THE WINGS OF NEUROPTERA 



153 



Figures 146 and 147 represent the wings of a normal Osmylus hyalinatus. 



An important fact bearing on this question of the switching of the base 

 of a vein from one support to another is that in the family Sympherobiidae 

 the base of vein R0+3 has been transferred to vein Ri. 



The third possible method of suppression of the dichotomy of the radial 

 sector would produce the result shown diagrammatically in the fifth 

 diagram (Fig. 138, 5), where veins R4 and R5 are represented as coalescing 

 completeh'. I know of no case where the suppression of the dichotomy has 



Fig. 146. — Wings of Osmylus hyalinatus with the radial sector 

 pcctinately branched. 



been attained by this method. I am now convinced that in the supposed 

 examples of this method, cited by Comstock and Needham, that is in 

 Corydalus and Chauliodes, the result has been attained by the splitting apart 

 of veins Ri and Rr„ instead of by their coalescence. 



The earlier conclusion was based on the fact that occasionally the first 

 branch of the radial sector in Corydalus is forked, as is the case in the wing 

 represented by Figure 148, i. The forked condition of this first branch 

 was believed to indicate that it is a compound vein in which the coalescence 

 of its two elements is not quite complete. My reasons for abandoning this 

 view are the following:. 



