THE NATURE OF THE GENE 387 



concurrence of two or more events. The argument is as follows. Sup- 

 pose we have a large number of particles and produce among them a 

 number of processes which can cause the particles to change. We can 

 plot curves relating the frequency of the processes and the frequency 

 of changes. If only one process is necessary to cause a change in a 

 particle, the curve will at first be Unear, falling off later when the 



DOSE 



Fig. 156. The Unitary Character of the Mutation Process. — The curves show 

 (diagrammatical ly) the relation to be expected between dosage and mutation 

 rate on the suppositions that 1, 2, or 3 ionizations are needed to cause a gene 

 to mutate. The dotted line indicates the maximum dosage which can be employed 

 in biological experiments. 



processes become so frequent that there is a considerable chance that 

 the same particle may be affected twice. On the other hand, if two or 

 more processes are required to cause a change the curve will at first rise 

 very slowly, since at low frequencies of the processes it is unlikely that 

 any one particle will be affected by the required number; at rather 

 higher frequencies the curve will rise more steeply, falling off finally 

 when many particles are affected by more than the necessary number 

 of processes. In X-ray mutation work, only the first part of the curve is 

 available, since at higher doses the flies are rendered sterile or killed. 

 Thus the statement that the curve relating mutation rate to dosage is a 

 straight line means that it corresponds to the early part of the curve 

 which is appropriate when a change is produced by a single process. A 

 mutation is therefore produced when a single thing happens to a 

 gene. 



