Habitat destruction is believed to be the number one cause of the decline in unionid mussels. Around the world, cities, towns and agriculture alter the structure of watersheds, and the Black River in Ohio may be a typical example. We investigated the diversity and abundance of unionid mussels in this watershed and compared results to urbanization locations, to site-specific appearance of the habitat and to a 1997 fish survey, as host species are another factor important to the distribution of unionid mussels. Although shells were found for 21 species, only 11 of these species were found alive. Seven of the species represented only by shells occurred only in the urbanized lower main stem of the river and less than five shells were found for each. Most of these shells were old and worn. Furthermore, the present assemblage in the main stem varied from shells obtained at a nearby archeological site, and from a voucher set of species obtained at the turn of the 20th Century. Mussel communities higher in the river and those in tributaries were less diverse, but abundance of the species present was higher than in the main stem. A lack of fish hosts may limit mussel diversity, as hosts for several species present in the main stem do not reside higher in the watershed. Overall, mussel assemblages in the Black River appear typical for the region with relatively abundant, but low diversity communities upstream of the cities that line Lake Erie's coast and diverse, but small and potentially threatened, populations in the urban regions.

We examined habitat selection by the painted turtle (Chrysemys picta marginata) and common map turtle (Graptemys geographica) in the Ottawa National Wildlife Refuge, Ohio, USA. Basking traps caught 198 (162 C. picta and 36 G. geographica) and 140 (106 C. picta, 33 G. geographica, 1 Trachemys scripta elegans) unique individuals in 2003 and 2004, respectively. Radiotransmitters were attached to selected adults of both species. Chrysemys picta were caught in marshes more than in channels in both years. Radiotelemetry results corresponded with trap data; C. picta were located 65.6% and 90% of the time in marshes in 2003 and 2004, respectively. With just four exceptions, G. geographica were captured only in channels in both years. In contrast, radiotelemetry data revealed that G. geographica were located 40.7% and 43.7% of the time in marshes in 2003 and 2004, respectively. Our results showed differences in habitat use for these two species, but different measures of habitat use (basking traps vs. radiotelemetry) provided different insights to behavior.

Population declines of amphibians and reptiles throughout the world have led to the initiation of projects to monitor their status and trends. Historical collections give an indication of which species occurred in an area at one time, although the ambiguity surrounding locations and environmental conditions associated with collection decreases the value of this information source. Resampling using the same general protocols can give valuable insights to changes in community structure. However, this is only feasible when sampling methodology and exact site locations are known. From 2002–2005 we resampled 12 sites in St. Marks National Wildlife Refuge in Florida's panhandle, an area in which intensive herpetological surveys were conducted in 1977–1979. We documented a general decrease in species richness among the diversely managed sites, changes in dominant species and diversity and an increasing trend toward homogeneity of the herpetofaunal community among habitats. Changes were attributed to four causes: 28-y of forest community succession, changes in management practices, non-detection of species due to variation in sampling conditions and a decrease in occupancy by four amphibians and three reptiles. The use of population and habitat-related indexes helped define possible influences on community change and can be used to target species for monitoring. Declines of these seven species are of concern, especially considering the protected status of the refuge and its increasing isolation as surrounding landscapes are converted to urbanized settings.

According to optimal offspring size theory, natural selection pressures balance the egg size/clutch size trade-off at a point where increases in maternal body size result in increases in clutch size but not increases in egg size. However, many turtle species show increasing egg size with increasing maternal body size. The anatomical-constraints hypothesis explains this pattern by hypothesizing that smaller females lay smaller-than-optimal eggs because of morphological constraints; larger eggs simply would not pass through the pelvic aperture and caudal gap of the shell. We examined relationships among female body size (measured as plastron length), clutch size and egg size for a population of common map turtles (Graptemys geographica) at Presque Isle State Park in Erie, Pennsylvania. Correlation analyses were conducted using log-transformed data in order to address questions of isometry and allometry. Clutch mass increased isometrically with plastron length. Egg size and clutch size were both significantly negatively allometric in their relationship with plastron length. It appears that larger females split the increased reproductive allocation made possible by increased maternal volume devoted to eggs between increasing both clutch size and egg size, consistent with predictions of the anatomical-constraints hypothesis.

We documented the nesting chronology, nest site selection and nest success of smallmouth bass Micropterus dolomieu in an upstream (4th order) and downstream (5th order) reach of Baron Fork Creek, Oklahoma. Males started nesting in mid-Apr. when water temperatures increased to 16.9 C upstream, and in late-Apr. when temperatures increased to 16.2 C downstream. Streamflows were low (77% upstream to 82% downstream of mean Apr. streamflow, and 12 and 18% of mean Jun. streamflow; 47 and 55 y of record), and decreased throughout the spawning period. Larger males nested first upstream, as has been observed in other populations, but not downstream. Upstream, progeny in 62 of 153 nests developed to swim-up stage. Downstream, progeny in 31 of 73 nests developed to swim-up. Nesting densities upstream (147/km) and downstream (100/km) were both higher than any densities previously reported. Males selected nest sites with intermediate water depths, low water velocity and near cover, behavior that is typical of smallmouth bass. Documented nest failures resulted from human disturbance, angling, and longear sunfish predation. Logistic exposure models showed that water velocity at the nest was negatively related and length of the guarding male was positively related to nest success upstream. Male length and number of degree days were both positively related to nest success downstream. Our results, and those of other studies, suggest that biological factors account for most nest failures during benign (stable, low flow) streamflow conditions, whereas nest failures attributed to substrate mobility or nest abandonment dominate when harsh streamflow conditions (spring floods) coincide with the spawning season.

In many species of anurans, tadpoles adjust their behavior in the presence of fish or fish cues, in many cases reducing their activity. Tadpoles may also adjust their activity or behavior in response to group size. We examined the effects of predator cues (bluegill sunfish, Lepomis macrochirus) and group size on the behavior of Wood Frog (Rana sylvatica) tadpoles in a factorial laboratory experiment. Wood Frog tadpoles were less active and used the vegetated side of the experimental aquaria more in the presence of fish. Group size alone had no effect on the behavior of the Wood Frog tadpoles in our experiments. Our experiment shows that cues from fish predators have a greater influence on the behavior of Wood Frog tadpoles than does group size, at least in the range of group sizes used in this experiment.

For the incised, sand-bed streams of north-central Mississippi, USA, fish predation is one plausible mechanism to explain both relatively low crayfish densities and differences in stream size occupied by various native crayfishes. I conducted two mesocosm experiments to test effects of a fish predator (channel catfish, Ictalurus punctatus) on the survival and size structure of native crayfishes (primarily Procambarus hayi and Orconectes chickasawae) in the presence and absence of shelter. I used predominantly the larger species, P. hayi, in the first experiment and the smaller species, O. chickasawae, in the second. Experiments lasted 19–21 d, and each consisted of four replicated treatments: crayfish without shelter, crayfish with shelter, crayfish and predator without shelter, crayfish and predator with shelter. In both experiments, catfish significantly reduced crayfish survival. Shelter significantly reduced catfish predation on the smaller, but not the larger, crayfish species. Comparisons between experiments showed that in tanks containing catfish, P. hayi had higher survival than O. chickasawae. In both experiments, the mean size of crayfish increased less in the presence than in the absence of catfish, and I argue that the effect is due largely to a reduction in crayfish growth. Channel catfish directly and indirectly influenced crayfish in experimental settings, with the degree of influence varying by crayfish species and presumably related to crayfish size. Thus, fish predation and shelter availability are likely important factors influencing densities of and macrohabitat use by these native crayfishes.

This research explored the distribution, morphological variation, and molecular systematics of Elimia comalensis (Gastropoda: Pleuroceridae) using geometric morphometrics (n  =  565) and mitochondrial DNA sequences (n  =  15). Elimia comalensis was originally documented as endemic to Comal Springs, Comal County, Texas, but recent collections found this species in multiple springs and drainages in Central Texas. Morphometric analyses showed a high amount of morphological overlap with no clear geographic patterning. Phylogenetic analysis of mt COI sequence data indicated that E. comalensis represents a single species, with no genetic divergence among isolated populations. We conclude that E. comalensis may be an unrecognized native exotic, a species endemic to one area that has been spread by humans and assumed to be part of the natural fauna.

Raccoons are habitat generalists, existing in a variety of landscapes. Although known to inhabit prairie landscapes, few studies have examined habitat selection of raccoons within these environs and relationships between landscape structure and habitat selection are poorly understood in prairie systems. We monitored 64 radio-marked raccoons on a prairie landscape in Mississippi during 1997–2001. We subsequently used space use estimates to generate predictive models of habitat use. Probability of use by raccoons increased with increasing patch size of rowcrops and forested areas and proximity to available water sources (lakes, streams). Space use also was positively related to increasing amounts of edge associated with rowcrop fields and forested habitat patches. Our findings suggest that in prairie landscapes, rowcrop fields providing quality forages, water sources and riparian forests and woodlots greatly influence space use, attributable to life history and ecological strategies satisfied by these landscape features.

The eastern spotted skunk (Spilogale putorius) is a rare species of conservation concern throughout much of its range, but effective management is hampered by a lack of information on appropriate survey strategies. We validated three commonly used techniques to identify the presence of eastern spotted skunks at four sites in Missouri and Arkansas where the species was known to occur. Live-capture with box-traps revealed a strong seasonal pattern in capture success in both states, with virtually all captures occurring between late Sept. and early May. This pattern of detection also occurred when surveys were conducted using non-invasive camera-traps and enclosed track-plates in Missouri. Track-plates were more efficient than camera-traps at detecting eastern spotted skunks, with a lower latency to initial detection (LTD) and higher probability of detection (POD). Our results indicate that the use of enclosed track-plates is a powerful non-invasive technique for detecting eastern spotted skunks when surveys are conducted between late Sept. and early May. Surveys conducted during late spring and summers are inappropriate given the high likelihood of not detecting the species despite its presence.

We radiotracked nine hoary bats (Lasiurus cinereus) and characterized 12 roosts during late spring and early summer in the Ouachita Mountains of central Arkansas. Hoary bats generally roosted on the easterly sides of tree canopies in the foliage of white oaks (Quercus alba), post oaks (Q. stellata) and shortleaf pines (Pinus echinata). Roost heights averaged 16.5 ± 2.2 m and all roost trees were >21 cm dbh and were taller and greater in diameter than random trees. Roost sites averaged 7.7 m²/ha basal area (BA) of overstory pines, 4.4 m²/ha BA of overstory hardwoods and 80% canopy cover. Roost sites had a greater number of understory stems than random sites. All roosts were located in stands dominated by mature (overstory >50 y old) overstory trees. Seven roosts were in relatively unmanaged mixed pine-hardwood and hardwood stands >50 y old and four roosts were in mature mixed pine-hardwoods stands that recently had been thinned and subjected to prescribed burning; one additional roost was located outside the study area in a mature shortleaf pine stand that was thinned approximately 10 y previously. Although not abundant during early summer in Arkansas (only 3% of our captures), hoary bats were confirmed to reproduce in the region.

Predation by coyotes (Canis latrans) has been documented as the main cause of mortality in most populations of swift foxes (Vulpes velox), although reasons for such high predation rates were often unclear. Additionally, coyotes kill but generally do not consume swift foxes, suggesting coyotes kill for reasons other than food. To better understand ecological relationships between these species, we studied dietary overlap of syntopic coyotes and swift foxes in northwestern Texas from 1998 to 2000. Both species consumed the same food items and had similar seasonal changes in diets, although the order of these items differed for each species. Overall, coyotes and swift foxes had high dietary overlap (Ro  =  0.856), although some dietary partitioning was evident based on food size categories. Dietary overlap was least in summer (Ro  =  0.714) and greatest in winter (Ro  =  0.859). Swift fox diets were dominated by small food items (i.e., rodents and insects), whereas coyote diets had nearly equal representation of all food classes. The similarity in diets between coyotes and swift foxes indicated the potential for resource competition between these species, although we did not determine food availability. Regardless, the killing and spatial displacement of swift foxes by coyotes throughout their distribution might be due to their high food resource overlap, especially because coyote populations tend to be limited by prey availability.

Coyotes (Canis latrans) are an opportunistic predator that have adapted to many human-modified environments. Conflicts between coyotes and humans are an increasing concern for managers in urban areas. We examined the spatial and temporal utilization and availability of natural and human-associated food for coyotes in the Chicago metropolitan area, Illinois, USA. We collected 1429 coyote scats from May 2000 to December 2002, and conducted prey surveys in 2002, in 4 sites that varied in their degree of urban development. Dominant food items included small rodents, white-tailed deer (Odocoileus virginianus), fruit, eastern cottontail (Sylvilagus floridanus) and birds. Their availability and occurrence in scats varied among sites and seasons. The occurrence of human-associated food items, which was only found in significant amounts in the most developed site, varied seasonally (2–25%). Because coyotes in less-developed areas had lower dietary diversity, these coyotes may have to venture into developed areas when there is a decline in the abundance of major prey species for that specific area.

It is well-known that coyotes maintain exclusive territories in which they are least vulnerable to mortality from humans. Studies also indicate that resident (i.e., non-dispersing) coyotes rarely leave these secure areas when unperturbed. Here, we investigated whether coyotes would maintain their fidelity to territories in the face of human pursuit. We used radiotelemetry to determine approximate territorial boundaries of six coyotes, and then attempted, via on-foot pursuit, to drive these individuals in a straight line, causing them to vacate their territories. Coyotes generally did not allow us to push them in a straight line beyond their territories, but instead doubled back as they approached territory boundaries. In contrast, a transient coyote fled from pursuit in a straight line for a distance equivalent to two or three territory diameters before we ceased pursuit. These findings suggest that coyotes perceive the security of their territories to be sufficiently great as to outweigh advantages of outdistancing their pursuer via straight-line flight.

An inventory and monitoring program for species of non-volant small mammals was undertaken in Voyageurs National Park on the Minnesota-Canadian border. Other observations of mammals were made in and around the Park (1967–2004). Here we report species range extensions, most noted since 1982, for four species of rodents present in and/or around the Park and Superior National Forest. They are the white-footed mouse (Peromyscus leucopus), rock vole (Microtus chrotorrhinus), Franklin's ground squirrel (Spermophilus franklinii) and eastern gray squirrel (Sciurus carolinensis). We also report the expansions within the state of the badger (Taxidea taxus) and the raccoon (Procyon lotor), previously known to occur further north in Canada. The results are discussed in the context of inventorying strategies, anthropogenic global climate change and the reservoirs for diseases transmissible to humans.

Because the relationship between lead pellet availability and ingestion by mourning doves (Zenaida macroura) remains uncertain, we conducted an experiment to determine if doves held in captivity freely ingest lead shotgun pellets, investigate the relationship between pellet density and ingestion and monitor physiological impacts of doves ingesting pellets. We conducted two trials of the experiment with <60 doves per trial. We randomly assigned 10 doves to one of six groups per trial; 10, 25, 50, 100, 200 pellets mixed with food and a control group with no pellets. We monitored ingestion by examining x-rays of doves 1-d post-treatment and monitored the effects of lead ingestion by measuring heterophil∶lymphocyte (H∶L) ratios, packed-cell volume (PCV), blood lead, liver lead and kidney lead. Pooled data from both trials showed 6 of 117 (5.1%) doves ingested lead pellets. Two mourning doves ingested multiple lead pellets in each of the treatments containing a mixture of 25, 100 and 200 lead pellets and food. Doves ingesting lead pellets had higher blood lead levels than before treatment (P  =  0.031). Post-treatment H∶L ratios, however, were not different compared to pre-treatment values (P  =  0.109). Although post-treatment PCV decreased for four of six doves ingesting lead pellets, overall they were not lower than their pre-treatment values (P  =  0.344). Liver (P < 0.0001) and kidney (P  =  0.0012) lead levels for doves ingesting pellets were higher than doves without ingested pellets. Our lead pellet ingestion rates were similar to previously reported ingestion rates from hunter-killed doves and our physiological measurements confirm earlier reports of a rapid and acute lead toxicosis. Similar to previous field research, we did not observe a relationship between pellet density in the food and ad libitum pellet ingestion. Although one approach would be to ban lead shot for mourning dove hunting on managed public hunting areas, further research is necessary to ensure that policy development and implementation have a consensus among stakeholders.

Many Great Lakes coastal wetlands have been impounded (diked) to provide protection from flooding and to manipulate water levels for vegetation management. Dikes change the hydrological regime by isolating the coastal wetland from natural lake processes. We evaluated the seed bank composition of seven pairs of diked and undiked coastal wetlands in Green Bay (Lake Michigan) and Saginaw Bay (Lake Huron) to assess the effects of dikes on vegetation dynamics. We also compared soil pH, organic matter, total N, P and K between diked and undiked wetlands. On average, substrates of diked wetlands contained almost three times as many seeds as in undiked wetlands and approximately 20% more species. Seed banks of diked wetlands contained greater numbers of seeds from mudflat aquatic species, whereas seed banks of undiked wetlands yielded greater numbers of seeds from sandy/shoreline species. Diked wetland soils were more acidic, had higher organic matter content, and higher total N, P and K levels compared with soils in undiked wetlands. Our results suggest that diked wetlands may be “traps” for high levels of nutrients, organic matter and seeds. At the sites in this study, construction of dikes appears to have transformed coastal wetlands into systems that function more like inland wetlands.

Invasive plants can have substantial negative impacts on native flora and fauna. As a result, ecological restoration often involves removal of invasive species. We examined the effects of the removal of Hedera helix (English ivy) on regeneration of native vegetation in the Piedmont of Georgia. Ivy was removed by hand or by herbicide from five 5 m × 5 m plots for each treatment and half of each plot was seeded with native seeds. We then counted the number of seedlings present in each plot bimonthly over a 5-mo period. Ivy removal by pulling resulted in the greatest density and diversity of seedlings. Furthermore, these plots exhibited increased seedling density and diversity due to seed addition. Spraying was effective in removal of the ivy but significantly lowered seedling density and diversity and hindered any seed addition efforts. Control plots in which ivy was not removed had no seedlings germinate. Our results suggest that the method of exotic plant removal and the addition of native seed can have profound effects on the regeneration of native vegetation and should be a major consideration for future exotic plant removal projects.

Olympic marmots (Marmota olympus) are reported to skip at least one year between reproductive efforts. We observed several female marmots weaning infants in consecutive years. There was no evidence that reproductive skipping was more common than annual reproduction. High spring food availability resulting from climate change may allow females to wean consecutive litters regularly.

The central mudminnow (Umbra limi Kirtland, 1841) is a small freshwater fish that inhabits slow-moving waters in central North America. In this study, we investigate shoaling behavior in the central mudminnow. In Experiment 1, a focal mudminnow was given the simultaneous choice to spend time near an adjacent aquarium with five conspecifics or near an empty aquarium. Mudminnows (n  =  22) clearly preferred to spend time near the aggregate of conspecifics, suggesting that mudminnows are a shoaling species. In Experiments 2 and 3, a focal mudminnow was given the simultaneous choice to spend time near an aquarium with a small shoal of conspecifics (n  =  3 for both experiments) or near an aquarium with a larger shoal of conspecifics (n  =  7 for Experiment 2; n  =  12 for Experiment 3). In these experiments mudminnows (n  =  22 for Experiments 2 and 3) hovered near both small and large shoals, but did not prefer to spend time near one shoal over the other. Together, these experiments suggest that mudminnow shoaling decisions are not influenced by shoal size, at least under the conditions of these experiments. Our investigation begins to define social behavior in this relatively common but rarely studied species.

Blue-spotted salamanders (Ambystoma laterale) are a widespread and relatively common species throughout northeastern North America. The distribution of this species is marked by a pair of peripherally isolated populations at the southwestern boundary of its range in Iowa, a state where these salamanders are endangered. Because small peripatric populations suffer greater risks of extinction, the genetic state of the isolates was compared to that of a reference population that appears to be in geographic contiguity with the primary distribution of the species. Five polymorphic microsatellite loci were used to examine population genetic structure. Whereas allelic richness exhibited by each locus was qualitatively similar across study populations, genetic data indicate that the scaled effective population sizes of the peripheral isolates were demonstrably smaller compared to that of the reference population. One of the Iowa isolates shows evidence of a recent bottleneck and of substantial inbreeding; this population may therefore be subject to a particularly heightened risk of extirpation.