Vegetation types in Cumbres de Majalca National Park were classified using cluster analysis from 195 line transects and 975 1-m² quadrats at 39 sites. Six plant associations emerged from oak-grassland and oak-conifer communities. Each association is described using cover, constancy, elevation, physiographic, and soil parameters. A vegetation profile and a map of the plant associations are presented.

We compared production and breakdown of Fremont cottonwood (Populus deltoides wislizenii) leaf litter at matched floodplain sites on the regulated Green River and unregulated Yampa River in semi-arid northwestern Colorado. Litter production under trees was similar at sites in 1999 (250 g/m², oven-dry) but lower in 2000 (215 and 130 g/m²), a drought year that also featured an outbreak of defoliating beetles at the Yampa River site. Our production values were similar to the few others reported for riparian forests within semi-arid or arid areas. Leaf litter in portions of the floodplain not inundated during the spring flood lost organic matter at the same rate as leaves placed in upland sites in 1998 and 2000: 35 to 50% of organic matter during an approximately 160-day spring and summer period. Inundated litter lost 55 to 90% of its organic matter during the same period. Organic matter loss from inundated leaves increased with duration of inundation and with deposition of fine sediment. Pooled across locations, leafpack data suggested that nitrogen concentration (mg N/kg organic matter) increased until about 65% of the initial organic matter was lost. This increase likely reflected the buildup of microbial decomposer populations. The role of insects and other macroinvertebrates in litter breakdown apparently was minor at both sites. Large spatial and temporal variation in litter dynamics in aridland floodplain settings is ensured by microtopographic variation in the alluvial surface coupled with year-to-year variation associated with most natural flood regimes. Factors reducing flood flow frequency or magnitude will reduce overall breakdown rates on the floodplain towards those found in drier upland environments.

Activity above ground by the white desertsnail, Eremarionta immaculata, in relation to air temperature and humidity (saturation deficit) was observed during daylight 2 to 5 days following rainfall of 12 mm on 11 January 2001. A maximum of 12 snails was observed active within a 15-m by 15-m quadrat near the type locality of the species in the Riverside Mountains, California. Number of active snails increased as air temperature (8 to 23°C) increased and as saturation deficit (2.5 to 16 mm Hg) decreased. Most variation in number of active snails was attributed to saturation deficit. Active snails preferred epiphyta (lichen and moss) as a substrate compared with other types of substrate (plant detritus and 4 size-classes of rocks) within the quadrat. Epiphyta might provide active snails with food or with a moister surface, increasing water absorption.

We examined 404 specimens representing 7 species of sceloporine lizards from Mexico (Sceloporus formosus, S. grammicus, S. megalepidurus, S. mucronatus, S. parvus, S. torquatus, and S. variabilis) for helminths. Two species of Cestoda (Oochoristica scelopori and Mesocestoides larvae) and 9 species of Nematoda (Atractis penneri, Physaloptera retusa, Spauligodon giganticus, S. oxkutzcabiensis, Strongyluris similis, Thubunaea intestinalis, Ascarops larvae, Physocephalus larvae, and Acuariidea larvae) were found. The mean number of helminth species for each species of lizard was 4.7 ± 1.5 SD (range = 3 to 7). Twenty-six new host records and 23 new locality records are reported.

Whiskered screech-owls (Otus trichopsis) and northern pygmy-owls (Glaucidium gnoma) delivered freshly caught Yarrow's spiny lizards (Sceloporus jarrovi) and striped plateau lizards (S. virgatus) to nestlings from dusk to dark in southeastern Arizona. This observation stimulated studies of the prey deliveries by the owls and lizard activity patterns, because the lizards are not known to be nocturnal. Lizards were more frequent prey of both owls than endothermic vertebrates but infrequent compared to arthropods, a pattern in the pygmy-owl that differs from its northern populations. Yarrow's spiny lizard, the most abundant and frequently captured lizard, was most active in the morning but also active in the evening. Striped plateau lizard, the second most abundant and depredated species, had morning and evening peaks of activity. Few lizards, including S. clarki and Urosaurus ornatus, but not Cnemidophorus exsanguis and C. sonorae, were active at or after dark, when relatively few were captured by the owls.

Between 1975 and 2000, 4,525 sightings of wintering bald eagles (Haliaeetus leucocephalus) were recorded at Mormon Lake in northern Arizona. Numbers of wintering eagles fluctuated little in the 20 years from 1975 through 1994 (5.5 ± 3.0 mean sightings per day). However, during the winters of 1995 through 1997 local record highs of 59 to 118 eagles increased mean sightings per day to 22.4 ± 9.6. This dramatic population increase led to a major change in social behavior favoring consistent communal roosting; maximum roost counts of 2 to 8 eagles scattered among 11 roosts in prior years shifted to maximums of 33 to 45 eagles regularly using 2 roosts during 1996 and 1997. Winter population averaged 58% adults and 42% immatures, but during 5 recent years of greatest numbers (≥40 eagles in 1989, 1995 through 1997, and 2000), the proportion of immatures increased to 58%. Local increases in the wintering eagle population at Mormon Lake were largely attributable to this greater proportion of immature bald eagles. Both age classes peaked in February, with adults more abundant during October through December and immatures more abundant from January through April. Weekly maximum counts for 1995 through 1997 indicated changing weather and prey conditions resulted in annual variation in local numbers and habitat use of wintering bald eagles.

Our study documents home range, habitat use, and diet of Gould's turkey (Meleagris gallopavo mexicana) in the Peloncillo Mountains of New Mexico and Arizona. Fieldwork began in 1989 and consisted of 2 field seasons from May to August, with periodic winter and spring forays. The study was conducted during a drought period with only 58% of average rainfall. Combined annual home range (mean = 4,385 ha, SE = 1,845) of radio-equipped Gould's hens was similar to home ranges of other wild turkey subspecies that inhabit arid regions of the western United States. Preferred Gould's turkey habitat consisted of pinyon-juniper woodland with an abundance of pinyon ricegrass (Piptochaetium fimbriatum). In addition, 3 riparian habitat types were used disproportionately to their availability. A food habits analysis showed that the diet of turkey in our study consisted primarily of juniper (Juniperus deppeana) and manzanita (Arctostaphylos pungens) fruit, though mustard forbs (Cruciferae spp.) and pinyon ricegrass also were utilized. Periodic drought combined with livestock grazing in the Peloncillo Mountains most likely is a major factor determining forage availability and subsequent habitat use by this turkey population. A combination of natural and anthropogenic factors might result in fluctuations in the Gould's turkey population.

The High Plains of Texas is one of the southernmost nontraditional breeding areas for many duck species in North America. Because of a paucity of information on breeding ducks there, we conducted roadside surveys of breeding ducks and their habitats during May and June from 1988 through 1992. Breeding pairs of 15 species were observed on 6 types of ponds (natural and man-made wetlands containing surface water). Mallard (Anas platyrhynchos) density ranged from 9.1 to 23.1 pairs/100 km², and density for all species combined ranged from 14.8 to 46.7 pairs/100 km² (all years and survey periods included). Occupancy rates were highest on playa lakes and impoundments, though all pond types had occupancy rates exceeding 26% (all surveys and years). Duck pairs per occupied pond were highest on playa lakes (>7 and >4 on May and June surveys, respectively), followed by impoundments (>5 and >2) and entrenched draws (>2 and >3). Although the density of breeding pairs in the High Plains of Texas (47 pairs/100 km²) is generally lower than in prominent nesting areas (e.g., >200 in the San Luis Valley, CO; >600 in central Montana; >2,000 in California; >4,000 in the Prairie Pothole Region), information reported here further confirms the use of the Playa Lakes Region by breeding ducks and illustrates its importance as a major habitat area for waterfowl in the Central Flyway.

A partial left dentary containing c1 and p2-m1 is described as the second reported specimen of Bassariscus ogallalae, a taxon erected nearly 70 years ago. The specimen is from an early late Miocene (late Barstovian/early Clarendonian) exposure of the Ogallala Group in Ellis County, Kansas. Comparison of B. ogallalae to specimens of other species of Bassariscus, including a relatively large sample of B. astutus, reveals that only 2 of the original 4 diagnostic characters remain valid. A relatively large series of measurements and associated ratios reveal no new meristic characters that can be used to distinguish B. ogallalae from other species of Bassariscus. This specimen extends the temporal range of Bassariscus in Kansas from the Recent to the early late Miocene.

Abert's squirrels (Sciurus aberti) are thought to depend on ponderosa pine (Pinus ponderosa) for food sources, cover, and nest sites. Records of Abert's squirrels using other food sources, forest types, and nest trees are rare. In the 1940s, Abert's squirrels were introduced to ponderosa pine forests on Mount Graham in the Pinaleño Mountains of southeastern Arizona. Since 1989, while studying Mount Graham red squirrels (Tamiasciurus hudsonicus grahamensis), we recorded Abert's squirrel sightings. In over 41,000 field-hours of studying Mount Graham red squirrels, we documented 498 Abert's squirrel sightings in both mixed-conifer and spruce-fir forests. Behaviors observed included feeding, collecting nest material, nest building, intraspecific chases, and interactions with the endangered red squirrels in this nontraditional habitat.

As a signatory of the Conservation Assessment and Conservation Strategy for Swift Fox, the state of New Mexico is required to manage swift foxes (Vulpes velox) with special diligence. However, basic ecological data for the swift fox in New Mexico are lacking. I examined swift fox demography, home range size, dispersal, den site selection, and diet in northeastern New Mexico. Juveniles comprised the most numerous age class. Body size measurements were generally as large or larger, and weights were smaller, than those reported from more northerly areas. Traumatic injury, presumably by coyotes (Canis latrans), was the primary cause of death. Annual survival rates for adults averaged 0.53. Of 36 swift foxes captured during the study, 4 remained alive on the study area, 21 died, and 11 left the study area by the end of 32 months of field work. The annual 95% MCP home range size estimate was 1,494.5 ha. Diet was dominated by invertebrates and mammals. For den sites, swift foxes preferred the vicinity of roads, areas with greater road density, low slope, hilltops, and sandy loam and clay soils.

I compared the presence and abundance of nest-sites made by harvester ants (Pogonomyrmex), the primary prey for the endangered Texas horned lizard (Phrynosoma cornutum), among restored grassland plots planted in different grass species and indigenous prairie. The restored plots had been seeded as part of the Conservation Reserve Program (CRP) as exotic monocultures of either Old World bluestem (Bothriochloa ischaemum) or weeping lovegrass (Eragrostis curvula), or as mixtures of native grasses (both with and without buffalograss, Buchloë dactyloides). On average, the fewest ant mounds were found on Old World bluestem plots, whereas the indigenous grassland had the highest density of harvester ant mounds. However, there were no significant differences between native and exotic CRP plantings. Results obtained from a simultaneous visual survey for Texas horned lizards corroborate these findings. Thus, there is no evidence that CRP plots planted in exotic grasses are significantly poorer habitat for Texas horned lizards in terms of ant abundance than native grass plantings.

I studied 102 Texas horned lizards (Phrynosoma cornutum) encountered along roads or fortuitously at 50 sites in northern, southern, eastern, western, and central Texas from May through September 1999. Female-to-male and juvenile-to-adult ratios were 58:44 and 56:46, respectively. Distribution of horned lizards by sites were 60%, 50%, 50%, 30%, and 0% for the western, southern, northern, central, and eastern portions of the state, respectively. Average number of lizards captured was greater in the southern region compared to the central and eastern regions. The size of juvenile lizards was similar by region and sex; however, adult lizards were heavier and larger in the southern region compared to the western region.

The reticulate collared lizard (Crotaphytus reticulatus) is a large, crotaphytid lizard restricted to the lower Rio Grande Valley of southern Texas and northern Mexico. Our objectives were to determine the foraging mode of this species and test predictions made by other investigators about foraging behavior. We conducted focal observations on 10 male and 10 female lizards in Falcon State Park, Starr County, Texas, to quantify the number of movements per minute (MPM), the proportion of time spent moving (PTM), and the proportion of attacks made on prey items while moving (PAM). We removed foraging movements followed by social displays from the data and recalculated and reanalyzed foraging activity values. With “display” movements removed, there was a significant difference between MPM values and PTM values for males and for the sexes combined, but there was no significant difference in MPM values or PTM values for females. The foraging measures calculated for C. reticulatus, with and without “display” movements removed from the calculations, suggest that it is a “sit-and-wait” ambush forager like its congener C. collaris.

Blood parasites have been known to cause morbidity and mortality in waterfowl, particularly in Canada geese (Branta canadensis). However, little is known about blood parasites infecting Ross' geese (Chen rossii) and greater white-fronted geese (Anser albifrons). This study examined wintering Ross' and white-fronted geese for blood parasites. Blood smears from 16 Ross' geese (13 juveniles, 3 adults) and 46 white-fronted geese (21 juveniles, 25 adults) collected in Kleberg County, Texas, during winter 1999–2000 were examined for blood parasites. Leucocytozoon simondi was found in 2 juvenile white-fronted geese; density of infection was <1 parasite/2,000 erythrocytes. Additionally, 3 adult and 1 juvenile white-fronted geese had microfilaria. No blood parasites were observed in Ross' geese. We concluded that low prevalence and density of L. simondi gametocytes circulating in host blood precluded or greatly reduced transmission of this parasite on the wintering grounds in southern Texas.

Changes in geographic distributions of 5 bird species endemic to the Great Plains of North America were examined over the last few decades based on the United States Breeding Bird Survey. Examining the mean latitude of individuals of each species, 3 species showed significant or near-significant northward shifts, and 1 a significant shift southward. Over all 5 species examined, colonization events were concentrated in the northern part of the distributions of the species; in 3 species, extinctions were concentrated in the southern part of the distributions of the species. The conclusion is that significant distributional changes have taken place, but they have been subtle, and might be associated with global climate change.

Breeding bird populations were studied in forests recently damaged by tornados and in adjacent undamaged forests in the Ozark National Forest, Arkansas. During 1999 and 2000, surveys were undertaken at 6 points in forest moderately damaged by a tornado in 1996 and at 6 points in nearby undamaged forest. An additional 18 counts, 6 each in undamaged, moderately damaged, and heavily damaged forest, were undertaken in 2000 in an area affected by a 1999 tornado. Typical forest species, such as red-eyed vireo and ovenbird, were significantly less abundant in tornado-damaged forest than in undamaged forests, while edge species, such as indigo bunting and white-eyed vireo, were more abundant in damaged forest than in undamaged forest. Surprisingly, abundances of some species, such as black-and-white warbler, did not differ significantly between damaged and undamaged forests. Species composition differed between heavily damaged forest and moderately damaged or undamaged forest, with a number of species occurring only in the heavily damaged forest type. The congeneric summer tanager and scarlet tanager seemed to show habitat segregation in the study sites, with summer tanagers occurring in tornado-damaged forest and scarlet tanagers occurring in undamaged forest.

The worm-eating warbler (Helmitheros vermivorus) is a ground-dwelling neotropical migrant that typically breeds within large sloping forest patches. On 7 July 2000 at 07:36 CDT, a female H. vermivorus was captured at an agricultural ditch in Poinsett County, Arkansas. The warbler was aged as after hatching year (ASY) and had a fully developed brood patch with complete feather loss. This occurrence was unusual because it was outside the reported breeding range for this species and the warbler was in a small (3.52 ha) forest patch in a flat landscape. Observations of worm-eating warblers in alluvial areas of Arkansas occur during the migratory period, but are relatively rare. The timing of this summer observation, presence of a brood patch, and lack of fat deposition in the furcular cavity indicated a recent breeding attempt; however, this does not preclude a dispersing individual. In either case, closer monitoring in lowland habitats during breeding and post-breeding periods might provide additional knowledge on the conservation status and habitat needs of this species.

On 1 August 2000, a house wren (Troglodytes aedon) was observed exhibiting intra-specific helping behavior. An unbanded individual was observed carrying a spider into and separately carrying a fecal sac from a nest box occupied by 2 color-banded adults. The feeding individual appeared to be a hatch-year bird using the aging criteria of tail-length, breast coloration, and rictus characteristics. There were no additional observations of feeding by any unbanded individuals during the 1-h sample or in 3 hourly samples over the next 3 days. It is not known if there was any genetic relationship between the house wrens occupying the nest box and the unbanded juvenile helper.

We describe the first record of a loggerhead shrike (Lanius ludovicianus) impaling a bat. The eastern red bat specimen (Lasiurus borealis) was found at Anahuac National Wildlife Refuge (Chambers County, Texas) and was collected the following day. We describe the habitat at the site where the bat was impaled and discuss the rarity of this event. This is the first time we have observed an impaled bat during approximately 200 h of casual observation. The prey typically impaled are orthopterans (approximately one-half of all impaled prey), frogs and toads, crayfish, birds, and, to a lesser extent, reptiles and rodents. Approximately 7% of the prey were completely consumed, 35% partially consumed, and more than half were not consumed.

Historically, black bears (Ursus americanus) occurred throughout the southern Great Plains, including Kansas, Oklahoma, and Texas. However, due to habitat loss and human persecution, black bears were extirpated throughout this region by the early 1900s. We report on the recent range expansion of this species in the southwestern Great Plains, based on observations and records from state and federal agencies. Recent records of black bears, most of which exhibited the cinnamon color phase, suggest an established and reproducing population in Cimarron County, Oklahoma, that we estimate to consist of 20 individuals. This population apparently expanded from those in southeastern Colorado and northeastern New Mexico by the early 1980s and increased considerably during the 1990s. Today, black bears are reported annually from throughout Cimarron County. Dispersing young animals have increasingly wandered into Texas County, Oklahoma, and nearby areas of northwestern Texas and southwestern Kansas. Because black bears are protected in all 3 states and human densities are relatively low in this region, their range might continue to slowly expand.

Because effects of flooding on small mammal populations are poorly understood, we examined small mammal response, measured by relative abundance and community diversity, to a flood in a regenerating bottomland hardwood forest. We trapped small mammals immediately prior to and 5 months following retreat of floodwaters in Yazoo County, Mississippi during 1994 and 1995. Cotton rats (Sigmodon hispidus) and marsh rice rats (Oryzomys palustris) were the dominant species captured throughout the study, but abundance declined markedly after flooding and did not reach pre-flood abundance by the end of the study. Extended flood duration and depth were believed to have caused the disappearance of cotton mice (Peromyscus gossypinus) and a decrease in community diversity.

We report the first record of Yucatan deer mouse, Peromyscus yucatanicus, in Guatemala. The Yucatan deer mouse was restricted to Yucatán Peninsula, where it is considered a common species in deciduous to semi-evergreen tropical forests, and secondary growth forests. Two Yucatan deer mice were collected at Laguna Flor de Luna, Laguna del Tigre National Park, Guatemala, on April 1999. This new record constitutes a range extension of 116 km south of the nearest previously known locality at Escárcega, Campeche, Mexico.

We made an intensive survey of small mammals in the Sierra Norte de Oaxaca region, in Oaxaca, México. We recorded the presence of 2 interesting species: water mouse (Rheomys mexicanus) and river otter (Lontra longicaudis annectens). The locality for the river otter represented the highest altitude recorded in México for this species. The locality for the water mouse is the fourth known in all of its range and the second record in Sierra Norte.