1921.] R. J. TiLYARD : Epiophlchia laidlawi. 105 



are : — the slightly petiolate wing-base ; the < -shaped nodus placed 

 about half-way along the wings; the form of the pterostigma, 

 strongly built, elongated, about thrice as long as broad ; the presence 

 of the oblique vein ; the absence of any straight or well formed 

 supplementary sectors except iU, 4 ; the positions of the points of 

 origin of Ms and M.. ; the origin of M^ directly from the subnodus ; 

 the great divergence of Cu.^ from Cll^ distally, with the strong 

 arching of the former, especially in the hindwing. All these charac- 

 ters are to be seen in the wings of the larva here under discussion. 



We may now reinforce the argument from the wing-venation 

 by considering the general build and facies of the larva. This is 

 undoubtedly Anisopterous. What non-Anisopterous types are 

 there known which could conceivably possess a stout larva of this 

 type ? Most certainly only those genera of stout, heavy build. 

 The only two non-Anisopterous types known which could possibly 

 satisfy these conditions are Philoganga and Epiophlehia. Philo- 

 ganga is such a huge, clumsily built insect for a Zygopterid that 

 it might well possess such a larva as this ; also, the localitv in 

 which the larva was found might well be a habitat for this 

 Oriental genus. Many details of the venation, however, preclude 

 our acceptance of this solution. Philoganga, for instance, to 

 mention only a few obvious characters, has no hypertrophied 

 antenodals ; its nodus is much closer to the base of the wing than 

 to the pterostigma ; it has no oblique vein between M^ and Ms ; 

 its quadrilateral is almost rectangular, and extraordinarily short, 

 nor does it differ much in fore and hind-wings ; its discoidal field 

 is much narrower than the space below it ; and Cmj, instead of 

 diverging from Cm, distally, converges towards it. Thus we may 

 with safety rule Philoganga out. 



There remains, then, Epiophlebia as the only possible known 

 genus in which the general build of the imaginal body would lead 

 us to expect a stoutly built larval type of the form we have here. 

 The general build of E pi ophlehi a, 2L\oavt from the wings, is distinctly 

 Gomphine ; the same may be said of the larva before us. 



To settle the question whether we ought to place this larva in 

 the genus Epiophlebia itself, or relegate it to a new genus in the 

 family Epiophlebiidae, we have to rely only upon the wing-vena- 

 tion, since the larva of the only known species of Epiophlebia has 

 not yet been discovered. Against the overwhelming array of 

 characters which we have marshalled, in which the wings of this 

 larva agree with those of Epiophlebia supe/sies, we can only mention 

 the following doubtful points : the peculiar condition of the anal 

 vein which appears to run continuously through to Cm^, receiving 

 the basal portion of that vein from above like an oblique vein ; 

 and the presence, in the hind- wing, of a double row of cells for five 

 cells' length, between Cit, and the posterior margin of the wing ; 

 whereas, in text-fig 5, the hind-wing of E. superstes is shown with 

 only one divided cell in this area. 



Regarding these points, it can easily be seen that, in the pro- 

 cess of expansion of the wing at metamorphosis, the slight angula- 



