High Degree of Multiple Paternity and Reproductive Skew in the Highly Fecund Live-Bearing Fish Poecilia gillii (Family Poeciliidae)

Abstract

Multiple paternity is a common phenomenon within the live-bearing fish family Poeciliidae. There is a great variety in brood sizes of at least two orders-of-magnitude across the family. However, little is known about the ramifications of this remarkable variation for the incidence and degree of multiple paternity and reproductive skew. Mollies (subgenus Mollienesia, genus Poecilia) produce some of the largest broods in the family Poeciliidae, making them an excellent model to study the effects of intra-specific variation in brood size on patterns of multiple paternity. We collected samples of the live-bearing fish Poecilia gillii from 9 locations in Costa Rica. We measured body size of 159 adult females, of which 72 were pregnant. These samples had a mean brood size of 47.2 ± 3.0 embryos, ranging from 4 to 130 embryos. We genotyped 196 field-collected specimens with 5 microsatellite markers to obtain location-specific allele frequencies. In addition, we randomly selected 31 pregnant females, genotyped all their embryos (N = 1346) and calculated two different parameters of multiple paternity: i.e., the minimum number of sires per litter using an exclusion-based method (GERUD) and the estimated number of sires per litter using a maximum likelihood approach (COLONY). Based on these two approaches, multiple paternity was detected in 22 and 27 (out of the 31) females, respectively, with the minimum number of sires ranging from 1 to 4 (mean ± SE: 2.1 ± 0.16 sires per female) and the estimated number of fathers ranging from 1 to 9 (mean ± SE: 4.2 ± 0.35 sires per female). The number of fathers per brood was positively correlated with brood size, but not with female size. Next, we calculated the reproductive skew per brood using the estimated number of sires, and found that in 21 out of the 27 multiply sired broods sires did not contribute equally to the offspring. Skew was not correlated with either female size, brood size or the number of sires per brood. Finally, we discuss several biological mechanisms that may influence multiple paternity and reproductive skew in P. gillii as well as in the Poeciliidae in general.