60 
Psyche 
[Vol. 95 
societies considered as stock colonies were kept in large plastic 
boxes and reared in conditions similar to those described in Passera 
et al (1988). 
Forty-seven experimental queenless units were constituted by 
splitting four queenright stock colonies from 10/ 12/85 to 23/5/ / 86. 
Each experimental unit consisted of about 650 workers and brood 
ranging from eggs to worker pupae. The brood was carefully moni- 
tored with the aim of avoiding the introduction of differentiated 
sexual larvae which are easily distinguished from worker larvae on 
the basis of size and/or color. Queen larvae never develop in queen- 
right colonies (personal observation) so they were not present in any 
of the stock colonies. Moreover their matt color and their large size 
contrast sharply with those of the worker larvae. Male larvae are 
periodically produced in such stock colonies (Passera et al, 1987) in 
the presence of the queens, so we selected periods when they were 
absent. The experimental units were then monitored daily for the 
presence of sexual forms. 
Results 
Males: Of the 47 experimental units, 44 (93.6%) produced male 
larvae. This is explained by the fact that male brood is always 
present in stock colonies but its development is dependent on the 
trophic status of the society. This trophic status depends on the 
worker/ larva ratio and/or queen number per colony (Passera et al, 
1988). Males pupated between the 7th and 95th day after the begin- 
ning of the experiment (N = 758; mean in days ± SD = 38 ± 15). So 
the period of emergence of males is rather broad. 
Queens: Thirty-seven of the 47 experimental units (78.7%) pro- 
duced queen larvae. In queenright societies queen larvae were never 
produced whatever the brood composition or trophic status. 
Queen production was not dependent on the season. In December 
the brood used in experimental units was a little overwintered but 
the brood used later in spring was supplied by stock colonies kept at 
a high temperature (28° C) in the laboratory for several months. In 
both cases the experimental units produced queen larvae. This 
differs from other ant species in which sexuals forms generally 
develop only from overwintered brood (Passera, 1984). 
Queen larvae differentiated 24 to 53 days after the queens were 
removed. The first queen larvae were produced only after 24 days 
