Limitation to reproductive performance in the dioecious endemic shrub Oreoherzogia glaucophylla (Sommier) W.Vent (Rhamnaceae)

Dioecy is a relatively rare sexual system occurring in approximately 7–10% of angiosperm species (1). As the consequence of separate sexes is the impossibility of selfing, the avoidance of inbreeding has traditionally been invoked as an important selective force in the evolution of dioecy (2). However, other correlations of dioecy with several ecological and life-history attributes have been found, including tropical distribution, woody growth form, abiotic pollination, small inconspicuous flowers and fleshy fruits (3). Despite the multiple ecological roles that dioceous species play in the landscape, experimental studies of their reproduction have not received considerable attention compared to hermaphroditic plants (4, 5). Here, we examined the effect of ecological and population traits (including topography, pollination environment, neighbouring competition and plant size) on reproductive performance of Oreoherzogia glaucophylla (Sommier) W.Vent, a dioceous shrub endemic to Apuan Alps and a restricted range of Tuscan Apennine (6, 7). To this end, 60 females, designated as focal females, were randomly selected in two contrasting environments located at the extremes of its altitudinal range for monitoring their reproductive performance between April and September 2015. Generalised linear mixed models revealed that elevation, aspect and plant size have no significant effect on fruit set. On the other hand, the distance to nearest male resulted the single most important factor affecting fruit set, with number of fruits significantly decreasing with increasing distance to nearest male. Neighbouring competition (presence of other females) and floral competition (number of flowers within the same inflorescence) also negatively affect the fruit set. By contrast, fruit set was not limited by all the above mentioned factors at the low altitude site, where they might be overridden by other undetected factors. In both sites, seed set was unaffected by all considered factors. That is to say, in any single flower, ovules fecundation is not limited by pollen load in the light of the low number of ovules (four) per ovary. Overall, the results of this study support the importance of pollen limitation on fruit set in O. glaucophylla, however, the effect of year on resource availability cannot be ruled out, especially on seed quality and viabilit