WHAT IS A GENE? 171 



For example, 123 mutants have been shown to result from alterations at 

 the r274 site of the rll-A cistron of phage T4. These can be divided into 

 three groups according to the frequency of reverted r among them: 

 alpha group, 0.2-0.9 x 10~*^; beta group, 0.3-4 x 10~'\ and gamma 

 group, 0.5-1 X 10~*^. These differences may be due to alternate struc- 

 tures of the site or to the fact that what seems to be one mutable site is 

 really composed of at least three mutons. We saw that the r274 site, 

 like others, could not be larger than 0.05 map units; if this is divided by 

 three, we obtain 0.02 units, which is indistinguishable from the estimated 

 genetic length of one nucleotide pair. 



In addition to differences in stability, sites within a single gene may 

 show other specific differences. For example, although suppressors of 

 a given locus are frequently nonspecific in their action, some are highly 

 specific. The 0^20 mutant in Aspergillus is the only one of the ad^ 

 group responsive to a suppressor in the left arm of the same chromosome. 

 Although many recessive mutants form no enzyme peculiar to the cistron, 

 some are "leaky, " their defect allowing the synthesis of normal enzyme 

 at a slow rate or of an enzyme with reduced activity. Still others form 

 enzymatically inactive proteins with serological relationships to the 

 enzyme. Frequently a completely nonfunctional mutant will revert to a 

 state of partial activity. Then again, some alleles, such as the h series 

 in T2, cause the phages containing them to be inactivated by heat at 

 different rates. 



Clearly, on the criteria of stability and function, a functional unit 

 may achieve many alternate states which we recognize as multiple alleles. 

 Each allelic stage represents a particular configuration of the gene in 

 terms of the mutons it contains. Since there may be hundreds of mutons 

 in a gene, the number of possible alleles must be very large. Many 

 alleles which are really different may be indistinguishable because of the 

 insensitivity of the tests we use to discriminate among them. When dis- 

 crimination can be achieved only by changing the genetic background 

 or by using special environments, the alleles are called isoalleles. A 

 most favorable case for the discovery of different alleles is found in the 

 incompatibility locus of some higher plants. Unless the male and female 

 gametes carry different alleles of this locus, fertilization will not succeed. 

 The more different incompatibility alleles, the more likely the success 

 of a random pollination. Clover, as an example, seems to have exploited 

 many of the changes that can occur at its incompatibility locus, for more 

 than 200 alleles have been discovered there. 



When a multiple allelic series in, say, Drosophila, is carefully studied 

 by recombination techniques, it is usually found that the locus is pseudo- 

 allelic. This simply means that it is a gene composed of a number of 



