Genetics of coxal chaetae in Drosophila Melanogaster

Abstract

Differences observed between the numbers of chaetae on the coxae of corresponding legs on the two sides of a fly are most probably due to developmental instability, as with sternopleurals. The metric used in the present study has therefore been taken as the sum of the chaetae on the coxae of the pair of corresponding legs. The average number is lowest on the rear (R), highest on the front (F) and intermediate on the middle (M) legs. The number of coxals on the M legs often approximates to the average of R and F in other cases being most commonly higher than this average: indeed at 25°C no case has been observed of M falling significantly below the average of R and F. A reduction of temperature to 18°C almost always raises the number of chaetae on the R legs, but usually, though not always, it reduces the numbers on the F legs. The number of chaetae on the M legs is always lower at 18°G than at 25°C except for the case of Australia 8. Chromosome assays based on sets of substitution lines made between the Wellington (W), Samarkand (S) and 6C/L (L) inbred lines, taken in pairs, show all the major chromosomes to be carrying genes affecting the numbers of coxal chaetae. A further analysis of the X chromosome using a marked tester chromosome (T) reveals at least four and possibly as many as seven loci affecting coxal chaeta number, with a concentration at the left end of the chromosome. Regression analysis of the type used in investigating genotype x environment interactions shows that the rise in chaeta number from R through M to F cannot be due to a uniformly proportionate enhancement of the effects of the genes concerned, and this is confirmed by a closer analysis of the differences between the homologous chromosomes revealed by the chromosome assay: a chromosome which gives a higher number of chaetae than its homologue on one pair of legs can give a lower number on a different pair. The genes must in fact show differential changes of activity from one pair of legs to another. The results can be accounted for by postulating three classes of gene: one class (α) being equally active on all three pairs of legs; a second (β) inactive on R but active on both M and F ; and a third class (γ) inactive on R and M but active on F. This hypothesis allows predictions to be made about the responses of coxal chaeta number to certain types of selection, the results of which will be the subject of a later publication. The differential changes in the action of the genes could perhaps be understood in terms of a uniform activity of the genes themselves being modified by appropriate variation in the availability in the cell of other materials necessary for the gene products to display their activity by chaeta production. A more attractive interpretation, however, is to be found in terms of regulatory, control and structural elements similar to the tripartite system postulated and discussed by Britten and Davidson.