The Hairston, Smith, and Slobodkin (1960) paper stimulated 40 years of interest on how communities are organized. Their “green world” perspective, in which terrestrial vegetation was considered to be resource rather than consumer limited, failed to acknowledge a significant role for mammals, especially ungulates. As such, its conclusions differed from comparable overviews developed for aquatic ecosystems in which the character of the producer level is determined by consumers, especially on marine intertidal shores. I argue here that omission of mammals as influential grazers distorted their perspective. Experimental studies published since 1960 identify mammals as strong interactors, that their grazing can dominate the rate of primary production, alter floristic composition, and change successional trajectories. When strong interactions have been documented in other ecosystems, they are often associated with a significant increase in indirect effects, alternative assemblage states, and even trophic cascades. Although these features will be more difficult to identify in terrestrial communities because of slow rates of assemblage response and experimental intractability of the interactors, they should be sought.

Food-handling ability was examined in Florida manatees (Trichechus manatus latirostris) feeding on different species of aquatic plants during controlled feeding trials. Subjects were presented with freshwater (Hydrilla verticillata, Myriophyllum spicatum, and Vallisneria americana) and marine species (Syringodium filiforme and Thalassia testudinum) of aquatic plants. Food handling was quantified by measuring length of cyclic movements (or feeding cycles) of perioral bristles that manatees use to introduce food into the mouth. Mean feeding-cycle lengths varied, depending on body size of the subject and species of plant being consumed. Rates of food introduction derived from mean feeding-cycle lengths were similar to chewing rates reported by other researchers. Manatees consumed plants with tubular stems and numerous branches faster than plants with flat blades. Food-handling time using perioral bristles differed, depending on the species of plants consumed and was reflected in the mean feeding-cycle length.

Hadromys humei, Hume's rat, has a restricted geographical distribution limited to northeastern India and southern China. During the middle Pleistocene, its distribution extended to the south, as it occurs in 10 fossil localities in Thailand. This southward extension of its distributional range is related to climatic changes that affected southeastern Asia during the Pleistocene.

Although Oreamnos americanus is absent from most Pacific Coast islands, including Vancouver Island, 12,000-year-old skeletal remains were recovered in 2 caves on northern Vancouver Island. The specimens may represent early postglacial immigrants or a relict population derived from a coastal glacial refugium. Limb bones of the fossils are within the size range of modern specimens, suggesting a postglacial origin. O. americanus probably became extinct on Vancouver Island during the early Holocene warming, but inadequacies in the prehistoric faunal record prohibit a determination of a terminal date. The modern distribution of O. americanus on Pacific Coast islands reflects both prehistoric extinctions and low colonization rates across water barriers.

Because recent studies have not demonstrated a strong relationship between rodent density and Sin Nombre virus (SNV) seroprevalence, there is speculation that seroprevalence may be related to other factors, including habitat quality and food availability. We evaluated densities of deer mice (Peromyscus maniculatus), plant cover and biomass, and terrestrial arthropod biomass at 2 sites in the southwestern United States to identify factors that may affect the seroprevalence rate of SNV within a rodent population. Seroprevalence differed significantly between years. Although interaction of deer-mouse density, plant cover and biomass, and arthropod biomass was not a strong predictor of seroprevalence (R² = 0.64, P = 0.04), we observed a significant contribution to a repeated-measures model by deer-mouse density (P = 0.02). Our data suggest that as rodent density increases, so does the seroprevalence rate within that population. Although not significantly correlated, we observed the lowest levels of arthropod biomass when seroprevalence was highest. Based on our results, evaluating changes in habitat quality and incorporating measurement of local ecological variables with studies of fluctuations in rodent density may aid in predicting human outbreaks of hantavirus disease.

Many mammalian herbivores remain active throughout winter. To satisfy daily energy needs, they ingest large quantities of cold food that subsequently must be warmed to body temperature. Some energy is inevitably lost during this process. Because the specific heat capacity of cellulose is only one-third that of liquid water, the quantity of energy that is lost depends primarily on the temperature and amount of water (free or contained in plants) that is ingested. Using the doubly labeled water method with meadow voles (Microtus pennsylvanicus) studied under field conditions, I show that the cost of heating ingested water represents 4.7–12.9% of the daily energy budget of nonreproductive individuals. Whether ingested water is liquid or frozen is critical because of the high cost of melting ice. I show that the fraction of individual energy budgets diverted to heating ingested food should be similar for small and large herbivores, and I explore some consequences of this result for our understanding of the winter ecology of mammalian herbivores.

Litter size strongly affects reproductive output and is therefore of central interest for the understanding of life-history evolution. I investigated the effect of variation in litter size (natural and manipulated) on the energetics of reproduction in the highly precocial guinea pig (Cavia porcellus). Unlike the situation in altricial species, increasing litter size in this precocial rodent influenced maternal performance (increase in litter mass and decrease in mass of individual offspring) as strongly during gestation (68 days) as during lactation (20 days). To cover the energy cost of offspring production, mothers increased their daily food intake. Their daily increase in metabolizable energy intake varied significantly with litter size during gestation but not during lactation. Instead, offspring in large litters consumed more solid food during lactation than did offspring in small litters. The efficiency of energy conversion into offspring production was not affected by litter size, neither during gestation nor during lactation. Furthermore, manipulation of litter size did not influence a mother's total energy cost of lactation. As a consequence, offspring in enlarged litters grew significantly slower than offspring in reduced litters despite similar body masses at birth and significantly greater consumption of solid food by nursing offspring in large litters. Results suggest that the effect of litter size on the energetics of reproduction in the precocial guinea pig differs from that of altricial rodents. Consequently, ecological implications of variation in litter size in the precocial guinea pig might also be very different to those in altricial species.

California ground squirrels (Spermophilus beecheyi) in the natural environment maintain a higher mean body temperature (T_b) compared with animals maintained in the laboratory. Our goal was to determine whether that increase was the result of an elevation in T_b set point analogous to a fever induced by acute stress. Body temperatures were measured by telemetry in a paired experiment on 10 animals maintained in the laboratory, with each animal receiving an injection of sterile pyrogen-free saline and 50 μg/kg endotoxin of E. coli. Body temperatures also were measured by telemetry on 8 pairs of free-living animals, with 1 member receiving an injection of sterile saline and the other animal receiving an injection of 50 μg/kg endotoxin of E. coli. Animals maintained in the laboratory exhibited a fever following injection of endotoxin, with significant elevation in T_b from 6 to 9 h after injection. Free-living animals that were injected with endotoxin did not exhibit an increase in T_b compared with saline-injected controls over the same time. Absence of a febrile response in free-living California ground squirrels supported the hypothesis that this large squirrel had a pre-existing elevation in thermoregulatory set point.

Estimates of postnatal growth rates and age based on data collected from free-ranging little brown myotis (Myotis lucifugus) captured sequentially at the same colony in the same year were used to compare longitudinal (mark–recapture) and cross-sectional (grab) sampling methods. Analyses of these data indicate that cross-sectional sampling significantly underestimates growth rates for length of forearm, body mass, and length of epiphyseal gap. Equations based on cross-sectional samples significantly overestimated ages of young, as compared with equations based on the longitudinal method. These results support the hypothesis that cross-sectional sampling is unreliable for deriving postnatal growth curves for free-ranging bats and emphasize the importance of using longitudinal data to derive growth rates and estimates of age during the postnatal period.

Presumably, reproductive female bats are more constrained by thermoregulatory and energy needs than are males and nonreproductive females. Constraints imposed on reproductive females may limit their geographic distribution relative to other bats. Such constraints likely increase with latitude and elevation. Males of 11 bat species that inhabit the Black Hills were captured more frequently than females, and reproductive females typically were encountered at low-elevational sites. To investigate the relationship between female distribution and elevation, we fitted a logistic regression model to evaluate the probability of reproductive-female capture as a function of elevation. Mist-net data from 1,197 captures of 7 species revealed that 75% of all captures were males. We found a significant inverse relationship between elevation and relative abundance of reproductive females. Relative abundance of reproductive females decreased as elevation increased. Reproductive females may be constrained from roosting and foraging in high-elevational habitats that impose thermoregulatory costs and decrease foraging efficiency. Failure to account for sex differences in distributional patterns along elevational gradients may significantly bias estimates of population size.

The maternity colony of Myotis formosus in Yun-lin County, southwestern Taiwan, began to form in late March, and the numbers peaked (about 200) in May. Most bats left the roost site between mid-August and early October for unknown hibernation sites. Females gave birth during May and June. Before parturition, females (n = 19) spent 563.9 min ± 47.28 SD outside the roost per night. Time away from the roost decreased significantly after they gave birth. Lactating females increased the total time that they spent outside the roost per night by 37.8% from early to late lactation. Females spent more time grooming young than self-grooming during the first 2 weeks of lactation. From the 3rd week on, more time was spent on self-grooming. Forearm length and body mass of newborns were 21.46 ± 1.90 mm (n = 13) and 3.71 ± 0.41 g (n = 12), about 42.8% and 24.1% of adult size, respectively. Young rarely moved during the 1st week after birth. They started to crawl at the 2nd week and to attempt short straight-line flight at the 3rd week. Females always rested next to their young while roosting in the first 2 weeks of lactation but separated from them from the 3rd week on. Behavioral changes in females coincided with the growth and development of their young.

Dominance ranks in male and female prairie voles (Microtus ochrogaster) were determined from 6 measurements that mimicked environmental situations that might be encountered by prairie voles in communal groups, including agonistic interactions resulting from competition for food and water and encounters in burrows. Male and female groups of 6 individuals each were tested against one another in pairwise encounters (i.e., dyads) for 5 of the measurements and together as a group in a 6th measurement. Two types of response variables, aggressive behaviors and possession time of a limiting resource, were collected during trials, and those data were used to determine cardinal ranks and principal component ranks for all animals within each group. Cardinal ranks and principal component ranks seldom yielded similar rankings for each animal across measurements. However, dominance measurements that were conducted in similar environmental contexts, regardless of the response variable recorded, ranked animals similarly. Our results suggest that individual dominance measurements assessed situation- or resource-specific responses. Our study demonstrates problems inherent in determining dominance rankings of individuals within groups, including choosing measurements, response variables, and statistical techniques. Researchers should avoid using a single measurement to represent social dominance until they have first demonstrated that a dominance relationship between 2 individuals has been learned (i.e., subsequent interactions show a reduced response rather than an escalation), that this relationship is relatively constant through time, and that the relationship is not context dependent. Such assessments of dominance status between all dyads then can be used to generate dominance rankings within social groups.

Mammalian social organization can vary over ecological time. We experimentally manipulated food resources in enclosed populations of prairie voles (Microtus ochrogaster) to test the hypothesis that food quality influences the potential for group formation. During each field season, populations were started by releasing 5 pairs of prairie voles into each of 8 0.1-ha enclosures. Populations were monitored for 18–19 weeks during each field season. One-half of the enclosures received supplemental food, and the other one-half were unsupplemented controls. Density of voles increased throughout each field season. There were significant increases through time in philopatry and number of groups in both the food-supplemented and unsupplemented treatments, but there were no differences between treatments. Groups formed early in the season, apparently before the need for thermoregulatory benefits. Results are consistent with the hypothesis that the social organization of prairie voles is not flexible in response to changes in food quality but that formation of groups might be a density-dependent response.

The relationships among testes mass, sperm size (as indicated by sperm tail length), and body mass of Old World rats and mice were investigated. A near isometric relationship between testes mass and body mass was found but not between body mass and sperm size. In a few lineages, testes mass deviated considerably from the regression line with relatively very small testes occurring in Bunomys fratrorum, Bandicota indica, and B. savilei of Asia, Aethomys ineptus of Africa, some Pseudomys, and all Notomys of Australia. By contrast, relatively large testes mass occurred in Apodemus, Berylmys, and Maxomys bartelsii of Asia, Pogonomys of New Guinea, and some Australasian Rattus, Melomys, and Mastacomys. Considerable variability in sperm size was also evident; some species that had relatively small testes also had relatively small spermatozoa. The reason(s) for interspecific variation in relative testes mass and sperm size is unknown, but the data provide an opportunity to test the hypothesis that differences in relative testes mass, and perhaps sperm size, relate to interspecific differences in the amount of intermale sperm competition and in breeding systems.

We examined temporal breeding patterns of bighorn sheep (Ovis canadensis) in the Peninsular Ranges of California to determine the degree of seasonality and identify potential selection for seasonal breeding in this low-latitude desert environment. During a 4-year period, births occurred during 7 months of the year, but 87% of young were born in February–April and 55% were born in March. Peak months of mating and parturition remained relatively constant across years and among different regions of the Peninsular Ranges, and young born in February through April had greater survival than those born later. Female age influenced lamb survival but not timing of parturition. Successful recruitment of young by an adult female had a weak but significant negative effect on the survival of that female's offspring the following year. Previous reproductive history of a female did not influence offspring production or timing of parturition. Months of peak parturition followed annual winter rains and, therefore, were likely to coincide with periods of high plant productivity. Furthermore, most young were born before the hot, arid summer months. We concluded that bighorn sheep in the Peninsular Ranges are seasonal breeders and that climate patterns likely act as ultimate factors in shaping the breeding season.

We examined relationships between mammalian assemblages and landscape context and habitat fragmentation in southeastern Australia. Data were gathered from spotlighting and hair sample surveys at 166 sites in 3 different spatial (landscape) contexts: remnant patches of native eucalypt forest surrounded by an extensive plantation of exotic radiata pine (Pinus radiata—86 sites), the radiata pine plantation (40 sites), and large areas of continuous native eucalypt forest that occurred at the margins of the plantation (40 sites). Continuous eucalypt forest supported more species than eucalypt patches, although some species were more common in the patch areas. All assemblages in the radiata pine sites were substantially impoverished. There was a significant patch size effect for the total mammalian assemblage and for terrestrial native mammals but not for arboreal marsupials. Bigger remnants supported an assemblage different from (and more species rich) that found in smaller remnants, particularly those <3 ha where many mammal species occurred less frequently. The landscape context and patch area effects recorded in this study have important implications for plantation design in southern Australia. Eucalypt remnants should be exempt from clearing during plantation development; larger remnants are the most important areas.

The Siberian flying squirrel (Pteromys volans) is a herbivorous, nocturnal, and arboreal rodent living in boreal coniferous forests. Home-range sizes, movements, and nest-use behavior of Pteromys were studied by radiotelemetry in southern Finland in 1996–1998. Thirty-seven animals were tracked. Average home-range size measured by 100% minimum convex polygons was 59.9 ha for males and 8.3 ha for females. Both sexes concentrated their activities in core areas that represented 9% and 11% of the home-range areas in males and females, respectively. Home ranges of males and females were several times larger than predicted according to body mass. Similarly, home ranges of Pteromys were much larger than in other gliding herbivores. Males especially showed great mobility; the average distance moved from the nest at night was 292 m, and the longest distances recorded were >2 km. A plausible explanation for the large home ranges and great mobility of Pteromys is its gliding ability; both sexes can reach distant parts of the home range for foraging, and males also can reach distant parts for receptive females. Pteromys had several nests, both cavities and dreys (nests in branches of trees), which they changed frequently.

We investigated how white-nosed coatis (Nasua narica) cope with the extreme seasonality of a Mexican tropical dry forest by studying their activity, home ranges, and habitat selection in relation to climatic seasonality. From November 1994 to March 1997, we radiotracked 7 solitary adult males and 11 bands of females and juveniles. Males extended their activity more into night hours, were more active in both the dry and the wet seasons, and traveled a greater daily distance during the wet season than groups of coatis. Average total home range was 383.0 ha ± 32.86 SE and did not differ between sexes. Home ranges differed seasonally only in groups that used areas during the dry season that were twice as large as those used during the wet season. Three major habitats that differed in phenology were used by coatis. Both males and groups preferred arroyo forest to dry forest and semideciduous forests. These results illustrate the importance of behavioral traits that permit coatis to have access to habitats where sparse resources (e.g., food and water) are more available as a mechanism to cope with climatic seasonality. Our study provides a basis for design of management and conservation strategies for the Chamela–Cuixmala Biosphere Reserve. The population of coatis in this reserve may be considered as a model to predict the type of behavioral responses that other populations of coatis may use to cope with climatic seasonality in other tropical dry forests throughout México and Central America.

During 1986–1991, carcasses of 619 arctic foxes (Alopex lagopus) collected from local trappers and at biological field camps on the Yukon-Kuskokwim Delta in western Alaska from November through May were analyzed to determine gastrointestinal contents, age, sex, and body condition. Prey in declining order of importance were small mammals (95% tundra voles, Microtus oeconomus), birds, marine mammals, and fishes. Foxes with small mammal remains in their stomachs were captured farther from the Bering Sea coast (X̄ = 5.2 km) than those without small-mammal remains (2.8 km); foxes consuming remains of marine mammals were closer to the coast (1.9 km) than others (4.9 km). Although eggshells had a poor likelihood of occurrence in stomachs, they were found in all months and years. In 1986 and 1987, foxes consumed fewer small mammals than in other years. Mean ages of foxes captured in 1986 (3.7 years) and 1987 (3.2) were greater than in all other years (1.5). Capture of adults was more common as winter progressed. Indexes of subcutaneous fat decreased annually in April–May and were highest in 1991, when occurrence of carrion of marine mammals was highest.

Craniometric characteristics of the mouse opossum, Marmosa robinsoni, in Venezuela were analyzed with respect to sex, climate, and macrohabitat. We assessed patterns of phenotypic variation by comparing 6 samples established on the basis of geographic and taxonomic criteria. For most parameters, males were larger than females, and there was geographic variation in skull size that was not related to geoclimatic factors but to the type of vegetation. Specimens inhabiting agricultural lands and disturbed forests were larger than those from cloud and gallery forests; the latter generally was associated with savannas. We suggest that large skull size is related to the higher productivity of secondary-growth vegetation in anthropogenic areas compared with mature forests. We conclude that the specimens studied should be considered as a single subspecies, which corresponds to M. r. robinsoni.

Cranial and external morphology of 54 specimens of rodents of the genus Akodon from southern Brazil was analyzed. The sample included 27 individuals with a karyotype of 2n = 44 and 13 with 2n = 46 and 14 nonkaryotyped animals. Principal components analysis separated individuals with 2n = 44 from those with 2n = 46. Individuals with 2n = 44 were distinguished from those with 2n = 46 by narrow and elongated molars, ectolophid present, narrow interorbital breadth, reduced tegmen tympani, and distal baculum with the central and lateral digit approximately equal in size and proportions, not enlarged in the extremity. Discriminant functions analyses including 6 A. sanctipaulensis individuals and holotypes of A. serrensis and A. s. leucogula resulted in 3 distinct groups. The 2n = 46 individuals were related to A. serrensis, but those with 2n = 44 were not assignable to any described species for Akodon. We describe and propose the recognition of a new species for the genus.

Nucleotide sequence analysis of the cytochrome-b gene of mtDNA and univariate and multivariate statistical analyses of cranial morphology were used to determine the evolutionary status of relictual, isolated populations of pocket gophers (genus Geomys) from the Ozark Mountains of Arkansas. Results of the mtDNA sequence analysis indicated that populations from the Ozark Mountains are most similar to G. bursarius missouriensis in Missouri and not to geographically proximate populations of G. breviceps. Morphological analyses were concordant with those results. Our findings confirm earlier work comparing similarities in allozymes and species-specific ectoparasites. Nucleotide sequence and morphological divergence between the Ozark populations and G. bursarius missouriensis are typical of differentiation between other subspecies of G. bursarius. We conclude that populations of pocket gophers isolated in the Ozark Mountains represent a discrete genetic entity and should be recognized as a new subspecies.

Since its description, phylogenetic affinities of the New Zealand short-tailed bat (Mystacina tuberculata) have been difficult to determine. Although previous morphological studies have aligned Mystacina with several microchiropteran families, the general consensus has been that it is nearest to Molossidae. In contrast, immunological and DNA-hybridization data support inclusion of Mystacina within the New World Noctilionoidea, although those data failed to provide resolution for relationships within the superfamily. Our purpose was to test the hypothesis that Mystacina is most closely related to Noctilionoidea rather than Molossidae and, if so, to elucidate affinities among Mystacina and other noctilionoid families. We performed a phylogenetic analysis of DNA sequences from 3 adjacent genes (12S rRNA, tRNA^Val, and 16S rRNA) in the mitochondrial genome from Mystacina and representatives of 8 microchiropteran families. Results from parsimony analysis agree with previous molecular studies that Mystacina is a member of Noctilionoidea. Additionally, this study provides resolution for relationships within the superfamily.

Taxonomic affinity of Peromyscus boylii sacarensis was examined using DNA sequences from the mitochondrial cytochrome-b gene. Maximum parsimony and genetic distance analyses were used to examine the phylogenetic relationship of this taxon to other closely related members of the P. boylii species group (P. beatae, P. boylii rowleyi, P. levipes ambiguus, P. l. levipes, and P. simulus). All analyses, despite differential weighting of transitions and nucleotide positions (1st, 2nd, and 3rd) or method of calculating genetic distances, resulted in similar topologies with P. b. sacarensis forming the sister group to P. beatae. Our data suggest that P. b. sacarensis should be aligned taxonomically with P. beatae instead of its current placement in P. boylii and that 3 genetic subunits exist within P. beatae.

Previous systematic studies, using morphometric, allozymic, and cytogenetic data, have produced conflicting results as to the relationship of Dipodomys elator to other species of kangaroo rats. In this study, phylogenetic relationships of D. elator and other closely related taxa (D. merriami, D. phillipsii, D. ordii, D. spectabilis, and D. nelsoni) were examined using nucleotide-sequence data (1,143 bp) from the mitochondrial cytochrome-b gene. With 1 exception, all parsimony and neighbor-joining analyses (regardless of differential weighting scheme or method of calculating genetic distances) indicated that D. elator is most closely related to D. phillipsii. As a result, it is recommended that D. elator and D. phillipsii be included within the phillipsii species group.

A series of reintroduction programs have reestablished fisher (Martes pennanti) populations to a large part of their former range. Horizontal starch-gel electrophoresis of 20 presumptive gene loci was used from 1996 to 1998 to investigate gene dynamics in 4 remnant fisher populations (sources) and 4 reintroduced fisher populations from the northeastern and Great Lakes regions of the United States. Mean multilocus heterozygosities, mean number of alleles per locus, and percentage polymorphic loci were 0.027–0.090, 1.2–1.6, and 10.0–30.0, respectively, in the 8 populations surveyed. Significant allelic frequency differentiation was detected among the 4 source populations and among the 4 reintroduced populations, but a significant proportion of the genetic variance was partitioned only among the 4 reintroduced populations. Pairwise comparisons between each source and its associated reintroduced population indicated that only older reintroductions have attained significant differentiation of allelic frequency from their sources. Significant heterozygotic deficiencies were detected for statewide populations and regional populations within states, suggesting that breeding biology of the fisher, presumably among females, is creating levels of fine-scale genetic structure within populations.