Preserving the earth's diversity of species requires that conservation biology turn much of its attention to reconciliation ecology, which is the science of sharing our habitats with wild things. Although many reconciliation projects are already working, we can extend our efforts with modern natural history research focused on species we aim to open our landscapes to. Some of this research will utilize what we already know about the principles of community ecology and niche organization and some will teach us more about those subjects. Perhaps most important will be a deeper understanding of the coevolution of niche apportionment. However, lacking perfect understanding is no reason to delay. The ecologist's motto ought to be, “Stop whining and try something.”

Garlic mustard (Alliaria petiolata) is an invasive, spring-germinating, obligately biennial plant. In a central Pennsylvania forest, two distinctive plant patch types exist simultaneously: patches with mixed juveniles and adults and patches that are exclusively juvenile. We experimentally tested two hypotheses to explain this peculiar distribution. The habitat limitation hypothesis proposes that garlic mustard, like other biennials, is limited to rarely available habitats and can only reach maturity in a subset of the patches where its seeds germinate. The intraspecific competition hypothesis proposes that juveniles only survive to maturity in patches lacking adults. We mapped the natural distribution of ten juvenile and ten mature patches and used transects to assess juvenile and adult density in 2 y. We found that patches cycled annually from juvenile to mixed, supporting the intraspecific competition hypothesis. We looked for biotic (interspecific competition) and abiotic (shading) factors that might contribute to habitat limitation, but none affected juvenile performance. We also looked for evidence of intraspecific competition by monitoring juvenile growth in juvenile, mixed and mixed with adults-removed patches. We found that juveniles from juvenile patches have a higher probability of survival than juveniles from mixed patches and adult removal improved juvenile survival in mixed patches. We conclude that the spatial and temporal segregation of juvenile and adult age classes is maintained by intraspecific competition.

The clonal shrub Cornus drummondii is rapidly displacing mesic grassland in the central U.S. due to fire suppression and changes in land use. Once established, this shrub is not readily eliminated by the return of frequent fire, leading to significant and perhaps irreversible shifts in tallgrass prairie structure and function. We assessed the impacts of C. drummondii encroachment on herbaceous aboveground net primary production (ANPP) and plant community structure in tallgrass prairie and the role this species plays in the conversion of grassland ecosystems to shrub/woodlands. We also removed established C. drummondii clones to assess the potential of the grassland ecosystem to recover after shrub dominance. Aboveground net primary production, vegetative cover and species richness and diversity (exclusive of C. drummondii) were significantly reduced beneath shrub islands relative to open grassland, with reductions in ANPP and richness of up to 94 and 45%, respectively. Forbs were the dominant growth form in the C. drummondii understory, and 10 species primarily associated with woodland habitats occurred only within shrub islands. Upon removal of C. drummondii, ANPP, richness and diversity recovered to grassland values within 2 y; however, forbs remained the dominant growth form, comprising 73% of total cover. These results indicate that C. drummondii exerts strong control over the structure and function of unburned mesic grassland ecosystems and that this shrub may be key in the conversion of grasslands to woodlands. Removal of C. drummondii resulted in some aspects of recovery, but the return of graminoid dominance was not attained after 2 y. This represents a legacy effect of C. drummondii of unknown duration. Prevention of woody species encroachment through frequent burning is a preferred management option for this ecosystem.

Seeds (true seed plus endocarp) of Viburnum acerifolium have underdeveloped linear embryos and deep simple epicotyl morphophysiological dormancy. Seeds mature and are dispersed in autumn, and radicles and epicotyls emerge the following autumn and late autumn-winter, respectively. Embryos in seeds incubated at 25/15 C became fully elongated in 16 wk, whereas those incubated at 5 C hardly grew at all. The optimum continuous temperature regime for both rate and percentage of radicle emergence was 25/15 C. At this regime, however, ≥20 wk were required for radicles to emerge in a high percentage of the seeds. GA₃ had little or no effect on radicle emergence. In seeds subjected to a winter → spring → summer → autumn temperature regime in controlled environments, radicles emerged in autumn in the 20/10 → 15/6 C portions of the sequence. In those subjected only to the summer → autumn portions of the sequence, radicles also emerged in autumn at 20/10 → 15/6 C. Thus, in nature, the cold winter season has no effect on dormancy-break. Radicles of seeds planted in a nonheated greenhouse in November 1998 emerged in October 1999, when mean maximum and minimum temperatures were about 19 and 9 C, respectively. The optimum continuous temperature regime for both rate and percentage of epicotyl emergence was 15/6 C. For seeds planted in the nonheated greenhouse in November 1998, rate of epicotyl emergence peaked in early January 2000, when mean maximum and minimum temperatures were about 10 and 5 C, respectively.

Low levels of fruit production are thought to threaten the persistence of Asclepias meadii (Mead's milkweed) populations. We evaluated four hypotheses explaining the low fruit production in one population (herbivory, pollen limitation, resource shortages and rainfall) by collecting within-year data on stem size, levels of fruit initiation and mature fruit production and by considering among-year data on total fruit production and annual rainfall. In 2002, severe herbivory resulted in the death of about 63% of the population's flowering stems. Managers could increase fruit production by protecting stems from mammal herbivory. A path analysis revealed that leaf width and the number of fruits initiated were the most important predictors of mature fruit production. In 1991–2002 the previous year's total annual precipitation significantly affected the total number of fruits produced only in years of burning. Increasing the persistence of these long-lived plants to years of greater rainfall could have the most dramatic effects on fruit production and on survival of the species.

Adult insects were collected using emergence traps from salt water pools, brackish water pools and vegetated marsh in the Little River Estuary, Wells, Maine, USA. Samples were continuously taken once every week from June through October of 2001 to examine the community composition, as well as the spatial and temporal patterns of insect emergence from a salt marsh system. Chironomus sp. and Tanytarsus spp. (Diptera: Chironomidae) chironomids were the most abundant insects emerging, the former representing 87% of all insects collected from brackish pools and the latter 55–80% of all insects collected from salt water pools and the vegetated marsh, respectively. Insect emergence was higher from brackish pools (1450 ± 172 individuals·m⁻²·y⁻¹) and vegetated marsh sites (1350 ± 851 individuals·m⁻²·y⁻¹), but not significantly different from salt water pools (289 ± 126 individuals·m⁻²·y⁻¹). Emergence of some species was restricted to certain marsh areas resulting in non-significant trends in patterns of community composition, diversity and evenness. More species (n = 15) were collected from the vegetated marsh compared to brackish (n = 13) and salt water pools (n = 11), while diversity and evenness were greater in salt water pools (H′ = 1.56 ± 0.4, 0.5 ± 0.2) compared to the vegetated marsh (H′ = 1.18 ± 0.8, 0.3 ± 0.2) and brackish pools (H′ = 0.72 ± 0.01, 0.2 ± 0.02). Temporal patterns of the two most abundant chironomids revealed increased numbers emerging in June (Tanytarsus spp. and Chironomus sp.) and again in September (Chironomus sp.). This suggests the importance of salt marsh insects as a food source to fish and birds, particularly during breeding and nesting seasons in the spring (June) or migratory season in the fall (September).

This study determined whether the presence of Eastern tent caterpillars on black cherry affects selection of the same tree by fall webworms in the same season. Five study sites were selected, each containing 30 to 50 black cherry trees, which were marked and measured. The total number of Eastern tent caterpillar nests was counted on each tree in June 2001. Fall webworm nests were recorded as they developed throughout the summer. A logistic regression showed a significant difference (P < 0.01) in the use of previously colonized vs. uncolonized trees by the fall webworm. Fall webworms were less likely to occur on trees that had been used by Eastern tent caterpillars earlier in the season. At the end of the study period, 8.5% of the trees had been used by both species and fall webworms alone had colonized 21% of the trees observed.

Female amphipods carry their eggs and offspring in a ventral brood pouch. Previous research has demonstrated that females of species inhabiting harsh environmental conditions expend additional energy in active parental care such as ventilating eggs by beating their pleopods. This study investigated whether similar forms of care could be detected in amphipods collected from spring fed ponds and streams. Two behaviors reported to function in parental care in other species were observed in ovigerous females, but were absent in nonovigerous females and males. Ovigerous females were also shown to beat their pleopods significantly longer than males or nonovigerous females, suggesting that this is another form of active parental care. The duration of pleopod beating varied considerably depending on both the season and the habitat from which the amphipods were collected. Differences were most pronounced for ovigerous females. Similarly, the duration of pleopod beating was significantly longer, especially for ovigerous females, when amphipods were placed in water with experimentally reduced levels of oxygen. These results suggest that even in habitats with high water quality, amphipods can show active parental care and that females have a phenotypically plastic response to environmental conditions in determining the amount of care provided. The potential fitness effect of such response may have important implications for water quality management.

Freshwater mussel beds of the lower 68 km of the Cache River, Arkansas, were delineated, sampled using dive techniques and a stratified random sampling methodology and analyzed for density and species richness. A total of 38 mussel beds were delineated, 14 major beds (Mbeds) and 24 minor beds (mbeds), and defined by areal extent and mussel density. Analysis of our sampling precision indicated 80% or better confidence levels for a majority of our sites and suggested that a sample size of 15 1-m² quadrats is sufficient to obtain 80% or better confidence. Our large river diver-assisted sampling methodology has been shown to be a useful and appropriate methodology for obtaining large geographic scale baseline distribution (bed and species), species richness, density and population and community numerical standing crop estimates information where tradeoffs are required in order to complete a project within time and budget constraints.

Freshwater fish assemblages are sensitive to changes in land use, but it is unclear how rapidly assemblages respond to such change or how closely they track physical changes in the stream environment. We monitored fish assemblages at the outflow of three watersheds on the outer coastal plain of Mississippi over a period of 8 y. The watersheds differed in patterns of land use, being predominantly forested, rapidly urbanizing, and long-urbanized, respectively. Watersheds were distinct in fish species composition, differing primarily in abundance of Etheostoma lynceum, E. stigmaeum, Luxilus chrysocephalus and Lythrurus roseipinnis. Species richness was consistently greatest at the predominantly forested site and least at the urbanized site. Fish assemblages were distinguishable in terms of substrate preference, water-speed, silt tolerance, and trophic habit. The stream draining the urbanizing watershed showed an increase in richness of species of high silt tolerance and a decrease in richness of those preferring a gravel substrate over the sampling period. Contrary to expectation, annual variation in composition was modest relative to differences among sites. Annual variation in richness and numbers was unrelated to measures of streamflow or seasonal precipitation. Changes in stream character related to land use change appear to have impacted the fish assemblages of these streams, but the transition to an urban fish assemblage in the urbanizing stream occurred before the study period, well before the obvious signs of physical degradation appeared at the site.

We compared the effectiveness and efficiency of three terrestrial salamander and gastropod trapping techniques: pitfall traps, ground searches and cover boards. The study was conducted on 18 stands with three management histories in the Umpqua National Forest, southern Oregon Cascades. A total of 648 pitfall traps were open for 28 consecutive days in fall 1999. Two hundred twelve amphibians (eight species) and 202 gastropods (six species) were captured. Also in fall 1999, 36 h of ground searches covering 3600 m² resulted in the detection of 19 amphibians (two species) and 130 gastropods (six species). Four cover boards (100 × 100 cm) in stacks of two were placed in each stand and checked four times in fall 1999 and once in spring 2000 after snow melt. Cover boards concealed no amphibians and only two gastropods (one species). Pitfall traps were more efficient at capturing amphibians than ground searches (0.41 vs. 0.25 captures per hour of effort), but less efficient at capturing gastropods than ground searches (0.39 vs. 1.73 captures per hour of effort). Cover boards as used were ineffective at capturing either amphibians or gastropods. Climatic conditions of the southern Oregon Cascades likely influenced the results.

We found more green tree frogs (Hyla cinerea) in canopy gaps than in closed canopy forest. Of the 331 green tree frogs observed, 88% were in canopy gaps. Likewise, higher numbers and biomasses of insects were captured in the open gap habitat. Flies were the most commonly collected insect group accounting for 54% of the total capture. These data suggest that one reason green tree frogs were more abundant in canopy gaps was the increased availability of prey and that small canopy gaps provide early successional habitats that are beneficial to green tree frog populations.

In the southeastern United States, coarse woody debris (CWD) typically harbors high densities of invertebrates. However, its importance as a foraging substrate for southeastern amphibians is relatively unknown. We examined effects of CWD manipulations on diet composition of southern toads (Bufo terrestris) in upland loblolly pine (Pinus taeda) stands in the Coastal Plain of South Carolina. Twelve 9.3-ha plots were assigned one of the following treatments: removal- all CWD ≥10 cm in diameter and ≥60 cm long removed; downed- five-fold increase in volume of down CWD; and unmanipulated control stands. We collected southern toads ≥4 cm snout-vent length (SVL) during 14 d sampling periods in June and October 2002, June 2003 and during a 28 d sampling period in April 2003. We collected 80, 36 and 35 southern toads in control, downed and removal treatments, respectively. We found no difference in relative abundance or frequency of invertebrate groups consumed among treatments (P > 0.05). Average body weight (g), SVL (cm) and stomach content weight (g wet) of individuals also were similar among treatments (P > 0.05). The role of CWD as a foraging substrate for southern toads in loblolly pine stands of the southeastern Coastal Plain may be negligible, at least in the early stages of decay.

We performed two experiments to address the questions of whether toad tadpoles (Bufo a. americanus): (1) gain an advantage from consuming conspecifics and (2) engage in active cannibalism as opposed to scavenging. Our results show that tadpoles fed algal mats (and associated debris found within them) from their natal pond and supplemented on the bodies of dead tadpoles for 28 d attained an average of 5 more stages of development than tadpoles fed only on algal mats, which suggests a strong advantage to eating conspecifics. No tadpoles fed algal mats died, which suggests that this is an adequate food source. In a series of trials used to address the second question, we found no indication that tadpoles engage in active cannibalism. We tested this under conditions of hunger and using injured tadpoles as potential prey and progressively older tadpoles as potential cannibals. In no cases did we observe any tadpole attacking living tadpoles (even if injured) within 1-h after being placed in experimental tanks. Although some tadpoles were missing from tanks after 3-d periods, our results suggest that some (e.g., injured) tadpoles may succumb and are eaten thereafter. Test tadpoles began consuming dead crushed tadpoles significantly more quickly than dead intact tadpoles, suggesting that cues used to induce feeding are chemical and perhaps a result of some bacterial decomposition. We suggest that toad tadpoles have not evolved active cannibalism because of energetic and other costs and because mortality rates of tadpoles are naturally high in shallow ephemeral breeding ponds. Living tadpoles thus have access to dead tadpoles as a food source in such circumstances without incurring some costs of cannibalism.

Stream salamanders are sensitive to acid mine drainage and may be sensitive to acidification and low acid neutralizing capacity (ANC) of a watershed. Streams in Shenandoah National Park, Virginia, are subject to episodic acidification from precipitation events. We surveyed 25 m by 2 m transects located on the stream bank adjacent to the water channel in Shenandoah National Park for salamanders using a stratified random sampling design based on elevation, aspect and bedrock geology. We investigated the relationships of four species (Eurycea bislineata, Desmognathus fuscus, D. monticola and Gyrinophilus porphyriticus) to habitat and water quality variables. We did not find overwhelming evidence that stream salamanders are affected by the acid-base status of streams in Shenandoah National Park. Desmognathus fuscus and D. monticola abundance was greater both in streams that had a higher potential to neutralize acidification, and in higher elevation (>700 m) streams. Neither abundance of E. bislineata nor species richness were related to any of the habitat variables. Our sampling method preferentially detected the adult age class of the study species and did not allow us to estimate population sizes. We suggest that continued monitoring of stream salamander populations in SNP will determine the effects of stream acidification on these taxa.

Habitat loss has been identified as a principle reason for decline of many water snakes, and surface mining for coal could potentially put Copperbelly Water Snakes (Nerodia erythrogaster neglecta) at risk due to the severity of land cover change that takes place once mining and reclamation are complete. We studied Copperbelly Water Snakes in riparian habitat impacted by adjacent surface mining in southern Indiana. Snakes were surveyed pre-mining (1992 and 1993), during mining (1994 to 1996) and post-mining (1997 to 2000). Abundance of Copperbelly Water Snakes (cws/km ± se) was higher after mining (5.43 ± 0.37) compared to abundance levels before (2.04 ± 0.74) or during mining (3.32 ± 0.76). No Copperbelly Water Snake was observed in reclaimed habitat immediately following reclamation, but use of reclaimed habitat by these snakes was evident in subsequent years (n = 89). Size class distribution of Copperbelly Water Snakes observed in reclaimed habitat largely paralleled that of the entire population of these snakes in the flood plain. Evidence of reproduction by Copperbelly Water Snakes, in the form of juveniles (<76 cm in length) and young-of-year (<30 cm in length), was observed throughout the study. These data indicate that the population of Copperbelly Water Snakes was reproductively active, sustained higher levels of abundance following completion of mining and reclamation and made frequent use of reclaimed habitat. The extensive use of constructed ponds and drainage ditches by these snakes suggests that reclamation following mining can be done in a manner that facilitates recovery of habitat for this species.

The carapace temperatures of ornate box turtles (Terrapene ornata ornata) were monitored by affixing temperature dataloggers during two winters, 2001–2002 and 2003–2004. The dataloggers recorded temperature every 3 h to the nearest 0.5 C. Date of burrowing in the fall, emergence time in the spring and temperatures 48 h prior to emergence were analyzed along with the entire temperature profile of the turtles during the winter. In 2003–2004 soil temperatures from the surface to 0.75 m depth were also monitored.

Most turtles did not experience freezing temperatures during either winter, but two turtles were below freezing for up to 54 consecutive days during winter 2001–2002.

During 2001–2002, eight of nine turtles began burrowing within 7 d of each other in the fall, and all nine emerged within 7 d of each other in the spring. During 2003–2004, fall burrowing for 16 turtles occurred within 14–21 d, and emergence happened within 15 d. Overwintering periods varied between 172–201 d, which was intermediate between dates noted in studies to the north, south and west for this species. During 2003–2004, it was inferred that turtles burrowed to a minimum of 0.5–0.75 m by comparing turtle temperatures to those recorded at different depths in the soil, although we recorded a previous depth of up to 1.67 m.

Emergence onto the surface in the spring did not relate to soil temperature, and there was no support for emergence triggered by 48 h of soil temperatures above 7 C (Grobman, 1990). It is suggested that spring emergence may be correlated with a more complex set of factors and that Terrapene ornata ornata can withstand prolonged freezing temperatures.

The Texas map turtle (Graptemys versa) is endemic to the Colorado River drainage in southcentral Texas. A study of its life history was undertaken using data collected in 1998–2000 from a population in the South Llano River, southernmost tributary of the Colorado drainage, and data from museum specimens that had been collected from the South Llano River in 1949. Compared to congeners, G. versa is a small-bodied species. Its small body size is, predictably, linked to relatively small clutch size, small egg size, rapid growth toward asymptotic size and early maturation. As many as four clutches may be laid during an active season, although the effects of follicular atresia on clutch frequency are not known. Both clutch size and egg width were positively correlated with female body size, with the former relationship having a log-log slope significantly less than the expected value of 3, probably due to the latter relationship. Analyses were consistent with the hypothesis of anatomical constraint on egg size, with at least smaller females laying eggs that are of less than optimal size. No differences were found in body size or clutch size between 1949 and 1998–2000 despite a large-scale change in diet associated with invasion of the river by Asian clams (Corbicula sp.). However, body size is substantially reduced in the South Llano River compared to other sections of the Colorado drainage, a finding mimicked by at least one other turtle species in the drainage, Pseudemys texana.

Knowledge of chick ecology is essential for understanding and managing populations of northern bobwhites (Colinus virginianus). We report on chick survival and growth in western Oklahoma during 1997–2002. We used Cox proportional hazard models to estimate chick survival rate. Covariates included mass at time of capture, Julian day of capture and year. Chick mass was the only significant covariate in the model. Survival depended on time since capture because survival increased with chick mass, which concomitantly increased with time since hatch. Because of the dependence of survival on chick mass, we modeled chick growth as a function of age using a logistic model (r² = 0.98). Growth-rate was highest at day 35 post-hatch, which was around the mean capture-to-death survival time for chicks (30.0 ± 4.4 d se). Our results suggest that the 30–35-d period post-hatching is a critical period in chicks' lives.

From 1999 to 2001 we monitored 12 coyotes (Canis latrans) in northwestern Texas to determine their home ranges and habitat use in a landscape interspersed with native prairie, farmland and Conservation Reserve Program (CRP) fields. Annual home range size was 10.1 km² for residents and 84.5 km² for transients. We determined habitat use at two spatial scales: within home ranges and within study area. Habitat use patterns were similar at both scales, as residents selected for native prairie and transients selected for CRP fields. Habitat use between residents and transients differed in both seasons, with residents selecting more native prairie, less farmland and less CRP (summer only) than transients. Habitat at natal den sites also differed from expected for residents, as most dens (8 of 10) were located in CRP fields. The CRP fields contained the only tall permanent vegetation on our study sites and appeared to provide important foraging habitat for transient coyotes, and denning habitat for resident coyotes.

Forest managers are increasingly expected to incorporate biodiversity objectives within forest landscapes devoted to timber production. However, reliable data on which to base management recommendations for bats within these systems are extremely limited. Although the red bat (Lasiurus borealis) is a widespread common species in temperate forests of North America, little is known of its ecology within intensively managed pine (Pinus spp.) forests of the southeastern United States. Therefore, we investigated size of foraging areas and habitat use by red bats during summer 2000 and 2001 in an intensively managed pine landscape in east-central Mississippi, USA. We captured bats using four-tier mist nets placed over water and attached radiotransmitters to red bats. Radiotagged red bats (n = 16) used habitat types randomly at the study area and foraging area scale. Mean size of foraging areas and mean maximum distance traveled between diurnal roosts and foraging locations were not different (P < 0.05) among adult male, adult female, juvenile male or juvenile females (n = 18). Most foraging areas contained a reliable source of water and all but one diurnal roost was located within foraging areas. Location of diurnal roosts may dictate location of foraging areas. Open canopy conditions in intensively managed pine stands (young, open canopy stands, thinned stands and riparian hardwood stands) likely provide appropriate foraging habitat for red bats. Landscape context may influence size of foraging areas and commuting distances of red bats. Provision of appropriate aged forest stands for diurnal roosts may be the best management action for red bats within intensively managed pine landscapes.

The rapid growth of coyote (Canis latrans) populations in Illinois since 1980 prompted a need for current demographic data to be used in population models for management. From 1994 to 1997 we examined 977 coyotes harvested by hunters and trappers throughout the state and compiled data on age structure, sex ratios, reproductive rates and physical condition. Statewide, 55% of harvested coyotes were juveniles, 20% were yearlings and 25% adults. The sex ratio did not differ from unity among juveniles and yearlings, but was skewed towards males in the adult class. The number of harvested animals decreased by 45–60% between successive classes from 0.5 through 2.5 y old, then by 30–40% through 9.5 y old. Statewide, 57% of females bred and breeders averaged 4.9 placental scars. Males were larger than females, averaging 14.1 and 12.1 kg whole body mass, respectively. Kidney fat reserves were highest in juveniles and adult females and lowest in yearlings. Femur marrow fat was generally high and did not differ among age-classes. Ovulating females were heavier than non-ovulators among juveniles and yearlings. We found that coyotes in Illinois are in good physical condition with high winter fat reserves and reproductive rates. However, reproductive rates are lower than they were in 1978–1979 when the population was rapidly expanding in the state.

We studied turtle populations inhabiting a canal and a lake (both man-made) within a heavily disturbed, urban setting. Six aquatic and semi-aquatic turtle species were collected in both habitats: spiny softshell turtle (Apolone spinifera), painted turtle (Chrysemys picta), common snapping turtle (Chelydra serpentina), common map turtle (Graptemys geographica), common musk turtle (Sternotherus odoratus) and red-eared slider (Trachemys scripta). While G. geographica was the most common species in the canal habitat, T. scripta was most common in the lake habitat. We describe patterns of sexual size dimorphism and sex ratios for the three most abundant species (G. geographica, T. scripta and S. odoratus). We discuss our data in light of problems facing turtle assemblages in urban settings.

Aspen stands and riparian areas are important to breeding birds in the southwestern U.S. because they provide resources such as food and shelter. We investigated how this importance varies throughout the year for both resident and migratory birds. We sampled birds in 96 sites, half in small isolated aspen stands and half in the ponderosa pine forest in northern Arizona during the summer of 1996, and a subset of those plots during fall of 1996 and the spring of 1997. Bird species richness and abundance varied seasonally. During the summer there were more birds and more bird species in aspen stands. This relationship appears to be driven by an affiliation between Neotropical migrants and aspen trees. During fall, residents were associated with riparian areas. We demonstrate the importance of small inclusions of aspen to Neotropical migrants in the Southwest during the breeding season and we show that preference for habitat types among migratory groups can vary seasonally.

Franklin's ground squirrel (Spermophilus franklinii) is declining in the eastern portion of its range, and this decline is often attributed to habitat fragmentation. However, the ability of S. franklinii to disperse across an agricultural landscape is not known. During spring 2002 we live trapped a small, apparently isolated, population of Franklin's ground squirrels in a 12-ha tallgrass prairie restoration located south of Urbana, Champaign County, Illinois. This prairie “island” was surrounded primarily by row-crop agriculture. We radio-tracked 14 juvenile Franklin's ground squirrels (seven males and seven females) throughout dispersal to determine how far dispersers traveled, the timing of dispersal, if dispersal distance differed between sexes and if the agricultural matrix surrounding the study site was a barrier to movements. Males dispersed farther than females, but individuals of both sexes moved ≥1 km from the study site. The farthest movement recorded was by a male who traveled 3.6 km. Dispersal was age-dependent for both sexes, occurring at 9–11 wk of age. Agricultural fields did not seem to hinder movement, probably because dispersal occurred in late July and August before row crops were harvested. Open areas such as roadways, however, may be barriers for some individuals.

Diet of the bobcat (Lynx rufus) in the mid-Atlantic region of the United States is poorly known. Age, sex and time of year are thought to influence bobcat prey selection, but accounts vary. We examined the contents of 85 bobcat stomachs taken from Pennsylvania during autumn and winter 2000–2002. White-tailed deer (Odocoileus virginianus) and rabbits (Sylvilagus sp.) occurred most frequently as prey. A larger percentage of female bobcats consumed lagomorphs (28%) than did males (12%; P = 0.04). More male bobcats consumed meso-mammals (14%), including raccoons and porcupines, than did females (2%; P = 0.04). Diet did not differ between bobcats <2 y and bobcats ≥2 y. Variety of prey eaten was larger in autumn than in winter. Sexual differences in diet may be related to sexual dimorphism in this species and may help to effect interspecific niche partitioning. Winter weather reduces the availability of certain prey items in this region and may decrease the bobcat's ability to travel and hunt effectively.