138 RADIATION GENETICS 



that are recessive lethals. Since about three-fourths of the induced gonial mutants 

 reported by Russell are recessive lethals, the yield in the backcross and sibmating 

 procedures may be further reduced by another factor of four. With the exception of 

 the fact that there is no restriction of the number of loci under test, little can be said for 

 these latter methods of genetic analysis. In any of these methods, the mutation rate 

 for either spermatogonial cells or post-gonial cells is isolated by analysis of matings only 

 in poststerile or presterile periods, respectively. Only the rate for oocytes can be 

 obtained for females. 



Recessive lethal mutations. — Of the three methods described above, only the specific- 

 locus test system will permit an estimate of the frequency of recessive lethal mutations. 

 Russell and Russell 1138 report that about 75 per cent of the induced mutations at the 

 seven loci are recessive lethals. This proportion varies among loci, according to 

 present reports, from about 50 to 100 per cent of the number of observed mutants. 

 The locus with the highest gonial mutation rate, the piebald locus, also happens to have 

 produced only lethal mutations. 



These data pertain to those mutants derived from the single dose tests with the 

 dose delivered at high intensity. Whether or not the same ratio of lethal to viable 

 mutants will occur among those induced by low-intensity continuous exposure remains 

 to be seen. Presumably, there should be no qualitative difference between mutations 

 induced at different rates of radiation dosage, even though the mutation rates them- 

 selves may vary (see below). However, the recovery process, acting upon premuta- 

 tional damage to reduce the mutation rate under continuous exposure as compared to 

 single-dose exposure, could conceivably act in a selective manner at the molecular level. 

 On the assumption that variation in genie action is associated with variation in the 

 molecular structure of the gene, it is then conceivable that some forms of genetic 

 damage may be more amenable to spontaneous recovery. Whether or not lethal 

 mutations are selectively acted against cannot be stated, but this would seem to be a 

 potentially important point to have clarified in a mammalian test system. 



An additional method that has been employed to determine the recessive lethal 

 mutation rate for autosomal genes can be designated as the linked-lethal procedure. 

 In its simplest form, a single recessive marker gene is carried homozygously in the 

 irradiated parent; the parent is outcrossed and the heterozygous progeny are inbred to 

 produce the normal 3 : 1 ratio in the F 2 . The absence of the marker in the segregating 

 generation is accepted as prima facie evidence of the induction of a lethal closely linked 

 to the marker. The linked-lethal procedure was tried by Snell, 1250 but without 

 success. His test entailed the screening for aberrant segregation ratios in F 3 progeny 

 produced by backcrossing an F 2 mouse heterozygous for a marker with its F x parent. 



An obvious deficiency of the method is in the limited length of the chromosome 

 under test. Crossing over will naturally occur with greater frequency as the map 

 distance between the marker and the lethal increases. Minor aberrations in the 

 segregation ratio would require extensive testing for proof of presence of a lethal 

 mutant, which could go far beyond its economic value in terms of information yield. 



