STADLER 



genie substance is reduplicated in each 

 cell generation. Its distinctive seg- 

 ments, in many known cases, deter- 

 mine v\ hether or not a specific chem- 

 ical reaction will occur, presumably, 

 in some cases at least, by determining 

 the production of a specific catalyst. 



The great bulk of the substance of 

 the cell apparently consists of ma- 

 terials produced by the aforemen- 

 tioned guided reactions. The nature 

 and behavior of these materials, so far 

 as we know them, do not require the 

 assumption that they have properties 

 essentially difiFerent from those of non- 

 living matter. 



The genie substance, on the con- 

 trary, appears to have properties quite 

 difiFerent from those with which we 

 are familiar from our knowledge of 

 the physical science of nonliving mat- 

 ter. Modern physical science gives us 

 no model to explain the reduplication 

 of the gene-string in each cell genera- 

 tion, or to explain the production of 

 eff^ective quantities of specific enzymes 

 or other agents by specific genes. The 

 precise pairing and interchange of seg- 

 ments by homologous gene-strings at 

 meiosis also suggest novel physical 

 properties of this form of matter. 

 These facts indicate that a knowledge 

 of the nature and properties of the 

 genie substance might give clues to the 

 distinctive physical mechanisms of life. 



The difficulties in the study of the 

 genie substance are obvious. It cannot 

 be isolated for chemical analysis or 

 pure culture. The possibility of direct 

 analv^sis of specific segments or in- 

 dividual genes is, of course, even more 

 remote. The properties of the genes 

 may be inferred only from the results 

 of their action. 



Furthermore, a critical study of the 

 eflFects of a single gene may be made 

 only by comparing individuals wholly 

 comparable in genotype except for a 

 difference in the one gene concerned. 



245 



This means that gene mutations are 

 essential for such comparisons, since 

 it is only by gene mutation that we 

 can identify individuals differing only 

 by the effects of a single gene. The 

 prospect of determining the properties 

 of the gene is, therefore, depend- 

 ent upon the development of valid 

 methods for the study of gene muta- 

 tion. 



It is appropriate to cite here the 

 monumental contributions of H. J. 

 Muller to the investigation of this 

 problem. More than 30 years ago he 

 recognized clearly the unique signifi- 

 cance of gene mutation in the study 

 of the physical nature of life (1) and 

 boldly attacked the imposing technical 

 problems that blocked its experimental 

 investigation. 



The difficulties of analysis that have 

 been mentioned are not different in 

 kind from those involved in other 

 problems in which the properties of 

 hypothetical elements must be in- 

 ferred from their effects— for example, 

 in the problems of molecular or atomic 

 structure. In such studies, the inves- 

 tigator proceeds by constructing the 

 simplest model that will fit the known 

 facts and then attempting to apply 

 every significant experimental test of 

 the predictions that may be made from 

 the model. By a series of successive 

 approximations, the model finally 

 evolves to a form that seems to pro- 

 vide the most plausible mechanism for 

 the behavior observed. The study of 

 the physical nature of the gene from 

 purely genetic evidence is closely 

 comparable to this. 



These difficulties of analysis are 

 mitigated in some degree by the pos- 

 sibility of parallel investigation of cer- 

 tain problems of mutation through di- 

 rect observation of the chromosomes. 

 Although the gene-string itself is be- 

 low the limit of microscopic visibility, 

 its behavior is such that it provides a 



