DOBZHANSKY, SPASSKY and TIDWELL ( 1963) have reported experiments on the effects of inbreeding on the egg to adult viability in progenies of Drosophila pseudoobscura females collected in several places in Arizona and California. As expected, the inbred progenies were less viable than the outbred ones, and the greater degree of inbreeding (brother-sister mating) gave a greater reduction of the viability than the lesser inbreeding (approximately equivalent to half-sib mating). Attempts were then made to estimate the genetic loads, in the populations examined, in terms of the A and B statistics suggested by MORTON, CROW, and MULLER (1956). Similar experiments, giving closely similar estimates of the A and B values, were reported by STONE, WILSON and GERSTENBERG (1963). DOBZHANSKY, SPASSKY and TIDWELL (1963) also approached the problem of the magnitude of the inbreeding effects by another method, previously used for studies on genetic loads in several species of Drosophila. This method consists in obtaining individuals homozygous for certain chromosomes, by means of a series of crosses using genetic markers and suppressors of recombination in the chromosomes under investigation. The estimates of the genetic load found by this method proved to be considerably higher than those obtained by studies of the viabilities of the inbred progenies. This discrepancy reinforced doubts previously expressed about the validity of the methods proposed for the evaluation of the magnitude of the genetic loads, and about the usefulness of these methods for discrimination between the mutational and the balanced components of these loads. These doubts in no way diminish the interest and importance of inbreeding studies. The experiments reported in the present article are made using Drosophila willistoni, a species in many respects different from Drosophila pseudoobscura. D. willistoni is a species very common in tropical America, and the previous work of PAVAN and associates ( 1951 ) has indicated that the populations carry very heavy genetic loads. As will be shown in the following pages, the results obtained are indeed different from those in D. pseudoobscura.
MATERIALS A N D METHODSSamples of natural populations of D. willistoni were collected in several localities in Venezuela,