PROBLEMS OF MEASUREMENT OF MUTATION RATES 73 



causes the disease called tuberculosis, under what condition it does 

 and does not lead to death, what are the changes which it causes in 

 organism, etc. The problems faced with mutations are fairly similar. 

 The first, and perhaps the simplest, question that is asked is how 

 frequent mutations are. Then we shall inquire at what stages of the 

 life cycle do mutations occur. Let me remind you of some of the figures 

 given by Dr. Atwood. Mutation frequencies per gene per generation 

 in man are usually of the order 10"^ or 10-^. In microorganisms the 

 orders of frequency seem to be 10~^ or 10"^. Little information con- 

 cerning the frequencies of mutations exist for organisms intermediate 

 between man and bacteria. Being an old-fashioned naturalist, I am 

 tempted to remind this group that such organisms are quite numerous. 

 A fair amount of data exist only in Drosophila. In about half a dozen 

 species of this genus of flies, the mutation rates for changes producing 

 lethal effects are pretty uniform. They are of the order of 10"^ per 

 gene per generation, very much like human mutations. Mutation fre- 

 quencies for other types of changes are little known. Really next to 

 nothing is known for mutations of the polygenic type, even in Dro- 

 sophila. It is, in my view, a remarkable fact that the mutation fre- 

 quencies seem to be so similar in organisms as different as man and 

 Drosophila. 



Now, where does mutation occur? This is a problem which is 

 important for purposes of analyzing the cause of mutation. Whether 

 we measure the frequency of mutation per generation or per unit time 

 will depend on what we wish to do with our data. I am not sure that 

 I understood correctly the drift of Dr. Atwood's presentation. I take it 

 that his argument tends, on the whole, to the conclusion that mutation 

 occurs at the time when cells divide. We then have to enter into the 

 analysis of the problem whether mutations accumulate with age and, if 

 so, what are the mechanisms of the age accumulation. Here it becomes 

 important to know how frequently mutations arise in males and in 

 females, whether they arise at meiosis or at mitosis, during the period 

 of rapid growth of the organism, after the rapid growth has been 

 completed, in the spermatozoa which are being produced in the testis, 

 in the spermatozoa during the storage period, as in Drosophila, etc. 



The most interesting and suggestive data which Dr. Atwood has 

 presented with respect to the presumed mutation rate in the erythro- 

 cytes seem to me very promising as a possible technique of going at 

 this problem in the future, and I take it his estimate of the mutation 

 rate obtained from the progression curve, 10'^, 10"^, per cell per hour, 

 has to be taken as the first and a very rough approximation. 



