No abstract

Impacts of felling, mulching, and burning on budgets of C, N, S, P, K, Ca, and Mg; rates of C0 2 evolution from the soil; soil seed storage; and plant growth were evaluated. The felled tropical evergreen forest was 8-9 yr old, interspersed with patches of 70-yr-old forest and had a leaf area index of 6 and aboveground biomass of 5.2 kg/rn 2 • Harvest of the largest trees removed 18% of the S, and > I 0% of all other elements except N. During the 11-wk mulching and drying period, 33% of the K and 13% of the P disappeared, but losses of other elements were <6%. Burn temperatures were >620°C in surface fuels, but soil temperatures were seldom > l00°C at I ern or >50°C at 2 ern. The burn volatilized 1600 g/rn 2 C, 49 g/rn 2 N, and 13 g/rn 2 S. Postburn wind and water erosion of ash, plus leaching, removed 34 g/rn 2 N, 20 g/rn 2 K, I g/ rn 2 P, 39 g/rn 2 Ca, and 7 g/rn 2 Mg, but insignificant amounts of C and S. After the burn and onset of the rains, 57% of the initial amount of N and 39% of the initial C still remained because of conservation of the organic-rich upper 3 ern of soil.Soil C0 2 evolution was greater from beneath the 11-wk-old slash (3.6 gC· rn-2 • d-1 ) than from beneath the forest (2.5 gC · rn-2 • d-1 ), probably because the slash conserved soil moisture better than the actively transpiring forest. After the burn both the burned field and forest soil evolved C0 2 at =4.5 gC· rn-2 ·d-•. At this rate, 154 d of decomposition and respiration would release as much C into the atmosphere as did the burn.Soil seed storage dropped from =8000 seeds/rn 2 (67 species) in the forest, to 6000 seeds/rn 2 (51 species) after II wk of mulching, to 3000 seeds/rn 2 (37 species) after the burn. The seeds not killed by the burn, the survival of mycorrhizal fungi, and the release of nutrients resulted in vigorous and diverse postburn regrowth.

Survival and height growth of tree seedlings and rooted cuttings introduced into artificially shaded and unshaded plots in a degraded dry forest were measured at intervals for nine months. Ten tree species were selected to represent a range of ecological characteristics of the dry–forest plant community on St. John, U.S. Virgin Islands. Of three propagule types – seeds, seedlings, and rooted cuttings – introduced to field plots, seedlings survived best (52%) over the initial nine‐month period. Cuttings of six species rooted successfully in a shadehouse, but only two of these species survived the nine–month field experiment. Seed germination was low, under 11% for eight of ten species tested, and four species did not germinate. Subsequent mortality of seedling recruits was moderately high. Plumeria alba was the only species for which seedling height growth was not significantly greater than cutting height growth. Shading treatment (25% of full sun) significantly increased seedling survivorship (p= 0.03) but suppressed growth slightly for some species. Shading enhanced survival of seedlings produced from broadcast seeds, but not seed germination. Mortality occurred during dry periods, apparently from drought stress. Results suggest (1) that seedling introductions are the preferred propagule type (over seeding or rooted cuttings) for ecological restoration of degraded tropical dry forests, and (2) that some level of shading is required to increase the survivorship of many dry‐forest species or to avert complete mortality of some species. This study suggests that early secondary dry forest may be best restored by underplanting within the existing vegetation. Sufficient shading suitable for growth of native dry‐forest trees may be attained using a nurse crop of fast‐growing leguminous trees.

To investigate the relationship between herbivory and floristic complexity, we measured losses to herbivores in four 0.1— to 4—yr—old tropical ecosystems: (1) unmanipulated successional vegetation, (2) successional vegetation with higher plant diversity than the unmanipulated succesion, (3) an ecosystem of investigator—controlled species composition, designed to imitate the physiognomy and species richness of the successional vegetation, and (4) monocultures of maize, cassava, and Cordia alliodora. We measured herbivory rates (loss of leaf area per day) on dominant plant species in each system and aggregated over species to estimate rates for plant communities. Although herbivory rates varied widely among species, losses to herbivores in terms of mass of leaf tissue lost per unit of ground area were approximately equal in the four systems, 71.5 to 78.5 g°m—2. yr—1. Ecosystems with greater plant species richness lost a lower proportion of available leaf area and exhibited lower temporal variability in herbivory. Species—rich ecosystems had relatively constant, predictable rates of herbivory due to counterbalancing of low rates on some species with high rates on others. The rate of herbivory on any species was strongly influenced by the nature of the surrounding vegetation. Although surrounding vegetation often conferred protection upon potential pest targets, in some cases a plant species experienced increased susceptibility to herbivores through association with other species.