We hypothesized that Strychnos nux-vomica and Strychnos potatorum in seasonal tropical ecosystems have dormant desiccation-tolerant seeds, while those of Strychnos benthamii growing in aseasonal wet habitats have non-dormant desiccation-sensitive seeds. Germination, imbibition, the effect of gibberellic acid on germination and changes in the embryo to seed length ratio (E:S) during incubation were determined for the three species. Seed storage behaviour was identified with the hundred seed test. Time taken for epicotyl emergence was recorded. Radicle emergence of S. nux-vomica, S. potatorum and S. benthamii at 25°C under light/dark conditions (12/12 h) was completed within 76, 49 and 11 d, respectively. S. nux-vomica and S. potatorum seeds incubated on GA3 germinated to a higher percentage than non-treated seeds. E:S of S. nux-vomica, S. potatorum and S. benthamii had increased by 38.2, 34.5 and 25.5%, respectively, at radicle emergence. Shoot emergence of S. nux-vomica, S. potatorum and S. benthamii was observed after 76, 74 and 45 d from radicle emergence, respectively. Thus, it can be concluded that the seeds of all three species have epicotyl morphophysiological dormancy. Hundred seed tests revealed that S. nux-vomica and S. potatorum seeds were desiccation-tolerant, while those of S. benthamii were desiccation-sensitive. Our study showed that species from seasonal habitats (S. nux-vomica and S. potatorum) have desiccation-tolerant morphophysiologically dormant seeds, while those from an aseasonal habitat (S. benthamii) have desiccation-sensitive morphophysiologically dormant seeds, revealing that their dormancy and desiccation tolerance behaviour are adaptations to their environment.

Lianas are woody climbing plants that begin their life cycles as seedlings rooted in the ground, but eventually rely on other plants for physical support in order to reach the top of the forest canopy (Holbrook & Putz 1996, Putz & Holbrook 1991). Lianas can negatively affect plants they climb by competing with them for common resources such as light, water and nutrients, and by causing them direct physical damage (Dillenburg et al. 1995, Pérez-Salicrup & Barker 2000, Stevens 1987, Whigham 1984). Yet, there is little documentation about the size at which liana individuals of different species begin to climb on other plants in nature. This information is important because the size at which a liana begins to climb on other plants will determine when lianas potentially start physically affecting their supporting plants. Furthermore, although the growth of liana seedlings might be determined by light (Sanches & Válio 2002), the availability of support will also largely influence the rate of growth of liana stems (Peñalosa 1982, 1983, 1985). Thus, information about the size at which liana species find support in the forest understorey will be useful in understanding future growth of liana individuals.