JAMES A. G. REHN 305 



In coloration this species varies as much as all the others of 

 this and the related vaga-zapoteca complexes of the genus. In 

 addition there is a considerable degree of vaiiation in the colora- 

 tion of the proximal section of the marginal field of the tegmina. 

 This may be broadly yellowish, may have this color weak along 

 the costal margin or may have no yellow at all. In no case have 

 we seen the 3'ellow as brilhant, as clear and as strongly contrasted 

 as in fimbriatn {=flavofasciata of Stal and Scudder). Structur- 

 ally this species differs from fimbriata in the distinctly greater 

 width of the proximal portion of the marginal field of the teg- 

 mina and the broader cerci, which have their distal margin trun- 

 cate or emarginate, instead of being of the tapering character of 

 fimbriata. 



Regarding the relationship of the eastern South American 

 forms of this species complex, it seems that desiliens indicates a 

 tendency toward the vaga-vitticeps-zapoteca complex, somewhat 

 away from the Jlavofasciata-fimbriata-infumata group. However, 

 this conclusion is based largely on the evidence of the extreme 

 desiliens condition as found in material from extreme eastern 

 Brazil, while these tendencies are not so decided in specimens 

 from the Rio de Janeiro region. This statement is made chiefl}' 

 on the basis of general coloration and not on genital characters, 

 as there desiliens holds a peculiar position on account of the more 

 tapering cerci. The importance of this feature is, however, not 

 as great as one would imagine from Scudder's comment on the 

 same; fimbriata approaches very close to desiliens in cereal form 

 and a sufficient series from localities in southeastern Brazil will 

 probabh^ show the typical forms connected up in this feature. 

 The relationship of desiliens, fimbriata and infumata as geo- 

 graphic races of the same species ma^- be proven when sufficient 

 material is in hand. Returning to the coloration, after due allow- 

 ance has been made for the recessive and intensive features of 

 the general color pattern, it would seem that the pattern is more 

 truly phylogenetic in certain species of this and numerous related, 

 considerably diversified genera, such as Melunoplus, Dichroplus, 

 etc., than is usually supposed to be the case. To properly appre- 

 ciate pattern, its fixed and transitory features in the scale of 

 individual, dimorphic and environmental variation must be 

 understood. When these are full>' weighed and given their 

 proper value we will probably find in color pattern clues as 



TRAXS. AM. ENT. SOC, XLII. 



