174 DORSET LEPIDOPTERA. 



I found them very hard to rear owing to the difficulty of preserving 

 the plants in anything like a sound condition, but some of the 

 larvse being nearly full-fed when I got them I luckily succeeded in 

 proving their identity by breeding one moth. It is interesting to 

 note that Pedicularis had been suggested as a probable food-plant 

 for EupceciUa geyeriana no less than 17 years ago by Mr. C. G. 

 Barrett, but in spite of this the larva had never before been detected. 

 From some plants which Mr. Cambridge afterwards sent to Mr. 

 Bankes one moth also emerged. The egg is no doubt laid on the 

 bud or in the flower of the food-plant in June, when the moths, 

 generally rather few in number, first make their appearance, and 

 the larva bores into the seed-capsule and feeds there. It 

 apparently moves from one capsule to another, as I occasionally 

 found a capsule with the seeds partly eaten and with a hole bored 

 in the outer covering not large enough for the exit of a full-grown 

 larva. When nearly full-fed it sometimes feeds upon the capsules 

 from the outside and eats the case as well as the seeds, as in one 

 instance I found that a larva had eaten about half the capsule itself 

 and was feeding upon the remainder. When full-fed (about July), 

 it descends to the earth and spins a thick silk cocoon at or just below 

 the surface. The bulk of the moths appear to emerge early in August, 

 which was the case with the specimen that I bred \ but here a 

 difficulty occurred to me, as it has doubtless to others, as to what 

 the larva could feed on in August certainly not the seed-capsules 

 of Pedicularis, for they are all gone, and Mr. Cambridge tells me 

 that the plants (which are biennials according to Sowerby) are 

 gone also. Under these circumstances it was with extreme 

 interest that I found that one of my larvse, which had spun itself 

 up in a strong cocoon, was still unchanged some time after the 

 emergence of my moth, which I ascertained by opening the end of 

 the cocoon, and on subsequent examinations on Oct. 13th, 1891, 

 and Feb. 9th, 1892, I found that the larva was still alive and 

 healthy and had not turned to a pupa. It is perhaps too much to 

 speak with certainty from this single instance, but it appears to 

 give us the key to the difficulty I have mentioned and to shew that 





