MULTIPLE CHROMOSOMES 541 



group in which the number of chromosomes is ahnost constant, 

 and yet, within the one species, there are wide departures from 

 the normal number. If such changes may occur without pro- 

 ducing any effect upon bodily characters that will serve to mark 

 individuals by their variations, then surely it may be argued that 

 there can be no direct relation between chromosomes and somatic 

 structures. This would indeed be a vital objection to the theory 

 if, under its terms, the existence of a fixed group of free chromo- 

 somes of unchanging behavior were postulated. Such how- 

 ever are not the conditions of the theory. If it can be shown 

 that the smaller numbers present in some individual are not 

 caused by the loss of any chromatic units, and further that all 

 the conditions in the group are consistent with the maintenance 

 of certain associations between chromosomes and their chance 

 combinations in fertilization, then the conditions of the theory 

 are not \aolated. In effect this position would extend the theory 

 beyond the observed conditions of the chromosomes to their 

 subdivisions. It would be most exact if it could be based upon 

 constancy of chromomeres, the limit of our observational analy- 

 sis, as has been done for certain chromosomes in Phrynotettix 

 by Wenrich ('16). 



The genus Hesperotettix, so far as our studies have gone, has 

 two species, brevipennis and festivus, in which the normal hap- 

 loid number in the male is twelve; two species, speciosus and 

 pratensis, in which the number is eleven; and one, viridis, in 

 which both of these numbers are represented and, in addition, 

 nine, ten and thirteen. There can be little doubt that the 

 occurrence of the typical family number in the genus is signifi- 

 cant of conformity to type; the presence within one species of 

 -this number, and of variations from it, most strongly indicates 

 that, whatever differences there are, they must not be due 

 to any fundamental disturbance of the chromosome organiza- 

 tion. This a priori argument, while in no sense conclusive, has 

 its value and must be considered. Fortunately however there 

 is strong objective evidence to support it. 



First of this evidence is the case where the number is reduced, 

 from twelve to eleven. This I have already considered at length 



JOURNAL OF MORPHOLOGY, VOL. 29, NO. 2 



