INDUCED CHROMOSOMAL ABERRATIONS IN ANIMALS 1189 



{•111 ones {w-r interval in the A', dp-h and cn-px intervals in the second, 

 ru-D interval in the third chromosome). Finally, crossing over near 

 the free ends of the chromosomes is again infrequent, resulting in the 

 genetic distances being relatively too small (the y-w interval in the X, 

 px-sp, and, probably, al-dp intervals in the second chromosome; in the 

 third chromosome the situation in this respect is not clear). The signifi- 

 cance of these regularities in the distribution of cross-overs along the 

 chromosomes is as yet a matter of speculation, but the importance of 

 these facts for any theory of crossing over is obvious. 



A knowledge of the cytological maps throws light on the phenomenon 

 of crowding of genes in certain regions of genetic maps. The genetic 

 maps of Drosophila melanogaster (and also of other species) show "clus- 

 ters" of known mutant genes located very close to each other, and rela- 

 tively long empty spaces in which genes seem to be scarce. At first 

 glance this phenomenon might suggest that genes in certain regions of 

 the chromosomes are rather more likely to mutate, while other regions 

 are more stable. The crowding of genes proves to be, however, an illu- 

 sion due to variable frequencies of crossing over. The regions of the 

 genetic maps showing clusters of genes are exactly those which are 

 relatively much longer cytologically than they are represented by the 

 genetic maps. Conversely, the spaces with apparently few genes corre- 

 spond to relatively very short sections of chromosomes. If the maps are 

 drawn on the cytological scale, the distribution of genes becomes more 

 or less uniform. Whether this distribution is completely random cannot 

 be decided at present, because of the incompleteness of the cytological 

 maps. The inert region in the X, which, in spite of its considerable 

 length, contains only a single known gene is certainly an exception to 

 this rule. It should also be remembered that the cytological maps now 

 available in Drosophila are constructed exclusively on the basis of studies 

 of metaphase chromosomes. It is entirely possible that if the chromo- 

 somes at other stages of the life cycle (especially at prophase) were 

 studied we might obtain cytological maps which would be somewhat 

 different from those known at present. 



Our knowledge of the cytological maps in Drosophila has entered into 

 a new period of development owing to the introduction by Painter 

 (95, 96, 97) of the method of studying chromosomes in the salivary gland 

 cells. As shown by a number of investigators, notably by Heitz and 

 Bauer (51), the chromosomes in the cells of the salivary glands in Diptera 

 grow to enormous dimensions (compared with the chromosomes in 

 oogonia or nerve cells), and become long and relatively narrow cylinders, 

 consisting of alternating light and dark disks or cross bands of various 

 thicknesses, and showing a number of constrictions and inflated places. 

 The bandings and other characteristics are constant in their relative 



