210 



the nature of gene action since Cue- 

 not's time. As a result of the investiga- 

 tions of the last ten years stemming 

 from the discovery of nutritional mu- 

 tants in Nemospora by Beadle and 

 Tatum (1941), one is now in a position 

 to scrutinize this supposition more 

 closely than was previously possible. 

 Specifically, we are in a better position 

 to trace the consequences of the hypo- 

 thesis and of its various alternatives, 

 and to appraise the evidence which 

 may have a bearing on it. In this paper 

 we propose to examine some of the 

 evidence, deriving from studies on 

 Neiirospora, and E. coli,^ which re- 

 lates to this problem. 



Before considering the experimental 

 findings, it may be useful to define 

 more explicitly the meaning of the one 

 gene-one enzyme hypothesis. The 

 concept is that of a gene whose sole 

 activity aside from self-duplication is 

 that of functioning in the synthesis of 

 a particular enzyme or enzyme pre- 

 cursor. It is not thereby implied that 

 genes at other loci may not also func- 

 tion directly in the formation of the 

 enzyme. This is a completely inde- 

 pendent problem with which we are 

 not concerned here, and regarding 

 which there is little evidence in Neu- 

 rospora one way or the other; all that 

 can be said with assurance is that if 

 two or more genes do, in fact, co- 

 operate in the production of a given 

 enzyme, then their respective contri- 

 butions must be different. Nor does 

 the one gene-one enzyme hypothesis 

 imply that the final phenotypic ex- 



1 The studies on E. colt reported in this 

 paper were supported by a Grant-in-Aid 

 from the American Cancer Society upon 

 recommendation of the Committee on 

 Growth of the National Research Council; 

 by a grant from the Rockefeller Foundation; 

 and by a contract between the Office of 

 Naval Research, Department of the Navy 

 and the California Institute of Technology 

 (NR 164010). 



HOROWITZ AND LEUPOLD 



pression of a mutation is necessarily 

 restricted to a particular structure or 

 function of the organism. The ulti- 

 mate effect of a mutation is the result 

 of an enormous magnification of the 

 initial gene change, brought about 

 through a system of reactions which, 

 originating at the gene, rapidly 

 branches out in various directions and 

 coalesces with similar networks deriv- 

 ing from other loci to form a reticu- 

 lum of as yet indeterminate extent and 

 complexity. It is impossible to decide 

 from the end-effects alone whether the 

 gene has one or many primary func- 

 tions, since on either assumption a 

 complex pattern of effects is expected 

 in most cases. In the biochemical mu- 

 tants of Neiirospora and other micro- 

 organisms, the end effects would, if 

 they could be analysed, undoubtedly 

 prove to be exceedingly numerous. A 

 mutation which induces a deficiency 

 of an amino acid, for example, must 

 secondarily affect the synthesis of 

 virtually every protein of the cell, and 

 an exhaustive enumeration of the end 

 effects might well include every struc- 

 ture and function of the organism. 



It turns out, however, that it is pos- 

 sible in such a case to prevent the sec- 

 ondary damage and the consequent 

 death of the mutant by supplying the 

 lacking amino acid. When given a suf- 

 ficient quantity of the amino acid the 

 mutant becomes normal in growth 

 rate, morphology, and fertility. It is 

 difficult to escape the conclusion that 

 the sole function of the gene in this 

 case is to play some essential role in 

 the synthesis of the amino acid. When 

 biochemical analysis of the mutant is 

 carried farther, it is discovered that 

 the field of action of the gene is even 

 more circumscribed than might have 

 been supposed: it is restricted to sen- 

 sibly a single chemical step of the 

 synthesis. Apparently a single reaction 

 is abolished in the mutant, while all 



