THE NATURE OF THE GENE 399 



of a neighbouring link, the same type mechanism may explain the 

 phenomenon of position effect. But in the latter case, the influence is 

 between neighbouring genes (i.e. breakage units) and extends over 

 distances about a hundred times as great as in the former case. No 

 direct analogy between mechanisms of two phenomena is possible; 

 and in fact no example of a direct chemical influence extending through- 

 out such a distance appears to be known in protein chemistry. 

 Certain of the properties of the genes give some hints as to the 



Fig. 160. Table of sizes. — The sizes are ^iven in mjii (= ^0A° = 10""^ mm.). 

 Where only one dimension is given, it is the diameter of a spherical unit. 



(Partly after Stanley.) 



Vaccinia virus 175 



Rous sarcoma virus 100 



Tobacco mosaic virus 430 x 12-3 x 12-3 



Bushy stunt virus 28 



Haemocyanin molecule 59 x 13-2 x 13-2 



S13 Bacteriophage 10 (? shape) 



Repeat unit of virus crystal 15x15x7 



Haemoglobin molecule 2-8 X 0-6 X 0-6 



Protein fibre (repeat unit) 0-334 x 0-45 X 1-0 



Nucleic acid (repeat along fibre) 0-336 



Gene (estimated maximum dimensions) 100 x 20 X 20 



Sensitive volumes: 



gene mutations (Timofeeff-Ressovsky) c. 1 



,, (somatic, Haskins and Enzmann) 15 



cytological effects (Marschak) 5 



possible kind of units which may fill the gap between the o • 334 m/x 

 polypeptide links and the 100 m^u, genes. The most important is the 

 property of identical reproduction. Between two cell divisions, each 

 gene causes the formation of another gene exactly like it; if the gene 

 mutates into an abnormal form, it is the mutated gene which is 

 reduphcated. The gene, then, must in some way act as a model on 

 which the new gene is formed. This can only occur if chemical forces 

 originating in the radicals in the gene can extend far enough to influence 

 the nature of radicals formed in the equivalent places in the new gene. 

 The thickness which we can postulate for the gene is therefore Umited 

 by the distance through which we can imagine such chemical forces 

 extending. Probably the maximum estimate which is chemically reason- 

 able is about 10 m/x, which is the order of magnitude of the thickness 

 of the repeat units out of which protein crystals are built. This is of 

 the same order of magnitude as the estimate given above for the maxi- 



