PROBLEMS OF MEASUREMENT OF MUTATION RATES 65 



This seems to be caused by loss of some cytoplasmic material. It is 

 a change, it is inheritable, but it is in a completely different class. 



Benzer: I would be inclined to class that as a mutation of that 

 particular type. 



Zatnenhof: Another thing — you have to remove the mutagen, for 

 if the cell changes only while the mutagens are around, it may be just 

 a phenotype change. If you put 5-bromouracil in a medium, it pro- 

 duces some change in all cells, but it is not a mutation because the 

 cells return to normal upon removal of 5-bromouracil (45). 



Benzer: Of course. That is implied in the statement of its being 

 heritable. 



Novick: These are sufficient conditions, but it is not necessary that — - 



Benzer: I insist that they are necessary. Whether they should be 

 taken as sufficient is another matter. Personally, I would be inclined 

 to do so and let the precise subcategory await further investigation 

 in each case. 



Lederberg: Why do you object to tiiis being included for the case of 

 differentiation? I find that perfectly congenial. 



Auerbach: But if one has an operational method of classifying 

 things, why should one not use it? We have a method for determining 

 whether it is a change in the DNA information or whether it is not, 

 by making the additional postulate that it should cither segregate 

 or recombine. 



Lederberg: You can have nucleic mutations and nonnucleic muta- 

 tions on this fomiulation. 



Auerbach: As they are distinguishable operationally, do you not 

 want to distinguish between them? 



Lederberg : Yes, it is a classification. 



Neel: Are we agreed, then, that from here on a mutation is defined 

 as an event manifested by a heritable change in the organism? 



Auerbach: 1 should like to include in the definition that it shows that 

 it is affecting the nuclear information by either segregating according 

 to Mendel's law or by recombining. 



Stern: These definitions include chromosome aberrations. Mongolism 

 is a change, it is inheritable, and we now understand its genetic 

 nature. It is caused by an extra chromosome. Its frequency may be 

 positively correlated with the number of cell divisions since it orig- 

 inates during cell divisions. To include its origin in one definition of 

 mutation would lump together the kind of changes which affect the 

 molecular configuration of DNA and those which are simply gross 

 changes in the quantity of DNA material. Since we are aware of 



