BY R. J. TiLLYARD. 197 



trachea?, and the area whicli they supply, are completelv lost, the 

 petiolation of the iinaginal wing cutting out all the area below 

 Ac, and making A' itself form the hind border of the winir from 

 the base up to Ac. Finally, A^ runs distad as a fine trachea 

 towards Cu.,, but fails to reach more than half-way towards it. 

 It is then continued, in the imaginal venation, by a white band 

 (Ab), which runs on to join the first small descending trachea 

 from Cu.^(Cu.3i. The vein developed from this white band mav 

 be conveniently termed the anal hridfje ( Al)). 



There is only one interpretation to be given to all this, viz., 

 that (Jaliayrion is a reduced descendant from an original ancestor 

 fvhich had a fairly broad anal area supplied by a icell developed 

 four-branched anal trachea. Also, since a difference in the width 

 of this anal area, and in the amount of development of the four 

 branches of A, is still easily to be detected in the fore and hind 

 wings of the Caliayrion nymph, we must go further, and state 

 that this ancestor was of an Anisopterous type; not, of course, 

 necessarily wath triangles developed, but most certainly with 

 hindwings broader than forewings. 



Farther proof of this interesting point can be obtained by 

 studying the formation of the imaginal venation in the nympli of 

 Calingrion. The oiiginal posterior border of the wing is formed 

 at first, in the usual manner, as a pale band running far pos- 

 teriorly to Ac, towards which the remnants of two descend- 

 ing veins can still be seen developing as thin white bands along 

 A., and A3 resi^ectively. Further, the white bands which descend 

 from Cuo to the posterior border, are not only strong and clearly 

 to be seen, but they are of great length comjjared with their 

 remnants in the imago. In fact, the wdiole of this portion of the 

 wing is constrained to pass through, in its ontogenetic de^■elolJ- 

 meut, the past phylogenetic stages of its ancestry, before arri\ing 

 at its pr esent highly specialised petiolate form. Just in this case, 

 the application of the Biogenetic Law of Haeckel seems \ery 

 complete and exact. 



As metamorphosis approaches, the wing-rudiment becomes 

 drawn away, along its posterior margin, from the edge of the 

 wing-case; it shrinks in rapidly, but more especially towards the 



