The use of soil pollen to determine the sex of overhead individuals of a temperate dioecious shrub.

Abstract

In dioecious species, determining the sex of individual plants from one-time phenological observations is rarely feasible when some individuals capable of reproducing are not flowering or fruiting at the time of observation. Currently, sexing those individuals requires long-term phenological data on individuals and populations, but such data are rarely available or feasible to collect. We tested the hypothesis that differences in soil pollen concentrations beneath the crowns of female and male plants would exist and be sufficient to reliably determine the sex of the individual plant overhead in a dioecious species. We predicted that soil pollen concentrations beneath male plants would be significantly higher than beneath female plants because only males produce pollen and pollen should accumulate in the soil underneath the male plants over repeated flowering events. We collected samples from surface soil under both sexes of the insect-pollinated dioecious shrub, Aucuba japonica (Garryaceae). Pollen grains were present in surface soil in both Oe and A horizons, and mean pollen concentration under males was significantly higher than under females. Pollen concentrations beneath males were positively correlated with male plant height, potentially reflecting greater pollen production by larger individuals. Considering the small plant size and relatively low pollen production of A. japonica, this method may hold promise for sexing other dioecious species in the absence of direct phenological data. Our phenology-free and relatively low-cost method for sexing dioecious plants may be especially useful in tropical forests where many species are dioecious.