58 



THK BNTOIMOLOGIST. 



segmental divisions ; the legs and claspers are similar in colour 

 to the body. 



From observations I made last year concerning the deposition 

 of the eggs in a natural state, I felt convinced that some con- 

 nection existed between arion larvae and the common yellow ant 

 {Formica jiava) by the preference shown by the butterfly in 

 selecting the thyme growing on ant-hills for oviposition. There- 

 fore, the following extracts from my note-book may be of interest, 

 as they throw some light on the habits of both arion larvae and 

 of the ants. 



From July 5th to 17th inclusive found L. arion numerous. 

 During this period I watched several females depositing, and 

 on the last day saw four laying their eggs on thyme-blossom 

 on the top of a hill ; the thyme grew in patches among the 

 short turf (and on the ant-hills), which was composed of the 

 usual small plants which clothe the surface of the Cornish downs, 

 and with a few furze-bushes dotted about ; but the plants selected 

 were those growing in the open, and some distance from the 

 furze-bushes, therefore fully exposed to wind and rain. Under 

 every patch of thyme visited by the female butterflies I found 

 ants'^ nests. I also saw other females deposit on the thyme 

 growing on ant-hills, and also on the thyme upon the turf-walls, 

 where ants are likewise in abundance. Several captured females 

 deposited on plants potted up during the last three weeks of July. 

 On Aug. 8th I had larvae in four different stages, some only just 

 hatched, one over the third moult, and others fixed for third 

 moult, a large number after first and second moults; after 

 moulting they ate part, and in some cases nearly all, the cast 

 skins. 



On July 29th I found two of the larvae rolling about together 

 under the thyme blossom ; upon close examination I found the 

 smaller one had seized the larger with its jaws, which were 

 buried into its side, apparently sucking it. Upon pulling 

 them apart I placed the victim under the microscope, and found 

 a deep hole in its side, with the surrounding surface shrunken, 

 and liquid exuding from the wound. This conclusively proves 

 the cannibalistic habits of these larvae, which I had always sus- 

 pected, as on previous occasions large numbers of larvae had 

 disappeared in a mysterious manner. I then placed about fifty 

 larvae on as many sprigs of thyme, so as to keep them separate 

 and under very close observation. 



On Aug. 11th, many having passed their third moult, when 

 they cease feeding on thyme, I started investigating what relation 

 there might be between the larvae and ants, thinking in all 

 probability that they might feed either on the larvae or pupa? of 

 the latter. I at first selected one of the larvae after the second 

 moult to experiment with, as I found that after the third moult 

 they do not attack each other ; so that their cannibalistic habits 



