Behav Ecol Sociobiol (1995) 37:32%335 (() Springer-Verlag 1995 Benjamin P. Oldroyd • Adam J. Smolenski Jean-Marie Cornuet • Siriwat Wongsiri • Arnaud Estoup Thomas E. Rinderer • Ross H. Crozier Levels of polyandry and intracolonial genetic relationships in Apis florea Received: 4 January 1995/Accepted after revision: 18 March 1995 Abstract DNA was extracted from worker and drone pupae of each of five colonies of the dwarf honey bee Apis florea. Polymerase chain reactions (PCR) were conducted on DNA extracts using five sets of primers known to amplify microsatellite loci in A. metl(/er_a. Based on microsatellite allele distributions, queens of the five colonies mated with at least 5-14 drones. This is up to 3 times previous maximum estimates obtained from sperm counts. The discrepancy between sperm count and microsatellite estimates of the number of matings in A..florea suggests that despite direct injec- tion of semen into the spermatheacal duct, either A. floma drones inject only a small proportion of their semen, or queens are able to rapidly expel excess semen after mating. A model of sexual selection (first pro- posed by Koeniger and Koeniger) is discussed in which males attempt to gain reproductive dominance by increasing ejaculate volume and direct injection of sper- matozoa into the spermatheca, while queens attempt to maintain polyandry by retaining only a small frac- tion of each male's ejaculate. It is shown, at least in B. P. Oldroyd (~;~)1. A.J. Smolenski. J.-M, Cornuet - R. H. Crozier School of Genetics and Human Variation, La Trobe University, Bundoora, Victoria 3083, Australia J.-M. Cornuet Laboratoire de Neurobiologie Compar& des Invert6brds, INRA- CNRS, URA 1190, 91440 Bures-sur-Yvette, France S. Wongsiri Bee Biology Research Unit, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand A. Estoup Laboratoire de Populations, G~ndtique et Evolution, CNRS, 91198 Gif-sur-Yvene, France T.E. Rinderer Honey-Bee Breeding, Genetics and Physiology Research Laboratory, USDA-ARS, 1157 Ben Hur Road, Baton Rouge LA 70820, USA Present address: ~School of Biological Sciences, Macleay Building A I2, University of Sydney, N.S.W. 200& Australia this limited sample, that the effective number of mat- ings is lower in A. florea than in A. mell!Tbra. Key words Polyandry- Microsatellite - Multiple mating. Relatedness - Apis Introduction Haplo-diploidy produces high relatedness among female offspring of singly-mated hymenopteran queens. Evolution ofeusociality was probably facilitated by this high relatedness (Hamilton 1964; Pamilo 1991). However, in many of the advanced eusocial species, multiple mating (polyandry) occurs, which causes a reduction in average relatedness (Page and Metcalf 1982). Given appropriate cost-benefit coefficients (Crozier 1979), eusociality can be maintained despite multiple mating, but the reasons for the evolutionary shift are of great interest (Page 1980; Page and Metcalf 1982; Cole 1983; Crozier and Page 1985). It has been postulated that multiple mating is adap- tive because: 1. Genetically diverse colonies may be able to tolerate a wider range of environmental conditions, perhaps by increased polyethism (Oldroyd et al. 1992 a, b) or the potential for increased caste differentiation (Crozier and Page 1985), or by increased tolerance to pathogens (Sherman et al. 1988: Shykoff and Schmid-Hempel 1991a, b). 2. Multiple mating eliminates the possibility of a queen mating with a single drone carrying the same sex allele as herself (Page 1980; Crozier and Page 1985; Ratnieks 1990). Because of the method of sex deter- mination in Hymenoptera, such matings result in a 50% reduction in brood viability (Woyke 1963). 3. Multiple mating can reduce conflict between work- ers and queens over the preferred sex ratio (Moritz 1985: Pamilo 1991).