Patterns of mating in Persoonia mollis (Proteaceae) revealed by an analysis of paternity using AFLP: implications for conservation

A detailed characterisation of mating in natural populations is desirable for the better conservation of rare or threatened taxa. The PCR-based DNA-fingerprinting technique amplified fragment length polymorphism (AFLP) is a powerful new genetic marker for mating system analysis because it enables the unambiguous assignment of paternity to progeny. Mating patterns following natural pollination were characterised by paternity assignment using AFLP in a natural population of Persoonia mollis subsp. nectens (Proteaceae). This study serves as a comparison for future studies of the mating system of the threatened close relative P. mollis subsp. maxima. Twelve seeds from each of 21 plants (252 seeds in total) were analysed for paternity. Of these, 199 were assigned paternity unambiguously to one of the known potential sires. The remaining 53 seeds were sired by plants outside the known population. Three seeds were presumably selfed as they possessed only maternal alleles, giving a population outcrossing rate of 98.8%. Realised pollen flow distances showed a leptokurtic distribution, with a mean of 24.7 m (s.e. = 2.0; range = 0-137 m). Mean realised pollen dispersal distance to and from each plant varied markedly from 0 to 57.4 m, and was largely influenced by the immediate density of plants. The seeds of all plants had multiple sires, with an average of 5.3 sires for 10 seeds (s.e. = 0.3; range = 2-8). Individual paternal success varied from one to 24 seeds sired (mean = 9.1; s.e. = 1.2). Individual maternal success varied from a percentage fruit set of 5.2 to 37.9 (mean = 18.3; s.e. = 2.1). Male and female reproductive success, measured as the number of seeds sired and percentage fruit set, respectively, was positively correlated. Mean genetic dissimilarity among all seeds (29.1% of 151 loci polymorphic; s.e. = 0.05; n = 30 876) was not significantly different from the mean genetic dissimilarity among all adults (29.5%; s.e. = 0.5; n = 300), and indicates an overall absence of inbreeding. The active management and conservation implications of these more sensitive data include the avoidance of inbreeding in established and new populations and in ex situ collections, as well as the sensitive detection of changes in mating in disturbed populations which may indicate future genetic decline.