The progressive development of ten Holocene peatlands in northeast Indiana was determined by analysis of macroscopic subfossils recovered from sediment cores. All of the peatlands began as extremely mineral-rich lakes or ponds after retreat of Wisconsin-age glacial ice. The oldest basal date was 13,170 radiocarbon ¹⁴C y BP. The subfossil assemblage characterizing the lake stages included Chara sp., Ceratophyllum demersum, Najas flexilis and Potamogeton spp. The transition from lake to peatland was marked by a marsh flora dominated by Nuphar sp. Nymphaea sp., and, especially, Brasenia schreberi. Evidence for the development of a fen began early in the hydrosere (in the limnic phase) and colonization of the open water by a floating fen mat brought closure to the emergent marsh. Calcicolous mosses of the family Amblystegiaceae, including Calliergon stramineum, Calliergon trifarium, Campylium polygamum, Drepanocladus aduncus and Scorpidium scorpioides, dominated the fens. During the later stages of the fen phase an association of Calliergon trifarium and Meesia triquetra dominated the subfossil assemblage. Depending on the morphometry and geology of the respective basins, some of the fens became Sphagnum-dominated bogs. Subfossil assemblages of the bog phase included remains of ericaceous shrubs as well as sedges. Regardless of their developmental pathways, all of the peatlands exhibit an apparent trend towards senescence into lowland forests dominated by Acer rubrum.

We asked: Is Lycium californicum an ecotone species and what factors restrict it from overlapping more broadly with dominant plants of the salt marsh (downslope) and the coastal sage scrub (upslope)? Field sampling at Tijuana Estuary revealed that Lycium grows within a 1.1-m elevation range intermediate between that of Eriogonum fasciculatum (a subshrub of the coastal sage scrub) and Salicornia subterminalis (an obligate wetland subshrub of tidal marshes). Hence, Lycium was confined to the wetland-upland ecotone. In greenhouse experiments salt water wetting and soil moisture influenced these three species as follows: Salicornia required saturated soil to establish, restricting it to the wetland. Eriogonum failed to tolerate salt water wetting of the soil, both as seedlings (100% mortality) and as adults (100% mortality), likely restricting it to the upland. Lycium adults tolerated seawater wetting of the soil (100% survival) but seedlings did not (100% mortality). Both Lycium and Eriogonum should be able to establish seedlings near the wetland when soil salinity is low, but only Lycium would survive subsequent tidal surges (seawater wetting). Lycium showed evidence of drought tolerance but no requirement for groundwater (which would restrict it to the lowland); hence, its absence from the coastal sage scrub community may be due to competition with less salt-tolerant species. The few remaining populations of Lycium are restricted by the narrow range of suitable habitat and continuing pressure for coastal development. Thus, the transition from salt marsh to coastal sage scrub should be protected and restored to support this rare ecotone species.

We evaluated in situ the effects of long-term grazing (>100 y) on the morphological (i.e., shoot height, leaf blade length, width and angle) and physiological (i.e., gas exchange and water relations) responses of the dominant perennial grass species from arid lowlands and more mesic uplands of the Pryor Mountain Wild Horse Range (PMWHR) in Montana. Pseudoroegneria spicata, the most abundant grass in the lowland communities, had shorter vegetative shoot heights and leaf blade lengths and narrower leaves in plants from grazed than long-term ungrazed sites. Similarly, vegetative and reproductive shoot heights of Festuca idahoensis and Elymus lanceolatus, common upland grass species, were shorter in plants from grazed than ungrazed sites. Leaf lengths of these upland grasses also were shorter and less erect in plants from grazed sites than ungrazed sites. The physiological responses of the dominant grasses to grazing were not consistent between species or sampling dates. Overall, photosynthetic rates, stomatal conductances and xylem pressure potentials were the same in over 65% of the comparisons between plants from grazed and ungrazed sites and were higher in grazed sites in only 11 to 22% of the grazed-ungrazed comparisons. Collectively, our results indicate that long-term grazing of grasses by wild horses and other herbivores in the PMWHR has resulted in morphological modification, but has not substantively altered physiological function.

Heart-leaved birch (Betula cordifolia) and yellow birch (B. alleghaniensis) are commonly found growing during early and midsuccession after disturbance in northeastern forests. Betula cordifolia most often occurs at high elevations (above 600 m) and is considered less shade tolerant than B. alleghaniensis. Though shade-tolerance affects distribution of both birches, other physiological differences may also influence the success of each of these species in forest communities. In particular, on boulder-covered depositional zones of landslides common to steep slopes of the Appalachians, water relations may be important for birch survival. In Franconia Notch, NH, a suite of landslides occurred during this century; we investigated water relations of pairs of B. cordifolia and B. alleghaniensis trees rooted within two meters of one another on two of these landslides. Midday water potential of B. alleghaniensis was often significantly more positive (and never more negative) than that of B. cordifolia, whether trees were growing in a soil-rich lower depositional zone of a 1959 slide or on a boulder-covered steep section of an overlapping 1948 slide. Also, the magnitude of differences in water potential between the two birch species varied through the season and between sites. The hydraulic conductivity of stems of the two species was indistinguishable, but predawn water potential measurements indicated that B. cordifolia was rooted in zones with lower water availability. This differential rooting may contribute to differential water stress between young, co-occurring birches in the successional forest environment.

Seeds were collected from three red maple swamps (wet sites) and three uplands (dry sites) near Ithaca, New York, and the resulting seedlings were used in flood tolerance studies to investigate if red maple's broad habitat range is due to ecotypic differentiation. One-year-old seedlings were flooded while still dormant (spring flooding study) and net photosynthesis, growth and chlorophyll levels were measured at 1 mo intervals for 3 mo. Flooding reduced net photosynthesis, growth and chlorophyll levels in seedlings from both sites, but survival of both wet and dry site seedlings was near 100%. After 1 mo of flooding net photosynthesis of wet and dry site seedlings were similar, but after 3 mo, flooded wet site seedlings had higher photosynthetic rates than did dry site seedlings. Control wet site seedlings were significantly larger than dry site seedlings and had significantly higher photosynthetic rates and chlorophyll levels. These differences suggest either genetic variation between seedlings from the two habitats in response to the growing conditions or the influence of seed size differences and confound the spring flooding study results. Flooded wet site seedlings had higher, final, net photosynthetic rates than did dry site seedlings, but the response to flooding was greater for wet site seedlings than it was for dry site seedlings.

In a second study (summer flooding study), 1-y-old seedlings in full leaf were flooded for 22 d and then drained to determine if recovery from flooding stress differed for wet and dry site seedlings. Again, flooding decreased net photosynthesis for seedlings from both habitats but, when the trees were drained, net photosynthesis for wet site seedlings recovered more quickly and to a higher level than it did for dry site seedlings. Flooding also caused a drop in chlorophyll levels for seedlings from both habitats, but chlorophyll levels of seedlings from neither habitat recovered when the seedlings were drained. There were no significant differences between wet site and dry site control seedlings for net photosynthesis; therefore, the quicker and larger recovery of photosynthetic potential in wet site seedlings in the summer flooding study suggests that ecotypic differentiation has occurred and that genetic differences, in part, account for red maple's occurrence on contrasting edaphic sites.

We examined the dynamics of the tree-seedling bank in old-growth, subalpine Picea engelmannii—Abies lasiocarpa forests in south-central British Columbia by collecting all individuals <1.3 m tall within belt transects at two sites (n = ca. 500 per site) and determining their height, basal diameter and age. Seedlings were up to 149 y old. Regressions showed that spruce 1-m tall averaged 65-y old and fir 92-y old. Age structures indicated that recruitment into the seedling bank was variable with time, but that individuals of both species occurred in almost all 5-y age classes. Regressions of height and diameter vs. age had high predictability and showed that spruce grew more rapidly than fir. Allometric (height/diameter) relationships differed significantly between species; spruce had a much greater height/diameter ratio than fir. Seedlings accumulate gradually and persist for long periods in the seedling bank, thus forming a large pool of individuals that can potentially respond to opening of the canopy.

We compared abundance, habitat use and diurnal activity patterns of sympatric populations of fox squirrels (Sciurus niger) and gray squirrels (S. carolinensis) in forest-farmland interfaces in a four-county area of Pennsylvania from August 1995 to October 1996. The mean number of fox squirrels observed was significantly less (F = 21.34, P < 0.001) than that of gray squirrels. Fox squirrel abundances were significantly greater in spring than in summer or fall (F = 6.1, P < 0.01). Habitat use by fox squirrels and gray squirrels was differentiated primarily by the distance to a forest-farmland edge, with fox squirrels using areas closer to the forest edge. Habitat used by fox squirrels was also characterized by fewer short shrubs compared to available habitat. Habitat used by gray squirrels was farther from the edge and had fewer understory trees and logs and a greater basal area of snags compared to available habitat. Gray squirrels were more active in the morning compared to midday (P < 0.001), but activity of fox squirrels did not vary between time periods. We conclude that the relatively sparse populations of fox squirrels coexisting with the more abundant populations of gray squirrels on our study sites differ in habitat use principally with regard to proximity to the forest-farmland edge.

Understanding population genetics is important for increasing both our basic knowlege of wild species and our ability to conserve endangered species. In the Carnivora questions about genetic variability across the order also require population genetic information about more species. I present estimates of polymorphism and heterozygosity in the striped skunk, Mephitis mephitis, an abundant species throughout North America. In two populations in eastern Tennessee the average heterozygosity and polymorphism are, respectively: Cades Cove, 0.038 and 0.11; Knoxville, 0.040 and 0.11. These values are well within the ranges for mammals and are intermediate between values suggested by previous studies of skunks using small numbers of individuals. Possible explanations are discussed.

Agonistic behavior of heteromyid rodents has been studied using staged encounters in the laboratory, but there have been no previous attempts to test for a linear dominance hierarchy among individuals of the same species. Dominance hierarchies are important in learning about sociality, priority of access to resources and consequences for individual variation in fitness. We used standard laboratory methods to assess agonistic behavior of Merriam's kangaroo rats, Dipodomys merriami. Males exhibited a dominance hierarchy that was strongly linear. Dominance rank was not correlated with body mass, but dominant males lost greater percentages of their body mass during trials than did subordinate males. Males dominated females and females showed little agonistic behavior in intrasexual trials. The linear dominance hierarchy among males may have reflected individual variation in aggressive tendencies, but dominance rank was not correlated with individual variation in total amounts stored or proportions of seeds larderhoarded in food-hoarding trials. Different patterns of individual variation in males and females are promising topics for future research.

We monitored 63 buildings in Indiana serving as hibernacula for big brown bats for up to six winters. Fidelity of individuals to roosts averaged 21.4% in buildings with maternity colonies and 33.3% in buildings without maternity colonies, although this difference was not significant. There was much turnover, both within and among years, some bats being present one year, absent for a year or two and then returning. The sex ratio of hibernating bats was male biased (65.6% males) in buildings with maternity colonies and female biased in buildings lacking maternity colonies (65.2% female). Most big brown bats in Indiana hibernate in heated buildings rather than in caves or mines.

We investigated whether changes in chestnut oak (Quercus prinus L.) leaf quality caused by gypsy moth (Lymantria dispar L.) defoliation affected leaf breakdown rates in southern Appalachian streams of differing disturbance history. Breakdown rates of second-flush leaves produced after defoliation were compared to those of natural spring-flush leaves shed in autumn. Second-flush leaves broke down significantly faster than spring-flush leaves in three of the six streams tested. Initial fiber content and the ratio of fiber to protein were significantly higher in spring-flush leaves than in second-flush leaves, showing that initial differences in internal leaf constituents could explain the faster breakdown rates of second-flush leaves. Using changes in leaf toughness through time as a measure of microbial conditioning we found that the faster-decaying second-flush leaves also softened at a faster rate than the spring-flush leaves. In addition, both types of leaves incubated in three streams draining a recovering 14-y-old clear-cut catchment broke down significantly faster than leaves incubated in three streams draining a reference catchment. We attributed this increase in leaf breakdown to significantly higher abundance and density of leaf-shredding insects and greater microbial conditioning in leaf packs in the streams of the recovering clear-cut catchment. Overall, our results show that insect defoliation accelerates detritus processing in southern Appalachian streams and that this acceleration may be especially important in previously disturbed streams in which leaves are already processed faster.

Holocnemus pluchei spiders (Family Pholcidae) facultatively live in groups: sometimes they live alone and sometimes they share webs. In the field groups vary in size and composition and include spiders of all ages and either sex. Group membership is flexible and individuals move frequently among groups. To understand group formation and maintenance it is necessary to understand the costs of group membership. We used focal animal sampling to investigate the cost of group living for spiders of different ages across a range of group sizes. Both spider age and group size affected the costs incurred by group-living spiders. There was no variation among groups of different sizes in the percentage of time focal small or large spiders spent in costly behaviors (moving, web maintenance, bouncing or interactions with conspecifics), but medium-sized spiders spent more time engaged in costly behaviors with increasing group size. Medium and large spiders also had more interactions with greater numbers of different conspecifics when they were in groups larger than three, whereas small spiders interacted rarely with conspecifics regardless of group size. These results suggest that there are significant ontogenetic shifts in the costs of group living in H. pluchei.

Cycleptus meridionalis Burr and Mayden (Catostomidae) is a new species recently distinguished from blue sucker C. elongatus in large rivers draining to the northern Gulf of Mexico. Collections of this new species were taken for 12 mo from the Pearl and Pascagoula rivers, Mississippi, to document catch-per-unit-effort, relationships with physical-chemical conditions and spawning migration. We observed/collected 263 C. meridionalis in the Pearl River on 14 dates and 296 in the Pascagoula River on 19 dates. River stage was highest between January and late April in both rivers and there was a negative relationship between CPUE and river stage in both rivers. Highest CPUE of C. meridionalis was after May in both rivers when river stage decreased and we consistently collected more C. meridionalis in the Pearl than the Pascagoula river. Stepwise multiple regression indicated C. meridionalis were abundant in the Pascagoula River from summer through fall when conductivity was high and in fall and winter when water temperature was cooler and DO higher. In the Pearl River C. meridionalis were abundant when water was shallow (summer and fall) and rainfall was high (April and May).

Temporal patterns in the sex ratio were similar between rivers from March to May. Males were more abundant early in the migration, becoming relatively equal with females by May, particularly in the Pascagoula River. Although males were more abundant from late-May until August in the Pearl river, females were more abundant between June and January in the Pascagoula river. Breeding tubercles were noted as early as 24 October, increased in both rivers through 2 April and were generally absent for the remainder of the year. Tagging results indicated only a 1.8 and 1.9% recapture rate in the Pearl and Pascagoula rivers, respectively. Only fish tagged in summer were recaptured and they had moved upstream < 3.2 km. Our observations indicate that C. meridionalis is viable in both rivers but this status may change if future river modification occurs, as happened with C. elongatus in other large river systems.

Contrary to the reports of several authors, Acomus generosus, a sucker of the subgenus Pantosteus of Catostomus, may have been collected in Big Cottonwood Creek or Little Cottonwood Creek, Salt Lake County, Utah, and possibly in Provo River, Utah County, Utah. Multiple Pantosteus species may exist in Provo River. Synonymies of the suckers that have been collected in Provo River and “Cottonwood Creek” are provided. A comparison of their morphologies shows several distinctions. There is a need for further systematic (including up-to-date biochemical) analyses of these catostomids.

We used RNA-DNA ratios in both field and laboratory experiments to examine the effects of hypoxia on short-term growth of bluegill Lepomis macrochirus collected from the lower Atchafalaya River Basin, Louisiana. In the field experiment, RNA-DNA ratios of bluegill (17.8–52.3 g) from hypoxic habitats (DO < 2.0 mg L⁻¹; n = 26) were significantly lower than ratios of bluegill from normoxic habitats (DO > 4.0 mg L⁻¹; n = 31). In each of two laboratory experiments 40 bluegill (14.54–76.70 g) were individually placed in aquaria to test the effects of hypoxia on RNA-DNA ratios. Additionally, bluegill were fed at different rates in the first experiment to determine the effects of feeding level on RNA-DNA ratios. Results of the first experiment confirmed the sensitivity of RNA-DNA ratios to short-term changes in growth, as fed bluegill had significantly higher ratios than starved bluegill. In both experiments RNA-DNA ratios were not significantly different in bluegill subjected to hypoxia. Results of the study indicate that laboratory experiments did not adequately simulate increased bioenergetic demands and fluctuating DO levels found in hypoxic areas of the Basin.

We compared the physiological status and biochemical composition of unionid mussels (Amblema plicata) from a riverine population with and without attached zebra mussels (Dreissena polymorpha). Unionid specimens were collected from Lake Pepin, Mississippi River, WI and MN, where both infested and noninfested unionids were experiencing local increases in water clarity. Infested specimens had higher ammonia excretion rates, lower respiration to nitrogen excretion ratios and lower clearance rates than noninfested specimens. Infested specimens also had lower carbohydrate and protein contents. Zebra mussel infested unionids had depleted energy stores and we hypothesize that they were starving, relative to noninfested individuals from the same location. This comparison helps to clarify the nature of zebra mussel impacts on unionids by showing that direct attachment is a very important component of the effect of zebra mussel populations on unionids.

We studied the ecology of the checkered garter snake, Thamnophis marcianus, in a desert grassland in southeastern Arizona. The adult sex ratio was 0.67 M:F and varied significantly among seasons. As with other species of Thamnophis, adult females were significantly larger than males in SVL and body mass, and females also had longer jaws. Activity occurred both during the day and nighttime, but was confined to aquatic areas or their immediate vicinity. Mating occurred in late March and females gave birth to a single brood of an average of 15 offspring in late May and early June. The timing of birth in this population was among the earliest on record for any live-bearing snake in North America and was much earlier than for other checkered garter snake populations. In comparison with populations of T. marcianus from northern and southern Texas, females from Arizona had larger maternal body sizes and their offspring were larger as well; conversely, we found no significant differences in brood size among localities once maternal body size was taken into account. The implications of early timing of birth and geographic variation in reproductive traits are discussed.

From 1992 to 1995 we used radiotelemetry to monitor winter habitat selection and survival of female ring-necked pheasants (Phasianus colchicus) in southeastern North Dakota. We captured 100 birds at nine sites in six study blocks centered on cattail-dominated (Typha spp.) semipermanent wetlands. Pheasants showed nonrandom habitat use at two hierarchical scales. At the second-order scale (23-km² blocks) semipermanent wetlands were preferred during two winters in which habitat selection could be assessed (1992–1993 and 1994–1995). An additional second-order preference for grass-covered uplands was shown during the mild 1994–1995 winter. At the third-order scale (home-range) pheasants preferred the edges of wetlands in 1992–1993 and 1994–1995. The central portions of wetlands were preferred in 1992–1993 and used proportionately in 1994–1995. Seasonal wetlands were avoided at the third order scale during 1992–1993 and 1994–1995. The average winter survival rate was 0.41, with rates ranging from 0.04–0.86 and differing significantly among winters. Survival was lower during early winter and midwinter periods for birds weighing less than 1090 g and for birds captured in semipermanent wetlands under private ownership. A 1 C increase in the mean weekly maximum temperature decreased the probability of death by 0.06 and a 2.5 cm increase in new snow raised the probability of death by 0.08.

Taxonomists describe panicles of green and yellow foxtail (Setaria viridis (L.) and S. glauca (L.) Beauv., respectively) as dense, contracted, narrow and cylindrical. However, a 6-y survey near Hays, Kansas documented inflorescences of green and yellow foxtail that were divided into two to four or more elongated segments. Those panicles were forked or digitate rather than a single compact cylinder. Anomalous panicles were always found on plants having several normal heads. Branched morphology was infrequent (<1 head in 10,000), but was observed in 1993 and 1995–1998. Before 1998 Kansas collections of foxtail specimens with abnormal panicles were limited to a 1600 × 3200 m area in central Ellis County. Observations during 1998 increased the known geographic distribution in Kansas to a 3200 × 3600 m area. Seed collected from branched panicles of both species and grown in greenhouse conditions did not produce plants with any aberrant inflorescences. Occurrence of deviant green or yellow foxtail panicles did not correlate with herbicide usage or drift.

We gathered dietary data for several harvestman species in a soybean field and adjacent hedgerow habitats to assess the extent of their polyphagy. A total of 1032 harvestmen, predominantly Leiobunum spp., were observed during almost 50 search hours over two seasons. More harvestmen and increased foraging activity were observed in the hedgerow than in the soybean field, and in both habitats harvestmen were more active at night. Earthworms (Oligochaeta: Lumbricidae) were the prey items most frequently recorded as being consumed by all observed harvestman species in both habitats (47.1%), while fruit and other plant material constituted 18.1% of the records. Feeding trials conducted with female harvestmen, Hadrobunus maculosus (Wood), revealed that these arachnids were not capable of subduing live earthworms, suggesting that consumption of these in the field might be limited to scavenging. Implications of harvestman foraging for local food web dynamics are discussed.