THE NATURE OF THE GENE 393 



the same two constituents, which perhaps makes it likely that the 

 resting stage chromosomes are built up 'in the same way, and that the 

 chromosome constitution is constant throughout the division cycle. 



The chemical make-up of protein is not yet fully understood. ^ It is 

 known that some of the chemically rather inert proteins such as hair 

 and silk are formed from fibrous elements consisting of chains of 

 "polypeptide links," each Hnk having the constitution -CO-CHR-NH-, 

 where R is a group (the side chain) which may be a simple hydro- 

 carbon, an alcohol, or a base such as arginine. In the fibrous proteins 

 we have mentioned, the links are arranged in linear chains, the chains 

 being connected together by means of the side chains. There are, 

 however, other types of proteins in which the molecules seem to be 

 spherical rather than elongated; these are known as the globular 

 proteins, and there are others intermediate between the fibrous and 

 globular types. We do not know to which of these types the chromo- 

 some proteins belong, since the protein isolated from sperm (clupein) 

 has not been examined from this point of view. The thread-like appear- 

 ance of the extended chromosome, and the two-ended nature of the 

 chromomeres (p. 367), suggests that the chromosome consists of 

 protein fibres arranged more or less parallel to its length. But this 

 does not by any means necessitate the assumption that the chromosome 

 protein is itself fibrous, since the orders of magnitude are quite different. 

 The polypeptide Hnks in a fibrous protein chain are about 0-35 m/x 

 long by o -45 mju, thick by i mfx wide in the direction of the side chains. 

 The visible chromonema is a few hundred m/x thick, while Muller's 

 estimate, based on its length in salivary gland chromosomes, gives it 

 a width of about 20 m/x. Fibres as large as this can just as well be 

 formed from globular as from strictly fibrous proteins, since the units 

 (molecules or repeat cells in crystals) of the former are about 6 m/x in 

 diameter, and cases are known, for instance in some of the virus 

 proteins, in which these units unite to form fibres a few tens of m/x 

 thick. 



Studies on the extensibihty, and particularly the reversible extensi- 

 bility, of the chromosomes give some, and could probably give much 

 more, information about globular or fibrous nature of the chromosome 

 proteins. In completely fibrous proteins the polypeptide chains he 

 fairly parallel and are more or less unfolded; they can only be stretched 

 by actual straining of the chemical bonds. It is probable, however, 



^ Cf. Astbury 1933, 1936, 1938, Bergmann and Niemann I937j Crow- 

 foot et al. 1938, Wrinch 1938, Cold Spring Harbor Symp. 1938. 



