Published online by Cambridge University Press: 14 April 2009
1. Lines with ten pairs of parents and selected at an intensity of 20% were exposed to 1000 r. of X-rays for 0, 2, 10 or 30 generations.
Lines which received some irradiation generally gave greater response than the unirradiated controls. The phenotypic variance in the irradiated lines was much higher than in the controls. There was little difference in behaviour between lines receiving ten generations of irradiation and those irradiated every generation. Lines receiving only two generations of irradiation had lower variances than the other irradiated lines, but in one of three replicates the response was greater than the corresponding continuously irradiated line.
3. Lethal frequencies were much higher in irradiated than unirradiated lines. Particular chromosome II and III lethals were at high frequencies in most of the irradiated lines but in only two out of five controls.
4. On relaxation, the mean of the irradiated lines generally declined considerably, but in the unirradiated lines there was only a very small regression.
5. It appears that most of the extra response and increased variance in the irradiated lines were caused by a few genes with large effect on bristle number.