THE NATURE OF THE GENE 379 



and the difference is probably one of the degree or kind of spiralization, 

 but in some species hybrids, chromosomes may appear up to ten times 

 larger than they normally do, and this increase is almost certainly due 

 to the accretion of extra chromatic material on to their surface. More- 

 over, it is probable that the inert regions of the chromosome are not 

 tighly coiled at metaphase, yet they appear, with most stains, to be of 

 the same thickness as the active parts ; this thickness may be due to the 

 condensation of material on to the surface. There are then considerable 

 grounds for supposing that a metaphase chromosome contains other 

 material in addition to the chromonema, and that the estimate of the 

 thickness of the chromonema given above is considerably too large. 



Various other estimates have been made of the sizes of genes, but 

 they rest on even less secure assumptions.^ One attempt was based on 

 an estimate of number of genes in an organism, which is an interesting 

 question for its own sake. The total length of the cross-over maps in 

 D. melanogaster is about 285 units. The minimum cross-over distance 

 measured in breeding experiments is about o • 2 units, whence one can 

 deduce the total number of genes as about 1,425.^ Dividing this into the 

 total length of the active chromonemata, one can arrive at an estimate 

 of the length associated with a single gene. But the estimate of gene- 

 number is almost certainly too small. There are about 5,000 to 6,000 

 separate bands in the salivary chromosome complement of D. melano- 

 gaster,^ and we have seen that it is possible that each band contains 

 more than one gene. In other organisms the number of chromomeres 

 cannot be found so accurately. In liliaceous plants there may be up to 

 2,000 in the haploid complement, but each may contain rather more 

 genes than a salivary band. Perhaps both Drosophila and UHes contain 

 about 10,000 genes. 



The two best estimates we have obtained above are maximum 

 estimates for the dimensions of parts of the chromonema ^ in the first 

 place, a length of about 125 m/x seems to be associated with each gene, 

 and secondly, the maximum thickness of the thread is about 20 m/x. If 

 we can accept these two estimates, arrived at in different ways, as of 

 about equal accuracy, it is perhaps significant that even the units of the 

 chromonema are elongated, fibre-like bodies (cf. p. 393). But we must 

 note that, quite apart from the possibility that these figures may over- 

 estimate the size of the chromonema-unit-particle, there is no proof 

 that the whole of this particle is occupied by the gene. It is possible to 

 assume that the gene is a very much more minute body embedded 



^ Cf. Gowen and Gay 1933. 2 Muller 1926. =* Cf. Bridges 1935, 1938. 



