16 HEREDITY. 



the name of potential variability, and for so far as this goes, 

 the group is amenable to change by selection or otherwise. But 

 in those cases where we are sure, that the origin of the group 

 insures a purity for one genotype, an absence of potential 

 variability, selection has been shown to be ineffectual. 



For this reason, no theory of the nature of the genes needs 

 to make a provision for qualitative variability of the genes 

 themselves. And this point was, we think, the only justifica- 

 tion for a supposition that the genes are vital, protoplasmatic. 



Protoplasm is clearly an emulsion, and it must be ultimately 

 made up of a number of non-living substances, the combina- 

 tion of which makes it living. One of us has compared the 

 attitude of the vitalist who reasons that every constituent of 

 protoplasm which is an integral part of it, and which shows one 

 or more properties of protoplasm, must itself be protoplas- 

 matic and living, to the attitude of a philosophically inclined 

 eater of plum-pudding, who would argue that the round, sweet 

 things he could dissect out of his helping, and which looked like 

 raisins could not be raisins, as he found them in his plum-pud- 

 ding, and forming an integral part of it, they must consist of 

 plum-pudding. 



Quantitative propagation combined to qualitative stability 

 is not exclusively a property of protoplasmatic bodies multi- 

 plying by bi-partition. Those chemical substances which have 

 autokatalytical properties till both requirements, they propa- 

 gate themselves, that is, suitable materials are changed into a 

 new substance under the influence of the presence of that sub- 

 stance. Also, they remain qualitatively unchanged. Some 

 years ago one of us therefore published the hypothesis that 

 genes are relatively simple chemical substances, non-living 

 things, having autokatalytical properties. 



Is this theory compatible with the facts known about genes 

 and the action of genes? In the first place it does not admit of 

 variation within the genes, or even within these groups of organ- 

 isms which are known to be pure for all their genes, groups 

 without potential variability. It is for this reason, that we have 



