Life Histories in the Hcpialid Group of Lrpiiloptcm. 423 



entrance of the "tube." The length varies to about six 

 inches, and tlioy are either vertical, oblique, or curved ; 

 obstacles probably determine the shape of the " tube," 

 which is practically of the same circumference as the larva ; 

 a large cavity is constructed at the lower end (P. ccrvinata), 

 within which the larva can curl up comfortably. I have 

 never observed the larva3 feeding, but believe the pabulum 

 to be green stuff above surface, not roots ; I have removed 

 green stuff (grass, etc.) as far as two inches below the 

 surface from inside the " tube." I am inclined to think 

 that during wet weather these larvae are often drowned ; 

 on one occasion I found several larva? killed by a white 

 fungus, which formed a complete shroud round the dead 

 larva. 



The larva of Ilcpialus lupalinus, Linne, constructs a 

 "long vertical tube" (" Entom. Record.," vol. iii, 13.124) 

 like our New Zealand Porinas ; this probably feeds above 

 the surface ; other Ilcpialus, however, are root feeders. I 

 remember havinor du» roots of dock with numerous larva? 

 (II. humuli) sticking into them. 



Charagia live in the wood of living trees. The burrows 

 are commenced at an upward inclination and then verti- 

 cally downward (C. vircsccns) ; the entrance is invariably 

 covered by an operculum of silk — Zcuzera is an interesting 

 parallel. Before pupation, the larva constructs a trap- 

 door at the top of the vertical burrow — the only parallel 

 known to me is the hinged covers of the trap-door spiders. 

 Mr. H. Illidge, in his paper on the life history of the 

 timber moths ("Proc. Roy. Soc. Queensland," vol. xiv), 

 believes that the sap constitutes an important item of 

 food ; when the wood ceases to live the larvse die. It may 

 be remarked that the larva^ of C. vircsccns do a great deal 

 of damage to the trees they frequent, not so much indi- 

 vidually as collectively; in one extreme case, a section 

 G feet long 8i- inches in diameter contained two old 

 empty burrows, six inhabited by nearly full-fed larvas, two 

 about half-grown, and four small ; usually, however, three 

 or four burrows are contained in one tree-trunk. There 

 is the difference in habits between European Ilcpialus 

 (subterranean) and Australasian Charagict (wood feeders). 

 The period of larval existence varies from one to three 

 years in the IlcpjialUhv. It is equally difHcult to examine 

 the structural characters of the larva? in the case of wood 

 feeders: the wood has to be split, at considerable risk 



