JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecology. Abstract. We examined the changes in leaf phenolic chemistry and insect herbivory from saplings of two temperate deciduous species, Liriodendron tulipifera (tulip poplar) and Cornusflorida (dogwood), planted in five microenvironments in Gilmer County, Georgia, USA. The experimental design permitted comparisons between saplings grown in an open field, under shade cloth within the field, on the edge between field and forest, in forest understory, and within canopy gaps established within the forest. Half of the trees in each microenvironment were fertilized. Leaves from each tree were sampled at the end of the growing season in 1989 and 1990 and analyzed for toughness, percent dry mass, total phenolics, hydrolyzable tannins, condensed tannins, and insect herbivory (percent leaf area damaged).The shade-tolerant dogwood saplings contained higher levels of total phenolics and hydrolyzable tannins than the shade-intolerant tulip poplar saplings. Dogwood generally had lower levels of herbivory. These results support earlier studies suggesting that slowgrowing, shade-tolerant species tend to have higher levels of phenolics and experience lower levels of herbivory than fast growing, shade-intolerant species. However, dogwood leaves contained lower levels of condensed tannins and were as tough as tulip poplar leaves.Sunlight availability had a significant positive influence on levels of phenolics in both species. Leaf phenolics generally increased with greater insolation from forest to field and when sunlight was greater within field or forest habitats. However, the levels of tannins in dogwood saplings only dropped significantly in the deep shade of the forest. The similar levels of dogwood phenolics in most microenvironments are indicative of the relatively high photosynthetic efficiency of this species in reduced light environments. Overall, these results are consistent with carbon/nutrient balance theory that predicts trade-offs in the allocation of photosynthate from defense to growth as light declines.Levels of insect herbivory and total phenolics were inversely related for dogwood. However, the relationship with tannins was less apparent. Herbivory on tulip poplar was unrelated to changes in phenolics, possibly reflecting the greater chemical diversity of that species. Fertilization increased the biomass of both species, but had no apparent influence on levels of leaf phenolics or insect herbivory. The lack of a fertilization effect was unexpected in light of previous suggestions that fertilization results in reduced phenolics and increased herbivory.

Four sizes of forest opening (0.016, 0.08, 0.4, 0.4, and 2.0 ha; two replicates each) were established in a Southern Appalachian forest to examine the effects of disturbance size on earl successional community structure and function. Solar radiation, soil temperature, and air temperature were all higher in large openings than small openings and increased from edge to center of disturbance patches. Aboveground net primary productivity (NPP) was 3—4 times as highe in larger (2.0 ha) as small (0.016 ha) openings, presumably in response to greater light availability in large patches. Stump and root sprouts of tree species accounted for the largest fraction of NPP in all patch sizes. Herbs, vines, shrubs, advance regeneration trees, and tree seedlings had progressively smaller NPP, respectively. Vegetation biomass reached 0.7—2.6% of undistributed forest levels and aboveground NPP reached 17—58% of forest levels by the 2nd yr after cutting. Plant species richness was generally higher in large than small patches. Tree species composition shifted considerably followed disturbance. Liriodendron tulipifera was important before and after logging. Large canopy dominants such as oaks and hickories were relatively unimportant sources of sprouts during early revegetation. Instead, minor canopy and understory species such as Robinia pseudoacacia, Halesia carolina, Acer rubrum, Cornus florida, and Magnolia fraseri were the major sprouters in all patch sizes. The N—fixing black locust (Robinia) was much more important in large than small openings. Disturbance size within the Southern Appalachians thus affects microenvironment, species composition, and NPP during early revegetation.

We examined the seasonal changes in plant chemical defenses of key tree species (Robinia pseudoacacia, Liriodendron tulipifera, Acer rubrum, and Cornus florida) regenerating during early succession in different patch sizes within Southern Appalachian forests. Trees of each species were sampled from five sizes of forest openings (0.016, 0.08, 0.4, 2.0, and 10 ha; two replicates each) and from the understory of surrounding forests. Leaves of rapidly growing, pioneer species (Robinia and Liriodendron) were slightly less tough and contained lower levels of hydrolyzable tannins and total phenolics, but higher condensed tannin concentrations than the slower growing, more shade—tolerant species (Acer and Cornus). These results generally support earlier findings indicating the evolutionary development of greater levels of constitutive phenolics in slow—growing than fast—growing species. The differential use of condensed and hydrolyzable tannins by slow— and fast—growing species may reflect adaptive trade—offs in the allocation of different types of phenolics. Regenerating trees often had tougher leaves and contained higher levels of phenolics in large than small openings by midsummer. The differences in phenolics were more pronounced for condensed tannins and when comparing the smallest canopy openings with larger patch sizes. Plant phenolics were also generally lower in understory trees than in conspecific sprouts in openings. These findings may reflect phenotypic carbon/nutrient adjustments as light availability changes across the patch—size gradient. The greater sunlight in large than small openings apparently promotes excess carbon production and a metabolic buildup of phenolic compounds. Independent measures of phenolic compounds peaked at different times during the growing season. Hydrolyzable tannins reached peak levels quite early whereas condensed tannins increased later in the growing season. Total phenolics peaked in midsummer. Herbivore damage was consistently high (5—9%) on Robinia sprouts in all openings. Insect herbivory on the remaining three species (1—5%) was significantly lower in large than small patch sizes. Tree species regenerating in the high sunlight of larger openings in Southern Appalachian forests thus tend to have tougher leaves, contain greater phenolics, and experience less herbivore damage than conspecifics in small openings.

Five sizes of canopy openings (0.016 ha to 10 ha) were established in the Southern Appalachian Mountains in early 1982 to examine the initial patterns of plant and arthropod establishment across a size range of forest disturbances. Vegetation standing crop after the first growing season was considerably higher in large than small openings in apparent response to greater resource release (e.g., sunlight) in larger openings. Woody stump and root sprouts were the dominant mode of revegetation in each patch size. Forest dominants such as Quercus rubra, Q. prinus and Carya spp. were less important as sprouters in openings than several minor forest components (e.g., Robinia pseudo-acacia, Acer rubrum, Halesia carolina and Cornus florida). Arthropod abundance and community composition varied across the size range of forest openings. Arthropods from the surrounding forest readily utilized the smallest canopy openings (0.016 ha). All feeding guilds were well represented in these small openings and herbivore biomass and load (mg of herbivores/g of foliage) were much higher than in larger patches. In contrast, arthropod abundance and species richness were significantly lower in mid-size than smaller patches. The relatively sparse cover and high sunlight in mid-size openings may have promoted surface heat buildups or soil surface/litter moisture deficits which restricted arthropod entry from the surrounding forest. Arthropod abundance and species richness were higher in large than mid-size patches. The greater vegetation cover in larger openings may have minimized the deleterious effects on arthropod populations. However, the absence of population increases among these arthropod species maintained herbivore loads at very low levels in large patches. Our results suggest that arthropod abundance and diversity in sprout-dominated forest openings are highly dependent on the extent of environmental differences between patch and surrounding forest.

A shallow, 5.4—ha dystrophic Carolina bay wetland was studied between 1974 and 1978 to test the premise that biomass and production are constrained by the stagnant hydrology and dilute, acidic chemistry of bay wetlands. Our objectives were to evaluate: (1) surface and subsurface hydrology, (2) sources of production, (3) community change along a depth gradient, and (4) seasonal community patterns. The hydrology study compared surface water levels to groundwater levels in four adjacent wells. A cylinder enclosure and total harvest procedure and 24—h dissolved oxygen curves were used for 2 yr in a 1—ha sampling area to measure spatial and temporal biomass patterns and organic production of community components. Surface hydrology was dependent on seasonal and annual precipitation patterns. Lateral, episodic groundwater exchanges accounted for the dilute surface chemistry and apparent differences in surface gains and losses. Substrate exposure and fire oxidation in dry years promoted low detritus standing crops and a dark, clay—loam mineral soil. A depth gradient, marked by a peripheral grass/sedge zone and central water—lily zone, produced strong spatial patterns for most community components. Aquatic macrophytes had low shoot biomass and a high root to shoot ratio. Algae were light limited and even less productive. Purple photosynthetic bacteria approached algal productivity levels and may have important chemistry and trophic support functions. As predicted from the stagnant, dystrophic conditions, autotroph biomass and production were low. However, the warm climate and periodic fire rejuvenation may stimulate higher production than that generally found in northern bog wetlands. Secondary production was also low. Animal biomass was dominated by insects (especially odonates) and salamanders. High turnover of prey (largely midge larvae and microcrustaceans) was implied by the predator—dominated community structure. This study of a wetland ecosystem lacking several common dystrophic bog features (peat deposit, high dissolved organic matter, acidophilic mosses) enlarges our perspective of the dystrophic condition.