Pierre-Michel Forget

SEED FATE : PREDATION, DISPERSAL AND SEEDLING ESTABLISHMENT

Effects on plant demography - a temperate perspective The fate of seed banks: factors influencing seed survival for light-demanding species in moist tropical forests Different plant species, seasons and spatial distributions Seed predation, seed dispersal and habitat fragmentation Selection, predation and dispersal of seeds by tree squirrels in temperate and boreal forests Effects of large seeds on rodent populations Diplochory and the evolution of seed dispersal The role of dung beetles as secondary seed dispersers Observing seed removal: remote video monitoring

Evidence for secondary seed dispersal by rodents in Panama

SummaryThe data presented show thatVirola nobilis (Myristicaceae), a bird/mammal-dispersed tree species in Panama, may also be dispersed by a terrestrial rodent, the agouti (Dasyprocta punctata). Using a thread-marking method, we observed that agoutis scatterhoardedV. nobilis seeds that they found both singly or in clumps. Seed removal and seed burial rates were strongly affected by features of forest habitats, such asV. nobilis tree richness (rich vs poor) and/or forest age (old vs young), but not by seed dispersal treatment (scattered vs clumped). Predation (mostly post-dispersal) of unburied seeds by weevils was independent of habitat and dispersal treatment. Seeds artificially buried in aVirola-rich area were more likely to escape predation and become established than unburied seeds under natural conditions. The food reward for agoutis is in the germinating seedlings. The seed dispersal syndrome ofV. nobilis involves long- and short-distance dispersers which both appear important for tree recruitment.

Seed-dispersal of Vouacapoua americana (Caesalpiniaceae) by caviomorph rodents in French Guiana

A possible mutualistic dispersal system between a large-seeded tree of French Guiana, Vouacapoua americana (Caesalpiniaceae), and caviomorph rodents, Myoprocta exilis and Dasyprocta leporina , is described. Mast fruiting of Vouacapoua at the beginning of the wet season coincides with scatter-hoarding seed dispersal. During the wet season, almost 100% of marked seeds on three sites were removed: nearly 70% were buried and the rest were eaten by mammals. Unburied seeds were attacked by insects and/or lost their ability to germinate. Rodents preferred ungerminated seeds, and had no interest in germinated seeds. Seeds were buried individually near natural objects such as palms, branches, logs, lianas, roots and trees. After predation by rodents, seedling distribution did not differ from seed distribution. Most seeds were transported less than 5 m from the feeding plots but some were carried as far as 22.4 m. Between 40 and 85% of dispersed seeds were retrieved during the following month by rodents and eaten. The disinterest of caviomorph rodents in germinated seeds, because of rapid exhaustion of endosperm reserves, prevents feeding from hoarded Vouacapoua during the long dry season when resources are scarce. Seedlings emerging from forgotten or abandoned cached seeds appear to increase the recruitment of Vouacapoua americana .

Meta-Analysis of the Effects of Human Disturbance on Seed Dispersal by Animals

Animal-mediated seed dispersal is important for sustaining biological diversity in forest ecosystems, particularly in the tropics. Forest fragmentation, hunting, and selective logging modify forests in myriad ways and their effects on animal-mediated seed dispersal have been examined in many case studies. However, the overall effects of different types of human disturbance on animal-mediated seed dispersal are still unknown. We identified 35 articles that provided 83 comparisons of animal-mediated seed dispersal between disturbed and undisturbed forests; all comparisons except one were conducted in tropical or subtropical ecosystems. We assessed the effects of forest fragmentation, hunting, and selective logging on seed dispersal of fleshy-fruited tree species. We carried out a meta-analysis to test whether forest fragmentation, hunting, and selective logging affected 3 components of animal-mediated seed dispersal: frugivore visitation rate, number of seeds removed, and distance of seed dispersal. Forest fragmentation, hunting, and selective logging did not affect visitation rate and were marginally associated with a reduction in seed-dispersal distance. Hunting and selective logging, but not fragmentation, were associated with a large reduction in the number of seeds removed. Fewer seeds of large-seeded than of small-seeded tree species were removed in hunted or selectively logged forests. A plausible explanation for the consistently negative effects of hunting and selective logging on large-seeded plant species is that large frugivores, as the predominant seed dispersers for large-seeded plant species, are the first animals to be extirpated from hunted or logged forests. The reduction in forest area after fragmentation appeared to have weaker effects on frugivore communities and animal-mediated seed dispersal than hunting and selective logging. The differential effects of hunting and selective logging on large- and small-seeded tree species underpinned case studies that showed disrupted plant-frugivore interactions could trigger a homogenization of seed traits in tree communities in hunted or logged tropical forests.

Seed removal and seed fate in Gustavia superba (Lecythidaceae)

The fate of the large (mean 12.6 g) seeds of Gustavia superba, an understory tree, was analyzed on Barro Colorado Island (BCI), Panama. Eight hundred thread-marked seeds were placed on the ground at two contrasting 1-ha forest sites (Gustavia-rich vs Gustavia-poor) during two periods (June and July) of the fruiting season of G. superba. These months correspond to maximum food availability on BCI. On average, 85.5 percent of the seeds were removed within 28 days, 47.5 percent and 3.8 percent of them being found scatterhoarded (buried) by agouti (Dasyprocta punctata) and gnawed by rodents, respectively, within 10 m of their origins. The effect of site and the interaction of site with month significantly affected seed removal rates, but not scatterhoarding rates. Proportions of seeds removed were greater where G. superba trees were rare. Because freshly fallen seeds were not infested by bruchid weevils, unburied and germinating seeds provide an abundant short-term food supply for terrestrial mammals. Burying seeds allowed agoutis to later consume cotyledons of germinating seeds from mid-August throughout October, when food is scarce on BCI. Overall seed dispersal effectiveness (% seed dispersal multiplied by % seedling survival) contrasts dramatically between forest sites, being 10.1 percent and 0.75 percent at Gustavia-rich and Gustavia-poor areas, respectively. This result suggests that predator-disperser satiation occurred, maybe due to greater G. superba seed availability and other alternative food supply, allowing greater G. superba seedling survival at the Gustavia-rich area.

Above-ground biomass and productivity in a rain forest of eastern South America

Abstract: The dynamics of tropical forest woody plants was studied at the Nouragues Field Station, central French Guiana. Stem density, basal area, above-ground biomass and above-ground net primary productivity, including the contribution of litterfall, were estimated from two large permanent census plots of 12 and 10 ha, established on contrasting soil types, and censused twice, first in 1992?1994, then again in 2000?2002. Mean stem density was 512 stems ha?1 and basal area, 30m2 ha?1. Stem mortality rate ranged between 1.51% and 2.06% y?1. In both plots, stem density decreased over the study period. Using a correlation between wood density and wood hardness directly measured by a Pilodyn wood tester,we found that the mean wood densitywas 0.63 g cm?3, 12% smaller than the mean of wood density estimated from the literature values for the species occurring in our plot. Above-ground biomass ranged from 356 to 398Mgha?1 (oven-dry mass), and it increased over the census period. Leaf biomass was 6.47Mg ha?1. Our total estimate of aboveground net primary productivity was 8.81 MgC ha?1 y?1 (in carbon units), not accounting for loss to herbivory, branchfalls, or biogenic volatile organic compounds, whichmay altogether account for an additional 1MgC ha?1 y?1. Coarse wood productivity (stem growth plus recruitment) contributed to 4.16 MgC ha?1 y?1. Litterfall contributed to 4.65MgC ha?1 y?1 with 3.16 MgC ha?1 y?1 due to leaves, 1.10 MgC ha?1 y?1 to twigs, and 0.39MgC ha?1 y?1 to fruits and flowers. The increase in above-ground biomass for both trees and lianas is consistentwith the hypothesis of a shift in the functioning of Amazonian rain forests driven by environmental changes, although alternative hypotheses such as a recovery from past disturbances cannot be ruled out at our site, as suggested by the observed decrease in stem density. Key Words: above-ground biomass, carbon, French Guiana, net primary productivity, tropical forest

Phylogenetic density dependence and environmental filtering predict seedling mortality in a tropical forest

Negative density dependence (NDD) and environmental filtering (EF) shape community assembly, but their relative importance is poorly understood. Recent studies have shown that seedlings mortality risk is positively related to the phylogenetic relatedness of neighbours. However, natural enemies, whose depredations often cause NDD, respond to functional traits of hosts rather than phylogenetic relatedness per se. To understand the roles of NDD and EF in community assembly, we assessed the effects on seedling mortality of functional similarity, phylogenetic relatedness and stem density of neighbouring seedlings and adults in a species-rich tropical forest. Mortality risks increased for common species when their functional traits departed substantially from the neighbourhood mean, and for all species when surrounded by close relatives. This indicates that NDD affects community assembly more broadly than does EF, and leads to the tentative conclusion that natural enemies respond to phylogenetically correlated traits. Our results affirm the prominence of NDD in structuring species-rich communities.

Predation by Rodents and Bruchid Beetles on Seeds of Scheelea Palms on Barro Colorado Island, Panama

Predation by rodents on seeds of Scheelea zonensis was studied on Barro Colorado Island, Panama, during the latter half of this palms fruiting season. The number of Scheelea bearing fruit declined steadily during this period. The proportion of both intact and bruchid-infested seeds among those accumulated under fruiting Scheelea declined, while the proportion of gnawed seeds increased. By early October, 87 percent of the Scheelea seeds sampled had been gnawed by rodents, and few intact or bruchid-infested seeds remained below parent palms. From August through October, agoutis, Dasyprocta punctata, visited Scheelea less and less frequently, while visits by squirrels, Sciurus granatensis, did not dedine. To learn the fate of late-falling Scheelea seeds, we placed small piles of thread-marked seeds, each similar to a small Scheelea fruit crop, in the forest during three successive months, a total of 400 seed each month. Of these 1200 thread-marked seeds, 57.2 percent were retrieved within 20 m of their piles 7 days after placement. Of the retrieved seeds, 22.2 percent were gnawed within 1 m of their pile, 70.6 percent were carried between 1 and 20 m before gnawing, and 7.2 percent were cached (not gnawed, but buried in the ground or covered with litter). More seeds were gnawed, and fewer cached, in September and October than in August. On Barro Colorado Island, intense seed predation by rodents on Scheelea correlates with a previously documented forestwide low in availability of fruits of all kinds, and with a decline in egg laying by bruchids.

Causes and Consequences of a Tropical Forest Gold Rush in the Guiana Shield, South America

Abstract Statistical and spatial analyses of both historical time series and remotely sensed data show a link between the spatial distribution and growth of gold production across the Guiana Shield in northeast Amazonia. Results indicate that an exponential rise in production across an expanding area is primarily a delayed response to the 1971−1978 market flotation of international gold prices. The subsequent 10-fold (2-fold) average nominal (real) price increase has provided a compelling economic incentive to mass exploitation of lower-grade gold deposits. The ground-based and remotely sensed distributions of mining activity are strongly attached to these deposits that dominate the regions gold geology. The presence of these gold-bearing formations in conservation and sustainable timber zones has sparked social conflict and environmental degradation across the region. Left unmanaged, more than a quarter-million square-kilometer area of tropical forest zoned for protection and sustainable management could ultimately be compromised by the price-driven boom in gold mining through poorly integrated resource use planning, lack of reclamation effort, and control of illegal operations. Serious public health issues propagated through the unregulated mining environment further erode the financial benefits achieved through gold extraction. This study demonstrates in part how international economic policies successfully stabilizing more conspicuous centers of the global economy can have unintended but profound environmental and social impacts on remote commodity frontiers.

Pre- and post-dispersal seed predation in Tachigali versicolor (Caesalpiniaceae): effects of timing of fruiting and variation among trees

Fruiting phenology, habitat types and proximity to conspecific fruiting adults may influence the degree of escape from pre- and post-dispersal seed predators. Successful predator escape by means of satiation is considered to be especially important for selection of masting and monocarpic reproduction in trees, such as exhibited by Tachigali versicolor, a tropical canopy tree. How pre- and post-dispersal predation rates varied with dispersal timing and among trees in T. versicolor was examined in young and old forests on Barro Colorado Island in Panama during a 4-mo period. Seeds were collected from above-ground traps to assess predispersal predation by bruchid beetles, and from quadrats on the ground to record predation and removal by terrestrial mammals. Proportion of seeds aborted varied greatly among trees (range 6-30 %, mean 16 %), and was especially high for trees on the edge of the island with fruiting conspecifics nearby during the early part of fruiting season. The proportion of seeds killed by bruchid beetles varied less among trees (14-25 %, mean 20 %), and remained constant throughout the fruiting season. Seeds on the ground were attacked mostly by rodents, and possibly by deer (26 % of all seeds and 43 % of intact dispersed seeds). The post- dispersal predation level was higher in the young forest than in the old forest (61 and 26 % of intact dispersed seeds, respectively), and was unaffected by the prox- imity of fruiting conspecifics. Temporal satiation of seed predators was evident only for post-dispersal mammalian predators in the old forest.