PARENTAL EFFECTS ON PROGENY PHENOTYPE IN PLANTS: DISTINGUISHING GENETIC AND ENVIRONMENTAL CAUSES.

The experimental measurement of additive genetic variation in plant populations is complicated by the potential for non-Mendelian inheritance. Maternal influences on progeny phenotype resulting from the cytoplasmic inheritance of plastids or RNA transcripts and effects of the maternal environment have consequently been the focus of much research. To exclude or to control for these sources of variation, breeding designs (e.g., cross-factored, nested, or diallel) in which genetically unrelated pollen donors are mated to maternal genotypes have been adopted. Using these designs, some empirical studies have detected statistically significant differences among pollen donors in the mean performance of their pollen (the mature male gametophytes) or in the mean phenotype of their progeny. These statistical effects of pollen-donor identity on pollen performance or progeny phenotype have frequently been interpreted as evidence for additive genetic variance among pollen donors, although patrilineal cytoplasmic inheritance or effects of the paternal environment on pollen performance or gene expression are recognized as alternative explanations. We note that environment-specific selection among developing gametophytes-in which the environment experienced by developing pollen grains (or ovules) provides a selective force causing the differential survival of gametophyte genotypes (analagous to meiotic drive)-is an additional process that may cause genetically based paternal (or maternal) effects on gametophyte performance. If genes selected during this process are expressed in the sporophyte (postfertilization), this process could also influence the phenotype of the diploid progeny. Here, we review the potential causes of statistically significant differences in mean phenotype among the gametophytes or progeny of maternal (seed-bearing) or paternal (pollen-donating) parental plants. We suggest an experimental approach that permits the detection or elimination of selection among developing gametophytes as one such cause. Specifically, the replication of homozygous parental genotypes within and across environments allows the detection and measurement of paternal and maternal environmentally induced effects on gametophyte or offspring phenotype, while eliminating meiotic drive as a source of the phenotypic variation.