224 The Nature of and Changes in Genes 



chromatic region, probably located around the centromere. Where 

 the prochromosomes are more numerous than the metaphase 

 chromosomes, the chromosomes undoubtedly have more than 

 one large heterochromatic region. In the cold-treated plants of 

 Paris and Trillium, a number of heterochromatic regions of 

 various sizes are found in the metaphase chromosomes, and this 

 number is in close agreement with the number of condensed 

 chromocenters observed in the resting stage. When the number 

 of chromocenters or prochromosomes is fewer than the number 

 of chromosomes or of heterochromatic regions, it is undoubtedly 

 because two or more of these regions have fused together in 

 the resting cell. 



Salivary Chromosome Bands 



It is interesting to speculate on the possibility that each band 

 in a salivary gland chromosome is a gene, but, unfortunately, 

 the experimental evidence on this point is not yet so clear as is 

 desirable. There are, however, several lines of evidence that 

 point to the view that a band is a gene locus. For one thing, 

 it has been shown that genes are found in greatest numbers in 

 the deeply staining euchromatic regions of the salivary gland 

 chromosomes, whereas very few are found in heterochromatic 

 regions. Similarly, most of the bands are in these euchromatic 

 regions and very few are located in regions predominantly het- 

 erochromatic. The frequent association of missing bands with 

 changes in the phenotype of the affected fly also lends support 

 to this view. A study of notch deficiencies caused by X-radia- 

 tion indicates that there is a field or area of the chromosome 

 which will produce a notch phenotype if a break is induced 

 within it. The position in the chroniosome of a number of such 

 breaks within the area or field indicates that a certain band is 

 approximately in the center of this field and provides additional 

 evidence in support of the position that bands are associated 

 with loci. If this association can be assumed as a working 

 hypothesis, some notion of the maximum size of a gene can be 

 obtained. Measurements show that these bands vary between 

 0.2 lUfi and 1 m/t in thickness. If a band is a gene, or if it 

 contains the gene, the gene must be no greater in width than 

 the width of a band. 



