Phenocopy and Norm of Reaction 263 



Rapoport (1947) found that among a large number of chemicals 

 tested certain ones always produced the phenocopy of one definite 

 mutant, and this occurred in 100 per cent of the treated cases. The 

 experiments have been repeated in different places and found to be 

 correct in a general way. But my own ( unpublished ) work shows that 

 there are more details with which an interpretation must reckon. 

 Thus, in some cases, it is clearly a single metal ion which is respon- 

 sible, since all inorganic and organic compounds tested always pro- 

 duced the same phenocopy (e.g., arsenic). We might conclude that a 

 single enzymatic process is inhibited by the arsenic ion, which is one 

 of its known actions. But such known inhibitions work upon very 

 basic enzymatic processes which occur all the time in all cells. For 

 example, arsenate uncouples fermentation and phosphorylation (Mey- 

 erhof). Luciferin and lucif erase in distilled water do not produce 

 light, but they do when magnesium or manganese salts are added 

 (Harvey). Numerous other examples are available to the biochemist. 

 Thus it is not imaginable that arsenic could affect only a single, very 

 specific synthesis which is the chemical basis for the morphogenetic 

 effect. We would have to make some extra hypothesis, based upon 

 the fact that all these phenocopies appear only when the treatment is 

 almost lethal. We might assume that the change of development 

 leading to the phenocopy is simultaneously such as to countermand 

 the lethal effects of arsenic; this means in the end a selection of those 

 individuals which produce, by starting on a side path of development, 

 a condition of arsenic tolerance. This tentative explanation is not very 

 satisfactory, but it shows at least that the chemical agents for the 

 production of the specific phenocopy do not necessarily act by a direct 

 biochemical change. 



Other ions tested (unpublished work), for example, boron, have 

 different effects in different compounds; hence a simple inhibition of 

 an enzymatic process by the ion is improbable. Also, in this case, 

 definite phenocopies appear, not a single one only, but different ones 

 according to the intensity of treatment. Again it is very improbable 

 that a specific enzymatic action is changed by the agent. As we have 

 already seen, boron interferes with carbohydrate metabolism by 

 means of a riboflavin-containing enzyme, which can hardly be con- 

 cerned with a single developmental feature like the development of 

 the antenna. I think that this situation is rather different from that in 

 chicken. Here the phenocopy means that a growth process affecting 

 major parts of the body is interfered with, which could be accounted 

 for by a generalized biochemical effect (meaning one involved in all 



