Parentage analysis in plants: Mating systems, gene flow, and relative fertilities

This chapter deals with the reconstruction of mating and dispersal processe s in natural and experimental plant populations. Within the past fifteen ye ars, the study of mating and dispersal processes in plants has been greatly expanded due to (i) the development of methods for assessing parentage of individual progeny; and (ii) an increase in the diversity and availability of molecular genetic markers. Most attention has been focused on the use of maternal progeny arrays to estimate shortterm rates of pollen gene flow be tween populations and to examine patterns of pollen dispersal and male fert ility variation within populations. The earliest studies relied solely on e xclusion methods, but it was almost immediately realized that many gene flo w events were not being detected, and that unique parentage could not be de termined for a large proportion of the non-immigrant offspring in most natu ral populations. As a consequence, methods for maximum-likelihood estimatio n of gene flow and male fertilities have been introduced. Less effort has b een devoted to the inference of parentage using dispersed seeds and seedlin gs, but all of the currently available models can be adapted to this situat ion as well. I begin this chapter with a short review of single-locus and m ultilocus methods for estimating the proportions of self-fertilization and outcrossing events for individuals and populations. This provides a motivat ion for the broader question of parentage analysis, focusing first on the u se of multilocus data to estimate levels of pollen gene flow. I then review the methods available for estimating relative fertilities of individual pa rents within populations, focusing on the strengths and weaknesses of each method. I conclude with some statements about currently unanswered problems with parentage models and some suggestions about novel or insufficiently e xplored uses to which parentage analysis may be put in the future.