SCIENTIFIC NOTES. 13 



did not contain a flowering stem, and not one was discovered in the 

 roots, although occasionally when the stem was rather short the larva 

 ate the central portion down as far as the crown of the root, no 

 further. The larvfe were frequently found in the upper part of large 

 shoots, a foot or more above the ground. 



" The locality was low-lying and damp, with a very small stream 

 of water, near an area of bog, and on the borders of a wood. During 

 the remainder of the month other larvse were found in the same plant 

 in similar localities, all damp and low-lying, and no doubt sometimes 

 flooded in winter. Just a few larvae were also discovered in Iris stems 

 growing on the extreme edge of a salt marsh, which was, I was 

 informed, often flooded during the winter months, but I doubt whether 

 the salt water would actually reach the plants in which the larvas 

 were found, although it would seem quite possible. In this latter 

 place, and in others near the coast, larvae of H. micacea were found in 

 the roots of Iris on July 16th and onwards, the larvae of this species 

 being always in some portion of the roots, and eating only so far as 

 the base of the shoots. 



" I ought not to attempt to describe the larvfe of H. crinanensis, 

 as I have no notes of my observations. But I remember that it was 

 of a dirty-white colour, with some longitudinal stripes of a pale 

 pinkish, or flesh-colour, which became less pronounced when the larva 

 became full grown. There were also black dots on the segments. 

 Several of the larvae were ' stung.' 



" I experienced no difficulty in feeding the larvse. I cut fine, 

 healthy shoots of Iris, with a fair portion of root attached, in order 

 that they should remain fresh as long as possible. I then cut off the 

 ends of the shoots and formed a hollow in the centre. I placed the 

 head of the larva towards this hollow, at the same time making him 

 very uncomfortable by tickling his tail. He immediately made for 

 the hollow, and satisfying himself that it was his proper food-plant, 

 at once settled down and made himself at home. It was not even 

 necessary to change the stems, as I placed only one larva in each, and 

 ihere he remained until full-fed. 



" When full grown the larvae leave the stems and pupate in earth. 

 I lost a number at this period, as they burrowed about a good deal in 

 the earth, and those which had already formed their cocoons were 

 often disturbed by later arrivals. This naturally often ends in mal- 

 formed pup^ and death. I was rather surprised at this as I gave 

 them plenty of earth and they were not overcrowded. 



" The imagines emerged from July Slst to August 10th, and show 

 an interesting amount of variation of colour." 



Mr. Sabine also kindly sent me further specimens of H. crinanensis, 

 taken in another locality in 1914, on " Ragwort " flowers, "on the 

 lower slopes of a limestone mountain, the nearest fresh water being 

 about a mile away." This does not militate against the conclusions 

 drawn from all former experiences that this species is always connected 

 with fresh-water streams. These examples may have flown to the 

 higher ground. 



Mr. Sabine's discovery is an interesting one. Our previous idea, 

 that the food-plant of H . crinanensis might be Scabious or some other 

 stream frequenting plant, is finally answered. It is for friends who 

 have found the insects to confirm the observation by stating whether 

 Iris occurs in the localities where it occurred. 



