Search

There is mounting evidence that Drosophila populations are mlore polymorphic at the level of genetic loci than was formerly thought by many geneticists. There have been many experimental demonstrations of the extensive chromosomal polymorphis^ms, of the presence of considerable quantities of hidden deleterious and lethal alleles in the populations, and of the extensive complexity of the coadapted genetic systems integrated through natural selection.' Nevertheless, the proportion of polymorphisms among the majority of all loci is largely unknown because of inadequate screening techniques. Furthermore, the mechanisms for the maintenance of polymorphisms -remain elusive to experimental verification. The ideal situation of screening populations for genetic comparisons at and between individual loci has been brought a step closer to reality by electrophoretic separation of enzyme molecules. Though enzyme differences are not necessarily related with genetic differences on a one-to-one basis, most may be. At any rate, zymogram analysis of individual flies for electrophoretic mobility or qualitative activity differences gives a good first approximation of the underlying genetic situation for a different class of variation than has previously been studied directly. However, as Lewontin and Hubby3 have pointed out, the randomness of the samples of loci studied by this technique is critical for broader generalizations which might be made from the findings in this class of loci. Sample characteristics are yet to be determined. Lewontin and Hubby3 have found about 40 per cent of all electrophoretically separable enzyme systems and protein bands of no known enzymatic activity of D. pseudoobscura to be polymorphic, having two or more electrophoretic forms of the same protein. One expects the difference in molecular forms to represent at least as much genetic difference, so that the variation of molecular types reflects similar variation of genotypes. If only a moderate bias is involved in estimating this proportion for the genome, then there exists a major discrepancy between the data and the expectation based upon the theory of segregational load. According to the theory, a population may bear, without being jeopardized by reduced average fitness, only a relatively small portion of loci in heterotic polymorphic conditions. In order to establish the meaning of the apparent inconsistency between the theory and observations, it must be established whether or not these isozyme polymorphisms are maintained more or less independently by some form of selection. It is very important to know if these variations are random representatives of a large number of such loci. It has been demonstrated in a laboratory population of D. busckii4 that the polymorphism at the "leucine aminopeptidase" locus is maintained by heterotic selection, though this was not the case for an esterase locus in D. inelanogaster and D. simulans.5