Anim. Behav., 1996, 51, 445–449 Queen number influences the primary sex ratio in the Argentine ant, Linepithema humile (= Iridomyrmex humilis) LAURENT KELLER*², GRÉGOIRE L’HOSTE*, FRANÇOIS BALLOUX* & OLIVIER PLUMEY* *Institut de Z oologie et d’Ecologie Animale, University of Lausanne ² Z oologisches Institut, Bern University ( R eceived 14 M arch 1995; initial acceptance 9 June 1995; final acceptance 14 July 1995; M S. number: 4879) Abstract. Hymenoptera are characterized by a haplodiploid sex determination system where females are diploid and males haploid. Because females can decide whether they fertilize their eggs they have control over the sex of their offspring. Females are thus expected to lay haploid and diploid eggs in a ratio that maximizes their fitness. A surprising finding of recent studies on the primary sex ratio of ants has been that queens lay male eggs outside the time of production of sexuals, a period during which all haploid eggs are usually destroyed by workers. It has been suggested that the production of a high proportion of haploid eggs by queens might be a mechanism to achieve higher reproductive success when the colony loses all its queens and rears new sexuals from the brood present within the colony. Here it is shown that, as predicted by this hypothesis, the proportion of haploid eggs laid by L inepithema humile queens is higher in colonies containing several queens than in single-queen colonies. Because L . humile sexuals mate in the nest, the females that produce a higher proportion of haploid eggs should have higher fitness when the colony sex ratio of the brood is female-biased and sons of several queens compete to fertilize the new queens produced.  1996 The Association for the Study of Animal Behaviour Study of sex ratio in social Hymenoptera (ants, wasps and bees) has provided important insights into sex-ratio theory (Fisher 1930), inclusive fit- ness theory (Hamilton 1964a, b) and parent– offspring conflict (Trivers 1974). The haplodiploid sex determination system, by which females are diploid and males haploid (arising by partheno- genesis), results in asymmetries of genetic related- ness between colony members. Such asymmetries of within-colony relatedness leads to a queen– worker conflict over relative allocation of resources in the two sexes (Trivers & Hare 1976; Charnov 1978; Nonacs 1986; Pamilo 1991). Because queens are equally related to both sexes of reproductives reared in their colony, their fit- ness is maximized by an even sex-ratio investment. In contrast, workers are more related to their sisters than to their brothers, so that their inclusive fitness is maximized by an increased investment in female brood. Consequently, the population investment sex ratio is predicted to be even under queen control, and to be female-biased under worker control. Several studies suggest that the population sex-ratio investment is female- biased in single-queen (monogynous) colonies, supporting at least partial worker control over sex-ratio investment (e.g. Trivers & Hare 1976; Nonacs 1986; Boomsma 1989; Bourke 1989; Mueller 1991; Crozier & Pamilo 1992). An implicit assumption of hypotheses and empirical studies concerning the conflict between queens and workers in highly eusocial species is the ability of the workers to distinguish between male and female offspring (Nonacs 1993; Aron et al. 1994, 1995). Queens are in control of the proportion of haploid (male) and diploid (female) eggs they lay. Thus, workers can shift the sex ratio towards females by eliminating male brood or by increasing the resources invested per female repro- ductive. Comparing the primary sex ratio (at the egg stage) and secondary sex ratio (at the adult Correspondence: L. Keller, Bern University, Zoolo- gisches Institut, Ethologische Station Hasli, Wohlen- strasse 50a, 3032 Hinterkappelen, Switzerland (email: lkeller@ulys.unil.ch). 0003–3472/96/020445+ 05 $12.00/0  1996 The Association for the Study of Animal Behaviour 445