In the eastern United States, the mimetic viceroy butterfly (Limenitis archippus) exhibits clinal variation in wing color, ranging from a tawny orange phenotype (L. a. archippus) in the N to a dark mahogany phenotype (L. a. floridensis) in Florida. Geographic distributions of these two subspecies are roughly coincident with the ranges of the viceroy's two eastern mimicry models: the monarch (Danaus plexippus) in the N, and the queen (D. gilippus) in the S. This coincidence has historically been attributed to “model-switching”: presumably, southern viceroys have switched from mimicking the monarch to mimicking the locally predominant queen, due to selective pressure exerted by visually foraging predators. As an initial test of this hypothesis, I sought evidence of selective predation on light and dark viceroys by captive red-winged blackbirds previously exposed to either monarchs or queens. Results were consistent with the model-switching hypothesis: queen-conditioned birds preferentially avoided the dark, queen-like L. a. floridensis, whereas birds exposed to monarchs avoided (to a lesser degree) the light L. a. archippus phenotype. I propose that this differential predation, while perhaps asymmetrical, demonstrates the selective mechanism responsible for the evolution of regional viceroy races, and that geographic model-switching explains the large-scale modern-day correlation between Danaus biogeography and viceroy wing color.

We quantified the frequency and seasonal variation of ant-plant associations mediated by plant-derived food sources, and the intensity of their use by ants. Thirteen ant species and 42 plant species were related in 135 pairs of associations. The plant families Leguminosae and the Cactaceae were those most visited by ants. The ants Camponotus rubrithorax (28 species) and Crematogaster opaca (16 species) used the most plant species as food sources. No obligatory ant-plant associations were found. Ant species studied did not use all food resources available to them, resulting in a significant differential choice of food resources; using nectar from plant reproductive structures and floral nectar more and extrafloral nectar less. Ant-plant asociations were more frequent during the warm, humid months, when weather conditions are milder, most plants are producing new vegetative growth, many are in flower or fruit and other insects increase their activities. Thus, ants can diversify their foraging activities. Ants are less active during the coldest and driest time of the year. The generalized linear model fitted to the ant-plant interactions curve explained 80% of the variation. This variation is explained by temperature alone; precipitation and the interaction of the two factors were not significant.

The sedentary terrestrial larvae of the tiger beetle, Cicindela togata, inhabit areas that are often flooded for days or weeks. We tested the ability of these larvae to survive immersion and anoxia. Maximum survival time of immersed, anoxic C. togata was 6 days at 25 C. Time to 50% mortality (LT₅₀) in these conditions was 85.9 ± 23.5 h. Survival times were more than eight times longer than those of similarly treated larvae of Tenebrio molitor (LT₅₀ 10.1 ± 3.2 h). Similar or somewhat longer survival times were observed in larvae of C. togata exposed to an anoxic nitrogen atmosphere (LT₅₀ 102 ± 31 h) and of T. molitor (LT₅₀14.4 ± 6.5 h). At 10 C, LT₅₀ of C. togata in anoxic atmospheres exceeded 10 days. Tiger beetle larvae are physiologically capable of surviving several days of inundation during floods without mechanisms to prevent burrow flooding. By entering a quiescent state, C. togata larvae survive much longer periods of anoxia than has been previously reported for terrestrial insect larvae.

The Old World species, Drosophila subobscura, has recently invaded North America and become sympatric with the native obscura-group species. This study investigates the summer diurnal activity patterns of two northwestern North American populations of Drosophila subobscura and compares them to the documented Old World trend as well as that of its native North American congener, D. pseudoobscura. Both species peak in activity at temperatures between 15 and 20 C, and activity in both species increases with decreasing sun angles, thus causing high levels of activity near sunrise and sunset. However, D. subobscura becomes more abundant than D. pseudoobscura at lower temperatures and sun angles. No differences in diurnal activity were noted between Old World and New World D. subobscura. Drosophila subobscura might have contributed to the reduced abundance of a native species, D. persimilis, in areas of recent sympatry, though this causal connection is not certain.

The purpose of this study was to interpret patterns of branching in natural populations of Verbascum thapsus L. (Scrophulariaceae) and to investigate whether the suppression of lateral meristems by apical dominance imposes a potential fitness cost in this species. Disruption of apical dominance by removal (clipping) of shoot apices resulted in shorter plants and greater branching but did not affect total biomass, number of fruits, number of seeds per plant or percent germination of seeds. Seeds from clipped plants, however, were approximately 30% heavier than those from unclipped plants, indicating that apical dominance imposes a potential fitness cost to V. thapsus. Plants with shoot apices intact were mostly unbranched but were more likely to be branched if they were taller. The fruits on the main stalks of these taller and more branched plants were significantly more heavily infested by a curculionid weevil (Gymnaetron tetrum Fab.). Lateral branches of these naturally branched plants, however, were less infested than the main stalks. These results indicate that the weevils are attracted to larger plants and that weevil damage may induce branching. An alternative hypothesis could not be discounted, i.e., that larger plants are more branched because they have greater resource availability, thus allowing release of lateral meristems from inhibition by apical dominance.

The surface cover of three species of small, free-floating, aquatic plants in a beaver swamp was monitored for 1 yr. Simultaneous experiments at the same site quantified competitive relationships among these species in different seasons. In the summer competition experiment, Salvinia minima grew rapidly and had negative effects on the relative change in cover of both Azolla caroliniana and Spirodela punctata. Relative change in cover of S. minima increased when it was grown with S. punctata, but only when A. caroliniana was not present. A negative correlation in the swamp between cover of S. minima and S. punctata, the two most abundant species during the summer, is consistent with the strong competitive effects of S. minima in the summer competition experiment. During the autumn competition experiment, S. minima had a negative effect on A. caroliniana's relative increase in cover. This effect, however, was not competitive and was caused by an herbivore of S. minima that switched to A. caroliniana as S. minima's growth slowed in the autumn. Total cover of small, floating, aquatic plants was lowest in the autumn. Salvinia minima, the best competitor for surface space during the summer experiment, is also the largest species, the most difficult to sink and the most buoyant once submerged.

Competition has an important, but limited, role in this community. The effects of competition on surface area were seasonal, and the striking gains in cover of Salvinia minima during the summer were reversed by its relative intolerance of winter conditions and its higher reduction in cover during floods. Apart from late summer, cover of S. punctata was by far the highest throughout the study period, despite S. minima's competitive ability.

Shrub thicket (Myrica cerifera) gaps were sampled on a Virginia barrier island to elucidate successional mechanisms in the development of maritime forests. Microclimate, edaphic characteristics and spatial heterogeneity within gaps, as well as within intact thicket understories, were compared for young and aging thickets. Midday photosynthetic photon flux density in gaps ranged from 5–1000 µmol m⁻² s⁻¹, but at the gap center photon flux density was up to 15% higher for the aging thicket. Soil temperatures at the surface were 9 C warmer in aging thicket gaps compared to young gaps, and 6 to 14 C warmer when comparing gaps to the understories for the young and aging thickets, respectively. Edaphic characteristics differed little with thicket age; however, gaps in the aging thicket had greater structural diversity. Species richness was 2.5 to 3.7 times greater within gaps than in intact thicket understories, with highest richness within aging thicket gaps. Relative to intact shrub thickets, gaps enhance environmental variability through greater structural diversity. Gaps may facilitate the establishment of later seral species.

Due to the extensive removal of the forest cover of the southeastern piedmont during the 19th century and to the lack of systematic presettlement records for most of this region, there has been little basis for relating the piedmont's maturing postsettlement secondary forest to its pre-European condition. This study compares species composition patterns between pre-European and present periods for a portion of the Georgia piedmont that had a systematic presettlement land survey. Detrended correspondence analysis of presettlement, immature postsettlement, and mature postsettlement forests identifies a primary gradient that distinguishes between the vegetation of these periods on the basis of habitat moisture preferences and fire tolerance of species. A secondary gradient emphasizes life history characteristics of species that typically differentiate immature and mature postsettlement forests; presettlement forests were not dominated by the late successional species typical of mature postsettlement forests, but had abundant disturbance-favored taxa. Changes in the abundance of individual species from presettlement to mature postsettlement forests occurred across all habitat types. The xerophytic, fire-tolerant taxa that dominated presettlement forests (e.g., Pinus species, Quercus stellata, Q. velutina) are less important in mature postsettlement forests. Dominants of mature postsettlement forests (e.g., Q. alba, Liriodendron tulipifera, Carya species) are primarily mesophytic, fire-intolerant species that were far less prominent in presettlement forests. Marked contrasts in composition between presettlement and mature postsettlement forests of the piedmont have been produced by changes in the prevalent disturbance regime from fire-dominated dynamics to gap-phase processes.

We employed a bioassay using soybean seeds and native prairie bush clover (Lespedeza capitata) seeds to demonstrate that legumes may be rare throughout secondary succession in nitrogen-poor grasslands due to a lack of suitable rhizobia and consequently lower growth rates.

The Poor Fork of the Cumberland River has been interpreted as an area of secondary contact and hybridization between Etheostoma nigrum and the closely related E. susanae. I examined mitochondrial DNA (mtDNA) diversity among E. nigrum populations in the adjacent Kentucky River drainage, E. susanae populations in the Cumberland River drainage, and the population of E. nigrum occurring in the putative hybrid zone of the Poor Fork. Fifteen restriction enzymes revealed the presence of four mtDNA haplotypes, two each in the Kentucky and Cumberland rivers. Etheostoma nigrum from the Kentucky River had two haplotypes (K1 and K2), which differed by a single restriction site. All E. susanae from the Cumberland River had a distinctive haplotype (S1), while E. nigrum from the upper Poor Fork had a unique haplotype (CN1). Johnny darters from the upper Poor Fork of the Cumberland River were phenotypically and genetically more similar to E. nigrum populations in tributaries of the upper Kentucky River than to E. susanae, and were probably introduced by stream capture. Implications of this distribution pattern include: (1) that more than one event or mechanism was responsible for the isolation of fishes in the upper Cumberland River, and (2) that conservation efforts for E. susanae should focus on habitats occurring immediately above Cumberland Falls rather than in the contact zone in the Poor Fork.

We observed wild banana slugs (Ariolimax columbianus) eating fruits of several Pacific Northwest plant species. Slime trails and direct observations indicated that slugs are capable of reaching the fruits of many wild plants. This motivated us to test the hypothesis that slugs may act as seed dispersers, provided that they defecate viable seeds. We fed captive slugs the fruits of Rubus spectabilis, R. discolor, Vaccinium ovatum, V. parvifolium, Gaultheria shallon and Disporum smithii to determine the effects of slug ingestion on seed germination. At least some seeds of each species germinated after the fruits were consumed by the slugs, but the effects on germination were species-specific. Seeds of Rubus spectabilis were less likely to germinate after passage through the guts of slugs, and we found significant evidence that the two fruit color morphs reacted differently over time. Disporum smithii seeds did not statistically differ in germination behavior between treatments, although the trend suggested possible germination enhancement following rasping of the seeds by slugs. All other species of seeds tested germinated following consumption by slugs, but results could not be tested statistically. Gut passage times of R. discolor seeds were determined (x̄ = 25.4 h, se = 3.6 h). We conclude that despite the short distances slugs are likely to disperse seeds, their generalist habits and ubiquity suggest that they may have complex and ecologically significant effects on seed dispersal in Pacific Northwest forests.

To better understand the role of prairie landscape configuration on the population dynamics and conservation of greater prairie-chickens, we compared population trends and breeding ecology of prairie-chickens in prairie mosaic and contiguous prairie landscapes in southwestern Missouri. Over 27 yr, the contiguous prairie landscape supported a stable population, whereas the prairie-chicken population in the prairie mosaic landscape declined. In 1986 and 1987, less than one-third of greater prairie-chicken nests occurred in native prairie habitats in the prairie mosaic landscape, but 90% of nests in the contiguous prairie area were in native prairie. Greater prairie-chicken nests established in agricultural habitats had substantially lower nest success than nests in native prairie or mixed native-exotic grass pastures. Nest success was higher in the contiguous prairie landscape than in the prairie mosaic area in 1 of 2 yr. Habitat use by broods differed between the areas: in the contiguous prairie landscape we detected females with broods most often in native prairie, but brood females in the prairie mosaic were most often detected in agricultural habitats. Females with broods exhibited greater daily movement and had larger home ranges in the prairie mosaic than in the contiguous prairie landscape. In Missouri landscapes with ≤15% prairie composition, contiguous tracts of prairie, at least 65 ha, offer greater potential for greater prairie-chicken conservation, than smaller, scattered prairie tracts.

Recently, possible biological effects of electromagnetic fields (EMFs) produced by high-voltage transmission lines have come under intense scrutiny, mostly with respect to human health. Because little has been done to assess possible effects of EMFs on the biology of free-ranging animals living within such fields, this study monitored the breeding biology of birds using nest boxes placed under transmission lines and in reference areas. Tree swallows (Iridoprocne bicolor) had significantly lower reproductive success (number fledged/clutch size) under the powerlines than in the reference sites. No such effect was apparent for house wrens (Troglodytes aedon). These results suggest that habitats under high-voltage lines may be suboptimal for tree swallows.

Coevolutionary models of the interactions between fruiting plants and avian seed dispersers have been influenced by the assumption that regurgitation and defecation of seeds have diffferent effects on seed coats, and consequently seed germination. We evaluated how the manner of seed processing affects seed germination by feeding fruits of three bird-dispersed shrubs, spicebush (Lindera benzoin), chokecherry (Prunus virginiana), and arrowwood (Viburnum dentatum), to captive cedar waxwings (Bombycilla cedrorum) and thrushes (Turdus migratorius, Hylocichla mustelina, Catharus guttatus and C. minimus). Cedar waxwings defecate all seeds, whereas thrushes regurgitate most seeds. For all three shrub species and all five bird species, there were no differences in germination success between seeds manually cleaned of pulp, and cleaned, bird-passed seeds, regardless of whether seeds were regurgitated or defecated. However, seeds of Lindera and Prunus that were defecated by cedar waxwings and planted with feces, mimicking the depositional environment of defecated seeds in nature, suffered reduced germination relative to cleaned seeds. Thus, our results do not suggest that whether seeds are regurgitated or defecated is an important component of dispersal quality because of direct effects on the seed coat. Instead, seed processing modes differed in their effects on seed germination because of the feces associated with defecated seeds. Removal of fruit pulp from seeds by frugivores, a fundamental consequence of dispersal in animal guts, was critical for germination, especially for seeds within the lipid-rich fruits of Lindera and Viburnum. This suggests that for some fruiting plants, frugivores provide an essential service by freeing seeds from fruit pulp, in addition to their role in seed dispersal.

We assessed shorebird use of artificial wetlands within the Mississippi Alluvial Valley during the winters of 1991–1992 and 1992–1993 and during the autumn of 1994. On agricultural fields managed to provide habitat for waterfowl from November to March, mean shorebird density was 58.6 birds/100 ha, but shorebird densities were greater on soybean fields than on rice or moist-soil fields. Killdeer (Charadrius vociferus) and common snipe (Gallinago gallinago) were common throughout winter, but shorebird abundance and species richness along survey routes increased from November through April. During the late summer and autumn, wetlands on public lands in the Mississippi Alluvial Valley are managed by the U.S. Fish and Wildlife Service specifically to provide foraging habitat for shorebirds. From August through October 1994, we observed 14,564 individual shorebirds of 22 species using these anthropogenic wetlands. Mean shorebird density on wetlands managed by flooding previously dry, disked fields was 695 birds/100 ha, whereas mean density on wetlands managed by drawing down water reservoirs was 1224 birds/100 ha. We recommend increased shallow-water flooding of agricultural fields, particularly soybean fields, during winter to provide habitat for wintering and early spring migrant shorebirds. More importantly, we recommend continued water management on public wetlands from July through October, preferably by drawing down water reservoirs, to provide foraging habitat for southward migrating shorebirds.

Population trends for seven genera of rodents were monitored by snap-trap capture rates on a study area (700 km²) in Webb County, Texas, during winter from 1976 to 1986. Population irruptions occurred in 1982 (46-fold) and 1986 (13-fold) for Sigmodon, and in 1982 (17-fold) for Reithrodontomys. Annual variability in abundance was positively correlated between the following pairs of genera: Sigmodon-Reithrodontomys, Peromyscus-Onychomys, Onychomys-Reithrodontomys, and Dipodomys-Chaetodipus. Annual abundance of Sigmodon in winter was positively correlated with rainfall in the prior growing season. Analysis of the distribution of the composite rodent community among five vegetative habitats revealed greater abundance of rodents in the most dense and diverse shrub community (high-density shrub/mixed-grass habitat).

Spatial data were collected in a population of individually marked nine-banded armadillos (Dasypus novemcinctus) at a site in Florida from 1992–1995. Approximately ⅓ of the animals (118/313) were sighted more than once. Data on distances moved between successive sightings showed that armadillos moved <200 m both within and between years. However, animals moved significantly farther between than within years. The average distance between sightings was significantly longer for adults than for juveniles within, but not between, years. There were no sex differences in distances moved within or between years for either adults or juveniles. Analyses of distances to nearest neighbors showed that adults were closer to one another than to juveniles and that juveniles of the same sex were closer to one another than to juveniles of the opposite sex. This latter result probably reflects maintenance of proximity among littermates.

White-footed mice (Peromyscus leucopus) and deer mice (P. maniculatus) were studied under both allotopic and syntopic conditions in northeastern Kansas. These species typically occupy different habitats with the slightly larger P. leucopus inhabiting wooded areas, and the smaller P. maniculatus inhabiting grasslands. We recorded external measurements of each species in distinct habitats where only one species was typically present. Results were compared to measurements obtained from a grassland area recently invaded by woody plants where both species occur in syntopy. Under allotopic conditions, the two species were morphologically distinct and individuals could be readily assigned to one of the two species using a combination of measurements. Under syntopic conditions, however, many individual Peromyscus were intermediate in external measurements when compared to the allotopic populations, and could not easily be classified into one of the two species. Morphologically intermediate animals in the syntopic population could be the result of smaller than average adult P. leucopus occupying the area. The early successional habitat of the syntopic population is not typical for P. leucopus, but the recent invasion of woody plants may provide adequate habitat for smaller individuals, which may have been pushed out of the surrounding woodlands by larger individuals.

Baylisascaris procyonis, the common large roundworm of raccoons (Procyon lotor), causes clinical neurologic disease in many species of mammals and birds. Infective eggs of B. procyonis are present at raccoon latrine sites, and these sites may be important in the transmission of this parasite to syntopic small vertebrates in forested areas. We located raccoon latrines in forested sites in Indiana, sampled soil and fecal material from these locations, and examined these samples for the presence of Baylisascaris procyonis eggs. We also quantified the structural characteristics of raccoon latrines in wooded areas, compared their characteristics with randomly located sites, and classified sites based on structural features using stepwise discriminant function analysis. B. procyonis eggs were present at 14% of the raccoon latrines sampled. Latrine sites differed from randomly located sites and exhibited characteristics generally associated with treefall gaps. Most latrines were located either on logs (49%) or at the base of large trees (37%). Structural features surrounding latrines often are important travel routes or foraging areas for various small vertebrates. The visitation of mammals and birds to sites exhibiting these structural features may result in infection with B. procyonis. In this way, Baylisascaris procyonis could have long-term impacts on populations of native mammals and birds.

Five hundred seventy-two Sceloporus occidentalis from Los Angeles County, California, were examined for helminths: 200 from the San Gabriel Mountains (montane, ca. 1584 m elevation) and 372 from the Puente Hills (lowland, ca. 150 m elevation). The component helminth community of the montane fence lizard population consisted of four species: two species of cestodes, Mesocestoides sp. and Oochoristica scelopori and two species of nematodes, Physaloptera retusa and Spauligodon giganticus. The component helminth community of the lowland fence lizard population contained three species: Mesocestoides sp., Physaloptera retusa and unidentified nematode acuariid larvae. The two helminth communities are dissimilar in composition.

Observations and experiments were conducted on natural mortality of fish at a small Michigan lake during spring. During an extensive fish die-off in 1989, only 22% (202) of the dead fish were found even though an estimated 10% (918) of the fish population >150 mm TL died. Scavengers, principally yellow bullheads (Ameiurus natalis) and turtles feeding in the littoral zone, claimed approximately 60% of the dead fish that year and 82% in another year with a smaller die-off. The remainder decomposed in 6–34 days. The long-standing mystery of why so few large dead fish are noticed in lakes, despite high annual natural mortality rates of abundant fish populations, was adequately explained by unseen decomposition in deep water and the ability of resident scavengers to keep up with the supply of dead fish except during unusually concentrated fish die-offs.