THREE DECADES PROGRESS IN MICROBIOLOGY 



reactions will suffice to prevent the synthesis of the compound in 

 question. Since there were already several indications that each part- 

 icular step-reaction is determined by one gene, it might be expected 

 that mutants failing in the synthesis of one and the same essential 

 compound may differ mutually in the nature of the step-reaction 

 blocked. 



This expectation has now been amply confirmed in an investigation 

 of a great number of mutants deficient in their ability to synthesize 

 the amino acid arginine. Amongst these not less than 7 different types 

 have been recognized. Independent biochemical evidence had made 

 it probable that the synthesis of arginine would proceed in this way 

 that first a still unknown series of reactions would lead to the forma- 

 tion of ornithine which then was converted into citrulline which then 

 in a final reaction would yield arginine. It was therefore tested in how 

 far either ornithine or citrulline could meet the growth requirements 

 of the various 'arginine-dencient' strains. It was then found that for 

 one mutant these compounds could not be used as a substitute for 

 arginine itself. Evidently here the final step-reaction had been blocked. 

 For a second type of mutants citrulline was acceptable, but ornithine 

 not, indicating a blocking of the ornithine-citrulline step-reaction. For 

 the remaining strains both ornithine and citrulline were suitable 

 substitutes for arginine which must imply that one of the step-reactions 

 in the long chain leading to ornithine synthesis had been blocked. 

 Since these strains differed mutually, it was obvious that here too in 

 each of the strains different step-reactions were inactivated. 



In the first place these results form a strong argument for the one- 

 gene-one-enzyme concept which in slightly different terms was al- 

 ready so passionately advocated by my famous predecessor Beijerinck 

 as long ago as 191 7. But it is also clear that the production and in- 

 vestigation of biochemical mutants form a powerful tool for the ana- 

 lysis of anabolic processes in the microbial cell. It will be clear that 

 in many cases it will be possible to submit the correctness of assump- 

 tions regarding the way in which various cell constituents are synthe- 

 sized to an experimental test by studying the behaviour of special 

 mutants towards supposed intermediate products. 



It is in this way that Tatum, Bonner and Beadle have made it very 

 probable that the ultimate step in tryptophan synthesis is the conden- 

 sation of indole and serine, and that moreover anthranilic acid is an 



389 



