This commentary originally was presented to recognize receipt of the Joseph Grinnell Award for Excellence in Education at the 85th Annual Meeting of the American Society of Mammalogists in Arcata, California, in June 2004. Natural history is the multidisciplinary description of nature, and naturalists are those who study nature. In its own right, natural history is a relevant discipline, despite varying degrees of focus by professional biologists and academic institutions over the past 100 years, and it is a critical contemporary discipline relative to global crises in the conservation of biodiversity. Many scholars have written that natural history has fallen out of favor at American universities. I review this perspective within the context of 20th century developments in society and trends in American higher education. My conclusion is that a narrowing of the context of natural history in the 20th century has diminished its significance. However, there is compelling evidence that, if we broaden our approach and horizons, natural history scholarship can play a pivotal role in American science and education in the 21st century. Institutions of higher education that emphasize natural history in their curriculum will enhance not only their academic profile but also students' appreciation of the importance of natural history throughout their lives.

Here we describe a new species of Anoura from the Andes of Ecuador, distinguished from all other species of Anoura by its elongated tubelike lower lip and its much longer tongue. Based on size and other characteristics, Anoura n. sp. appears most similar to Anoura caudifer, but averages approximately 10% larger and has a wider uropatagium in the shape of an inverted V. Anoura n. sp. has been collected infrequently in cloud forests on the eastern and western slopes of the Andes throughout Ecuador, where it co-occurs with the more abundant A. geoffroyi and A. caudifer. Comparison with the diets of sympatric species of Anoura suggests that Anoura n. sp. visits flowers that have longer corollas, as might be expected given its much longer tongue.

Integrated analyses of morphological, chromosomal, and molecular data resulted in the description of a new species of the genus Akodon, endemic to the Northern Campos in southern Misiones, Argentina. The new species presents a unique combination of characters, including a narrow zygomatic plate almost without a free upper border, short rostrum, short tail (40% of the head–body length), small and flat auditory bullae, diploid complement of 2n = 36, and several molecular synapomorphies. Phylogenetic analysis, based on cytochrome-b gene sequences, indicates that the Brazilian species Akodon lindberghi is sister to the new species, although this relationship is weakly supported. We comment on the conservation significance of our study, considering that these perisylvan grasslands are suffering substantial human disturbance by agriculture and burning practices.

The taxonomy and dental evolution of cotton rats, genus Sigmodon, from Pliocene and Pleistocene deposits in the Meade Basin of southwestern Kansas are examined. Prosigmodon is here subsumed under Sigmodon. Morphological and multivariate statistical analyses identify 3 Sigmodon lineages and 5 immigration events in the Meade Basin. The 1st immigration event occurred during the early Blancan and the latest during the Holocene. The analyses of dental characters indicate directional trends in almost all features, representing both size and shape. A comparison of information from the cotton rat fossil record with a recent molecular phylogeny suggests the molecular phylogeny has set the origin of modern cotton rat groups too early. North American species groups probably do not predate the late Pliocene.

Genetic variation of caribou (Rangifer tarandus) at 18 microsatellite DNA loci and the cytochrome-b gene of mitochondrial DNA (mtDNA) was quantified in 11 herds of 3 North American subspecies: Alaskan barren ground caribou (R. t. granti), Canadian barren ground caribou (R. t. groenlandicus), and woodland caribou (R. t. caribou). Phylogenetic analysis of 1,194 nucleotides of cytochrome-b sequence resulted in a clade of 52 genotypes in R. t. granti, R. t. groenlandicus, and in 1 herd of R. t. caribou, and a clade of 7 genotypes in R. t. caribou. mtDNA sequence divergence is approximately 1% between these clades and 0.3–0.6% within these clades. The subspecies do not have monophyletic mtDNA, but do have different frequencies of mtDNA genotypes. Microsatellite allele frequencies also are differentiated between the woodland (R. t. caribou) and barren ground (R. t. granti and R. t. groenlandicus) subspecies. An exception is the George River herd in Labrador, which is classified as R. t. caribou but has mtDNA and microsatellite allele frequencies intermediate between the other herds of R. t. caribou and R. t. groenlandicus. Within subspecies, there is relatively low differentiation of microsatellite allele frequencies and mtDNA genotypes among herds of R. t. granti and R. t. groenlandicus, and relatively high differentiation of microsatellite alleles and mtDNA genotypes among herds of R. t. caribou in 4 geographically separate areas in Canada. The extent of differentiation of mtDNA genotype frequencies and microsatellite allele frequencies within and among each subspecies reflects past and present gene flow among herds. Issues related to subspecies, populations, ecotypes, and herds are discussed.

Understanding factors that influence birth mass of mammals provides insights to nutritional trade-offs made by females to optimize their reproduction, growth, and survival. I evaluated variation in birth mass of caribou (Rangifer tarandus) in central Alaska relative to maternal characteristics (age, body mass, cohort, and nutritional condition as influenced by winter severity) during 11 years with substantial variation in winter snowfall. Snowfall during gestation was the predominant factor explaining variation in birth masses, influencing birth mass inversely and through interactions with maternal age and lactation status. Maternal age effects were noted for females ≤ 5 years old, declining in magnitude with each successive age class. Birth mass as a proportion of autumn maternal mass was inversely related to winter snowfall, even though there was no decrease in masses of adult females in late winter associated with severe winters. I found no evidence of a hypothesized intergenerational effect of lower birth masses for offspring of females born after severe winters. Caribou produce relatively small offspring but provide exceptional lactation support for those that survive. Conservative maternal investment before parturition may represent an optimal reproductive strategy given that caribou experience stochastic variation in winter severity during gestation, uncertainty of environmental conditions surrounding the birth season, and intense predation on neonates.

Mammalian herbivores from northern environments often store fat in summer to decrease the energetic stress of winter. We measured mass gain during the spring-to-fall season in a Quebec population of North American porcupines (Erethizon dorsatum) and determined how it is affected by the age, sex, and reproductive status of individuals. Porcupines gained mass without interruption from midspring to late summer, but mass gains were more important in mid- to late spring and late summer than at other times. This likely reflected local variations in the nutrient content of forage. Adult males lost mass during the rutting period of early fall. All age–sex classes except lactating females gained about 3 kg, or 40% of their spring body mass. This is not markedly different from mass gains observed in other nonhibernating mammalian herbivores. We generate the hypothesis that in this population some ecological, behavioral, or physiological mechanism limits the rate of mass increase to about 21 g per day.

Body mass is commonly used as an indicator of habitat quality and body condition in small-mammal populations. We examined the effects of consecutive days of livetrapping on body mass of 2 federally listed subspecies of beach mice on the eastern coast of Florida: the Anastasia Island beach mouse (Peromyscus polionotus phasma) and the southeastern beach mouse (P. p. niveiventris). Mean mass loss (approximately 0.5 g/recaptured day) was additive and occurred for Anastasia Island beach mice over 2 consecutive days and for southeastern beach mice over 3 consecutive days. Researchers should exercise caution when their trapping protocols call for consecutive nights of trapping.

This study addressed behavioral responses by black-tailed prairie dogs (Cynomys ludovicianus) to human intrusion in urban and rural environments in Boulder, Colorado. We expected that if prairie dogs habituate to repeated disturbances, they should allow a recurring human intruder to approach closer over time before sounding an alarm bark or initiating concealment. We also predicted that urban colonies could be approached more closely than rural colonies before displaying an avoidance response. Four colonies (2 rural and 2 urban) were approached >100 times over a 7-month period. Rather than exhibiting habituation, prairie dogs demonstrated increased responsiveness in concealment behavior, retreating to their burrows earlier, with recurring disturbances. Barking distances did not change consistently with repeated intrusion, but, over time, prairie dogs barked less frequently when performing their avoidance response, a result with implications for prairie dog management. Rural colonies had higher initial concealment distances, and these distances increased more rapidly with repeated intrusion than did concealment distances in urban colonies. Thus, rural prairie dogs may be more sensitive to human intrusion than urban prairie dogs.

Mating systems within the genus Peromyscus have traditionally been inferred from field observations of spatial relationships of males and females and from laboratory behavioral studies. The majority of species within Peromyscus are assumed to be promiscuous, but rarely have these conclusions been verified with molecular data. Genotypes constructed from 4 microsatellite loci were used to determine paternity in 10 litters of the canyon mouse (Peromyscus crinitus). We applied 3 criteria for paternity assignment: direct observation of paternal alleles, correlation of sampled male and offspring genotypes (mothers known) including spatial relationship data, and application of a likelihood estimator. Multiple paternity occurred in a minimum of 20% of the litters, yet we found no evidence that would indicate that any putative father sired 2 or more litters. Dispersal of males between the time of copulation and parturition was minimal (7.5–70 m) in most cases, but 1 male was captured in a rock outcropping separate from that of his sired litter (174 m). Our use of 3 different criteria allowed a greater number of male paternity assignments to be inferred than would have been possible when using software program CERVUS alone, and increased confidence in these results.

Communication among animals should use signals that are most efficient in their particular habitat. Here, we report data from 3 populations of Indo-Pacific bottlenose dolphins (Tursiops aduncus) in Japan that produce whistles transmitted efficiently through environmental ambient noise. We compared the characteristics of the ambient noise in the dolphins' habitats and the whistles produced. In habitats with less ambient noise, dolphins produced whistles at varying frequencies with greater modulations; when ambient noise was greater, dolphins produced whistles of lower frequencies with fewer frequency modulations. Examination of our results suggests that communication signals are adaptive and are selected to avoid the masking of signals and the attenuation of higher-frequency signals. Thus, ambient noise may drive the variation in whistles of Indo-Pacific bottlenose dolphin populations.

We examined social group formation, movements and denning relative to other group members, home-range use, and the response to social group disruption among 188 radiocollared swift foxes (Vulpes velox) in Colorado. We found that during the initial stages of pair formation mated foxes shared dens more frequently than during the remainder of their pair-bond. The average distance between mated pairs was influenced by season and time of day, with the greatest proximity in the breeding season (727.2 m ± 42.3 SE), and during diurnal hours (463.7 ± 34.7 m). Female foxes spent more time in the core area of the group home range (60.71% of locations versus 54.66% for males) and males spent more time on the range boundary (19.34% of locations versus 15.61% for females). Home-range use was influenced by season, with females spending more time in the core area in the breeding and pup-rearing seasons than in the dispersal season. Males also spent the highest proportion of their time in the core area during the breeding season, but used the boundary area more frequently than the core area in the pup-rearing season. A sex difference also was found in the response of a swift fox to the death or disappearance of its mate. All females maintained their territory in the event of mate death or disappearance; however, 50% of males emigrated from their range when the female mate died or disappeared. These differences in space use between social group members provide important insights into the territorial behavior and mating system of the swift fox and indicate that the 2 sexes likely play different roles in care of young and home-range defense.

The selection of a suitable nest (den) site should enhance individual survival and reproduction. We examined the effects of forage quality, vegetative cover, presence of preexisting underground nests, vole population density, and presence of conspecific nests on nest-site selection of prairie voles (Microtus ochrogaster). Vegetative characteristics did not predict the presence of nests; the only significant predictor was population density. Additionally, the presence of other currently occupied nests did not seem to affect placement of nests. In contrast to our expectation, voles did not nest in areas with preexisting underground nests. The presence of preexisting underground nests may have reduced the effects of vegetation on nest-site selection.

We studied the influence of preferred food, forbs, and vegetative cover on survival and reproduction at different population densities of Microtus ochrogaster and M. pennsylvanicus in alfalfa, bluegrass, and tallgrass prairie habitats in east-central Illinois for 25 years. Population densities of M. ochrogaster were greatest in alfalfa, least in tallgrass, and intermediate in bluegrass, whereas those of M. pennsylvanicus were greatest in tallgrass, least in alfalfa, and intermediate in bluegrass. For both species, preferred food availability was greatest in alfalfa, intermediate in bluegrass, and least in tallgrass. Vegetative cover was relatively sparse in alfalfa, especially in winter, and dense throughout the year in bluegrass and tallgrass. Variation in survival emerged as the most important factor explaining population differences between the 2 species. Reproduction had little differential impact on abundance of either species in any of the 3 habitats. Survival of M. ochrogaster was higher in alfalfa than in bluegrass or tallgrass; survival of M. pennsylvanicus was higher in tallgrass than in alfalfa or bluegrass. Differential survival among habitats and between species was influenced primarily by amount of vegetative cover. We suggest that M. ochrogaster is less susceptible than M. pennsylvanicus to predation by raptors and large carnivores (predators that hunt from above vegetative cover), whereas M. pennsylvanicus is less susceptible than M. ochrogaster to snakes and small carnivores (predators that hunt under vegetative cover).

We examined how corridors and patch shape affect foraging by the oldfield mouse (Peromyscus polionotus) by deploying foraging trays and live traps in experimental landscapes with 3 different patch types: patches connected with a corridor, unconnected patches with projecting corridorlike portions (“winged” patches), and unconnected rectangular patches. Corridors did not lead to different levels of activity of P. polionotus among the 3 patch types. Rather, corridors influenced activity by changing patch shape: foraging in seed trays and total number of captures of P. polionotus tended to be greater at the patch center than at the patch edge, but only in connected and winged patches where corridors or wings increased the amount of patch edge relative to the amount of core habitat in the patch. P. polionotus avoided open microhabitats near the patch edge in winged and connected patches, but not open microhabitats near the patch interior, suggesting that predation risk caused shifts in foraging near edges in connected and winged patches. Foraging in corridors and wings was generally low, suggesting that both are high-risk habitats where predation risk is not ameliorated by proximity to vegetative cover. By changing patch shape, corridors caused changes in within-patch activity of P. polionotus, changing foraging patterns and potentially altering the dynamics of P. polionotus and the seeds they consume.

Exhaustive analyses of plant–frugivore systems have revealed that few, if any, of these plant–animal interactions are tightly coevolved. Such lack of coevolutionary evidence could be related to frugivores selecting plants based on environmental cues, rather than on plant phenotypic traits. To evaluate this hypothesis, I examined whether the pattern and extent of fruit predation by long-tailed field mice (Apodemus sylvaticus) on the perennial herb Helleborus foetidus was directly related to mouse abundance, to environmental cues, to plant phenotypic traits, or to a combination of these. Thus, I estimated mouse relative abundance (through livetrapping) and percentages of fruit predation by mice, and quantified plant environmental and phenotypic traits in 9 populations of H. foetidus in southeastern Spain during 2 years (overall, 254 plants). I found substantial variation among populations in mouse trapping success (ranging from 0.0 to 21.8 individuals/100 trap nights), size of fruit crops (8.6–28.9 fruits per plant), and percentages of fruit predation by mice (0.0–93.3%). However, no simple relationship was found between mouse abundance (as estimated by trapping success) and strength of fruit predation. None of the 4 measured plant phenotypic traits (e.g., number of fruits or plant size) had a significant effect on mouse foraging. Conversely, 2 of the 6 environmental traits considered (substrate and distance to nearest tree) influenced fruit predation by mice. Plants located on rocky substrates and nearby trees experienced higher percentages of predation and this result was rather consistent across the 2 years and the 2 levels considered (inter- and intrapopulation). Even though mice could have selected other plant phenotypic traits not accounted for (e.g., chemical traits), such hypothetical phenotypic selection appears to be inconsequential as a source of individual variation in H. foetidus maternal fitness because its effects could have been “diluted” by the overwhelming influence of environmental factors. Thus, the results support the initial hypothesis that the net outcomes of the interaction between H. foetidus and mice in southeastern Spain may not have strong coevolutionary consequences.

We studied home-range size and patterns of range use of long-legged bats, Macrophyllum macrophyllum (Phyllostomidae), at Barro Colorado Nature Monument, Panama, by using radiotelemetry. Movements of 4 males and 5 females fitted with radiotransmitters were monitored for 4–7 entire nights each between April and July 2002. M. macrophyllum had an extensive range compared to other similar-sized phyllostomid bats. Median home-range size for the 9 individuals tracked was 23.9 ha (range 7.3–150.7 ha). With a median of 17.3 ha (range = 7.3–24.9 ha), home ranges of males were somewhat, but not significantly, smaller than those of females (44.4 ha, range = 16.3–150.7 ha). The bats foraged exclusively over water, whereby hunting activity was restricted to the immediate shoreline of Gatun Lake. Foraging areas ranged from 2.7 to 96.1 ha, with a median of 12.3 ha for both sexes. In general, there was high night-to-night consistency in the use of foraging areas. Most bats had multiple foraging and core areas. Core-use areas corresponded to approximately 35% of the bats' foraging areas and were larger in females (median = 5.3 ha, range = 1.1–54.1 ha) than in males (3.3 ha, range = 2.6–8.7 ha). Maximum range span varied from 0.5 to 7.5 km, whereby males in general and 1 tracked harem male in particular foraged much closer to the day roost than did females. Individual M. macrophyllum traveled an estimated 35–47 km on a nightly basis, and females moved significantly greater distances than did males. Our results do not provide support for the proposition that small bats typically have small home ranges. In contrast, we conclude that foraging strategy, diet, dispersion of food resources, and wing morphology are generally better predictors of home-range size. Furthermore, reproductive condition, as well as intraspecific competition linked to colony size or social status, also strongly affect the spacing behavior and movement patterns of these bats. Our results suggest that, although M. macrophyllum employs a gleaning foraging strategy, it closely parallels aerial insectivorous bats in terms of range size and movement distances.

This study documents the diet of Mexican free-tailed bats (Tadarida brasiliensis mexicana) at Carlsbad Cavern, Carlsbad Caverns National Park, Eddy County, New Mexico, and provides information on seasonal variation in food habits of this species throughout its summer residence. Diet was determined from 1,303 fecal samples. Eleven orders and 38 families of insects, unidentified insects, 2 orders of Arachnida (Araneae and Acari), bat hair, and mist net were consumed, with Lepidoptera (moths) and Coleoptera (beetles) occurring at greatest percentage volumes and percentage frequencies in the diet. Diet varied significantly throughout the season. Twenty-two food categories exhibited statistically significant variation among sampling sessions. Amounts of Coleoptera, Lepidoptera, Formicidae, Psyllidae, Hymenoptera, and Diptera consumed showed the greatest differences among sampling sessions.

Evidence on whether the northern short-tailed shrew (Blarina brevicauda) is a significant predator or a source of bias in trap-capture success of small mammals is contradictory. We compared subsequent captures of 2 murid rodent species (Clethrionomys gapperi and Peromyscus maniculatus) after capture of either murid species or B. brevicauda by using 4 years of capture–release data from New Brunswick, Canada. Capture success for P. maniculatus in a trap after occupancy by B. brevicauda was 65% lower than expected, and for C. gapperi, 67% lower than expected. The response by C. gapperi and P. maniculatus to odors of a different species was similar to their response to odor of B. brevicauda. Although we conclude that examination of odor-response data does not permit a conclusion as to whether B. brevicauda is a significant predator, the ubiquitous distribution of B. brevicauda implies that their influence on trap-capture success is significant, particularly in forests where small-rodent species are of similar weight to B. brevicauda.

Female caribou subsist primarily on lichens and some senescent browse during winter when demands for fetal growth add to costs of thermoregulation and mobility. Lichens, although potentially high in digestible energy, contain less protein than required for maintenance by most north-temperate ungulates. To understand the adaptations of caribou to the nutritional constraints of their primary food resource, we fed captive female caribou a sequence of 3 diets designed to resemble decreasing quality of forages during early, mid-, and late winter, respectively: high energy–high protein (HIGH), medium energy–low protein (MEDIUM), and low energy–low protein (LOW). In vitro digestibility of dry matter declined from 94% (HIGH) in November, to 66% (MEDIUM) in December and January, and to 53% (LOW) from February to April. Dietary protein averaged 19.8% in November and 4.3% from December to April. We used measures of body condition, stable isotopic signatures, and concentrations of nitrogen (N) metabolites to define protein dynamics in the animals. Subcutaneous rump fat declined between October and April from 2.3 cm ± 0.3 SE to <0.5 cm as intake of digestible energy declined from 44.0 ± 2.0 MJ/day to 16.3 ± 3.2 MJ/day. In erythrocytes, increasing enrichment of carbon (¹³C) throughout winter suggested that caribou reused body lipids, and increases in ¹⁵N during January and February indicated that they also recycled amino-N. Urinary N was primarily urea with an isotopic signature that tracked dietary ¹⁵N through late winter. Plasma urea-N declined from 44.0 ± 2.6 mg/dl to 8.5 ± 1.2 mg/dl as nitrogen intake declined from 91.5 ± 5.3 g N/day to 14.1 ± 0.9 g N/day. Examination of these data suggests that caribou catabolized dietary C and N in preference to endogenous fat reserves and body protein. Female caribou appear to tolerate low intakes of protein and energy in winter by minimizing net loss of body protein and reapportioning body reserves to support fetal growth.

Animal dispersal patterns influence gene flow, disease spread, population dynamics, spread of invasive species, and establishment of rare or endangered species. Although differences in dispersal distances among taxa have been reported, few studies have described plasticity of dispersal distance among populations of a single species. In 2002–2003, we radiomarked 308 juvenile (7- to 10-month-old), male white-tailed deer (Odocoileus virginianus) in 2 study areas in Pennsylvania. By using a meta-analysis approach, we compared dispersal rates and distances from these populations together with published reports of 10 other nonmigratory populations of white-tailed deer. Population density did not influence dispersal rate or dispersal distance, nor did forest cover influence dispersal rate. However, average (r² = 0.94, P < 0.001, d.f. = 9) and maximum (r² = 0.86, P = 0.001, d.f. = 7) dispersal distances of juvenile male deer were greater in habitats with less forest cover. Hence, dispersal behavior of this habitat generalist varies, and use of landscape data to predict population-specific dispersal distances may aid efforts to model population spread, gene flow, or disease transmission.

Examination of processes influencing forage habitat (patch) use is needed to understand factors driving spatial variation in habitat selection (use in relation to availability). I hypothesized that grass abundance was positively associated with use by Roosevelt elk (Cervus elaphus roosevelti) within patches and that elk use of 2 types of patches (clear-cuts and meadows) was supplemental, not complementary. Supplemental use of patches occurs when no single patch provides enough forage (grass in this study) to sustain the population and, hence, grass in one patch substitutes for grass in other patches. Complementary patch use occurs when a resource in one patch is nonsubstitutable with resources in other habitats. Predictions were generated, tested, and upheld for supplemental patch use. Elk use was higher in meadows, which had more grass, than in clear-cuts and elk use of clear-cuts was positively associated with grass biomass. I detected positive relationships among clear-cut size, meadow size, and elk use. Also, elk use of meadows was associated with proximity and size of the next meadow. Spatial variation in patch selection by this population of Roosevelt elk was influenced by grass biomass within patches and attributes among patches. I identified forces influencing patch selection because patch use was supplemental. Consequently, forage attributes within patches were connected to use among patches.