Some populations of white mangrove (Combretaceae: Laguncularia racemosa) are androdioecious, a rare breeding system with two breeding types: plants with male flowers and plants with hermaphroditic flowers. Theoretically, male plants require at least a two-fold fitness advantage relative to hermaphrodites to be maintained in androdioecious populations. If males flower earlier than hermaphrodites, they may attract more pollinators and sire proportionately more seedlings early in the flowering season. Because white mangrove is semi-viviparous (seeds germinate and seedlings continue to grow inside the fruits until dispersal), seedlings sired earlier in the season will develop for a longer period. These seedlings may receive greater maternal investment and be larger, which commonly results in greater seedling success. To test these hypotheses, flower production was estimated for male and hermaphrodite plants in one Florida population during June, July, and August. Pollinator visitation rates were estimated for the two breeding types in June and July. Open-pollinated flowers were marked and assigned to one of four treatments based on month and rachis position (June basal, July distal, July basal, or August distal). Marked flowers were covered in August, and fruits were collected in November. Fruits were weighed before planting in mudflat conditions to estimate seedling emergence success.

Results/Conclusions

Results show that males flowered earlier, with 85% of the male plants flowering in the June census versus only 39% of the hermaphrodites. Peak flower production for both male and hermaphrodite plants occurred in July. Male plants produced significantly more flowers on average than hermaphrodites throughout the season, and the relative ratio of male versus hermaphrodite flowers produced per plant increased over time (4:1–20:1). Pollinator visitation rates to males and hermaphrodites were lower in June than in July. June basal fruit mass was significantly lower than July basal fruit mass, but fruit mass did not differ significantly among other treatments. Fruit mass differed significantly among trees. Greater fruit mass was generally correlated with seedling emergence, but the relationship was not significant. Seedling emergence differed significantly among months, but the pattern was opposite of the prediction: seedlings emerged most frequently from August fruits and least frequently from June fruits. One possible explanation is that because pollinator visitation rates were lower in June than in July, many of the early fruits may have developed from self-pollinated flowers. If so, inbreeding depression could explain the lower seedling emergence from fruits produced earlier in the season. This will be tested in future studies.