THE MUTABLE UNIT OF HEREDITY 



57 



are cases of deletion where the actual substance of a gene may be absent. 

 Occasionally, deletions of parts of chromosomes may be so large as to be 

 visible with the microscope. Unless another complete chromosome is 

 also present in the same cell, large deletions are usually lethal, probably 

 because they include at least one gene essential for the life of the cell. 

 But not all lethals are deletions. This histidine-requiring mutant re- 

 ferred to in Table 2.3 is a lethal in an environment not containing that 

 amino acid; the fact that it back-mutates shows it is not a deletion. 

 Lethals may be maintained for study either if there are other genes in the 

 cell which can perform the same function or if the cell can otherwise be 

 supplied with the needed requirement. Mutation to lethal conditions 

 has provided methods for measuring mutation rates that are not in- 

 fluenced by the subjective errors involved in scoring visible traits. 



CONCLUSION 



The properties of the mutable unit described in this chapter are drawn 

 largely from examples of mutation in bacteria. As stated at the outset, 

 bacterial populations provide convenient systems for the analysis of this 

 process, but the legitimate question may be asked whether the results 

 found among them can be generalized. Certainly nothing is known from 

 studies of higher organisms or of other microorganisms that contradicts 

 the findings in bacteria. It is perfectly reasonable to expect that in the 

 course of evolution the genetic systems of bacteria and other organisms 

 have come to differ. But there also are common denominators — for 

 example, the nucleotides they contain. Recent work in biochemistry 

 and genetics has greatly strengthened our evidence for the basic unity of 



FIGURE 2.8. A kernel of 

 maize possessing the genes 

 a and Dt. The fact that the 

 spots are small and of approx- 

 imately equal size shows that 

 the mutations from a —> A in- 

 duced by the Dt gene all oc- 

 curred at about the same time 

 late in development (after 

 Rhoades, 1936, J. Genetics, 

 33:347). 



