304 ERNEST WARREN. 



habit and the production of relatively large larvae. Out of 

 scores of larvte collected at different seasons of the year from 

 a considerable number of termite nests the smallest specimens 

 found measured at least '75 mm. in length, and Ave may 

 plausibly suppose that very minute lai'vas would not have 

 sufficient strength to obtain the requisite food in the nest of 

 the termite. 



It is less easy to interpret the reason for the great hyper- 

 trophy of the testes in the male. The vesiculge seminales 

 hold a very large amount of semen — sufficientj in fact, to fill 

 scoi-es of spermatheca3. In the majority of insects the male 

 copulates once only and then dies. It is doubtful as to how 

 far male beetles may be capable of repeated copulations ; but 

 there can be no doubt at all that in the present case the 

 production of semen is very greatly in excess of actual 

 requirements. The meagre material collected tends to point 

 to a predominance of females (6 undoubted females and 3 

 undoubted males), but the amount of semen present in one 

 male would suffice for many females. A possible view is that 

 there is no use for the excess of semen, but that the stimulus 

 for the hypertrophy of the sexual organs has been transmitted 

 to the male from the female, where the necessity for the 

 production of relatively large larvae involves the hypertrophy 

 of the organs. In this connection it may be mentioned that 

 the eggs of the remarkable physogastric termitophilous 

 dipteron, Termitoxenia, are likewise of considerable size. 



Dispersal. — The means by which dispersal is effected in 

 the case of the highly modified termitophilous creatures are 

 altogether obscure. These termitophiles are sometimes so 

 greatly specialised that they would be unable to live for any 

 length of time outside the termite nest. The covering of 

 cuticle is frequently very thin, and rapid desiccation occurs 

 in a dry atmosphere. 



On one occasion I observed a young Paracorotoca larva 

 clinging to the side of the abdomen of a termite Avorker, and 

 if Avorkers sometimes enter a neighbouring nest in place of 

 their OAvn, it is possible for larvjB to be transported in this 



