1228 



BIOLOGICAL EFFECTS OF RADIATION 



r^^^Ji r^ Or^^.^-^ 



10 20 30 40 50 60 70 

 a 



10 20 30 40 50 60 10 

 d 



10 20 30 40 50 60 70 10 20 30 40 50 60 70 



demonstration would materially change the interpretation of certain 

 experiments. Patterson (138) has indeed shown in tests of a series of 

 chosen genes that on the average seven-eighths of the apparent mutations 

 at a locus probably are deficiencies and also are recessive lethals. 



In Table 8, the data of a number of investigators have been brought 

 together, and a comparison is available of the relative frequencies of 

 lethal and viable mutations. It may be remarked that the agreement 

 between the different investigators is surprisingly good. It was then 

 necessary to show that under given conditions the same picture would 

 present itself when lethal mutations were studied, as was found with the 

 viable mutations. MuUer (107) accordingly in his studies of spontaneous 



lethal mutation carried out tests of 

 the lethals, designed to determine 

 whether their distribution on the 

 chromosome was like that of the 

 viable mutations. This he also did 

 for the lethals of his X-ray experi- 

 ment (105) and similar tests have 

 ^ ^ been carried out by Harris (72), 



n i-i PI PI Ohver (125), and Gowen and Gay 



llHIrTln J~rl l^;TVf>>^h (52). If against the genetic map as 



an abscissa the frequency of muta- 

 tion per unit is plotted, a marked 

 clustering of mutations in certain 

 regions is detected. This is due to 

 a discrepancy between the cross-over 

 map and the cytological map (28, 

 29, 32, 120) which is dependent on 

 regional differences in the frequency 

 of crossing over. Figure 5 shows 

 the results of Oliver's (125) experiments. It is clear that a clustering 

 exists, such as is found in the visible mutations for the X. This 

 shows then that, in a lethal-mutation experiment, the same population 

 of genes furnishes the mutations as in an experiment with viable 

 mutations. It should be clearly understood, however, that it does 

 not necessarily follow that the same genes, or type of mutation 

 process, are involved in both cases. For the observed clustering probably 

 depends merely on the differences in the numbers of genes between two 

 points in different parts of the genetic map. 



Whatever the relation between lethals and viable mutations, the bulk 

 of the quantitative data available concerns the former. A discussion of 

 the experimental attack on the nature of the mutation process is therefore 

 to a large extent an account of experiments with the rate of lethal 

 mutation. 



10 20 30 40 50 60 70 

 c 



10 20 30 40 50 60 70 

 f 



Fig. 5. — The distribution of lethal 

 mutations in the X-chromosome of 

 Droaophila melanogaster . The chromo- 

 some, whose genetic map is the abscissa, 

 is divided into regions five units in length. 

 The ordinates are numbers of lethals: 

 (a) 47 lethals, dosage 385 r; {h) bl lethals, 

 dosage 770 r; (c) 46 lethals, dosage 1540 

 r; (d) 46 lethals, dosage 3080 r; (e) 37 

 lethals, dosage 6160 r; (/) 233 lethals, 

 total. {After Oliver, 125.) 



