1. The rate of production by X-rays of new genetic variation in two quantitative characters in Drosophila melanogaster (sternital and sternopleural bristles) has been investigated, using ‘plateaued’ populations which had reached the limit under artificial selection and, for sternital bristles only, populations which had been made genetically invariant by inbreeding. The genetic variation was always measured by the response of the population to selection. The X-rays dose given in any generation was always 1800 r. to adults.2. Seven plateaued lines had eight cycles of alternate irradiation and selection, each with its non-irradiated control. All the responses were small but in three lines they were significantly greater after irradiation.3. Selection was applied to three different inbred lines, genetically marked to detect contamination, after varying periods of irradiation. At the same time, the inbred lines and lines derived from them which had been mass mated in bottles were selected. The irradiated populations showed a greater response. The new genetic variance produced by the irradiation was approximately 10−5 units/r. The estimate of the dose required to introduce new variation equal to that in a standard outbred population was 500,000 r.4. The effective population size was an important factor in the interpretation of some of these results on the long-term effects of radiation. By observing the variation between replicate lines in the frequency of a gene with a visible effect under these culture conditions (i.e. in a single culture bottle) the effective population size was estimated at sixty. Outbred populations kept under these conditions for many generations showed a reduction of genetic variability in agreement with this value.5. To investigate the possibility that the deleterious genes produced by irradiation would interfere with the response to artificial selection, a standard outbred population was irradiated and selected. In spite of the observed high frequency of recessive lethals produced, the response to selection was very similar to that of the standard population.
The effects of X-rays on quantitative characters
