BY R. J. TILLYARD. 193 



had expected, as will be seen by comparing Plate xii., figs. 5-8, 

 also Plate xiii., figs. 2-3. While Argiolestes shows a typical 

 Zygopterid tracheation, with no sign whatever of either bridge 

 01' oblique vein, Synlestes exliibits a very long bridge, which 

 would have appeared as of typical Lestine form, had it not 

 become very cleverly masked, in the imago, by becoming hitched 

 on to Mo close to its origin, instead of joining up to M1.2 in the 

 usual manner. Also, it is very remarkable to see that, in spite 

 of the excessive obliquity of trachea Rs at its crossing from Mo, 

 the corresponding cross-vein in the imago is so little oblique in 

 direction (Plate xii., fig.6,0) that nobody would notice it at all 

 unless he had the larval wing to guide him. In fact, from my 

 long series of Synlestes iveyersi, I can select a fair number of 

 specimens in which the obliquity of this vein is completely lost. 

 To such unprecedented lengths, then, can convergences go, in 

 the formation of apparently similar and closely allied types of 

 imaginal wing- venation, that we can now lay down only one safe 

 rule for the study of the phylogeny of the Zygopt^ra; that is — 

 Never he sure of the homologies of the parts., in any genus oj 

 Zygopteva., until you have studied the tracheation of the larval 

 tvi7ig. 



Having thus shown that Synlestes has no real atiinit}^ to the 

 Podagrionini, towards which group it is a pure and very cleverly 

 masked convergence, Ave must next enquire whether any other 

 genera, now^ included in that tribe, ought to be taken out and 

 placed with Synlestes. To this, bearing in mind the rule I have 

 just laid down, we must give a very guarded reply. Without 

 having seen even the insect itself, and simply from the photo- 

 graphs of wing-venation sent to me by my friend. Dr. F. Ris, I 

 am able to state my a ery strong conviction that the genus Chlo- 

 rolestes will be found, when its larval wings are examined, to 

 possess a l>ridge and oblique vein of exactly similar form to those 

 of Synlestes. In the imaginal wing, the oblique vein can be 

 detected some four to six cells distad from the bifurcation of 

 M2 from Mj, much more clearly than it can be usually seen in 

 Synlestes itself. Chlorolestes, like Synlestes, has the distal angle 

 of its quadiilateral very acute, and Cuj in both genera arches 

 strongly upward away from Cu.j. 



