Abstract
BACKGROUND AND AIMS: Physiological epicotyl dormancy in which the epicotyl elongates inside the seed before the shoot emerges has been reported for only a few tropical rainforest species, all of which are trees that produce recalcitrant seeds. In studies on seeds of Fabaceae in Sri Lanka, we observed a considerable time delay in shoot emergence following root emergence in seeds of the introduced caesalpinioid legumes Brownea coccinea and Cynometra cauliflora. Thus, our aim was to determine if seeds of these two tropical rainforest trees have physiological epicotyl dormancy, and also if they are recalcitrant, i.e. desiccation sensitive. METHODOLOGY: Fresh seeds were (i) dried to various moisture levels, and (ii) stored at -1 and 5 °C to determine loss (or not) of viability and thus type of seed storage behaviour (orthodox, recalcitrant or intermediate). To identify the kind of dormancy, we tested the effect of scarification on imbibition and monitored radicle emergence and epicotyl growth (inside the seed) and emergence. PRINCIPAL RESULTS: FRESH SEEDS OF BOTH SPECIES HAD HIGH MOISTURE CONTENT (MC): 50 % for C. cauliflora and 30 % for B. coccinea. Further, all seeds of C. cauliflora and the majority of those of B. coccinea lost viability when dried to 15 % MC; most seeds of both species also lost viability during storage at -1 or 5 °C. Intact seeds of both species were water permeable, and radicles emerged in a high percentage of them in <30 days. However, shoot emergence lagged behind root emergence by 77 ± 14 days in B. coccinea and by 38 ± 4 days in C. cauliflora. Further, plumule growth inside seeds of C. cauliflora began almost immediately after radicle emergence but not until ∼30-35 days in B. coccinea seeds. CONCLUSIONS: Seeds of both species are recalcitrant and have physiological epicotyl dormancy. The kind of physiological epicotyl dormancy in seeds of C. cauliflora has not been described previously; the formula is C(nd) (root)-[Formula: see text] (epicotyl).