Petersen, S.M. and P.B. Drewa (Case Western Reserve University, Cleveland, OH 44106). Did lightning-initiated growing season fires characterize oak-dominated ecosystems of southern Ohio? J. Torrey Bot. Soc. 133(2): 217–224. 2006.—The persistence of oak-dominated ecosystems throughout the Central Hardwoods Region has been attributed to chronic, non-catastrophic fire regimes prior to Euro-American settlement. The nature of these fires is not well understood. We used lightning strike, thunderstorm activity, and precipitation data to elucidate the seasonal timing of natural fire regimes for southern Ohio. Our results suggest that fires are most likely to be initiated during the growing season (especially August) when there is a high frequency of lightning strikes and thunderstorm activity that occurs simultaneously with dry environmental conditions. We hypothesize that the use of prescribed growing season fires would effectively deter encroachment of shade tolerant hardwood vegetation and foster regeneration of oaks. Such fires would provide an evolutionary basis for the conservation of oak-dominated ecosystems in southern Ohio.

Pavlovic, N., R. Grundel, and W. Sluis (U.S. Geological Survey, Great Lakes Science Center, 1100 N. Mineral Springs Rd., Porter, IN 46304). J. Torrey Bot. Soc. 133(2): 225–239. 2006.—Frequency of groundlayer plants was measured across oak woodland canopy gaps at three sites in northwest Indiana to examine how vegetation varied with gap size, direction along the gap edge, and microhabitat. Microhabitats were defined as under the canopy adjacent to the gap, along the gap edge, and within the gap. Gap-sites consisted of gaps plus adjacent tree canopy. Gaps were classified as small (16 ± 1 m²), medium (97 ± 8), and large (310 ± 32). Neither richness nor diversity differed among microhabitats, gap sizes, or edges. Similarity between microhabitats wthin a gap-site increased as the distance between plots decreased and as the difference in PAR decreased, the latter explaining twice the variation in percent dissimilarity compared to Mg concentration, A horizon depth, and litter cover. Diervilla lonicera, Frageria virginiana, Helianthus divaricatus, Polygonatum pubescens, Quercus velutina, Smilacena stellata, and Tradescantia ohiensis decreased, while Tephrosia virginiana and legumes increased in frequency, from canopy to gap, and C₄ grasses peaked at the gap edge, independent of gap size. Additional species frequency varied across the microhabitat gradient within specific sites. Sorghastrum nutans was three times more frequent in gaps at large sites than elsewhere. The vegetation in medium-sized gap-sites was more variable than within small and large gap-sites, suggesting greater environmental heterogeneity at that scale. Within gap-sites, vegetation was more heterogeneous within edges and canopies than in gaps. Edges were more similar in composition to gaps than to canopy groundlayer within gap-sites. Few species varied significantly in frequency around the gap edge. The oak woodland groundlayer on sandy substrates can be characterized as a mosaic of forb dominated vegetation that varies across light gradients associated with canopy gaps, transitioning to islands of grassland vegetation when gaps exceed 160 m².

Gilliam, F. S., A. W. Hockenberry (Department of Biological Sciences, Marshall University, Huntington, WV, 25755-2510), and M. B. Adams (U.S.D.A. Forest Service, Timber and Watershed Laboratory, Parsons, WV, 26287-0404). Effects of atmospheric nitrogen deposition on the herbaceous layer of a central Appalachian hardwood forest. J. Torrey Bot. Soc. 133(2): 240–254. 2006.—Additions of nitrogen (N) have been shown to alter species diversity of plant communities, with most experimental studies having been carried out in communities dominated by herbaceous species. We examined seasonal and inter-annual patterns of change in the herbaceous layer of two watersheds of a central Appalachian hardwood forest that differed in experimental N treatment. This study was carried out at the Fernow Experimental Forest, West Virginia, using two adjacent watersheds: WS4 (mature, second-growth hardwood stand, untreated reference), and WS3 (∼25-yr old, treated with 35 kg N ha⁻¹ yr⁻¹ as (NH₄)₂SO₄ yr⁻¹). Seven circular 0.04-ha sample plots were established in each watershed to represent its full range of elevation and slope aspect. The herbaceous layer was sampled by identifying and visually estimating cover (%) of all vascular plants ≤ 1 m in height within 10 1-m² circular sub-plots in each sample plot, for a total of 140 1-m² sub-plots. Sampling was carried out in mid-July of 1991 and repeated at approximately the same time in 1992. In 1994, these same plots were sampled each month from May to October. Seasonal patterns of herb layer dynamics were assessed for the complete 1994 data set, whereas inter-annual variability was based on plot data from 1991, 1992, and the July sample of 1994. There were no significant differences between watersheds for any sample year for any of the other herb layer characteristics measured, including herb layer cover, species richness, evenness, and diversity. Cover on WS4 decreased significantly from 1991 to 1992, followed by no change to 1994. By contrast, herb layer cover did not vary significantly across years on WS3. Cover of the herbaceous layer of both watersheds increased from early in the growing season to the middle of the growing season, decreasing thereafter, with no significant differences between WS3 and WS4 for any of the monthly cover means in 1994. Similar seasonal patterns found for herb layer cover—and lack of significant differences between watersheds—were also evident for species diversity and richness. By contrast, there was little seasonal change in herb layer species evenness, which was nearly identical between watersheds for all months except October. Seasonal patterns for individual species/species groups were closely similar between watersheds, especially for Viola rotundifolia and Viola spp. Species richness and species diversity were linearly related to herb layer cover for both WS3 and WS4, suggesting that spatial and temporal increases in cover were more related to recruitment of herb layer species than to growth of existing species. Results of this study indicate that there have been negligible responses of the herb layer to 6 yr of N additions to WS3.

Honu, Y. A. K. and D. J. Gibson (Southern Illinois University at Carbondale, Department of Plant Biology, Center for Ecology, 1125 Lincoln Drive, Carbondale, Illinois 62901-6509, USA). Microhabitat factors and the distribution of exotic species across forest edges in temperate deciduous forest of southern Illinois, USA. J. Torrey Bot. Soc. 133(2): 255–266. 2006.—The study of microhabitat factors has attracted the attention of ecologists for about half a century. We examined crop-forest, access road-forest, and hay field-forest edge types in temperate deciduous forest in Southern Illinois to obtain information that may contribute to the control and management of exotic species. Forest edges have different abiotic factors compared to the interior, a phenomenon known as an edge effect. To investigate microhabitat factors, three 90 m transects were established from the forest boundary into the interior at each of the three edges in July 2002. Twelve 3.14 m² circular plots at 5 m intervals from each other for the first 20 m and 10 m apart thereafter were established along each transect. A 550 cm³ soil core was extracted from the six cm A-horizon in each plot along the three transects in August 2002 and analyzed for soil texture, pH., and major soil nutrients. To quantify the edge effect, the Distance of Edge Influence (DEI) was calculated by contrasting the value of microhabitat factors in plots ≤ 50 m from the forest boundary with the value of microhabitat factors in plots > 50 m from the boundary (interior plots) using a permutation approach. The DEI is calculated as the distance where two or more consecutive plots from the forest boundary have values that differ significantly from the expected value of the interior plots. Percent clay and sand generally declined across the crop-forest and access road-forest edges into the interior while the opposite relationship was observed at the hay field-forest edge. Percent silt showed a pattern that was opposite to those exhibited by percent sand and clay at all the three edge types. The DEI of the percent sand, clay, and silt varied between 15 to 50 m at the three edges. Seven exotic species were present in the vegetation and the relationship of the abundance of three of them to 11 microhabitat factors was tested. Lonicera japonica and Allium vineale were absent when canopy openness was < 15 % while the presence of Cardamine hirsuta was independent of canopy openness. Management strategies for the control of the invasive L. japonica and A. vineale should consider canopy closure among other factors. By contrast, management prescriptions for the exotic C. hirsuta may pose a challenge to forest managers and conservation biologists as a result of its insensitiveness to the 11 microhabitat factors including light (canopy openness) measured in this study. For conservation purposes, a buffer strip of 50 m around a conservation area would eliminate most of these microhabitat edge effects.

Corbett, E. A. (Department of Biological Sciences, Southeastern Oklahoma State University, Durant, OK 74701-0609) and R. C. Anderson (Behavior, Ecology, Evolution and Systematics Section, Department of Biological Sciences 4120, Illinois State University, Normal, IL 61790-4120). Landscape analysis of Illinois and Wisconsin remnant prairies. J. Torrey Bot. Soc. 133(2): 267–279. 2006.—We examined Illinois and Wisconsin remnant prairie data to determine regional patterns of species composition as influenced by landforms with different topographic positions and soil properties resulting from glacial history. Three data sets were used, including 84 sites from Wisconsin sampled in the 1950s; 216 sites from Illinois sampled as part of the Illinois Natural Areas Inventory (INAI) in the 1970s; and a subset of 29 INAI sites resampled in 1998. These data sets are the best characterization of remnant prairie vegetation in the two states after most of the original prairies were lost. Data were analyzed using detrended correspondence analysis (DCA), principal components analysis (PCA), and canonical correspondence analysis (CCA). CCA was used only for the INAI sites resampled in 1998 that had a complete set of soils data. Results from all analyses were similar. We determined that an interaction between topographic position and soil texture, which affected moisture availability, was most important in influencing species composition and abundance in Illinois and Wisconsin remnant prairies. Separation of dominant prairie species in Illinois and Wisconsin followed similar patterns, if differences in community designation for similar vegetation types were considered. However, Wisconsin and Illinois sites separated on DCA axis 2, which was negatively correlated with calcium and phosphorus. The first DCA ordination axis for INAI prairie sites was correlated with a topographically-based Integrated Moisture Index (IMI). For the 1998 INAI data, PCA was used to reduce soil variables to fewer dimensions and resulting axis 1 scores were used in stepwise multiple regression analysis to determine relationships between environmental variables and DCA ordination axis scores. The IMI was the first variable entered in the stepwise multiple regression model; however, PCA axis 1 stand scores based on soil variables did not meet criteria for entry into the model. For all data sets, including the CCA ordination of the resampled INAI sites, there was a secondary separation at the dry end of the moisture gradient between hill prairies and dry sand prairies, which was related to differences in soil texture and availability of calcium and phosphorus. In general, soil nutrients and texture were less important than topographically controlled moisture availability in determining prairie species composition and abundance at this scale of analysis.

J. K. Bush, F. A. Richter (Department of Earth and Environmental Sciences, The University of Texas at San Antonio, San Antonio, Texas 78249), and O. W. Van Auken (Department of Biology, The University of Texas at San Antonio, San Antonio, Texas 78249). Two decades of vegetation change on terraces of a south Texas river. J. Torrey Bot. Soc. 133(2): 280–288. 2006.—A chronosequence of stands on floodplain terraces of the San Antonio River in south Texas was sampled in 1983 and resampled in 2003. The youngest resampled stand, an early successional stage originally 19 yrs of age, was an open Acacia farnesiana (huisache) woodland. After 20 yrs this stand was a mid- successional A. farnesiana—Celtis laevigata (Texas sugarberry) woodland, and demonstrated a decrease in A. farnesiana density from 427 plants ha⁻¹ to 167 plants ha⁻¹ and an increase of C. laevigata density from zero plants ha⁻¹ to 860 plants ha⁻¹. The 25 yr old stands were originally mid-successional A. farnesiana woodlands. After 20 yrs, A. farnesiana density decreased by 462 plants ha⁻¹ and C. laevigata density increased by 1031 plants ha⁻¹. Acacia farnesiana basal area decreased by 9.4 m² ha⁻¹, while C. laevigata basal area increased by 5.8 m² ha⁻¹. The 27 and 29 yr old stands were originally A. farnesiana woodlands, but after 20 yrs were Celtis woodlands. The greatest decrease in A. farnesiana basal area (12.0 m² ha⁻¹) and the greatest increase in C. laevigata basal area (15.9 m² ha⁻¹) occurred in the now 47 yr old stands. The 32 yr old mid-successional stand was originally an A. farnesiana—C. laevigata woodland, and at 52 yrs was a C. laevigata woodland. Both density and basal area of A. farnesiana decreased in the successional sequence, while C. laevigata density first increased and then decreased, while basal area increased. Significant changes have occurred over the last two decades, however it is predicted that further changes will occur before the communities become mature. A major co-dominant in the mature stands, Ulmus crassifolia, was not found in the 52 yr old stand. In addition, several other mature community species including U. americana, Crataegus sp., and Ilex decidua were not found in the 52 yr old stand.

Craine S. I. and Orians C. M. (Tufts University, Medford, Mass.). Effects of flooding on pitch pine (Pinus rigida Mill.) growth and survivorship. J. Torrey Bot. Soc. 133(2): 289–296. 2006.—Pitch pines (Pinus rigida Mill.) colonize the shores of Cape Cod coastal plain ponds in periods of low water and, once established, inhibit the growth of resident herbaceous species, including rare endemics. The effects of flooding on P. rigida may vary with plant age and may be affected by previous exposure to flooded conditions. To determine the effects of flooding on pitch pine survivorship, we performed three sets of experiments. First, we examined the effects of relative flood depth—root flooding versus total submersion—on three-month-old seedlings. Second, we determined whether prior exposure to root flooding enhanced subsequent survivorship in flooded conditions one year later. Lastly, we compared the effects of root flooding on pines of different ages—three months, 15 months, and five years. We found that three-month-old P. rigida succumbed to total submersion in two to four weeks but tolerated root flooding for up to one year. Fifteen-month-old seedlings that had survived eight weeks of root flooding one year earlier were significantly more flood-tolerant than members of the same cohort that had never been flooded. Fifteen-month-old seedlings, as well as five-year-old saplings, suffered significantly more mortality after four months of root flooding than did three-month-old seedlings. The ability of young P. rigida, normally an upland plant, to survive in waterlogged soil appears to be due, in part, to morphologic modifications acquired in response to flooded conditions, including hypertrophy of lenticels on the stem and plagiotropism. Both these acquired modifications are known to help oxygenate root tissues through internal channels, avoiding cellular anoxia despite reduced oxygen availability in the rhizosphere. Overall, our results suggest that periodic flooding has historically played a role in preventing the colonization of P. rigida on coastal plain pond shores and that reduced flooding intensity may lead to more frequent encroachment of P. rigida in this habitat.

Heyel, S. M. and F. P. Day. (Department of Biological Sciences, Old Dominion University, Norfolk, VA 23529). Long-term residual effects of nitrogen addition on a barrier island dune ecosystem. J. Torrey Bot. Soc. 133(2): 297–303. 2006.—In 1991, large plots established on the dunes on Hog Island, part of the Virginia Coast Reserve Long Term Ecological Research site, were fertilized with nitrogen to examine plant community response to nitrogen addition. Nitrogen fertilization significantly increased biomass, and the fertilized plots continued to exhibit enhanced production for 9 years following fertilization.

The current study sampled aboveground biomass, belowground biomass, and nitrogen standing crops of the experimental plots on a mid-island dune to determine the mechanism of long- term nitrogen retention in this nitrogen-limited system. Aboveground and belowground biomass were significantly greater in the fertilized plots than in the control plots in 2000. Aboveground biomass declined significantly in control and fertilized plots from 1991 to 2000, while belowground biomass exhibited a significant increase in both control and fertilized plots. In particular, biomass of standing dead, litter, and live roots, and nitrogen standing crop in all plant components were substantially greater in fertilized plots in 2000. Nitrogen concentrations were higher in the litter in the fertilized plots. These data suggest that the retention of nitrogen within the fertilized plots has been primarily facilitated by increased biomass, predominantly in roots, and increased pools of plant litter.

Liu, H. (Department of Biology, University of South Florida, 4202 East Fowler Ave., SCA 110, Tampa, FL 33620), R. W. Pemberton (U.S. Department of Agriculture, Agricultural Research Service, Invasive Plant Research Laboratory, 3225 College Ave., Ft. Lauderdale, Florida 33314), and P. Stiling (Department of Biology, University of South Florida, 4202 East Fowler Ave., SCA 110, Tampa, FL 33620). Native and introduced pollinators promote an invasive woody vine (Paederia foetida L) in Florida. J. Torrey Bot. Soc. 133(2): 303– 311. 2006.—Little is known about the role of mutualistic relationships with local species in promoting or limiting introduced invasive plant species. In this study we carried out controlled pollination treatments and floral visitor watches at three sites varying in degree of human disturbance in central Florida to determine the breeding system and potential pollinators of Paederia foetida L (skunk vine). A native of eastern Asia, P. foetida is invasive in natural and human created habitats in the southern United States and Hawaii. Our data suggested that populations of P. foetida in Florida were self-incompatible and relied on both small-bodied halictid bees native to Florida and the introduced European honey bee for pollination. Visitation by native bees was more frequent in less disturbed sites, while the opposite was true for the honey bees. Pollinator limitation was evident at the relatively undisturbed and the highly disturbed sites, but not at the intermediate disturbed site. Mutualistic relationships with native pollinators as well as the introduced honey bee probably contributed to the establishment and spread of skunk vine in Florida.

Cortés-Palomec, A. and H. E. Ballard (Department of Environmental and Plant Biology, Ohio University, Athens, Ohio, 45701). Influence of annual fluctuations in environmental conditions on chasmogamous flower production in Viola striata. J. Torrey Bot. Soc. 133(2): 312–320. 2006.—The ecological factors affecting chasmogamous flower production by Viola striata, a perennial herb widely distributed in temperate forests of the northern and eastern United States and southern Canada, were studied in a natural setting. Forty semi-permanent quadrats were established at two sites in southern Ohio where populations of V. striata are abundant. Investigations were conducted from April to mid-June in 2002 and 2003. The number of chasmogamous flowers produced per individual per quadrat was recorded and the percentage of flowers becoming fruits per plant was measured as an indication of reproductive success. For each quadrat, canopy openness and soil properties (N, K, P, Ca, Mg, pH, and moisture) were documented and their effects on chasmogamous flower production were evaluated. Chasmogamous flower production was correlated in some, but not in all cases with light availability one to two weeks before the population reached the peak of blooming. Multiple correlations showed that phosphorus and calcium were important in all significant cases while light, magnesium, nitrogen and pH were important in some cases. Chasmogamous flower success was highly variable ranging from 1 to 100 % of success in fruit set possibly due to differential pollinator activity.

Boerner, R. E. J. Department of Evolution, Ecology, and Organismal Biology, Ohio State University 318 West 12th Avenue, Columbus, OH 43210 USA. Unraveling the Gordian Knot: Interactions among vegetation, topography, and soil properties in the central and southern Appalachians. J. Torrey Bot. Soc. 133: 321–361. 2006.—This review focuses on the development of our knowledge of the interactions among vegetation, soil forming processes, topography, and human actions in shaping the forests of the central and southern Appalachians. It proceeds from a review of the studies that led to and cemented the topographic paradigm in forest ecology in the 1940's to a review of the studies of soil development in the region during the 1960's and 1970's. This is followed by a review of how the incorporation of the emerging understanding of edaphic and biogeochemical factors affected studies of vegetation in the Appalachians over the last three decades, with emphasis on studies which have been particularly useful in beginning to unravel the interactions among microenvironment, vegetation, and soil. I then consider a suite of natural factors (storm damage, canopy herbivory, fire) and human disturbances (agricultural intensification, atmospheric deposition) that impact the interactions between forest vegetation and the soil system, and speculate on how viewing the Appalachian ecosystems from a more integrated landscape perspective might affect future attempts to unravel the Gordian Knot.

Stalter, R. (Department of Biological Sciences, St. John's University, Jamaica, NY 11439) and E. E. Lamont (Honorary Research Associate, Institute of Systematic Botany, The New York Botanical Garden, Bronx, NY 10458). The historical and extant flora of Sable Island, Nova Scotia, Canada. J. Torrey Bot. Soc. 133: 362–374. 2006.—Sable Island, Nova Scotia, Canada, was surveyed for vascular plants in August 2002. The island's past and present flora comprises 61 families, 156 genera, and 224 species, of which 145 species (65%) are native and 79 species (35%) are non-native. The most species-rich families are Poaceae, Asteraceae, Cyperaceae, Polygonaceae, Caryophyllaceae, and Rosaceae. The largest genera are Juncus and Carex. Floristic studies of five earlier investigators are presented and discussed, including Macoun, conducted in 1899; Güssow, in 1911; St. John, in 1913; Erskine, in 1952; and Catling et al., in 1981. During the past 100 years, the island's flora has remained relatively stable; the number of native species has changed little and most plants introduced to the island have not become established.