Germination Response in a Large-Seeded Neotropical Tree Species, Gustavia superba (Lecythidaceae)

Abstract

This six-month study examined percent germination, days until germination, and early seedling survival of Gustavia superba (Lecythidaceae) through field experiments in a tropical moist forest in Panama. The study had three goals: (1) to examine whether germination is a factor determining density of seedlings at three forest sites with different ages; (2) to test whether burial of seeds affects germination; and (3) to test if microsite conditions associated with light gap, gap-edge, or forest understory influence germination. The experiment, which was designed to fit a three-way ANOVA model, used three study sites, three microsites, two burial treatments, four replicates and 25 seeds in each replicate (1800 seeds total). Although seedling densities of G. superba at the forest sites varied inversely with age of forest, there was no evidence that germination success contributes to this pattern. Results indicated that this species germinates readily under most conditions: the overall mean was 85.4 percent and only six of 72 plots had less than 70 percent germination. Germination percent was reduced and mean number of days until germination was increased in the largest light gap. However, percent germination and days until germination did not differ among sites when only seeds in the understory microsite were compared. Burial of seeds and microsite conditions had minor effects on germination. Seedling mortality occurring soon after germination was not large (5.7%). It is concluded that G. superba is not a tropical species whose seeds germinate in response to a light gap but its seeds do germinate successfully under a range of microsite conditions. ALTHOUGH PLANT POPULATION SIZES can be affected by mortality at many stages, the seed to seedling transition generally is considered to be one of the most vulnerable. Mortality agents include pathogens, insect and vertebrate predators of seeds, as well as abiotic factors affecting germination levels. All these factors are intertwined in affecting seedling establishment. As a result, controlled experiments are required to identify the separate contribution of each factor. The focus of this paper will be to discuss the role of germination per se in seedling establishment of Gustavia superba (Kunth) Berg (Lecythidaceae), a