecology study of the Chisos Mountains, Dick-Peddie and Alberico (1977), presented a classification of vegetation associations over the higher elevations of Big Bend National Park. Five classes of plant assemblages were described.
The authors included two communities that were not described in previous classifications. One is labeled Disturbance Scrub. It was not interpreted by them as a natural community, but rather as either a disclimax or seral plant assemblage. The community is dominated by shrubs such as Aloysia wrightii, Acacia constricta, Aloysia gratissima, and Acacia greggii. Also important are the succulents and semi-succulents Agave lecheguilla, and Opuntia phaeacantha, and considerable grass cover of Bouteloua gracilis and Bouteloua curtipendula.
The structural composition of the Moist Woodland association consists of four community types or stands--ponderosa pine, Arizona cypress, douglas-fir, and talus assemblages. All of these except talus are closed-canopied forests. The talus community had not been described in any prior classification. It is generally depauparate of taxa, consisting of scattered trees of Populus tremuloides and Prunus serotina, the shrub Rhus aromatica, and the semi-succulent Nolina erumpens. This classification was the most comprehensive for describing the vegetation types in the Chisos Mountains.
Classifications of Leopold
Some of the most recent classifications have been developed by Leopold and his associates. In his study on the ecology of the desert mule deer, Odocoileus hemionus, Leopold (1984a) concentrated upon identifying plant associations found in the desert shrubland formation of Big Bend, areas in which mule deer are primarily distributed. These areas were subdivided into three general categories defined by community structure--creosote-dominant, other shrub-dominant, and no shrub-dominant vegetation covers. The first category appeared to correspond to Shrub Desert/Desert Plains connotations, the second to Succulent Desert/Grasslands/Foothills headings, and the third to Sotol-Grass/Grassland subdivisions.
Waggoner et al. (1984) expanded Leopold's classification to include additional vegetation types for mapping vegetation cover types over the entire park. The term "cover type" or "cover class" was popularized by Gardner and Wieslander (1957) to imply vegetation classes that could be mapped through the interpretation of aerial photographs, or from any other source of remotely-sensed data (Strahler, 1980). A working draft of a 1:100,000-scale map was constructed for the entire park by Leopold (1984b), but a final version was not completed. Waggoner et al. used aircraft-based, digital multi-spectral scanner data in an attempt to map the vegetation cover types of the park, but encountered problems with georeferencing the data (Nyquist, 1984).
The most variable and difficult association for Leopold to identify was Creosotebush-Lechuguilla-Grass. It intergrades with many other associations and may be densely or sparsely vegetated. One identifying characteristic is the abundance of Larrea tridentata with other shrubs, excluding Viguiera stenoloba or Flourensia cernua. The succulent Agave lecheguilla and grasses Erioneuron pulchellum and Bouteloua breviseta are other important constituents of this assemblage. This vegetation type generally occurs over level areas throughout the desert plains.
The Creosotebush-Lechuguilla-Prickly Pear association resembles the Creosotebush-Lechuguilla community in possessing a relatively low plant density and diversity. High-constancy plant species membership also is similar, with the primary distinction being the conspicuous presence of Opuntia rufida. The most extensive areas of this vegetation type occur southeast of Dugout Wells toward Rio Grande Village.
Leopold discussed in depth Brushy Wash as occurring within the small drainages interspersing the uplands of each of his previously described assemblages. Its vegetation is generally denser and has a greater species diversity than surrounding upland vegetation. Species constituents include those of the adjacent assemblage, but usually possess some increasingly mesic plants as well.
Waggoner et al. (1984) classed four riparian assemblages of vegetation along the Rio Grande--Arundo-Phragmites, Mesquite Thicket, Tamarisk, and Bermuda Grass. These cover types are generally intricately distributed along the river, although homogeneous stands of the first two are occasionally found.
Five cover types encompass the area of the Chisos Mountains. Only the Desert Grassland had not been previously described. Visiting sites labeled Desert Grassland, I found it difficult to distinguish most of them from Mixed Scrub, although their shrub component appeared to decrease slightly.
Compared with the other categorizations, the classification system of Waggoner et al. possesses a balanced number of categories within each of the major subregions of Big Bend National Park. This trait makes it a fairly comprehensive system for mapping vegetation at a park-wide level. The only important category omitted from the system is the moist woodland vegetation types described by Dick-Peddie and Alberico (1977).
Classification of Plumb
This classification was developed for mapping vegetation distributions parkwide at the medium mapping scales of 1:24,000 and 1:100,000. None of the prior classifications were wholly appropriate, because each system possessed one or more vegetation categories that were either too detailed or too coarse for meeting this map scale criterion. In addition, some of the classifications omitted vegetation types that occur within the park. Therefore this classification system was developed for mapping the vegetation of the park at these scales.
Categories were defined as vegetation cover types. They were not termed vegetation associations because the species composition of any given category may vary within its geographic distribution. If such variation is significant, then each variant should be considered as a distinct plant association. It was also possible that single associations may have been distributed over more than one cover type. The term association had been misused in the prior classifications, except by Whitson (1970). Many of the categories within each of these classifications were described as having areally wide amplitudes in their plant compositions. This characteristic is contradictory to the modern concept of a vegetation association. More detailed research is required before a category of vegetation can be confidently termed a plant association.
Each cover type was characterized by the nature of recognizable and distinctive lifeform and floristic components of vegetation. Five different lifeforms were considered--trees, shrubs, succulents and semi-succulents, grasses, and forbs. Details on the methodology used to construct this classification can be found in Plumb (1988, 1991).
The cover types were grouped into four environmental subregions of Big Bend National Park riparian areas; desert plains and badlands; desert mountains, foothills, and mesas; and high Chisos Mountains. These subregions were modified from the descriptions of Denyes (1956) and Wauer (1970). Figure 2 shows their relationships with the individual cover types and with the categories of the other classifications.
Several modifications from the other classifications were incorporated in this system. The most important element of the riparian cover types was the addition of the Mixed Riparian category. This cover type was included in the scheme to represent a complex of the individual riparian classes, which were frequently intricate in their spatial patterns. Mapping scales as large as 1:24,000 were insufficient to depict their distributions in various reaches along the Rio Grande. For such instances, Mixed Riparian was used to generalize this vegetation complex.
The plant assemblages of the desert plains corresponded well with the creosote-dominant vegetation types of Leopold (1984a). One exception was the Creosote-Grass cover type, which had not been described in any of the preceding classifications. This community is similar in part to the Creosote-Lechuguilla-Grass category of Leopold (1984a). Creosote-Grass and Creosote-Lechuguilla replaced this vegetation type, which was omitted from the Plumb classification because it primarily possessed a limited distribution as a transition between other, more extensive, assemblages. Creosote-Grass may have been largely absent during the time periods in which the earlier systems were developed, thus it would not have been considered. Grass cover in some areas previously dominated by Larrea may have increased greatly over the past decades, due to changing climatic trends, natural recovery occurring in previously disturbed areas, or a combination of these and other mechanisms (Wondzell 1984:11-22).
Other previously described vegetation types have been included in this classification, but were omitted from most of the other systems. One additional category was the Creosote-Yucca-Grass cover type, distributed in the desert plains. It was previously mentioned only by Taylor et al. (1944) and by Warnock (1970). Sotol-Nolina was first included as a vegetation type by Denyes (1956). Denyes (1956), Wauer (1970), and Dick-Peddie and Alberico (1977) also had Oak Scrub, or Chaparral, in their classifications. Lastly, most of the preceding studies had the Oak-Ponderosa Pine-Cypress cover type or variants thereof as a vegetation category in their systems.
Investigators who work in the Big Bend region and have a need for vegetation classification have several systems from which to choose. No one system is universally the best. Researchers should select the one that most appropriately suits their purposes. If vegetation classification is central to the research, subsequent critical study may yield an updated classification that refines an existing one. New classifications still need to be developed for research themes with purposes that have been inadequately addressed for the region. For instance, more work needs to be accomplished within Big Bend regarding the formidable task of examining vegetation associations and their synecologies. Vegetation classifications will continue to evolve with new purposes, goals, ecological knowledge, and methodological approaches.
The author thanks the National Park Service and the Big Bend Natural History Association for supporting vegetation classification and mapping research activities within Big Bend National Park.
Correll, D. S., and M. C. Johnston. 1970. Manual of the vascular plants of Texas. Texas Research Foundation, Renner, Texas, xv + 1881 pp.
Denyes, H. A. 1956. Communities of Brewster County, Texas, with special reference to the distribution of the mammals. Amer. Midland Nat., 55:289-320.
Dick-Peddie, W. A., and M. S. Alberico. 1977. Fire ecology study of the Chisos Mountains, Big Bend National Park, Texas. Unpublished report, Big Bend National Park library, iii + 54 pp.
Gardner, R. A., and A. E. Wieslander. 1957. The soil-vegetation survey in California. Proc. Soil Sci. Soc. Amer., 21:103-105.
Gould, F. W. 1975. The grasses of Texas. Texas A & M Univ. Press, College Station, Texas, viii + 653 pp.
Leopold, B. D. 1984a. Ecology of the desert mule deer in Big Bend National Park, Texas. Unpublished dissertation, Univ. of Arizona, Tucson, x + 172 pp.
______. 1984b. Vegetation map of Big Bend National Park, Texas. Unpublished draft, National Park Service, Denver Service Center, Denver, Colorado, 1 p.
McDougall, W. B., and O. E. Sperry. 1951. Plants of Big Bend National Park. U.S. Government Printing Office, Washington D. C., xi + 209 pp.
Nyquist, G. 1984. Vegetation/land cover mapping of Big Bend National Park using aircraft multispectral scanner data. P. 25, in Big Bend National Park 1984 Research Newsletter, Big Bend Nat. Hist. Assoc. 27 pp.
Plumb, G. A. 1988. An algorithmic approach to automated vegetation mapping of Big Bend National Park, Texas. Unpublished dissertation, Univ. Kansas, Lawrence, xxi + 499 pp.
______. 1991. Assessing vegetation types of Big Bend National Park, Texas for image-based mapping. Vegetatio, 94:115-124.
Strahler, A. H. 1980. The use of prior probabilities in maximum likelihood classification of remotely sensed data. Remote Sens. Environ., 10:135-163.
Taylor, W. P., W. B. McDougall, and W. B. Davis. 1944. Preliminary report of an ecological survey of Big Bend National Park. Unpublished report, Big Bend National Park library, ix + 55 pp.
Vankat, J. L. 1979. The natural vegetation of North America. John Wiley & Sons, New York, ix + 261 pp.
Waggoner G., T. H. Mace, B. D. Leopold, and V. Davila. 1984. Vegetation cover type classification for Big Bend National Park. Unpublished research, personal communication with M. Nyquist, Denver Service Center, National Park Service, 28 June 1985.
Warnock, B. H. 1970. Wildflowers of the Big Bend country, Texas. Sul Ross State Univ., Alpine, Texas, xix + 157 pp.
Warnock, B. H., and W. H. Kittams. 1970. Vegetation map: plant communities of Big Bend National Park. Sul Ross State University and National Park Service, Alpine, Texas, 1 p.
Wauer, R. H. 1971. Ecological distribution of birds in the Chisos Mountains, Texas. Southwestern Nat., 16:1-29.
______. 1973. Birds of Big Bend National Park and vicinity. Univ. Texas Press, Austin, xv + 223 pp.
Whitson, P. D. 1970. Dynamics of the shrub desert formation-succulent desert formation transition in Big Bend National Park, Texas. Unpublished dissertation, Univ. Oklahoma, Norman, ix + 104 pp.
Wondzell, S. M. 1984. Recovery of desert grasslands in Big Bend National Park following 36 years of protection from grazing by domestic livestock. Unpublished thesis, New Mexico State Univ., Las Cruces, xii + 94 pp.
GREGORY A. PLUMB
Department of Geography, University of Oklahoma, Norman, Oklahoma 73019
|Printer friendly Cite/link Email Feedback|
|Author:||Plumb, Gregory A.|
|Publication:||The Texas Journal of Science|
|Date:||Nov 1, 1992|
|Previous Article:||Winter breeding by Geomys breviceps.|
|Next Article:||Morphometric variation, measurement error, and fluctuating asymmetry in the red fig-eating bat (Stenoderma rufum).|