J. Anirn. Breed. Genet. 113 (1996), 93-97 zyxwvu 0 1996 Blackwell Wissenschafts-Verlag, Berlin ISSN 0931-2668 zyxwvutsrq Ms. received: 13.6.199J 'Seccidn de Genitica Animal, INIA, Madrid and 'Departamento de Genitica, Facultad de Ciencias Biologicas, Universidad Complutense, Madrid, Spain zyxw Changes in the additive variance of a fitness-related trait with inbreeding in zyxw Tribolium castaneum By R. G. RUANO', L. S. SILVELA', C. LOPEZ-FANJUL' and M. A. TORO' Introduction In the absence of selection, the heterozygosity of finite populations is expected to decrease with time, until it finally vanishes as a result of fixation. However, the additive variance of a quantitative trait will only decline monotonically with inbreeding if all contributing loci show additive gene action, as only in this case are heterozygosity and variance linearly related. Otherwise, the additive variance may initially increase with inbreeding up to a maximum. Theoretically, this has been shown for recessive alleles, at low initial frequency (ROBERSTON 1952), and with additive zyxwvu x additive epistasis (GOODNIGHT 1988). Whatever the gene action, the additive variance will eventually decrease towards zero, but this will occur at high degrees of inbreeding, usually higher than those commonly found in natural or domestic animal populations. The behaviour of the additive variance in populations is particularly interesting, since the response to selection will be dependent on its magnitude. The role of genetic drift in changing the mean of a population has been emphasized by WRIGHT (1978). With non- additive (particularly epistatic) gene action, transient increases in the additive variance may occur, contributing to larger responses to selection. Precise experimental evaluations of previous predictions are scarce, as most experiments lacked adequate replication (LYNCH 1988). An increase in the additive variance, following a population bottleneck, has been reported for several morphological traits in the house fly (BRYANT et al. 1986). In Drosophila melunogaster, the original additive variance of egg-to- pupa viability (a trait strongly depressed by inbreeding) increased by a factor of five in lines 25-50 % inbred (LOPEZ-FANJUL and VILLAVERDE 1989; GARCIA et al. 1994). In Tribolium custuneum, similar increases in the additive variance of egg-to-larva viability have been reported by FERNANDEZ et al. (1995). In this experiment, a trait closely related to fitness (TWL) has been scored in lines of T. custuneum. Changes in the mean and the additive variance, after one generation of brother x sister mating, were analysed. Materials and methods Base population, culture conditions, and trait scored The Consejo population was captured near Madrid in 1964, and has since been maintained in the INIA laboratory in 10 bottles. A circular mating scheme was used, to ensure a sufficiently large population (about 1000 parents per generation). All beetles in this experi- ment were kept in the same dark incubator, at 70 % relative humidity and 33 "C. The powdered culture medium consisted of 95 % whole-wheat flour and 5 % dried brewer's yeast by weight. Handling was performed at room temperature. U.S. Copyright Clearance Center Code Statement: 0931-2668/96/1302-0093 $1 1.50/0