1. S EASONAL V EGETATION C ORRELATIONS Winter: Lechuguilla-Sotol Series, Blackbrush Series, and Ashe Juniper-Oak Series (Table 2.) Spring: Lechuguilla-Sotol Series Summer: No vegetation series correlations were observed during the summer months, which may be a result of low trap success. Alternatively this species has been documented to switch to 40% animal matter consumption in the summer months which may decrease it’s dependence on vegetation based resources (Baccus et al. 2009). Fall: Blackbrush Series, Guajillo Series B LACKBRUSH S ERIES The Blackbrush vegetation series consists of two dominant species, Blackbrush (Acacia rigidula) and Texas persimmon (Diospyros texanus), with both exhibiting the most direct impact on that habitat (Table 1.). The Texas persimmon bears fruits that are a common food item for a variety of Texas wildlife. The fruit could potentially serve as a food source for P. pectoralis during the fall and winter months when insect activity is low. Baccus et al. (2009) observed a diet consisting 88.4% of fruits and seeds in winter, and 86.2% of fruits and seeds in fall, with a few individuals consuming Texas persimmon fruits in the winter months. L ECHUGUILLA -S OTOL S ERIES The association of P. pectoralis with the Lechuguilla-Sotol Series is a novel vegetation affinity, with none of the dominant or indicator species within this series explicitly known to be foraged upon by Peromyscus mice. Possible resources provided by this vegetation series include (Table 1.) Wright’s threeawn (Aristida purpurea), foliage of Guajillo (Senegalia berlandieri), Rio Grande Stickpea (Calliandra conferta), or even the offshoots, seeds or seedlings of the Lechuguilla (Agave lechuguilla) may be a food source as they are known to be eaten by deer and javelina (Pecari tajacu). A SHE J UNIPER -O AK S ERIES An association between P. pectoralis and Ashe juniper is well supported in the literature. Baccus et al. (2009) observed in winter only Ashe Juniper (Juniperus ashei) berries and green sumac berries were the important food staples of the study population. Our data suggest that when this food resource is available (Ashe juniper berries), the abundance indices of P. pectoralis should increase in areas with higher percentages of Ashe juniper. G UAJILLO S ERIES The Guajillo series is the most common vegetation series, covering 33% of DRSNA – BSU (Table 1.). The commonality of this habitat type may perhaps be causing this association by shear dominance of the present plant communities, as this is another undocumented habitat affinity for this species. Although possible resources may be provided by the associated indicator species listed in Table 1. R ESOURCE D EPENDENT S EASONAL S ELECTION Our data suggest that vegetative habitat selection by P. pectoralis is seasonal, and most likely dependent on the degree of resource availability within each vegetation type. This highly mobile species is capable of dispersing to an area of greater resource availability, and furthermore the fluctuation of abundance indices within the various habitat types described can be explained though the paradigm of trophic ecology. However fine-scale microhabitat analyses of P. pectoralis would provide greater insights into these vegetative habitat preferences, taking into account abiotic as well as biotic habitat variables. A BSTRACT : The purpose of this study conducted at the Devils River State Natural Area – Big Satan Unit (DRSNA - BSU), in Val Verde County, Texas, has been to delineate the seasonal vegetative resource selection of the White-ankled mouse (Peromyscus pectoralis), comparing trap-line indices of abundance, to geospatial plant community data. Trapping data were collected from trap- lines over a 21 month period (Feb 2013 - October 2014), resulting in 10,607 trap-nights and 185 P. pectoralis captures. Using ArcGIS software, trap-lines were buffered by the known approximate convex-polygon home range of P. pectoralis (A = 3,340m 2 ), and spatially overlaid with the known active vegetation series of DRSNA – BSU. The percentage of each habitat within the buffered trap- line area was tested against the abundance indices of P. pectoralis using multiple linear regression modeling. Positive correlations were observed in winter and fall among the Lechuguilla-Sotol (P<0.001), Blackbrush (P<0.007), Guajillo (P<0.05), and Ashe Juniper-Oak vegetation series (P<0.03). Data suggests that vegetative habitat selection by P. pectoralis varies seasonally, and is likely dependent on the degree of seasonal resource availability within each vegetation type. I NTRODUCTION M ATERIALS & M ETHODS R ESULTS & C ONCLUSIONS U NDERSTANDING H ABITAT R EQUIREMENTS A strong correspondence between small mammal assemblages and vegetation composition is well documented in research within various environments, and it is well known that small mammals typically will occur within, and move among, habitat patches distinguished on the basis of floristic composition (Monjeau et al. 2011). Habitat data collected by Mullican & Baccus (1990) suggests that the pattern of use of microhabitats by P. pectoralis is related to inherent selection rather than to interspecific competition. Furthermore, working under the assumption that P. pectoralis’ spatial distribution is associated with its habitat requirements, identifying those habitat requirements is an effective means to understanding and accurately predicting the distribution of this species within a mosaic landscape. S PECIES D ISTRIBUTION The distribution of Peromyscus pectoralis is broadly within central-northern Mexico continuing into the state of Texas (Figure 1.), where the species can be found in the Edwards Plateau regions, central Great Plains, the northern extent of the south Texas plains, and the Chihuahua desert regions of west Texas (Griffith et al. 2004). Based on capture location data, the most documented abiotic habitat affinity of P. pectoralis is an association with rocky areas including cliffs, limestone outcrops, or talus slopes with some form of woody vegetation (Schmidly 1972; Schmidly 1974; Kilpatrick and Caire 1973; Baccus and Horton 1984; Etheredge et al. 1989). L ACK OF R ESEARCH ON H ABITAT S UITABILITY A substantial amount of ecological information exists in the literature for Mexican populations of this species; however there have been limited habitat/resource selection studies of P. pectoralis in Texas populations. These Texas habitat studies have additionally been limited locally to areas of central, west-central, northern, and far west Texas (Figure 2). There has to-date been a lack of ecological studies conducted on this species in southwestern Texas, including a deficit in habitat suitability studies conducted in a highly mosaic region. Because of the paucity of habitat studies in this region, an in-depth habitat suitability study was conducted, comparing vegetation series to trapping data. S TUDY S ITE C HOICE To conduct a study of this nature, three study site criteria must be met: (1) a large sample size of P. pectoralis will be needed, (2) a mosaic of habitat types must be present, and (3) a study area must be chosen that is rocky throughout, to control for the known affinity for rocky areas. These three criteria are met at the Devils River State Natural Area – Big Satan Unit (DRSNA – BSU), which is located at the confluence of three different level III ecoregions which has created a mixing of floral communities (Figure 2.). Peromyscus pectoralis has been captured within the following habitats: Curly Mesquite-Sideoats Grama, Lechuguilla-Sotol, Blackbrush, Ceniza, Ashe Juniper-Oak, and Guajillo Series. These vegetation series were identified by Keith (2011) and named by their dominant plant species (Figure 3.). The dominant and indicator species within the vegetation series used for analysis are described in more detail in Table 1. R EFERENCES C LINT N. M ORGAN AND R OBERT C. D OWLER D EPARTMENT OF B IOLOGY, A NGELO S TATE U NIVERSITY, S AN A NGELO, TX 1.B ACCUS, J. T., J. M. H ARDWICK, D.G. H UFFMAN, A ND M. A. K AINER. 2009. Seasonal trophic ecology of the white-ankled mouse, Peromyscus pectoralis (Rodentia: Muridae) in central Texas. Texas Journal of Science 61:97- 118. 2.B ACCUS, J.T., AND J.K. H ORTON. 1984. Habitat utilization by Peromyscus pectoralis in central Texas. Pp. 7-26 in Festschrift for Walter W. Dalquest in honor of his sixty-sixth birthday (N.V. Horner, ed.). Midwestern State University, Wichita Falls, Texas. 3.E THEREDGE, D. R., M. D. E NGSTROM, AND R.C. S TONE. 1989. Habitat discrimination between sympatric populations of Peromyscus attwateri and Peromyscus pectoralis in West-Central Texas. Journal of Mammalogy 70:300- 307. 4.G RIFFITH, G.E., S.A. B RYCE, J.M. O MERNIK, J.A. C OMSTOCK, A.C. R OGERS, B. H ARRISON, S.L. H ATCH, AND D. B EZANSON. 2004. Ecoregions of Texas (color poster with map, descriptive text, and photographs): Reston, Virginia, U.S. Geological Survey (map scale 1:2,500,000). 5.K EITH, E.L. 2011. Plant community, fuel model, and rare species assessment and baseline establishment of twenty vegetation monitoring plots at Devils River State Natural Area- Devils River Ranch unit in Val Verde County, Texas. Raven Environmental Services, Huntsville, Texas. 6.K ILPATRICK, C. W., AND W. C AIRE. 1973. First Record of the Encinal Mouse, Peromyscus pectoralis, for Oklahoma, and Additional Records for North-Central Texas. The Southwestern Naturalist, (3). 351. 7.M ONJEAU, J. A., R OTELA, C. H., L AMFRI, M., M ÁRQUEZ, J., S CAVUZZO, C. M., S TANULESCU, M., AND... E. G. R IAL. 2011. Estimating habitat suitability for potential hantavirus reservoirs in north-western Patagonia using satellite imagery: Searching for the best predictive tools. Mammalian Biology, 76(4): 409-416. 8.M ULLICAN, T. R., AND J. T. B ACCUS 1990. Horizontal and vertical movements of the white-ankled mouse (Peromyscus pectoralis) in central Texas. Journal of Mammalogy 71:378-381. 9.O MERNIK, J. M. 1987. Ecoregions of the conterminous United States. Annals of The Association of American Geographers 77:118-125. 10.P ARMENTER, R. R., Y ATES, T. L., A NDERSON, D. R., B URNHAM, K. P., D UNNUM, J. L., F RANKLIN, A. B., AND G. C. W HITE. 2003. Small-mammal density estimation: a field comparison of grid-based vs. web-based density estimators. Ecological Monographs, 73(1), 1. 11. P ETERSON, A.T. 2001. Predicting species’ geographic distribution based on ecological niche modeling. Condor, 103, 599–605. 12. S CHMIDLY, D. J. 1972. Geographic Variation in the White-Ankled Mouse, Peromyscus pectoralis. The Southwestern Naturalist, (2). 113. 13. S CHMIDLY, D. J. 1974. Peromyscus pectoralis. Mammalian Species 49:1-3. A CKNOWLEDGMENTS We would first like to thank the Texas Parks and Wildlife Department for funding associated with the survey of mammals, reptiles, and amphibians of Devils River State Natural Area-Big Satan Unit. Further funds were provided by the CITR foundation of Angelo State University for a Graduate Research Fellowship to C. N. Morgan. We also thank the Park Superintendent of DRSNA – BSU, Joe Joplin, as well as Mark Lockwood and David Riskind of Texas Parks and Wildlife Department for providing logistic support. We’d also like to thank Dr. Nicholas Negovetich for his assistance with the statistical analysis. A special thanks to the many research assistants and colleagues for field work assistance, including Austin Osmanski, Grayson Allred, and countless other field assistants, colleagues, and friends who have made fulfilling this research objective possible. F IGURE 2. Map above displays the Level III ecoregions of Texas (Omnerik 1987) as well as the locations of other P. pectoralis habitat studies conducted in TX. F IGURE 1. Map above displays the known distribution of P. pectoralis within the state of Texas. R ESUMEN : El objetivo de este estudio realizado en el Parque de Área Natural Estatal Devils River – Big Satan Unit (DRSNA - BSU), ha sido para determinar la selección de recursos vegetativos a través de las estaciones por el Ratón tobillo blanco, (Peromyscus pectoralis), vía la comparación de índices de abundancia (usando trampas de línea) con datos de vegetación geoespaciales del área. Se realizaron muestreos durante un periodo de 21 meses (febrero de 2013 a octubre de 2014), lo cual equivalió a 10,607 noches de trampa y resulto en 185 capturas de P. pectoralis. Una área equivalente al ámbito de hogar de P. pectoralis (A = 3,340m 2 ), basado en la distancia alrededor de la trampa de línea, fue espacialmente montada sobre las series de vegetación de DRSNA-BSU usando ARC-GIS. Utilizando un modelo de regresión lineal múltiple, el porcentaje de cada hábitat dentro el ámbito de hogarse estimó contra los índices de abundancia de P. pectoralis. Se observó una correlación positiva entre el invierno e otoño y varias series de vegetación: Lechuguilla-Sotol (P<0.001), Blackbrush (P<0.007), Guajillo (P<0.05), and Ashe Juniper-Oak (P<0.03). Los datos sugieren que la selección de hábitat vegetativo de P. pectoralis varía estación almente y es probable que dependen del grado de disponibilidad de recursos temporales dentro de cada tipo de vegetación. T ABLE 2. Table below displays the results of the best habitat model for the regression analysis, performed with vegetation series as independent predictor variables. T ABLE 1. Table below displays the dominant and indicator species commonly found within each of the vegetation series used in our analysis, identified by Keith (2011). F IGURE 4. Trap capture and effort data displaying A) the seasonal capture totals of Peromyscus pectoralis at DRSNA – BSU, B) the seasonal trap-number and trap-night data totals, and C) the seasonal abundance index averages. AB C F IGURE 3. Map of the Devils River State Natural Area – Big Satan Unit displaying the 15 active vegetation series present, as well as the developed and disturbed areas identified by Keith (2011). F IGURE 5. Map of the Devils River State Natural Area – Big Satan Unit displaying the trap-line vegetation buffers generated to obtain vegetation series percentages within the buffered area. Outlines of trap-lines are color-coded seasonally: Winter = yellow, Spring = red, Summer = purple, Fall = light blue.