We analyzed pairwise species associations based on presence/absence data from transects of perennial vegetation at 129 sites located in the northern Chihuahuan Desert. The sites were chosen to represent the range of commonly encountered desert vegetation types from grassland to shrubland. Twenty-four significant associations were found, all of which were positive (i.e., the species pairs co-occurred more frequently than would be expected by chance). Some of the species associations identified by our analysis have been noted in the literature previously, but others have not and warrant further investigation. Given the nature and scale of the data used in our analysis, the associations we uncovered most likely represent species with similarities in ecological requirements or tolerances, but might or might not indicate strong interaction at the scale of individual plants.

Seed germination of 4 species of Mammillaria (Cactaceae) from the Tehuacán-Cui-catlán Valley in central México were compared at 4 light treatments (red, far-red, white light, and darkness) at 25°C and at 2 light treatments (white light and darkness) at 2 alternating temperatures (15/30°C and 20/35°C). Seeds were positive photoblastic, but germinated at far-red light. For all species, best germination results were obtained at 25°C with white and red light, and there were no significant differences among treatments. Higher germination percentages were obtained at 25°C than at alternating temperatures. None of the species exhibited any morphophysiological mechanisms of dormancy.

The spotted gar, Lepisosteus oculatus, population in the Lake Pontchartrain estuary in southeastern Louisiana was sampled monthly to characterize its reproduction, age, and growth from November 1998 through October 1999. Males ranged from 285 to 568 mm standard length (SL) and 0 to 8 years; females ranged from 395 to 606 mm SL and 0 to 10 years. Both sexes were mature before age 2. Spawning was from February to June, and fecundity was highest during October (mean number of eggs = 13,798 ± 7,654 SE) and lowest during June (mean number of eggs = 1,772 ± 392 SE). Ova diameters varied among months and among individuals within a month. Mean ova diameter during the spawning season was 3.02 mm ± 0.02 SE. Males grew significantly faster than females, but had a shorter lifespan. Condition was generally lowest during fall and varied by sex among seasons.

In the southwestern United States and northern Mexico, obligate brood parasitic bronzed cowbirds (Molothrus aeneus) and brown-headed cowbirds (M. ater) are sympatric and parasitize some of the same host species. Host species selection is driven, in part, by cowbird breeding habitat preferences, yet little is known about bronzed cowbird habitat selection, and no study has quantified bronzed cowbird and brown-headed cowbird habitat selection in sympatry. During 1998 and 1999, I compared the abundance and distribution of brown-headed cowbirds and bronzed cowbirds and the hosts they parasitize across 151 point-count locations in 10 major habitat types in the Huachuca Mountains and San Pedro River valley of southeastern Arizona. These sympatric brood parasites overlapped extensively in 4 of 10 habitats surveyed (pine-oak forest, urban, montane riparian, and foothill riparian). Passerine species richness was a better predictor of brown-headed cowbird and bronzed cowbird presence than host abundance. When combined with distance to foraging area and habitat type, passerine species richness was the best predictor of cowbird presence in logistic regression models.

The distribution of the Altamira oriole (Icterus gularis) in the United States is restricted to the Lower Rio Grande Valley of southernmost Texas. Our objective was to assess the current breeding status of this species at and near Santa Ana National Wildlife Refuge (SANWR), a principal nesting area in the region. During each breeding season from 1997 to 1999, we attempted to find and monitor all Altamira oriole nests in and within 0.5 km of SANWR to determine annual breeding-pair abundance, annual nest-site distribution, and nest distribution among tree species. We estimated that there were 6 or 7, 8 or 9, and 6 or 7 breeding pairs in 1997, 1998, and 1999, respectively. Of the 26 nests found, 12 (46%) were next to the Rio Grande and 7 (27%) were at the edge of a major wetland within the refuge. Only 1 of the other 7 nests was more than 115 m inside the refuge. Nests (n = 26) were distributed among Salix nigra (39%), Fraxinus berlandieriana (23%), Acacia minuata (19%), Celtis laevigata (8%), Ulmus crassifolia (8%), and Chloroleucon ebano (4%). Comparisons of results with historical accounts suggest a decline in annual breeding abundance and nest-site availability at SANWR since the early 1970s.

We assessed population trends of the endangered golden-cheeked warbler (Dendroica chrysoparia) at Fort Hood, Texas, using point-count data from 1992 through 2001. We assessed the effect of a large-scale fire in 1996 on these population trends and the effect of military training activities on relative abundance of golden-cheeked warblers. Results indicated that the population has increased steadily at Fort Hood during the 10-yr period. The 1996 fire might have caused a decrease in detections from 1996 through 2001, but excluding any localized effects of the fire, population size continued to increase. Analyses of data from areas with and without military training detected no effect of these activities on the relative abundance of golden-cheeked warblers. Habitat protection and a cowbird-control program might have contributed to the increasing population of golden-cheeked warblers at Fort Hood during the period of study. Data on population trends, demography, and factors limiting population size are needed from other areas to understand more fully the threats to and management needs of the golden-cheeked warbler.

Diets of double-crested cormorants (Phalacrocorax auritus) wintering on the central Texas coast were determined from the contents of 51 stomachs from Lavaca Bay and 28 stomachs from upper Laguna Madre. Diets consisted primarily of striped mullet (Mugil cephalus) and Gulf menhaden (Brevoortia patronus) in Lavaca Bay and striped mullet and Gulf toadfish (Opsanus beta) in upper Laguna Madre. Birds seemed to prefer striped mullet in both bays, and these fish were significantly larger than the other fish consumed. There were no differences between bays in the numbers of fish in bird stomachs; however, fish consumed by birds in upper Laguna Madre were larger than those taken by birds in Lavaca Bay. Based on the results of this study, there is no evidence that double-crested cormorants pose a significant threat to commercial or sport fisheries on the Texas Gulf Coast.

We examined morphological characteristics, abundance, survival rates, and causes of mortality for striped skunks (Mephitis mephitis) in the Southern High Plains region of Texas during 1994 and 1995. We captured 69 male skunks 133 times and 35 female skunks 57 times on 4 12.8-km² plots. Estimates of abundance on plots ranged from 0.10 to 0.71 skunks/km². Annual survival rates were 0.40 in 1994 and 0.48 in 1995. Survival curves did not differ significantly between years. From the 26 recovered carcasses, 50% of skunks died from human-caused trauma. Shooting of skunks around occupied houses was the largest single cause of mortality.

We found 44 Corynorhinus rafinesquii and 1 Myotis austroriparius tree roosts during the summer of 2000 on D'Arbonne National Wildlife Refuge in northeastern Louisiana. Corynorhinus rafinesquii and M. austroriparius are species of concern throughout their range. Little information exists on tree roosts of either species. The roost trees found in our study were hollow Nyssa aquatica with triangular-shaped basal openings, and all were located within a 5-ha tupelo stand. A high density of large-diameter cavity trees and a closed canopy characterized this stand. A canal bisected the stand and provided a source of permanent water. Roost trees and non-roost trees could not be compared due to biases in sampling technique. More research directed at roost selection factors by these bats is needed to form conservation plans; however, preliminary information indicates many, large-diameter cavity trees, especially Nyssa tree species, within bottomland hardwood forests are important as tree roosts for these bats.

We examined factors influencing foraging behavior of lesser long-nosed bats, Leptonycteris curasoae, in southeastern Arizona. When L. curasoae are present in this region, their diet is restricted to nectar and pollen from 1 species, Agave palmeri, which has heightened conservation concerns for the plant as forage for this endangered bat. We found that visitation rates of L. curasoae to individual A. palmeri near a roost were high (mean = 273 ± 17 visits per plant per hour) and varied with the spatial distribution and morphological characteristics of individual plants. Specifically, visitation rates varied with time of night, distance and orientation from the bat roost, and number and relative vertical position of flowers along the inflorescence. We suggest that both spatial distribution and temporal variation in flowering chronology be considered when developing strategies to manage A. palmeri to support L. curasoae.

Mammals recovered from archaeological investigations at Bighorn Cave represent the first mid-Holocene to late-Holocene faunal record from the Black Mountains of northwestern Arizona. With 2 exceptions, identified taxa have modern geographic distributions that include the location of the cave. A beaver (Castor canadensis) incisor from the lower shelter of the cave was possibly introduced into the deposit by humans. Microtus teeth from Bighorn Cave represent an extralocal record, and several explanations for their presence at Bighorn Cave are discussed. A new radiocarbon date presented here indicates that post-depositional mixing occurred in portions of the deposit.

First record of the robber fly, Microstylum morosum (Diptera: Asilidae), is reported for Arkansas (Clark County). Individuals of this species were observed, and a voucher specimen collected, from a blackland prairie in the southwestern corner of the state. Previously, M. morosum was only known from Kansas, Oklahoma, Texas, New Mexico, Arizona, and Colorado. The nearest occurrence to the reported Arkansas record is approximately 200 miles to the west in Texas.

Cyprinodon rubrofluviatilis (Red River pupfish) is established in a stream tributary to the Cimarron River in Oklahoma. Further work is needed to determine the extent of the distribution of this nonnative species in the Arkansas River system.

On 30 September 2000, 8 specimens of Fundulus olivaceus were collected in Shoal Creek, Cherokee County, Kansas. These specimens represent the first record of F. olivaceus in Kansas.

Grass carp, Ctenopharyngodon idella, is an invasive cyprinid that has been stocked throughout the USA for biological control of aquatic macrophytes. Natural reproduction was expected to be limited in the USA due to a lack of long, unimpounded rivers with sufficient flow needed for reproduction. However, natural reproduction has been documented in the Mississippi, Missouri, and Trinity (Texas) rivers. In 2 years of shoreline seining in Lake Texoma (Oklahoma-Texas), we found evidence of natural reproduction by grass carp in the Red and Washita rivers. In 1999 and 2000, we collected 79 juvenile grass carp from 6 of 41 sites sampled in the reservoir. Juvenile length ranged from 15.2 to 60.0 mm total length, with 2 apparent size groups (modes of 18 and 41 mm). Most grass carp juveniles were collected in 2000 in the upper Washita River arm of Lake Texoma. We hypothesize that tributary conditions and littoral habitats in Lake Texoma influenced grass carp abundance and survival.

Prior to 1987, the Pacific treefrog (Pseudacris regilla) was known from only 4 sites on the lower Colorado River, which borders Nevada, California, and Arizona. We conducted surveys on the lower Colorado River from Princess Cove on Lake Mohave to the Bill Williams River confluence and upstream on the Bill Williams River to Planet Ranch from 1987 through 2002, and we consulted museums and herpetologists working in the area for P. regilla localities. We found this frog to be well distributed along 90 km of the lower Colorado River and its backwaters from Davis Camp, just below Davis Dam, to Castle Rock in upper Lake Havasu. We report 33 new localities and confirmed presence at 1 historical site. We found P. regilla primarily in cattail and bulrush marshes along the main channel and in backwaters. What limits its distribution on the lower Colorado River is unknown, because cattail and bulrush marshes occur both north and south of the current range. Pseudacris regilla populations below Davis Dam are probably currently disjunct from those on the Overton Arm of Lake Mead and in the Las Vegas Valley. In Arizona, this species also has been recorded as an apparent introduction at Middle Spring and a nearby stock tank in the Virgin Mountains, Mohave County, and at 2 central Arizona plant nurseries. The species persisted for at least 19 years and successfully bred at 1 nursery, where it was reportedly introduced by frogs hitchhiking on ornamental plants imported from San Diego.

According to Dr. Philip Tibbs, past Secretary of the ICZN, and at the time speaking for the Commission, the name Eumeces epipleurotus Cope (1880) has priority over E. gaigeae Taylor (1935). E. epipleurotis is now considered a subspecies of E. multivirgatus.

The gray catbird (Dumetella carolinensis) has been documented visiting and breaking the eggs of artificial nests, but the implications of such observations are unclear because there is little cost in depredating an undefended nest. During the summer of 2001 at Konza Prairie Biological Station, Kansas, we videotaped a gray catbird that broke and consumed at least 1 egg in a brown thrasher (Toxostoma rufum) nest. Our observation was consistent with egg predation because the catbird consumed the contents of the damaged egg after breaking it. The large difference in body mass suggests that a catbird (37 g) destroying eggs in a thrasher (69 g) nest might risk injury if caught in the act of predation and might explain why egg predation by catbirds has been poorly documented. Our observation indicated that the catbird should be considered as an egg predator of natural nests and that single-egg predation of songbird nests should not be attributed to egg removal by female brown-headed cowbirds (Molothrus ater) without additional evidence.

We recorded for the first time in Cuatro Cienegas Basin, Coahuila, Mexico, 43 species, 33 of them represented by specimens and 10 only by visual observations. These records increase the species known for the basin by 38.4%. The fieldwork was conducted during 1996 and 1997. Two species are new records for the state of Coahuila: Larus atricilla and Camptostoma imberbe.

Interspecific and intraspecific relationships have received much attention from ecologists and play an important role in the structure of wildlife communities, but researchers have rarely attempted to study interactions between phylogenetically unrelated organisms. We observed 6 interactions between badgers (Taxidea taxus) and red-tailed hawks (Buteo jamaicensis) between February and June 1999 on the Barry M. Goldwater Tactical Range (32°53′06″N, 112°43′24″W) near Ajo, Arizona. Each interaction lasted 5 to 20 minutes. Interactions were characterized by a red-tailed hawk following a badger as it searched for prey. On 2 occasions, red-tailed hawks apparently attempted to take prey items escaping from a badger. Badgers and red-tailed hawks feed primarily on small mammals, suggesting possible competition between the 2 species. Further research is needed to determine if badgers and red-tailed hawks in the Sonoran Desert are competing for a limiting resource or if one species (presumably the red-tailed hawk) is benefiting from the interaction without affecting the other species (badger).

The eastern fox squirrel (Sciurus niger) has continued to move westward along riparian corridors in the southwestern United States. This species is now established along the Arkansas River and its major tributaries in southeastern Colorado. In addition, a population of S. niger exists in northern New Mexico, and this population might be the result of natural expansion from populations along the Purgatoire River in southeastern Colorado.

We report on intrasexual spatial patterns and dynamic interactions of northern raccoons in eastern Kansas. In general, extensive spatial overlap occurred among female and male dyads in each season and among age classes. The dynamic interaction test revealed positive interaction among 5 of 9 male dyads, and no evidence of negative interaction. Some male dyads denned and traveled together for intervals of varying lengths (2 to 16 days) in each season of the year. There was no evidence of dynamic interaction among female dyads. We provide additional evidence that social behavior of male raccoons exhibits geographic variation, and some level of sociality might be more common than currently described for this species.

This paper reports the first record of the bats Enchisthenes hartii and Natalus stramineus in the state of Hidalgo and of the rodent Nyctomys sumichrasti in the state of Guerrero, Mexico. These records increase the known range of these species. External and cranial measurements and reproductive condition of the specimens are reported. We also include the habitat where they were captured and the presence therein of other small mammals.

The population dynamics of Peromyscus levipes were studied in a deciduous forest in southeastern Morelos, Mexico. This study was performed at 2 sites, one having relatively little disturbance (low-disturbance site) and the other having extensive disturbance (high-disturbance site). Sampling was conducted from February 1991 through March 1992. At the low-disturbance site, the density in February 1991 was 39 individuals/ha. Although population size fluctuated temporally, it generally declined throughout the study until March 1992 (16 individuals/ha). Sex ratios of the populations were not significantly different. The reproductive pattern exhibited by P. levipes in the low-disturbance site was seasonally polyestrous. In the high-disturbance site, we caught only 6 individuals.

We report the first record of the Inyo shrew, Sorex tenellus, from the central Great Basin. A single specimen was captured in Great Basin National Park at 3,000 m elevation in habitat dominated by Picea engelmanni. This constitutes a northeastern range extension of 300 km into east-central Nevada.