Morph-specific mating patterns in a low-density population of a heterodichogamous tree, Juglans ailantifolia

Abstract

Heterodichogamy, including protandrous (PA) and protogynous (PG) morphs, is considered a mechanism to avoid selfing and promote disassortative mating. Although morphotypes are usually present in a population at a 1:1 ratio, this ratio may be biased in a low-density population by demographic stochasticity, resulting in a deficiency of mating partners in a neighbourhood dominated by a single morph. In this study, we determined morph ratio of the heterodichogamous tree species, Juglans ailantifolia by observing flowering in a low-density population during 2 years. The morph ratio (PG: PA) of 2.56:1 deviated significantly deviated from 1:1. We genotyped 59 reproductive trees and 405 offspring derived from eight PG-mother and three PA-mother trees with 11 microsatellite markers. Paternity analysis was conducted to clarify the effects of mother morph on the proportion of intra-morph mating. Then, we applied the Bayesian mixed effect mating model (MEMM) to clarify mating system, pollen dispersal, and individual fecundity of PG- and PA-mother trees. We found that the selfing rate and the distance of pollen dispersal were not clearly different between PG- and PA-mother trees. In contrast, the proportion of intra-morph mating was higher in the majority-morph (PG) mother trees than in the minor-morph (PA) mother trees. The MEMM indicated that mean dispersal distance of PG-mother trees was larger than that of PA-mother trees with large variance. Furthermore, we observed individuals with unusually high intra-morph fecundity for majority-morph (PG) trees. These findings indicate that intra-morph mating may occur when majority-morph mothers suffer a deficiency of potential inter-morph mates.