BIOSYNTHESIS OF PROTEINS 26l 



^vhich these organs carry out in the ^vhole organism. The 

 effect of natural selection is, therefore, only exerted on them 

 as parts of the whole living thing. It is impossible, unthink- 

 able, to imagine the evolution of isolated organs, the " eyes 

 needing bro^vs " of Empedocles, because for them alone the 

 function to which their structure is adapted has no meaning. 



The same applies to the catalytic powers of enzymes. For 

 example, the po^ver of carboxylase to decarboxylate pyruvic 

 acid is of no significance for the enzyme itself, but is only 

 significant for the organism in which the reaction in question 

 occurs. This reaction is only one link in a long chain of 

 energy-yielding transformations of sugar, and if the rate at 

 which it proceeds is co-ordinated with the rates of other 

 metabolic reactions, this will give an advantage to the par- 

 ticular living body in the process of natural selection. It is 

 therefore hard to agree with the recent suggestions of M. 

 Calvin"' about the possibility of the primary origin of 

 enzymes by the gradual natural selection, from among a 

 tremendous number of randomly changing organic mole- 

 cules, of particles with structures which were more and more 

 successfully adapted to the carrying out of particular catalytic 

 fimctions. 



For example, Calvin considers that in the primaeval 

 solution of organic substances there occurred a selection 

 of molecules with a continually increasing amount of car- 

 boxylase activity, like that brought about by Langenbeck by 

 artificial selection of his carboxylase models: methylamine, 

 glycine, phenylaminoacetic acid, aminooxindole, etc. 



According to Calvin, the natural selection of molecules 

 having a gradually increasing carboxylase activity in the 

 primaeval hydrosphere took place because this activity was 

 associated with autocatalysis and accordingly the more effici- 

 ently the molecule of a catalyst could decompose pyruvic 

 acid, the faster the catalyst itself would be formed. 



The models which have so far been constructed by Langen- 

 beck do not, hoAvever, bear this out. Their carboxylase 

 activity increases, btit no autocatalysis can be observed. The 

 artificial models have no such property nor, it would seem, 

 has the coenzyme of carboxylase, vitamin Bj, nor carboxylase 

 itself. One would expect that, if the primary origin of 



