THE NATURE OF THE GENE 39I 



mutant is not a mere absence (p. 368). The fact that the mutation rates 

 may be of the same order of magnitude demonstrates the further fact 

 that these genes at least cannot be aggregates of identical parts only one 

 of which has to change in order to produce the mutation; it is easy to 

 see that if the gene was made up in this way, it would be much easier to 

 produce an ionization in any one of the parts, to give a forward muta- 

 tion, than to hit exactly the one part which had been altered, as would 

 be necessary to change the mutant back into normal. 



The relation between dosage and induced mutation rate is accurately 

 known for a few particular mutation steps. Since the relation is Hnear, 

 we can give for each such step a constant (the mutation constant) 



Fig. 158. Mutation Constants and Sensitive Volume for some Mutation Steps 



in D. melanogaster. — The mutation constant is given in mutations per r unit. (From 

 Timofeeff-Ressovsky; the sensitive volume of m — > + is v/rongly given in the 

 original as 1 0.) 



Mutation 

 Step 



H >we 



we -> 4- 



which expresses the chance that a mutation will be produced by a unit 

 dose of ionization. This can also be expressed in another way: we can 

 calculate a volume such that ions are produced in it at exactly the 

 same rate as the mutation is produced in the gene. This volume is 

 known as the "sensitive volume." 



The relation between the sensitive volume and the size of the gene 

 cannot be assumed to be straightforward. If every ion produced in the 

 sensitive volume caused the particular step of mutation we are con- 

 cerned with, the sensitive volume could perhaps be taken as a measure 

 of that part of the gene which is altered by the mutation. But actually we 

 cannot be certain that every ionization is effective, and even if this 

 could be decided, it is possible that the disturbance due to the ioniza- 

 tion might be conducted to some smaller region within the sensitive 

 volume before it produced its definitive effect; and finally, it may be 

 only a small part of the gene which is altered in any one mutation-step. 

 The sensitive volume, in fact, cannot be taken to give any information 

 about the size of the gene; it is simply another method of expressing 



