Patterns of Allozyme Variation in Three Costa Rican Species of Piper

Abstract

Allozyme variation was studied in sympatric populations of three species of Piper (Piperaceae) in Parque Nacional Santa Rosa, northwestern Costa Rica. Levels of intrapopulational genetic diversity for all three species are low relative to other plants (proportion of polymorphic loci = 2/21 for P. amalago, 1/24 for P. jacquemontianum, and 0/20 for P. pseudo-fuligineum). Genotype frequencies at one polymorphic locus (b-Aatl in P. amalago) are significantly heterogeneous among subpopulations within the Santa Rosa population. Within subpopulations of P. amalago, heterozygote frequencies are consistent with the estimated effective outcrossing rate (0.58) at one locus (b-Aatl) but are in excess at the other locus (Pgi2). Nei's genetic distances between species are relatively high compared to other pairs of congeneric plants, ranging from 0.454 to 0.988. These high genetic distances are consistent with the idea that this large genus of flowering plants is relatively old. PATTERNS OF ALLOZYME-GENETIC VARIATION have been examined for a large number of plant species and have provided important insights into a number of evolutionary and taxonomic questions (Brown 1979, Hamrick et al. 1979, Gottlieb 1981, Loveless & Hamrick 1984). However, these studies are strongly biased toward herbaceous plants and temperate climates. Despite the immediate and practical need for increased understanding of the genetic structure and dynamics of tropical plant species (Frankel & Soule 1981, Schonewald-Cox et al. 1983), few such species have been assayed for levels and patterns of allozyme variation. In the present paper we present the results of a study of allozyme variation in three shrubby species of Piper (Piperaceae) which occur sympatrically within Parque Nacional Santa Rosa in northwestern Costa Rica. Our interest in this genus is motivated by two considerations. First, seed dispersal in Santa Rosa is mediated exdusively by frugivorous bats which may transport fruits relatively long distances (Heithaus & Fleming 1978). The generalization that gene flow is insufficient to prevent random genetic divergence of conspecific plant populations (Levin & Kerster 1974, Levin 1981) may not hold for such species (Loveless & Hamrick 1984), so it is of interest to ascertain whether spatial genetic structure is present within Santa Rosa. Second, Piper is one of the most speciose genera of flowering plants (Gentry 1982), suggesting it may be relatively old. The three species studied here are not considered to be closely related within the genus (Burger 1971), so it is of interest to determine whether the genetic distances among these species are consistent with a relatively ancient divergence. Thus we have addressed three specific questions in this study: (1) how much allozyme variation is present in these three species of Piper; (2) is there significant allozyme divergence between subpopulations within Santa Rosa; and (3) how do the genetic distances between these species compare with genetic distances between other congeners? MATERIALS AND METHODS Individuals of Piper amalago L., P. pseudo-fuligineum C.DC., and P. jacquemontianum Kunth were sampled from nine sites, or subpopulations, within Parque Nacional Santa Rosa, in northwestern Costa Rica during June and July of 1983 (Fig. 1, Table 1). Fleming (1985) has described the study site and natural history of these and two other sympatric species of Santa Rosa Pipers. Briefly, the three species investigated electrophoretically are common but patchily-distributed shrubs that require habitat disturbance to become established. Their flowers are visited by generalist, pollen-collecting insects, and exdosure experiments indicate that insect visitation is necessary for high seed set (Fleming, unpubl. data). Areas sampled ranged in size from 0.1 to 0.5 ha. Although connections between plants were not observed in the field, it is possible that some of the individuals sampled were ramets of the same done which are no longer connected. This would decrease the effective sample size per subpopulation and bias the sample toward larger dones. While this possibility exists, the electrophoretic results (below) reveal multiple genotypes per site for both P. amalago and P. jacquemontianum. The spatial distribution of genotypes among mapped individuals of P. amalago at several sites indicates that clones, if present, must be small relative to the area sampled (Heywood & Fleming, unpubl. data). Young leaves from established plants in the field served i Received 30 July 1984, revision accepted 23 April 1985. 208 BIOTROPICA 18(3): 208-213 1986 This content downloaded from 157.55.39.251 on Fri, 13 May 2016 05:20:25 UTC All use subject to http://about.jstor.org/terms