Studies in quantitative inheritance: v. chromosome analyses of crosses between selected and unselected lines of different body size in drosophila and meleanogaster

Abstract

1. By a special chromosome assay technique, it is possible to prepare genotypes consisting of various combinations of chromosomes from two inbred lines, and thus to study the effect of individual chromosome substitutions in different genetic backgrounds. •

2 . This method produces 15 genotypes for any pair of inbred lines, A and B, consisting of the parent lines and their Fl, together with the 12 genotypes carrying one or two chromosome pairs heterozygous for A and B and the remaining chromosomes either all A or all B. It has been used in an analysis of the heterosis for size and egg production in a number of crosses between unselected inbred lines. •

3. All pairs of lines show heterosis for each character on crossing. A least squares test shows that in many cases the separate effects of making the three chromosomes heterozygous do not combine additively in the triple heterozygote, so that interactions between genes on non-homologous chromosomes often have marked effects on all characters. Similar interactions between linked genes doubtless also occur, and it is likely that epistatic gene effects are generally of importance in quantitative characters. •

4. By studying the set of genotypes consisting of a pure line A and all genotypes obtained by making one or more of the major chromosome pairs heterozygous for line B, we can divide the interaction into three portions consisting of (a) interaction between chromosomes in a partially heterozygous background, ( b ) deviation of the triple heterozygote A /B, and (c) deviation of the homozygote A from their expected values as estimated from substitution effects in a partially heterozygous background. This test has been applied to each set of genotypes for all three characters. •

5. With a few exceptions, interactions among partial heterozygotes were slight, while the triple heterozygote tended to be a little greater than expectation. But the main source of interaction was located in the inbred line homozygotes, which were rather consistently lower in mean than their expected values. Generally, the lines with the lowest means showed the most interaction of this kind, so that the highest level of interaction in the homozygote tended to occur where the inbreeding decline was greatest. •

6. Difficulties of interpretation arise from the fact that the amount of interaction shown by a given inbred line could be very different when it was tested against chromosomes of tr.o different lines, and it is not yet clear how consistent such estimates are likely to be. •

7. In the series examined, the environmental variance declines with increasing level of heterozygosity for all three characters, though not necessarily in a linear manner. An attempt was made to separate the effects of the mean and the level of heterozygosity on the environmental variance, using a partial regression analysis. The mean and the level of heterozygosity (length of metaphase chromosome heterozygous) are closely correlated in these series, and both are closely correlated with the variance, so that a complete separation of their effects was impossible. But the analysis suggests that it is the mean and the particular gene combination responsible for it, rather than level of heterozygosity pera se, which is the main factor in determining the variance. •

8. estimates of chromosome substitution effects averaged for a number of inbred lines should give a measure of the relative activity of genes on the different chromosomes, activity depending on the number of genes on a chromosome affecting a given character and their mean effect. Estimates of relative activity, averaged over all lines, were obtained for changes in mean and variance of wing and thorax length and egg output. They were remarkably consistent in showing II to have about 60% and I only about 20% the activity of III. The three characters showed essentially the same distribution, of effects, which suggests that many genes are involved per chromosome. The activities of II and III are roughly in proportion to their metaphase chromosome lengths, but that of much less in proportion. This is thought to be a dosage compensation effect, arising from the fact that I is hemizygous in males. •

9. The general implications of these results are discussed and it is suggested: (a) Interpretation of the phenotypic variation in populations undergoing inbreeding has to take account of the increased effects of segregation at c& n td- loci as well as the greater sensitivity to environmental variation. (b) The relative. importance of environmental variation in different characters in out -bred populations may provide a reasonable guide to their resistance to inbreeding. (c) The increased sensitivity to external variation and the greater importance of interaction, which accompanies changes in the genetic constitution which lowers the level of performance, opens possibilities of investigating the nature of the environmental variation and the stages of development at which it is most effective. This is likely to throw further light on the attributes of the gene- controlled changes which lower performance as well as the effects of gene combinations, heterozygous or otherwise, which are involved ,in striking interactions.STUDIES IN QUANTITATIVE INHERITANCE I. THE EFFECTS OF SELECTION OF WING AND THORAX LENGTH IN DROSOPHILA MELANOGASTER BY FORBES W. ROBERTSON AND ERIC REEVE (With Fourteen Text- figures). From JOURNAL OF GENETICS, Vol . 50, No. 3, pp. 414 -448, FEBRUARY, 1952.] • •

Interactions Between Chromosomes from Large and Small Strains of Drosophila melanogaster by FORBES W. ROBERTSON. • •

Heterozygosity, Environmental Variation and Heterosis (Reprinted from Nature, Vol. 170, p. 296, August 16, 1952) • •

STUDIES IN QUANTITATIVE INHERITANCE IV. THE EFFECTS OF SUBSTITUTING CHROMOSOMES FROM SELECTED STRAINS IN DIFFERENT GENETIC BACKGROUNDS IN DROSOPHILA MELANOGASTER BY FORBES W. ROBERTSON AND E. C. R. REEVE. [FROM JOURNAL OF GENETICS, VOL. 51, No. 3, pp. 586 -610, JULY 1953.]