1911.] 51 



Only four of these larva3 emerfijed successfully on the following- 

 dates, May yOth, June 7th, June 10th, Jime 11th, of 1910. Of the 

 rest, I preserved three in formalin at different periods of growth ; 

 several of the ova were destroyed by the emerging larva3, which are 

 most dangerous in this respect, freqiiently attacking the ova near 

 them before one has had time to realise they are clear of their egg 

 shells. One larva was attacked when full grown and on the point of 

 spinning and eaten by another and larger individual. One or two also 

 got disturbed in spinning, and attempted to pupate without covering, 

 a proceeding I have always found fatal to successfvil emergence. One 

 other imago emerged successfully, but was unable to cast its pupal 

 covering, and died in a crippled condition. 



G. dorsalis is, to my thinking, c[uite the handsomest British repre- 

 sentative of the family. The brilliant " apple-green " of the body 

 parts contrasts most vividly with the deep, almost velvety blackness of 

 the various markings. The eyes are a brilliant bronzy-green, wdth a 

 coppery sheen on them. The head parts are yellow. A dark ring runs 

 round the eyes, and there are two other distinct dark markings. The 

 antenna?, are yellowish, darker ringed. The first joint is yellow, the 

 rings becoming closer together towards the apex, which gives the 

 antennse the appearance of gradually deepening in colour. The pa/^i 

 are dark madder ringed with straw-colour. The tvhigs are very 

 iridescent and somewhat thickly clothed with hairs posteriorly. The 

 coloration of the venation is peculiar, and as far as I have observed 

 constant. It is some what Jcomplicated and best explained by the Plate. 

 The stigma is of a lovely soft shade of quiet olive, which gives the 

 finishing touch to an exquisite combination of colour. Most of the 

 imagines have a peculiar diamond- shaped mark, black, on the throat, 

 round wdiich is a suffusion of faint tiu'quoise blue. The legs are 

 green, black haired, the joints of the tarsi being much darkened with 

 brown. A point of distinction, and of great structural importance, 

 between C. dorsalis and C. perla (wdiicli species it most resembles in 

 Britain) is that the tarsal claws are simple in dorsalis, and much 

 dilated in jjerla. This point of difference is shown in the Plate, by a 

 drawing taken from a photograph, for which I am miich indebted to 

 Mr. A. E. Tonge, and also to Dr. Chapman, who kindly prepared the 

 slides from which the photographs were taken. Mr. Tonge has also 

 kindly " manipulated " a very imperfect negative of the larva of 

 C. dorsalis, which now gives some idea of the general appearance and 

 markings, and for this I also owe him many thanks, as otherwise my 

 paper would have had to appear without any representation of the 

 larva. 



£2 



