NATURE OF THE GENETIC EFFECTS 407 



wise, genetic schemes of this kind can be used for detecting steriles (muta- 

 tions causing sterihty) and visibles. / 



By the use of these methods it has been ascertained that in most cul- 

 tures of Drosophila approximately one X chromosome in 600, on the 

 average, contains a lethal gene that has arisen by "spontaneous" muta- 

 tion during the immediately preceding generation. In contrast to this, 

 if the heavy dose of 5000 r of ionizing radiation had been applied to the 

 father's mature spermatozoa, approximately 14 per cent of the X chromo- 

 somes in them (a frequency 85 times as high as the preceding one) would 

 have come to contain an induced lethal. According to the same data, 

 if the frequency of induced lethals is proportional to dose (as will be 

 shown to be the case in Chap. 8), it must take approximately 60 r, if 

 appHed to mature spermatozoa, to induce lethals at a frequency equal to 

 that with which they usually arise spontaneously in the course of one 

 generation, but the induced frequency would be added to the spontaneous 

 one, thus doubhng the rate of origination of the lethals. 



These methods have also shown that not fully lethal but detrimental 

 mutations having an effect marked enough to be detected by the methods 

 used have a frequency of origination some three or four times that of the 

 lethals, and that mutations in the autosomes arise (considered collec- 

 tively) some four or five times as often as in the X chromosome. If these 

 two' facts are taken into account at once, the conclusion is reached that if 

 all the mutations thus detectable are considered (visibles being here 

 neglected as of insignificant relative numbers), their frequency is some 

 20 to 30 times that of sex-linked lethals. Hence their usual spontaneous 

 frequency of origination per generation is in the neighborhood of one 

 among twenty germ cells, and their frequency of induction by 5000 r 

 applied to spermatozoa averages approximately four within each germ 

 cell. By appropriate appHcation of ultraviolet it is also possible to 

 induce mutations at a frequency about as high as this. 



These over-all frequencies of mutation can be translated into average 

 frequencies per individual gene by dividing them by the total number of 

 genes that participate in giving such mutations. However, the estimates 

 of gene number thus far made may be inaccurate by a factor of about 2. 

 Various methods indicate that in Drosophila there are some 5000 to 

 10,000 genes in a single (i.e., haploid) set of chromosomes. This would 

 (assuming the great majority of genes to be capable of giving mutations 

 of the already mentioned types) make the frequency of spontaneous 

 mutation per individual gene per generation 1 in 20 X 5,000 or 10,000, 

 that is, between 1 in 100,000 and 1 in 200,000; and that of mutation 

 induced by application of 5000 r to spermatozoa 4 X 1 in 5000 or 10,000, 

 that is, between 1 in 1250 and 1 in 2500. 



A more direct method, called the specific locus method, of ascertaining 

 the frequency of mutation per individual gene, is to take individuals 



