THE ORIGIN OF VARIATION 



c 



D 



INTERPRETATIONS OF STRUCTURE 



FiGtTRE 86. The Structure of the Salivary Gland Chromosomes 

 AS Interpreted by Kodani. Cross-hatched strands are the chromone- 

 mata, coiled in the plane of the band. The radial bars are the chromatic 

 hairs, which, with interstitial chromatin, fomi the bands. In the inter- 

 band spaces, the chromonemata are hydrated to reach the diameter of 

 the bands. (Kodani, M., /. Heredity, V. 33, 1942.) 



Within such a unit, rearrangements of moderate size may be recognized 

 in the saHvary gland chromosomes, and they constitute the typical position 

 effects whenever a mutant effect is associated with the rearrangement. 

 The next larger sections are perhaps represented by the alternating blocks 

 of euchromatin and heterochromatin, which may be equivalent to the 

 large chromomeres of diplotene chromosomes (much thickened and vis- 

 ibly double, synapsed chromosomes). Rearrangements in which the mu- 

 tant effect seems to be localized at a considerable distance from the break 

 may well function in terms of these units. In such cases the second break 

 is almost always in the heterochromatin. Finally, the chromosome as a 

 whole may function as a unit. 



In such a system, some rearrangements would be neutral; others, the 

 ordinary mutations of genetics, would produce moderate effects; while 

 still others might disturb the chromosomal functions so profoundly that, 

 if viable, it might constitute a completely different reaction system, a 

 new species or even higher group. Natural selection would now act di- 

 rectly upon the new species, usually destroying it, occasionally permitting 

 its increase. Goldschmidt demonstrated that such so-called systemic muta- 

 tions are consistent with the facts of embryology and physiological genet- 

 ics. But he acknowledged that systemic mutations have not actually been 



226 



