INTRODUCTION 



most cases, resistance may be developed in a series of individually small steps, 

 speaking for the cumulative action of a number of gene mutations as postulated 

 above. The evidence for this comes primarily from kinetic studies of increasing 

 resistance under selection (5, 107, 135) . It has been confirmed by direct crossing 

 experiments involving Chloromycetin resistant mutants of E. coli (6). Progeny 

 of crosses between a sensitive and a fully-adapted (many-step) resistant parent 

 showed a segregation of many intermediate grades of resistance, representing 

 the reassortment of the sensitive and resistant genes in a variety of combinations. 



Streptomycin-resistance shows some unique features in contrast to the 

 agents just summarized. In many species, a mutation conferring full resistance 

 occurs at a rate, about io~ 10 per fission, which overshadows the smaller step 

 mutations characteristic of resistance to other agents (5, 28, 100, 135). This 

 low rate is perhaps the smallest mutation rate to be accurately measured in 

 any organism. A more striking oddity is the mutation which over-adapts the cell 

 to streptomycin so that the resistant mutant is dependent upon streptomycin 

 for growth (95, 100). The function of streptomycin is possibly to regulate an 

 over-expanded enzyme system, analogous to the precedent of a sulfonamide- 

 dependent mutant of Neurospora (141). 



Amino acids and vitamins are not commonly thought of as antibiotics but 

 it has long been known that they sometimes interfere with, rather than promote, 

 bacterial growth (78, in, 117-Snell, 140). The very ubiquity of these com- 

 pounds imparts a special interest to them as possible natural regulators. Of equal 

 interest is the correlation between sensitivity to amino acids and virulence in 

 Salmonella and Brucella (12, 109) which may reflect a hitherto unsuspected 

 general principle. 



It is likely that a biochemical basis will be found for the effects of all types 

 of mutations (29), but microorganisms have been especially prolific in the 

 production of mutants with overt effects on metabolism. For this reason, micro- 

 bial and biochemical genetics are intimately associated and often confused. 

 Gene mutations affecting anabolic processes are usually detected as nutritional 

 or auxotrophic mutants, whose growth depends upon an external supply of 

 the missing metabolite. Auxotrophs have been used for the exploration of many 

 biosynthetic pathways, which are remarkably similar in bacteria, molds, and 

 mammals. Davis' paper (7) outlines this methodology, which is based on the 

 pioneer investigations of Tatum, Beadle, and other workers on the production 

 and characterization of auxotrophic mutants in fungi (30, 69, 117-Tatum) 

 and in bacteria (63, 126, 127, 112). Mutations leading to catabolic defects 

 have been especially useful in bacterial work, both as genetic markers and 

 in the analysis of fermentation pathways (53) . 



Mutations to auxotrophy do not lend themselves to quantitative estimation 

 of rates, despite more or less efficient selective methods for their isolation (88) . 

 On the other hand, mutations from auxotrophy to the nutritionally wild type 

 or prototrophic state are suitable for selective counting methods, but due care 



