a1 Agricultural Research Council Unit of Animal Genetics, Institute of Animal Genetics, Edinburgh, 9
1. Published data suggest that mean left-right asymmetry in number of sternopleural bristles of D. melanogaster declines when inbred lines are crossed, while the corresponding variance for sternite bristles remains unchanged. Some genetic tests were undertaken to analyse this difference in behaviour of the two characters.
2. A progeny test on a wild stock showed that a small amount of genetic variance in sternopleural asymmetry was present, equivalent to about 2% of the total phenotypic variance.
3. It was possible to increase and decrease the level of sternopleural asymmetry in two wild stocks by selection. These experiments gave an estimated heritability of some 2–3%, in close agreement with the progeny test. Change in asymmetry did not necessarily lead to a change in mean count.
4. Homozygous lines, consisting of individual third chromosomes from the Renfrew wild stock made homozygous in an inbred line genetic background, were intercrossed, and the average indices for a number of characters of eight inter crosses involving eight lines were compared with their mid-parent averages. Thorax length was 2% greater and its variance 32% less in the crosses; total sternopleural count and its variance did not change significantly, but the asymmetry variance declined by 18%. In contrast, the corresponding asymmetry or independent variance for numbers of sternite bristles was 6% higher in the crosses, although the total sternite count and its variance did not change. These results fit in with previous work.
5. Tests on a similar set of homozygous lines in which the third chromosomes came from the SP wild stock, and on some long inbred lines from the Pacific wild stock, gave discordant results. Of eight SP lines examined, four were homozygous for a gene polychaetoid, and four were homozygous for a genetic effect causing sockets without bristles to occur among the sternopleurals. Both types had much greater sternopleural variance and asymmetry than the Renfrew lines, and both indices declined sharply in intercrosses leaving these genetic effects heterozygous, but neither declined if they were left homozygous in the crosses. Similarly high sternopleural variances were found in the Pacific lines, but only the total variance declined in males and only the asymmetry variance declined in the females, when they were intercrossed. All the four Pacific lines tested appeared to be homozygous for a genetic effect which caused a variable number of dorso-central and scutellar bristles to be replaced by sockets without bristles, and an occasional extra scutellar bristle to appear. This effect was also probably responsible for the high sternopleural variances.
6. Males of the Pacific inbred lines and intercrosses were compared when reared on the normal live medium and on a synthetic diet in reduced concentration, which reduced body-size by 23% (thorax area). The inbred lines were reduced more than the F1's in total sternopleural count and its variance, but the F1's were reduced more in sternopleural asymmetry, by the restricted diet.
7. The problems of interpreting these experiments, in view of our ignorance of the biological functions and attributes of the sternopleural and sternite bristles, are discussed. It is concluded that we have no basis yet for deciding whether sternopleural bristle number is of adaptive significance, but this is considered improbable.
8. The experimental evidence suggests that sternopleural asymmetry cannot be considered a measure of general developmental stability, particularly as the level of asymmetry can be reduced by selection well below that of typical wild stocks.
9. The scaling problems arising when the mean asymmetry of lines with different mean counts are to be compared, are examined, and it is suggested that the ratio of asymmetry to total count does not eliminate scale effects.
10. Developmental and anatomical differences between the sternopleural and sternite bristles suggest a possible reason why they behave differently when inbred lines are intercrossed.
(Received October 03 1959)