MULTIPLE CHROMOSOMES 567 



reported by Miss Carothers, has similar conditions of less than 

 the typical family number and more than enough V's to com- 

 pensate. Despite these exceptions the va^ue of the evidence for 

 constancy of number and form is strong because of the few 

 departures from type. 



With regard to size as a test for homology there is much less 

 certainty. No chromosome is more definitely to be determined 

 than the accessory chromosome and this falls in no constant size 

 position, either for the family, genus or species. Robertson re- 

 ports it varying in size from number 1 to number 5 in the Tet- 

 tiginae and from number 8 to number 10 in the subfamily 

 Oedipodinae. It seems to vary from number 3 to number 5 in 

 the genus Hesperotettix and from number 4 to number 5 in the 

 species viridis. Variation in size of the individual chromosomes, 

 recognizable by other morphological characters, has been reported 

 by Wenrich ('16) after a most careful study of Phrynotettix. 

 In this latter case variation appears to be the result of actual loss 

 of certain parts of the chromosome, but in general it is due to the 

 extent of condensation and possibly to the amount of more 

 fluid substance held within the chromosome. All our observa- 

 tions indicate the existence of a fairly definite series of sizes 

 throughout the group, and the extreme members of the complex 

 can not be confused, but it is clear that no certain identification 

 of neighboring elements can be made by size alone. For this 

 reason, even if the achromatic bridges did mark the limits of 

 joined chromosomes, it would not be certain that specific num- 

 bers united as is indicated by Robertson. Here it may be noted 

 that in H. viridis chromosomes of nearly equal size, and the 

 largest in the series, are joined together. Practically the same 

 conditions are reported by Woolsey in different species of Jamai- 

 cana. Judging from her figures I should be inclined to say that 

 it is not improbable that neighboring chromosomes, and those 

 of the largest size, are united here. Measurement of the chromo- 

 somes as drawn is not a very accurate way of determining their 

 size, since the amount of foreshortening can not be known, but 

 it is so frequently the case that the element marked ' 14' in her 

 figures of the multiple, is clearly larger than the free number 

 '15,' that the possibility of confusion exists. 



