THE MUTABLE UNIT OF HEREDITY 43 



material which may change from one condition to another in a single 

 mutational step and which, when changed, is replicated in the changed 

 form. 



As satisfactory as this definition is conceptually, it demands much of 

 the investigator in practice. We know from the experiments with nitrous 

 acid that the change of a single base in the polynucleotide chain can 

 result in a mutation; therefore, we know that the mutable unit under 

 investigation in such experiments may be no larger than a single 

 nucleotide. But most experiments allow no such precise statements. 

 We may, for example, observe the following relationship in E. coli: 



2 X 10~*^ 



his'^ \ =^ his" 



4 X 10~^ 



The functional gene allowing some step in the synthesis of histidine 

 probably consists of a specific sequence of a large number of nucleotide 

 pairs in the bacterial DNA. A change in any one of these nucleotides, 

 or in certain critical ones, may change the gene so that it can no 

 longer operate in histidine formation. The rate of mutation of his'^'^ 

 h'is~ would then be the sum of the rates of change of many mutable 

 units because all his~ phenotypes seem alike. But a particular his' 

 mutant may have only one inappropriate nucleotide pair; the rate of back 

 mutation from his~ to his^ would then be the rate with which that 

 single nucleotide pair changed to the one needed to make the gene 

 functional again. Therefore we would be observing in one case the 

 behavior of many mutable units and in the other case the behavior of 

 only one. 



Phenotypic changes resulting from alterations within the gene itself 

 are referred to as point mutations. On the other hand, some variations 

 in phenotype are shown to be caused by rearrangements or loss of large 

 segments of the genetic material. But mutations observed only as 

 heritable changes in phenotype have many properties in common. We 

 will now take up some of these properties and leave for later chapters 

 further consideration of the intimate nature of the change in the genetic 

 material. 



THE RANDOMNESS OF NATURAL MUTATION 



We have noted a kind of unpredictability about the mutations in- 

 duced by nitrous acid. Geneticists have long noticed a similar fact 

 about natural (spontaneous) mutation, and have concluded that it is a 



