192 DEVELOPMENT OF THE WING-VENATION OF ODONATA, 



triangle " of early Anisopterid type. The greater breadth of the 

 hind wing is, in itself, evidence of a descent from Anisopterous 

 ancestors; while the fact, that the forewing-qiiadrilateral is 

 already of Lestine form, strongly suggests that the latter group 

 are highly reduced descendants of an originally weakly-triangled 

 Anisopterous stock. In that case, the sharply acute distal angle 

 of the quadrilateral is the last remaining piece of evidence of the 

 originally present triangle. It does not follow, of course, that 

 any of the Zygoptera which possess fairly regular quadrilaterals 

 were descended from similar ancestors; in fact, the regularity of 

 the quadrilateral is an almost certain sign that all these forms 

 branched off as reduced members from the main stock long before 

 the first beginnings of the formation of the A nisopterid triangle. 

 On this view, the Calopterygidce and their more highly reduced 

 descendants, such as the Podagrioniiii, are a far more archaic stock 

 than the Epiophlehia-Lestes line, which has a truly Anisopterous 

 ancestry. As regards the acutely-angled quadrilateral of the 

 Agrionini, we should hesitate to describe that also as the result 

 of a reduction from an early Anisopterous form, since we have, 

 at present, no other evidence in favour of that view, and a good 

 deal against it. It is probably much more truly a specialisation 

 of an originally regular quadrilateral. 



To turn now to our second point - can we indicate any other 

 Zygopterous forms which possess a bridge and oblique vein ? A 

 search through the whole series of known imaginal wings will 

 not reveal a single case with certainty; hence we must turn to 

 the study of the larval wing. Now there is, in Australia, a 

 genus (Synlestes) which is usually placed in the Podagrionini 

 with Argiolestes and its allies. I have, for years, been struck by 

 the similarity in habits, appearance, and, in particular, larval 

 form, between Synlestes and Lestes, and also by the wide diver- 

 gence in all points (except wing-venation) between Syiilestes and 

 Argiolestes. I, therefore, expected that the similarity in wing- 

 venation between these two latter genera must be due to con- 

 vergence. This year, I obtained nearly full-fed larvae of Sy7ilestes 

 weyersi, Argiolestes griseus, and A. icte7'omelas, and photographed 

 the larval wings. The result was even more astonishing than I 



