Pollinator-mediated competition between two co-flowering Neotropical mangrove species, Avicennia germinans (Avicenniaceae) and Laguncularia racemosa (Combretaceae).

Background and Aims Three ecological relationships are possible between co-flowering plant species; they may have no effect on one another, compete for pollination services, or facilitate one another by attracting more pollinators to the area. In this study, the pollinator-mediated relationship between two mangrove species with overlapping flowering phenologies was investigated in one south Florida community. Methods Pollinator observations were recorded between 0900 h and 1700 h during June and July, 2008–2010. Insect visitation rates to Avicennia germinans and Laguncularia racemosa were estimated from 522 observation intervals of 10 min during three phenological time periods, when each species flowered alone and when they co-flowered. The number of timed intervals varied between years due to differences in flowering phenology, from four to 42 for A. germinans and from nine to 94 for L. racemosa. Key Results Avicennia germinans began flowering first in all years, and insect visitation rates were significantly greater to A. germinans than to L. racemosa (P<0·001). Flowers of both species received visits from bees, wasps, flies and butterflies; Apis mellifera was the most common floral visitor to both species. Visitation rates to L. racemosa increased significantly when A. germinans stopped flowering (P<0·001). However, there was no significant change in visitation rates to A. germinans after L. racemosa began flowering (P=0·628). Conclusions When they co-flowered, A. germinans outcompeted L. racemosa for pollinators. Laguncularia racemosa hermaphrodites self-pollinate autogamously when not visited by insects, so reduced visitation to L. racemosa flowers reduced the frequency of outcrossing and increased the frequency of selfing. Reduced outcrossing limits male reproductive success in this androdioecious species, which could lead to changes in the breeding system. The degree of overlap in flowering phenologies varied between years, so the effect on the mating and breeding system may differ between years. [ABSTRACT FROM AUTHOR]