In many semiarid regions woodlands, savannas and grasslands form an unstable landscape mosaic; the physiognomy of a particular patch in the mosaic changes over time. To explain such temporal and spatial variation in tree abundance in semiarid landscapes, greater understanding of woodland tree population dynamics and factors that affect population dynamics of trees in woodlands is required. We examined the apparent failure of adult recruitment in Quercus buckleyi (Spanish oak) populations in woodlands on the Edwards Plateau, Texas. To verify adult recruitment failure, we quantified age structures of adult stems (stems taller than 150 cm) in five Q. buckleyi stands. We found that adult recruitment has been low to absent for 35 to 60 y in four of our five sites. At the fifth site, some stems recruited above 150 cm (the browseline) in the past 30 y. Most recruitment of adult stems occurred between 1900 and 1935, a period of low deer abundance. At the one site with recent adult stem recruitment, relatively low deer densities have been maintained since 1970 by hunting and by a perimeter fence that prevents immigration. Based on coincidence of adult stem recruitment with low deer populations, we hypothesize that intense browsing pressure or the interaction between fire suppression and intense browsing pressure is limiting adult recruitment of Q. buckleyi. If the documented patterns persist, many Q. buckleyi stands on the eastern Edwards Plateau are unlikely to replace themselves.

Assessments of ecosystem restorations are necessary to improve restoration practices and goals. Restoration assessments, whether quantitative or qualitative, are also a vital part of managing previously degraded ecosystems. This study examined some of the key structural and functional characteristics and processes of a tallgrass prairie restoration near Arlington, Wisconsin for 5 y, 19 to 24 y after beginning restoration from cultivation, including mean annual drainage, N and C leaching, soil organic matter, pH, extractable P and K, total N and C contents, above- and belowground net primary production, leaf area index, soil surface CO₂ flux and net N-mineralization. Total soil N and C contents of the prairie restoration were compared to other nearby prairie restorations, remnants and an adjacent agricultural field, all on similar soil, to determine the degree of change in ecosystem properties as a result of ecological restoration. Soil properties and processes and vegetation characteristics varied annually throughout the 5-y assessment period, but most soil properties showed no significant temporal trend. Only soil N content in the 0–30 cm layer increased significantly in the 5-y period, but the rate of N increase did not coincide with the rate typical of N inputs to a prairie. Results suggest that most soil properties have either already come to some equilibrium with the surrounding environment or their rates of change were too small to measure over 5 y. This study demonstrates the difficulties of ascribing changes in ecosystem properties to restoration. The spatial and temporal variability and slow rates of change make it difficult to discern differences between restored, disturbed and natural ecosystems.

Two alternative hypotheses may account for the low fruit:flower ratios regularly observed in many species of plants. First, a low fruit:flower ratio may be a passive “nonadaptive” consequence of external stressors that destroy flowers or limit their pollination. Alternatively, a low fruit:flower ratio may be a result of adaptations that enhance plant fitness, perhaps by allowing the plants to compensate for the very stressors that limit fruit set. We tested these nonadaptive and adaptive hypotheses using Solanum carolinense (horsenettle), an herbaceous plant that generally exhibits a low fruit:flower ratio. Horsenettle often suffers substantial flower loss to herbivores and fruit set can be pollen limited. We present experimental evidence that neither floral destruction by herbivores nor pollen limitation directly affected fruit:flower ratio. Instead, the low fruit:flower ratio was an adaptive consequence of the production of “surplus” flowers. These flowers served as reserve ovaries that enabled the plants to compensate fully for both floral herbivory and incomplete pollination.

This study examined variation in the total number of seed, the number and percent of viable seed produced per catkin in red alder (Alnus rubra) across the following five levels of plant organization: stands, individual trees within stands, branches within trees, catkin clusters within branches and catkins within catkin clusters. Catkins were collected from four stands and seed germinated in the laboratory. When all stands were analyzed together all levels of organization, except branches, accounted for significant variation in all response variables. High elevation stands had significantly higher viable seed production per catkin than low elevation stands. When stands were analyzed separately, branches, but not individual trees, generally accounted for signficant variation in seed production. Although branches within trees can generally be considered autonomous units, a large proportion of variation in production of viable seed occurred both within branches and catkin clusters suggesting that important variation can occur within branches.

Euthamia graminifolia and Solidago canadensis may be found growing in dense, essentially monospecific, stands. One hypothesis to explain this phenomenon is that both of these goldenrods, as do many related taxa in Asteraceae, release allelochemicals into the surrounding environment. To test this hypothesis, leaf and root/rhizome tissues of both species were screened for allelopathic activity. Aqueous extracts (50 g dried tissue in 2000 ml distilled water) were prepared for each tissue type for each species and a series of dilutions ranging from 0% to 100% of each extract was used for testing radish and lettuce seed germination and root growth. Leaf extracts from both species significantly inhibited seed germination and root growth. Root/rhizome extracts significantly inhibited root growth, but had no effect on seed germination. These results suggest that allelopathy may inhibit the growth and survival of competing species. In addition, both taxa may have a negative effect on yields of nearby crop species.

The relationship between seed weight, dispersal ability and seed “quality” was tested for the seeds of pale swallow-wort Vincetoxicum rossicum (Kleo.) Barb., an invasive alien vine. In a field release, small seeds traveled significantly farther than large seeds. In one of two greenhouse experiments, large seeds were significantly more likely to germinate than small seeds. Seed weight also had a significant effect on seedling size at the time of harvest, but only when the seeds were grown in competition with a mixture of grasses. Seeds containing more than one embryo produced greater total seedling weight in the absence of competition, but polyembryony was not advantageous when seedlings were grown with grasses. We argue that if long-distance dispersers are less likely to germinate or survive, measured seed dispersal curves could lead to overestimation of the rate of spread of invasive plants.

Gray squirrels (Sciurus carolinensis) are an important component of forest ecosystems in the eastern United States, but their ecology in subtropical habitats remains unstudied. This study was conducted on a population of gray squirrels in a 10 ha forest composed of hammock and upland pine habitats. The goals were to determine microhabitat selection, diet and seasonal activity patterns of squirrels in these two distinct habitats. Weekly walking surveys of the study site were conducted for 1 y. Gray squirrels showed a significant peak in activity during the fall (P < 0.001). Acorns were the primary food item and pine seeds (Pinus palustris, P. taeda) were of secondary importance. Use-availability analysis revealed that squirrels were observed significantly less often than expected in some tree species and significantly more often than expected in laurel oak (Quercus hemisphaerica), loblolly pine (P. taeda) and longleaf pine (P. palustris).

Previous studies have shown that bat activity is greater along forest-clearcut edges than in the center of clearcuts or in the forest interior. Residual patches of trees in logged areas may also provide habitat for bats. To investigate this, we monitored bat activity at three locations within cutblocks: along the outside edge of the forest cutblock, in the center of the clearcut portion of the cutblock and along the outside edge of the residual patches of trees, at the EMEND (Ecosystem Management by Emulating Natural Disturbance) study site in northern Alberta, during the summer of 2000. Our results indicate that small maneuverable species such as Myotis lucifugus and M. septentrionalis were equally active along the edge of residual patches and the forest edge of cutblocks and least active in the center of cutblocks. Larger species, such as Lasionycteris noctivagans, showed no preference. Thus, patches of residual trees provide commuting habitat, and potentially foraging habitat, for bats.

Little is known regarding the influence of sex- and age-specific factors on mineral lick use by white-tailed deer (Odocoileus virginianus). We recorded 1020 visits to mineral licks by deer between April and September, of which we could classify 922 by sex and age (yearling or adult). We observed a relationship between the frequency of mineral lick visitation and both sex and age of users. No relationship was found between visitation frequency and time interval of use. Adult females used licks more frequently than other sex/age classes. Visitation frequencies were similar for adult males and yearling males and females. We detected no relationship between lick sodium (Na) content and sex or age of visitors. Greater lick use by adult females may have reflected an increased need for Na associated with gestation and lactation. The majority of male visitors were yearlings, which require Na for antler production and rapid somatic growth. In areas where Na concentration in forage is below dietary requirements and Na-conserving mechanisms are compromised and/or additional production demands occur, available mineral licks may ameliorate the potential negative impact of limited Na on pregnancy/lactation.

We monitored 131 (99 male, 32 female) radiocollared raccoons (Procyon lotor) from January 1991 to December 1997 on the Tallahala Wildlife Management Area, Mississippi. We examined inter- and intrasexual spatial relationships and temporal interactions among adults. Adult males frequently maintained overlapping home ranges and core use areas and some males maintained spatial groups that overlapped minimally with adjacent groups or solitary males, suggesting territoriality among groups. Males arranged in spatial groups were often significantly positively associated with each other; however, we observed instances of males who remained solitary and maintained exclusive home ranges and core areas. Adult females maintained exclusive home ranges and core areas during winter, but several females shared home ranges during other seasons. However, these females did not forage or den together and were significantly negatively associated with each other within shared areas, indicating that movements by these individuals were independent. Home range and core area overlap differed among seasons for males and females with neighboring or overlapping home ranges. Our findings suggest that social behavior in raccoons varies within the same protion of the landscape, ranging from solitary individuals to male social groups.

Long-distance dispersal is an important phenomenon in both theoretical and applied population biology, yet empirical data are difficult to collect. We analyzed long-distance movement data for cotton rats (Sigmodon hispidus) and prairie voles (Microtus ochrogaster) for sex and seasonal biases by use of Fisher's exact tests, and the effects of density and potential competitors on movement by use of logistic regressions. These data consisted of marked arrivals to an area used for a long-term, capture-recapture study of small mammals. The immigrants arrived from other study plots in the vicinity and from patches of habitat trapped during a long-term investigation of patch dynamics. In cotton rats, but not prairie voles, long-distance movements were male biased and dominated by subadults. In cotton rats there was a marginal statistically significant seasonal difference between the relative proportion of immigrants and the population where they arrived (target population). Immigrants in the autumn composed a higher proportion of the target population than in other seasons. In prairie voles immigrants were a higher proportion of the target population in the summer and autumn than other seasons. There was no evidence that either abundance of conspecifics or other species influenced long-distance movements.

We compared survey techniques for estimating relative and absolute abundances of swift foxes (Vulpes velox) in New Mexico. For relative abundance surveys, the most efficient technique is collection of scats followed by verification of species depositing scats with DNA analysis. By collecting scats, the proportions of individual locations where swift foxes were detected were 61.9% and 67.7% during surveys in 2000 and 2001, which were greater than the proportions using scent stations (31.4%, 47.1%) or trapping (11.5%, 8.4%). By collecting scats, we detected swift foxes in 100% of the fox home ranges within the study area. If scent-station surveys are used instead, scent-station transects consisting of stations spaced at 1.6 km (1.0 mile) intervals and operated for three nights are the most practical. Searching for tracks, spotlighting and calling are much less efficient techniques. For absolute abundance surveys, trapping and resighting with cameras at bait stations was more accurate than counting unique microsatellite DNA genotypes from collected scats. Using trapping/resighting, we estimated the 95% confidence intervals for the swift fox population within the study area to be 17.8–30.0, 11.9–25.3 and 15.2–17.3 in the periods November 1999–January 2000, February 2000 and January–March 2001, respectively. We counted 63 and 27 unique genotypes in early 2000 and 2001, respectively. The numbers of unique genotypes, which were much greater than population estimates obtained from trapping and resighting, were overestimated because of the presence of transient swift foxes and poor quality DNA from scats leading to allelic drop-out and/or false alleles.

We examined the interactions among insectivorous birds, arthropods and white oak saplings (Quercus alba L.) in a temperate deciduous forest under ‘open’ and ‘closed’ canopy environments. For 2 y, we compared arthropod densities, leaf damage and sapling growth. Saplings from each canopy environment were assigned to one of four treatments: (1) reference, (2) bird exclosure, (3) insecticide and (4) exclosure insecticide. Sap-feeding insects were the most abundant arthropod feeding guild encountered and birds reduced sap-feeder densities in 1997, but not in 1998. Although there was no detectable influence of birds on leaf-chewer densities in either year, leaf damage to saplings was greater within bird exclosures than outside of bird exclosures in 1997. Insecticide significantly reduced arthropod densities and leaf damage to saplings, but there was no corresponding increase in sapling growth. Growth and biomass were greater for saplings in more open canopy environments for both years. Sap-feeder densities were higher on closed canopy than open canopy saplings in 1997, but canopy environment did not influence the effects of birds on lower trophic levels. Although previous studies have found birds to indirectly influence plant growth and biomass, birds did not significantly influence the growth or biomass of white oak saplings during our study.

The amphipod Gammarus pseudolimnaeus is abundant throughout most of the length of two small streams on the coastal plain of Virginia, but is absent from their upper reaches. The streams begin as very dilute springs but exhibit a downstream gradient of increasing dissolved ions and pH. Amphipods displaced upstream of their distribution limit in fine-mesh cages experienced rapid mortality compared to those maintained downstream of the limit, suggesting that neither interspecific competition nor predation explains the observed limit. General osmoregulatory failure during intermolt was not limiting; percent body-water content, a general gauge of osmotic stress, was not greater in amphipods closer to the distribution limit, nor did it increase in amphipods displaced above the limit. However, newly molted amphipods in water from above the distribution limit died more rapidly than those in water from below the limit. The addition of Ca ² to water from above the limit increased postmolt survival to levels observed in those maintained in water from below the limit; raising pH had no effect on postmolt mortality. Insufficient ambient calcium is a limiting factor in the distribution of G. pseudolimnaeus.

Surveys for larval mosquitoes were conducted in West Virginia from 1992 through 1997. A total of 26 species were identified with the frequency of occurrence for each species determined by elevation category, month of collection and association with one or more of 15 defined larval habitat types (7 sunlit and 8 shaded). Larvae of Anopheles punctipennis, Culex restuans, Cx. territans and Ochlerotatus triseriatus (in rank order) were the most commonly encountered species. Anopheles punctipennis and Cx. territans were more likely to be found at low (<457 m) elevations, whereas Cx. restuans and Oc. triseriatus were more frequently encountered at elevations of 457 to 914 m. Ochlerotatus canadensis first appeared in February, but after May this species was encountered in <1.0% of monthly collections. No other species (with the exception of Oc. triseriatus which first appeared in March) appeared before April. Anopheles punctipennis and Cx. territans larvae were found in the most habitat types (14 of 15) and both, along with Cx. restuans, were significantly more likely to be collected in sunlit conditions. Conversely, the likelihood of encountering Oc. triseriatus was significantly greater in shaded habitats. Precise frequencies of occurrence are given for all 26 mosquito species by elevation category, month of collection and each defined larval habitat type.

The ants Aphaenogaster cockerelli and Pogonomyrmex barbatus compete for seed resources in the Chihuahuan desert. Previous work showed that intraspecific competition in P. barbatus is more intense between near neighbors with overlapping foraging ranges and depends on colony age. Just before reaching reproductive maturity (3–4 y), colonies are more aggressive and persistent in intraspecific competition for foraging area than younger or older colonies. In this study we examine how interspecific interference behavior by A. cockerelli towards P. barbatus depends on the age and proximity of P. barbatus colonies. Before sunrise when P. barbatus colonies become active, A. cockerelli colonies completely plug the nest entrances of some P. barbatus colonies, thereby delaying the onset of P. barbatus foraging behavior. Pogonomyrmex barbatus colonies closer to A. cockerelli were plugged more frequently than more distant colonies. As distance from A. cockerelli nests increased, older P. barbatus colonies were plugged more frequently than younger ones. Our results suggest that the intensity of interspecific interference competitive interactions may depend on the proximity and age of competing colonies.

Species richness, frequency and density of litter-dwelling springtails (Collembola) were analyzed from woodland litter samples collected for 4 y, over a 12 y interval, from The East Woods of The Morton Arboretum, Lisle, Illinois. The 162 samples were divided equally between areas that had been previously burned and areas that had not been burned. There were significantly fewer springtail species in burned areas and 8 of the 10 most common species had lower frequencies in burned areas. The effect of fire varied for the 30 species identified in the study. Isotoma notabilis, Lepidocyrtus spp., Neanura muscorum and Tomocerus flavescens had a significantly higher frequency and density in unburned areas and Isotoma viridis and Xenylla grisea had a significantly higher frequency and density in burned areas. The mean dry weight of litter from 126 samples (equal number from burned and unburned areas) over 3 y was significantly heavier from the unburned areas. To maintain the richness of the litter invertebrates of forest ecosystems where prescribed burning is used, it is proposed that the scheduling of annual fires be changed to every 2 or 3 y. Before the use of fire, scattered refuge areas should also be established and maintained to enhance invertebrate survival.

The spatial relationships exhibited by individuals in a population may indicate their social organization. In territorial species interactions between individuals should lead to maximal spacing in a uniform pattern. Using nearest neighbor distances (NND), I tested for territoriality in the green frog (Rana clamitans) by determining if males were uniformly dispersed within breeding choruses. Observed dispersion patterns were not consistent with territoriality. Males were randomly dispersed on all seven nights during the breeding period, four of five nights during the nonbreeding period and uniformly dispersed on one night during the nonbreeding period. Dispersion did not differ between periods although density was higher and NND was smaller during the breeding period. There was also no correlation between male size and NND. The disparity of my results might be explained by differences in male density and resource distribution and its effect on habitat quality since these factors influence behavior and spacing. Alternatively, uniform spacing may not be a reliable characteristic of territoriality.

Bipalium adventitium Hyman, 1943, an exotic terrestrial flatworm that is predatory on earthworms, is reported from Illinois for the first time. No-choice laboratory feeding tests were performed to test the acceptability of several lumbricid earthworm species as prey for B. adventitium and to compare the mass gained by B. adventitium in feeding bouts on different species of lumbricid earthworms. All species of earthworms (Allolobophora chlorotica (Savigny, 1826), Aporrectodea rosea (Savigny, 1826), Ap. turgida (Eisen, 1874), Eisenia fetida (Savigny, 1826) and Octolasion tyrtaeum (Savigny, 1826)) presented to B. adventitium were attacked and consumed. Allolobophora chlorotica, Aporrectodea rosea and Eisenia fetida are added to the list of potential prey species of earthworms in North America. Bipalium adventitium readily attacked and fed on earthworms up to 55 times larger than themselves in mass and they gained 52% of their prefeeding live mass during feeding bouts. Prefeeding live mass of flatworms was a significant predictor of the mass gained by feeding flatworms, but prey mass was not, possibly because all flatworms fed on earthworms at least 1.9 times larger than themselves in mass. Earthworm prey species had no influence on the mass gained by feeding flatworms.

Seasonal prevalence and abundance of helminth species in an introduced population (Oahu, Hawaii) of the brown anole Anolis sagrei were examined. One species of Cestoda Oochoristica sp., four species of Nematoda Atractis scelopori, Physaloptera squamatae, Physocephalus sp. (larvae) and larval Acuariidae, one species of Acanthocephala Acanthocephalus bufonis and one species of Pentastomida Raillietiella frenatus were found. Seasonal prevalence of Atractis scelopori is correlated with the seasonal reproductive cycle of Anolis sagrei with highest prevalence found during periods of maximum reproductive activity.

Activity, reproduction and overwintering behavior were investigated in a radio-telemetry study of ornate box turtles (Terrapene ornata ornata) in the Sandhills region of Nebraska between 1997 and 1999. Based on 477 radio-locations where activity of turtles was known, turtles were active at 20% and inactive at 80%. We located one nest in 1997 and seven in 1998. Eggs were deposited in underground nests between 28 May and 12 June. Mean clutch size in nests was 2.6 eggs. Hatchlings in some nests overwintered under the nest cavities, whereas hatchlings in other nests may have emerged in the fall. Adult turtles spent winters buried underground. Overwintering sites were located and excavated to determine depth to overwintering turtles; mean depth in late March/early April was 54.2 cm. All nests and 17 of 18 overwintering sites were located in upland prairies; one overwintering site was located in a lowland meadow.