This vegetation study describes the plant associations of El Vizcaíno Biosphere Reserve (Baja California, Mexico) as determined by the Braun-Blanquet method and supported by cluster analysis. Twenty-two associations were identified, 7 of which are described for the first time (Camissonio crassifoliae-Encelietum ventori, Encelio phenocodontae-Atriplicetum polycarpae, Lycietum brevipedis, Errazurizio benthamii-Pachycormetum veatchianae, Merremio aureae-Lysilometum candidae, Opuntio taponae-Burseretum microphyllae, and Suaedo taxifoliae-Allenrolfeetum occidentalis). Ecological, biogeographical, bioclimatic, and floristic data are provided for each of these new associations.

A model of vicariance speciation has been proposed to explain the proliferation of psammophilic (sand loving) species in several endemic genera of scorpions. The current distribution of Paruroctonus utahensis (Williams, 1968) populations on isolated sand dunes seems to fit this model; therefore, this species was examined to test the model for psammophilic population distribution and haplotype variation. A portion of the mitochondrial rDNA gene was sequenced from 44 individuals representing 7 populations of P. utahensis. Three individuals of Paruroctonus gracilior (Hoffmann, 1931) and 1 individual of Paruroctonus boquillas Sissom and Henson, 1998, were used for outgroup comparison. A maximum parsimony heuristic search indicated that the basal populations were in the south and derived populations occurred in the northeastern portion of the range. A nested clade analysis found a statistically significant correlation between haplotypes and their geographical distribution at several clade levels. The biological cause of this association was best explained by allopatric fragmentation. Our data supported the vicariance model of isolation by fragmentation of a larger habitat to explain the variation seen among populations of P. utahensis.

The helmet and large spines of zooplankton inhibit predation by small fishes. Daphnia lumholtzi, a long-spined, helmeted cladoceran, has recently invaded Lake Texoma (Oklahoma–Texas) and much of southeastern North America. To determine if inland silversides (Menidia beryllina) were size-selective predators on D. lumholtzi, we conducted laboratory experiments using 3 size-classes of fish (22 to 35 mm, 46 to 60 mm, and 67 to 80 mm). Inland silversides were allowed to feed for 80 min on D. lumholtzi of various sizes. Prey selection differed among the 3 silverside size-classes. Small inland silversides ate more small D. lumholtzi (total length 0.99 to 1.75 mm) than did large and medium inland silversides. Large inland silversides ate more large (2.62 to 3.44 mm) and very large (3.5 to 5.25 mm) D. lumholtzi than did smaller inland silversides. To examine if these trends occur in nature, we compared the lengths of D. lumholtzi and native Daphnia species (D. parvula and D. galeata) eaten by large and small inland silversides collected from Lake Texoma. As in the laboratory experiment, small inland silversides ate smaller D. lumholtzi than did large inland silversides. Although small inland silversides ate smaller D. lumholtzi, D. lumholtzi might be an important food source for young-of-the-year inland silversides, because D. lumholtzi reaches peak abundances during summer when native daphnid densities are low in Lake Texoma.

Turtle assemblages in Oklahoma are potentially interesting because the species pool is a mixture of eastern and western species, several of which are approaching their longitudinal geographic range boundaries in Oklahoma. From 1996 to 2002, we captured 1,940 turtles in 12 assemblages in central Oklahoma. Pond area ranged from 0.04 to 26.1 ha. Species composition was a mixture of 2 ubiquitous and 3 rare species. Trachemys scripta and Chelydra serpentina occurred in 11 and 10 assemblages, respectively. The rare species included Pseudemys concinna (4 assemblages), Kinosternon flavescens (3), and Apalone spinifera (1). In ponds with multiple species, T. scripta was always the most abundant species (mean ± SD percentage of individuals captured = 80.0 ± 27.4%) and C. serpentina was always the second most abundant (18.5 ± 27.7%). Population density and biomass of T. scripta and C. serpentina varied an order of magnitude and approached maximum published values.

We characterized differences in escape behavior among syntopic parthenogenetic and gonochoristic whiptail lizards (Teiidae: Aspidoscelis) in western Texas by performing 2 experiments. The first experiment revealed that Aspidoscelis exsanguis (parthenogenetic) allowed a human simulated predator to approach significantly closer than Aspidoscelis neomexicana (parthenogenetic), and both species allowed a significantly closer approach than Aspidoscelis tigris (gonochoristic). Furthermore, A. neomexicana and A. exsanguis did not flee as far as A. tigris, but this difference in flight distance was not statistically significant. Temperature was not correlated with approach distance or flight distance for any of the above species. The second experiment determined that, when captured, Aspidoscelis tesselata (parthenogenetic) was more likely to release cloacal contents on a human captor than A. tigris (gonochoristic).

Reproductive and fat body cycles were studied in a population of viviparous lizards (Sceloporus grammicus) from the Teotihuacán area, state of Mexico between October 1984 and September 1985. Males reached sexual maturity at a snout-vent length (SVL) of 44.5 mm and at an age of 6 to 7 mo; females reached sexual maturity at 44.1 mm SVL and at an age of 8 to 9 mo. Testicular mass of adult males began to increase in size in May and peaked during July, August, and September. Maximum testicular growth was associated with precipitation, but not with temperature and photoperiod. Enlarged vitellogenic follicles appeared during October, and females contained embryos in stages 1 through 8 during November. Females had embryos in later embryonic stages 29 through 40 from January to April. Vitellogenesis and embryonic development were associated with increasing photoperiod, temperature, and precipitation. Mean litter size was 5.09 ± 0.24 (2 to 9). Mean RCM was 0.338 ± 0.024 (0.015 to 0.80). Litter size and RCM were positively correlated with female SVL.

We studied movements of 36 juvenile Mohave ground squirrels (Spermophilus mohavensis) during their first active season, until immergence into hibernation, a period that lasted from late May to late August. For all radiocollared individuals, the mean greatest distance between any 2 locations was 1,520 m for males (range 110 to 6,230 m) and 505 m for females (range 105 to 3,862 m). Eleven individuals (6 females, 5 males) with known litter affiliations were followed from their initial capture until inactive underground: males moved a mean distance of 2,920 m (range 0 to 6,230 m), and females moved a mean distance of 753 m (range 0 to 3,862 m) between their natal area and hibernation site. Ten individuals (8 males, 2 females) moved distances >1 km. Individuals making long-distance movements were likely to have traversed habitat that would be considered marginal for adult Mohave ground squirrels. Long-distance movement by juveniles might be critical for connecting local populations and recolonizing sites after local, drought-related extirpation.

We conducted a preliminary study using small field crews, a single Anabat II detector coupled with a laptop computer, and point transects to examine patterns of bat activity at a scale of interest to local resource managers. The study was conducted during summers of 1996–1998 in Bandelier National Monument in the Jemez Mountains of northern New Mexico, a landscape with distinct vegetation zones and high species richness of bats. We developed simple models that described general patterns of acoustic activity within 4 vegetation zones based primarily on nightly variation and a qualitative index of habitat complexity. Bat acoustic activity (number of bat passes&sol point) did not vary dramatically among a limited sample of transects within a vegetation zone during 1996. In 1997 and 1998, single transects within each vegetation zone were established, and bat activity did not vary annually within these zones. Acoustic activity differed among the 4 vegetation zones of interest, with the greatest activity occurring in riparian canyon bottomland, intermediate activity in coniferous forest and a 1977 burned zone, and lowest activity in piñon-juniper woodlands. We identified 68.5% of 2,529 bat passes recorded during point-transect surveys to species using an echolocation call reference library we established for the area and qualitative characteristics of bat calls. Bat species richness and composition differed among vegetation zones. Results of these efforts were consistent with general knowledge of where different bat species typically forage and with the natural history of bats of New Mexico, suggesting such a method might have value for drawing inferences about bat activity in different vegetation zones.

Seventeen middens of the bushy-tailed packrat (Neotoma cinerea), ranging in age from 3450 ± 40 to 160 ± 50 radiocarbon years before present (B.P.), were collected in 1996 from the Upper Gunnison Basin, Colorado. Eight of these subfossil middens contained plant remains from species that no longer occur near the locations of the middens. Radiocarbon dates on conifer remains indicated shifts in forest communities in the late Holocene at 3320, 560, and 240 B.P. Past climates were cooler than today at 3320 B.P., when mixed lodgepole (Pinus contorta) and ponderosa pine (P. ponderosa) forest expanded to lower elevations in the basin. Conditions became warmer between 1500 to 950 B.P., with additional cooling from 660 to 170 B.P., corresponding with cooling periods in the 1300s and during the Little Ice Age (500–150 B.P. or 1500–1850 CE). These data indicated that the late Holocene of Colorado was marked by episodes of climate change that previously have not been identified in the paleoclimatic record.

We investigated the hypothesis that the squirting of blood from orbital sinuses by Texas horned lizards (Phrynosoma cornutum) is an antipredator defense against some mammalian species. Coyotes (Canis latrans) were tested for the first time. As expected, Texas horned lizards squirted blood in response to coyote attacks, and coyotes exhibited startle response, avoidance response, or both as a result of these events. Whereas lizard carcasses mixed into normal food-mash elicited regurgitation by coyotes, possibly due to physical effects, blood of horned lizards similarly mixed into food did not. Coyote responses to simulated squirts of 5 compounds, including blood of horned lizards, into 3 potential sensory target areas (eyes, nose, and mouth) strongly suggested that aversive effects were mediated by receptors in the oral or nasal cavities. Coyote responses were more frequent to delivery of blood plasma and whole blood of Phrynosoma to buccal and nasal membranes than to delivery of plasma and blood from spiny lizards (Sceloporus jarrovii) or to delivery of a saline control to these membranes. We concluded that Texas horned lizards squirt blood from sinuses surrounding the eyes during attacks by canids, such as coyotes, and that this blood affects oral receptors, causing a negative response in coyote attack behavior that potentially increases survival of the lizards.

Track counting is a quick, easy, cost-effective technique to estimate the population density of white-tailed deer (Odocoileus virginianus). By employing the double-count procedure, I generated 2 models in which track counts were calibrated to estimate the density of deer in dry tropical forest in Chamela on the Mexican Pacific Coast. For both models, I calibrated a track index using density obtained from the line-transect method as a reference. The first model was based on simple linear regression and the second on the strip-transect method. I discuss the usefulness of these models for monitoring local populations as well as possible applications in other regions.

Maps of vertebrate distributions resulting from the Kansas Gap Analysis were compared with maps based on ecological-niche modeling of primary point-occurrence information. The latter maps were considerably more predictive of independent sets of test data than were the Gap maps, further supporting the idea that the method of Gap wildlife-habitat modeling has little analytical power. In general, the Kansas Gap maps were untested and unchecked, thus being heterogeneous, unreliable, and of little use for further analyses.

A recent survey of benthic macroinvertebrates of the Cuatro Cienegas basin found 118 species in the 21 sites collected. Four exotic macroinvertebrates that could threaten the native biota were found within or near the basin.

Sampling from April 2001 to March 2003 revealed San Felipe Creek, Val Verde County, Texas, still supported a breeding population of the threatened Devils River minnow (Dionda diaboli). The species was restricted to creek habitats and was not found in the outflow channels of San Felipe Springs. We determined that breeding populations of introduced tropical fish species Hypostomus (a South American armored catfish) and Oreochromis aureus (an African cichlid) also were present in San Felipe Creek. We obtained evidence to suggest that presence of exotic species, particularly Hypostomus, might have a negative effect on the Devils River minnow. We recommend periodic monitoring of fish populations in San Felipe Creek to document future changes in the population of Devils River minnow and other endemic species, and to facilitate design and implementation of conservation plans in San Felipe Creek.

We examined 50 amphibians of 4 species (Bufo marinus, B. nebulifer, Rana berlandieri, and Scaphiopus couchii) from 4 localities in the state of Nuevo León, Mexico. We collected 7 species of digeneans (Cephalogonimus americanus, Clinostomum sp. (metacercariae), Haematoloechus complexus, Halipegus occidualis, Langeronia cf. macrocirra, L. cf. jimenezi, and Megalodiscus americanus) and 1 monogenean (Pseudodiplorchis americana), 3 cestodes (Distoichometra bufonis, Ophiotaenia filaroides, and larvae of 1 proteocephalidean), and 4 nematodes (Cosmocercoides sp., Chabaudgolvania sp., Rhabdias fülleborni, and R. savagei). Eleven host records are new, and the geographic ranges of C. americanus, H. occidualis, P. americana, O. filaroides, Chabaudgolvania sp., Cosmocercoides sp., R. fülleborni, and R. savagei are expanded by these records.

We compared several life history characteristics between 2 syntopic color pattern classes (Conchas C-E and Conchas D) of parthenogenetic Aspidoscelis tesselata from Conchas Lake, New Mexico. These pattern classes lacked significant differences in mean clutch size, mean size of gravid females, SVL frequency distributions of gravid females, and SVL frequency distributions of complete samples. We then compared the Conchas Lake assemblage to the other 2 pattern class assemblages (Higbee, Colorado, and Sumner Lake, New Mexico) whose life history characteristics had been previously reported. Comparisons of life history attributes within and between the 3 assemblages augment morphological and genetic evidence that each population of pattern class D was derived by mutation from individuals of pattern classes C or C-E in different geographic areas.

We studied the reproductive characteristics, such as snout-vent length (SVL) at sexual maturity, clutch size, and egg volume, and morphological characteristics [head length (HL), head width (HW), femur length (FL), and tibia length (TL)] in Aspidoscelis deppii and A. guttata from Oaxaca, Mexico. In A. deppii, males were larger in SVL than females, whereas in A. guttata, females were larger than males. Males of A. deppii had larger morphological structures, such as HL, HW, FL, and TL, than females; in contrast, males and females of A. guttata were similar, except in TL, in which females were larger. Mean clutch size was 2.3 for A. deppii and 4.5 for A. guttata. Females of A. deppii had smaller egg volume than females of A. guttata. Both A. deppii and A. guttata respond to the environmental factors from this region, but in different ways.

Herpetofaunal diversity of the Fort Wolters Military Installation, Texas, was surveyed from September 2002 through August 2003. During the 1-y sampling period, all amphibian and reptile sampling methods combined resulted in 1,421 captures of 10 species of amphibians (all anurans) and 25 species of reptiles. These actions represent an effort of the Texas Army National Guard to inventory the flora and fauna of federal property and create a baseline for future monitoring and management.

Avian collisions with human-made objects are relatively common; however, documented avian collisions with naturally occurring objects are rare in scientific literature. On 22 April 2002, we observed a tree swallow (Tachycineta bicolor) entangled in the spines of a teddy bear cholla (Opuntia bigelovii) in the Tule Mountains, Cabeza Prieta National Wildlife Refuge, Arizona. Two possible causes of this incident include the swallow flying into the teddy bear cholla, possibly being blown by a gust of wind, or being placed there by a loggerhead shrike (Lanius ludovicanus). Because of the flying abilities of swallows and documented diets of loggerhead shrikes, either would be unusual.

In New Mexico, the tawny-bellied cotton rat (Sigmodon fulviventer) previously was known only from central and southwestern parts of the state. In central New Mexico, most records were from areas of tall grass and marshes associated with the middle Rio Grande valley. In 2003, we discovered S. fulviventer in grassy and marshy habitats >100 km west of the Rio Grande in west-central New Mexico. Because past surveys in this region did not report captures of Sigmodon, we suspect our distributional records represent recent westward expansion of S. fulviventer in the state.

We observed black bears (Ursus americanus) foraging on aggregations of army cutworm moths (Euxoa auxiliaris) (Lepidoptera: Noctuidae) in subalpine felsenmeers (block fields) in the Jemez Mountains of New Mexico. Moth aggregations serve as food for grizzly bears (Ursus arctos horribilis) in the northern Rocky Mountains of Wyoming and Montana. However, black bears have not been reported to use these aggregations, nor have such aggregations been documented to occur this far south in the Rocky Mountains.