We studied floral visitors and traits of claret cup cactus (Echinocereus coccineus) at six sites and three elevations along a 1200-m gradient in the Chiricahua Mountains, Arizona. Although elevation affected flowering time, it had little effect on the size distribution or taxonomic classes of visitors. Hummingbirds visited at all sites and relative visit rates of large (7 g) and small (3 g) hummingbirds were similar at each elevation. Males of large species visited more commonly than females, whereas the reverse was true for small hummingbirds. Halictid bees visited commonly at all elevations, but other insects were rare. The survey of visitors revealed no elevation-related patterns that would select for differentiation in floral traits. However, flower length and correlated traits (distance from nectar chamber to anthers and stigma) increased significantly with elevation, with maximum differences of 15–18%, indicating a possible advantage of large flower size at higher elevations. Flower width and nectar chamber size did not vary significantly. Nectar production rate tended to increase with elevation, but differences were nonsignificant and overshadowed by wide variation among plants at a site.

Claret cup flowers are huge and highly rewarding compared to sympatric hummingbird flowers from other plant families. Flower length and width averaged 68 mm and 30 mm, respectively, and daily nectar production averaged 30 mg sugar/flower. This set of traits represents a syndrome rare among hummingbird flowers in which the flower fits the head rather than just the bill, and birds are well rewarded for a risky or inconvenient visit.

As a result of modification/destruction of its floodplain habitats, Boltonia decurrens (Torr. & Gray) Wood now is listed as federally threatened. We investigated dormancy-breaking and germination requirements of achenes, monitored seasonal changes in germination responses of buried achenes and determined if buried achenes have the potential for long term viability. At maturity in autumn, 70–100% of the achenes germinated in light at high (30/15, 35/20 C), but not at low (15/6 C), alternating temperature regimes. Cold stratification during winter increased germination percentages, especially at low temperatures, and by spring 95–100% of the achenes germinated over a range of temperatures, including those (e.g., 15/6 C) that were inhibitory in autumn. Achenes required light for germination, but exposure to light during stratification did not promote germination in darkness at stimulated spring and/or summer habitat temperatures. Nearly 100% of the achenes buried for 88 mo in a nonheated greenhouse were viable. Thus, achenes have the capacity for long-term viability when buried in soil and subjected to seasonal temperature changes; however, soil seed bank studies have not been done at population sites. In both spring and autumn during the 88-mo burial period, exhumed achenes germinated to high percentages at 20/10, 25/15, 30/15 and 35/20 C, but germination at 15/6 C was high in spring and low in autumn. The implication of these results is that achenes could germinate in the field any time from late March to late October if disturbance resulted in their being exposed to light and soil moisture was not limiting.

The annual brome grasses, Bromus japonicus and B. tectorum, are common invaders of the Northern Great Plains. Our objective was to determine if these exotic plants were positively or negatively associated with particular plant species or functional types in a prairie/pine ecotone at Wind Cave National Park, South Dakota. We addressed this issue by sampling at two spatial scales—the landscape scale and the site scale. For the landscape we evaluated species associations across a 3800 ha ecotone using 90 transects. Annual bromes were positively associated with C₃ grasses, particularly Agropyron smithii and Stipa viridula, in addition to the shrub Symphoricarpos occidentalis. Annual bromes were negatively associated with trees, C₄ grasses, particularly A. gerardii, Bouteloua curtipendula and Schizachyrium scoparium, as well as the shrub Rhus aromatica. For the site scale we assessed relationships at a finer resolution within two 1 ha stands of grassland vegetation. Annual bromes were negatively associated with Poa pratensis in both stands. Results indicate that the bromes often dominate in portions of the landscape with C₃ grasses and Symphoricarpos occidentalis. Within individual sites, the bromes appear most limited by a competitive interaction with P. pratensis, resulting from phenological overlap.

Differences between reproductive and nonreproductive plants in terms of germination date, size before competition, size after competition and number of near neighbors (altered by seed predation by ants) of the annual Erodium cicutarium were examined. Seed production was best correlated with the size of the plant relative to its neighbors 5 wk before flowering began. Reproductive plants in the treatment with the highest plant density (initially 900 seeds m⁻² with no seed predation) were 30% larger than nonreproductive plants in that treatment, but 33 to 36% smaller than nonreproductive plants in the low density treatments (initially 400 seeds m⁻², with or without seed predation). Thus, reproductive status was linked to relative size of neighbors rather than absolute plant size. The need for a plant to be large relative to its neighbors before it reproduces could explain why similar numbers of plants reproduced in the populations (=replicate flats) despite very different densities and spatial heterogeneity.

A 31-y-old forest restoration on coal mine spoil was studied to assess soil chemical composition, percent change in tree density and percent similarity of the seed bank and ground-story vegetation. Seed bank and soil chemistry results were compared to an adjacent unvegetated spoil bank. Of fifteen tree species planted on a barren, coal, spoil bank in 1966, four increased in density (indicating recruitment since 1966), seven decreased in density to 50% and four decreased in density >50% after 31 y. Nine volunteer tree species were recorded on the reforested site since the plantings in 1966. March and July soil collections in 1997 contained seeds of 18 species each, 13 in both collections. Most of the species found in the seed bank were shade-intolerant. No significant difference was found for total seed density between the March (4257 ± 490 seeds/m²) and July (3589 ± 365 seeds/m²) seed banks. Percent similarity between the seed bank and ground-story vegetation from the March and July seed collections was 9.8% and 8.2%, respectively. No seeds were found in soil collections from the adjacent unvegetated spoil bank. Soil pH and ionic content between the reforested study area and the unvegetated spoil bank differed significantly. We believe that the 1966 plantings have provided a habitat for other species to establish by reducing erosion and improving edaphic conditions.

The Asian shrub Lonicera maackii (Amur honeysuckle) is now common in many secondary forests in southwestern Ohio and adjacent states. We found lower species richness and abundance in plots (0.5 m²) placed below crowns of L. maackii than in plots placed away: all species (53% lower richness and 63% lower cover), tree seedlings with canopy potential (−41% richness and −68% density) and seed bud bank (−34% richness and −33% density). Moreover, most individual taxa had lower abundance below L. maackii: 86% of herbs, 100% of trees and 56% of seed bud bank taxa. In addition, richness of all species and richness and density of tree seedlings decreased in forests with longer residence time of L. maackii.

The importance of red squirrels (Tamiasciurus hudsonicus) as predators of songbird nests in the boreal forest was determined by monitoring survival of artificial nests before and after squirrel removal. In eight conifer-dominated forest fragments surrounded by agricultural fields we placed artificial ground and shrub nests along the edge and interior of each fragment. Nests contained one quail and one clay egg, which were monitored for 12 d. We then removed squirrels from four forest fragments and repeated the experiment. Nests located in reference plots, where squirrel numbers were not manipulated, had similar survival rates among trials (2 ± 7% change between trials). In contrast, the percentage of quail eggs surviving 12 d increased 32 ± 9% after squirrels were removed. The survival of clay eggs increased 22 ± 10% after squirrel removal, although this was not significantly different from changes in clay egg survival between trials in reference plots (−2 ± 6%). The increase in nest survival when squirrels were removed was consistent for nests on edge and interior transects and ground and shrub nests. Unlike previous studies, we did not observe compensatory predation by other predators, although the relative frequency of mice destroying nests increased after squirrel removal.

Declining Allegheny woodrat (Neotoma magister) populations suggest a need for more basic ecological information about the species for proper management. Within the core of the Allegheny woodrat's distribution in the central Appalachians, food habits and food resource availability are poorly understood. We collected fecal material from known Allegheny woodrats between November 1997 and December 1998 and used microhistological techniques to describe seasonal food habits in the oak (Quercus spp.), dominated forests of the Ridge and Valley and the northern hardwood forests of the Allegheny Plateau physiographic provinces. We examined dietary differences among seasons within and between provinces. Green vegetation, hard mast, soft mast and fungi were present in Allegheny woodrat diets in both provinces in all seasons. Presence of fungi and soft mast in the diet was higher and more widespread seasonally in the Allegheny Plateau than the Ridge and Valley due, in part, to the more mesic forest conditions and more extensive early successional forest habitat in the Allegheny Plateau. Presence of hard mast in the diet tracked acorn production and availability in both provinces in 1997 and 1998. Significant acorn use on the Allegheny Plateau, where oak-dominated forest stands are rare, highlights the importance of hard mast to Allegheny woodrats. Based on food habits we describe, managers seeking to enhance Allegheny woodrat habitat need to provide a mix of habitat conditions containing abundant green vegetation and optimize production and availability of hard mast, soft mast and fungi.

Anthropogenic disturbance near rock outcrops occupied by Allegheny woodrats (Neotoma magister) is suspected of contributing to population declines in the northern and western peripheries of the species' distribution. We compared microhabitat characteristics among clearcut, diameter-limited and intact forest stands in the Allegheny mountains of West Virginia to determine effects of timber removal on Allegheny woodrat habitat selection. We measured microhabitat variables within 0.04-ha plots centered around foraging locations identified using radiotelemetry. Microhabitat variables also were measured in random plots within home ranges. We compared microhabitat variables among harvest methods, and between foraging and random plots within and across methods. Differences in microhabitats among timber harvest methods (F_{3, 14} = 20.31, P ≤ 0.001) were related to overstory characteristics, including total number of trees, canopy cover and overstory diversity. We observed no differences in microhabitat variables between foraging and random plots within any harvest method (F_{3, 82} = 1.20, P = 0.303). Differences between foraging and random plots across all methods (F_{1, 82} = 2.13, P = 0.036) related to understory diversity. Our results suggest that Allegheny woodrats tolerate a wide range of macrohabitat conditions, but habitat selection is primarily determined by microhabitat factors. Microhabitat selection may be related to the high mobility and generalist herbivore diet of this species.

Raccoons (Procyon lotor) are ecological generalists, existing in diverse landscapes. Although general habitat use patterns of raccoons have been extensively described, little research has examined raccoon habitat selection within landscapes managed intensively for wood fiber production. Furthermore, no published studies using radio-telemetry have detailed raccoon habitat selection at multiple spatial scales. We monitored 31 raccoons on a 2000 ha area characterized by short-rotation (<35 y) pine forests in central Mississippi during 1996–1997 and examined seasonal habitat selection at three spatial scales. Habitat selection at the home range scale differed (P = 0.004) between genders. Gender and season interacted to affect habitat selection at the core area scale of selection. Both male and female core areas contained greater proportions of mature hardwood habitats during breeding and young-rearing. Habitat use within home ranges, as determined by point locations, did not differ (P > 0.440) with gender or season. However, raccoons used habitats disproportionately (P = 0.016) relative to habitat composition of the home range. Our findings illustrate the importance of examining individual habitat selection at multiple scales, as raccoon habitat selection in our study varied by scale. Furthermore, our results indicate the importance of hardwood dominated habitats for raccoons existing in pine-dominated landscapes.

We sampled sixty bottomland forest patches in the six southwestern-most counties in Illinois to determine the current status of the cotton mouse (Peromyscus gossypinus). Identification of Peromyscus was based on a modified allozyme electrophoretic technique with the diagnostic GPI-1* locus. Allozymes were isolated from toe-clip samples, rather than liver, and run on a cellulose acetate medium. One hybrid Peromyscus gossypinus-leucopus and one small Peromyscus, carrying a cotton mouse allele at the GPI-1* locus, were identified from 384 individuals screened with this genetic marker. We suggest that cotton mice are an ephemeral species in southern Illinois, disperse into the area occasionally or only during extreme environmental changes and breed with available white-footed mice because of small population size and reduced mate choice.

Several studies support specific status for the prairie form of Sorex cinereus, designated as S. haydeni. Evidence for introgression between these species has been found in Alberta, despite significant sequence differences (>15%) in mitochondrial DNA. Using mtDNA and morphological criteria, we identified to species 94 masked shrews to assess the distribution of these two species in Minnesota and to examine the extent of introgression in zones of sympatry. Only four specimens scored genetically and morphologically as S. haydeni, indicating a restricted distribution. Four specimens had incongruent genetic and morphological identifications, suggesting introgression between the two species.

We used artificial nests modeled after those of Trachemys scripta, a wide-ranging freshwater turtle species, to evaluate the effects of deer feeders, habitat type and visual and olfactory cues on nest predation in northern Florida. Nests were placed at lake sites with and without deer feeders, and in three habitat types: road, edge and forest. Overall nest mortality due to predators was high (89%). Nest survival was 5.5 times higher at lakes without deer feeders than at those with feeders. Among habitat types, survival was highest at road nests (23%), while survival rates at forest (4%) and edge nests (6%) were lower than that at road nests. No significant difference in survival was detected with respect to the presence of visual or olfactory cues. Our results suggest that deer feeders reduce recruitment in freshwater turtle populations, and that generalizations regarding the negative impacts of roads should be made cautiously, in a taxon and site-specific fashion.

We conducted the first formal quantitative study of the herpetofauna of the Jasper-Pulaski, Willow Slough and LaSalle Fish and Wildlife areas (FWA) in northwest Indiana from 1994–1996 to gather baseline data necessary to determine distribution and status of species and to monitor long-term population trends. We compared our results with those of earlier collectors Chapman Grant, Paul Swanson and Sherman Minton. A total of 339 populations of 13 amphibian species and 78 populations of 22 reptile species were encountered from 1994–1996. Only 9 species were found at all three FWAs and only 11 were commonly encountered. Sixteen new site records and new county records were made while at least 7 species with validated records before 1972 were not found during this survey and may be locally extirpated. The relative abundance of species at Jasper-Pulaski FWA has changed between the 1930s and 1990s with declines of the thirteen most abundant species and increases in some formerly overlooked or rare species. Large-scale changes in habitat may be partly responsible for changes in species abundance. All of the state-listed species (Acris crepitans blanchardi, Rana blairi, R. pipiens, Thamnophis p. proximus, Clemmys guttata, Emydoidea blandingii, Kinosternon s. subrubrum, Terrapene o. ornata, Opheodrys vernalis and Sistrurus c. catenatus), except Ambystoma laterale, have clearly declined.

Toads of the genus Bufo co-occur with true frogs (family Ranidae) throughout their North American ranges. Yet, Bufo are rarely reported as prey for ranid frogs, perhaps due to dermal toxins that afford them protection from some predators. We report field observations from four different localities demonstrating that Oregon spotted frogs (Rana pretiosa) readily consume juvenile western toads (Bufo boreas) at breeding sites in Oregon. Unpalatability thought to deter predators of selected taxa and feeding mode may not protect juvenile stages of western toads from adult Oregon spotted frogs. Activity of juvenile western toads can elicit ambush behavior by Oregon spotted frog adults. Our review of published literature suggests that regular consumption of toadlets sets Oregon spotted frogs apart from most North American ranid frogs. Importance of the trophic context of juvenile western toads as a seasonally important resource to Oregon spotted frogs needs critical investigation.

Hosts for Strophitus undulatus (Bivalvia: Unionidae) were identified through laboratory infestations. Strophitus undulatus had a low degree of host specificity, transforming on 15 of 22 species examined, including three non-native species and one anadromous species. Suitable hosts included five cyprinid species, two salmonids, two centrarchids, two percids and Acipenser oxyrhynchus (Atlantic sturgeon), Ameiurus natalis (yellow bullhead), Cottus cognatus (slimy sculpin) and Notophthalmus viridescens viridescens (red-spotted newt). Ten hosts were previously unknown. No metamorphosis was observed in a control treatment without a potential host. Recovery of juveniles occurred 12–41 d after infestation at 13–18 C. Duration of glocidial attachment to the hosts declined with increasing water temperature.

Spine length, pelvic girdle morphology and body form vary markedly among populations of the brook stickleback (Culaea inconstans). Predation may drive the evolution of well developed defensive structures because these traits increase handling times and allow captured prey to escape from small predators. Previous studies indicate that C. inconstans from Macochee Creek (Champaign Co., Ohio), a tributary of the Mad River, appear to have spine lengths that approach the extremes for the species. However, it is unclear whether long spines are common throughout the Mad River drainage or are limited to this single tributary. Morphometric analysis revealed that C. inconstans from the Mad River drainage have longer spines than other populations in its southeastern range and that long spines are consistently found throughout the Mad River drainage. The selection pressures responsible for long spines in C. inconstans from the Mad River drainage are presently not clear.

Although it is known that monarch butterflies (Danaus plexippus) in North America east of the Rocky Mountains migrate each fall to overwintering sites in central Mexico, the migratory routes monarchs take have only been indirectly studied. Over three fall migration periods (1998–2000) we captured monarchs on the eastern shore of Virginia in an attempt to recover monarchs initially tagged farther north. Of 2190 monarch captures, 6 were previously tagged in other locations, allowing possible migration routes and rates of travel in varying wind conditions to be inferred. Monarchs reached our site fastest when they migrated with strong northerly winds. Only one monarch tagged at our study site was recovered at the overwintering site in Mexico. Compared to recovery rates from monarchs tagged in other locations in North America, this suggests that monarchs at our site are less likely to reach the Mexican overwintering site in certain years. We also report discovery of an annual monarch accumulation area on the extreme southern tip of the eastern shore of Virginia on the Delmarva Peninsula.

Puddling in female butterflies is rare. When it has been reported in Papilio females they were alone and usually worn, unlike males which cluster and are usually recently eclosed. A fresh female of P. canadensis was found puddling on fresh mammal dung with a group of nine P. canadensis males. To my knowledge, such an observation has never been reported for any species of the swallowtail butterflies (Papilionidae).

Using museum records, we examined sex ratios for 32 collections of eastern red bats (Lasiurus borealis) in the eastern United States. Results from logistic regression analyses indicate that female-dominated sex ratios are associated with warmer, mean high temperatures in June. Male-dominated collections, and those with approximately even sex ratios, occurred disproportionately in the Appalachian Highlands where monthly mean high temperatures were lower than in either the Central Lowlands-Interior Low Plateaus or Coastal Plain-Piedmont regions. Efforts to maximize conservation of summer roosting and foraging habitat that favors the female population segment of eastern red bats should be directed at areas where June mean high temperatures exceed 28.5 C.

We observed formation of female flowers and seedpods, subsequent to the production of male flowers and pollen, on branch cuttings from a male Populus deltoides var. wislizenii tree. Labile sex expression is rare in the genus Populus, and this is the only recorded instance for this variety of P. deltoides. Mature trees exhibited the same phenomenon in the field and some released copious amounts of seed during 1997. These observations may explain apparent sexual lability within four populations of P. deltoides along the Rio Grande. If protandrous hermaphrodites exist in this species, observations of sexual lability in these field populations could be explained by year to year variation in the timing of census observations; i.e., earlier season observations would score these individuals as males. We conclude that P. deltoides var. wislizenii is a subdioecious species with occasional bisexual individual plants.