MUTATIONS 



observed by Bcnzer in the r 11 region of phage T4. For example, inter- 

 crosses involving 52 leucine mutants showed that the mutations had 

 occurred at different sites of the locus (12). Occasional repeats have 

 been found, but even those are not frequent. Certainly their frequency 

 is not as high as that reported in phage. What I have in mind is the 

 question of whether hot spot behavior is characteristic for phage, but 

 not of organisms on a higher level of organization, and this would in- 

 clude bacteria. Is there any evidence for hot spots in organisms higher 

 than phage? 



Benzer: Such information is very difficult to get in higher organisms 

 because you cannot easily distinguish between mutations at the same 

 site or at closely adjacent sites. 



Demerec: In bacteria, we should be able to detect them. 



Lederberg: I would like to answer the question from our experience. 

 I think it depends on which gene you're looking at. In the gal region, 

 some 100 or 200 loci have been examined, and there are no known 

 recurrences there. In the lac region, there are at least five or six recur- 

 rences at one particular locus, in a sample of 60 or 70 mutants ; so that 

 is not a very hot spot. But there is certainly a difference in the behavior 

 of the material at these two loci. 



Atwood: I wanted to mention the well known causes of mistakes 

 in human mutation rates. An equilibrium is commonly assumed be- 

 tween gain of mutant alleles by mutation and loss by selection; the 

 rate determinations usually involve three factors. The frequency of 

 affected individuals is one; then, a relative fertility factor, and then 

 another factor which is really the ratio of the alleles per person sub- 

 ject to selection to the alleles per person subject to mutation. For an 

 autosomal recessive, this is unity; for an autosomal dominant, %, and 

 for a sex-linked recessive, %. 



Ignoring these constants we could be off by a factor of 3 at most, 

 so they may seem a needless embellishment when you consider the 

 difficulty of measuring relative fertility, ascertaining affected individ- 

 uals without bias, and evaluating the assumption that the population is 

 in equilibrium. 



One can guess about how far off these estimates could be because of 

 such uncertainties. Where inbreeding has decreased so that gene fre- 

 quencies are not at equilibrium, the error would almost surely be less 

 than tenfold. Take an extreme case, for example of one per cent in- 

 breeding with an equilibrium gene frequency of 10"^. The homozygotes 

 would then be about 10"^. If inbreeding suddenly stopped, the homo- 

 zygotes would drop temporarily to 10"®. 



Another factor is the selection of unusually high rates for study. 



