8 BRITISH LEPIDOPTERA. 



Habits of larva. — Sepp worked out (Ned. Insecten, ii., 

 no. 52) the life-history of this species over 100 years ago. He observed 

 that the larvae hatched on July 7th, 6 days after the eggs were laid, 

 moulted for the 1st time 3 days after this (July 10th), moulted a 

 2nd time 4 days later (July 14th), moulted a 3rd time 4 days later 

 (July 1 8th), and a few days later for the 4th and last time, changing 

 colour when wandering off for pupation. Clifford notes (Ent., ii., p. 298) 

 the extreme rapidity of growth after the 2nd moult, there being only 4 

 days between the 2nd and 3rd, and 4 between the 3rd and 4th moults ; he 

 further states that after the larva has formed its cocoon it continues in 

 the larval state nearly a fortnight before the pupal state is assumed. 

 Prideaux says that the young larvae can suspend themselves by a thread 

 when disturbed. Some that left the eggs on July 1 6th, 1901, were 

 observed to moult on July 19th and on July 29th, turning a very dark 

 green colour after the last moult (almost black), and, being fullfed, com- 

 menced to spin on August 3rd and pupated on August 8th (Ransom). 

 Bacot observes that the larvae were exceedingly abundant at Frinton 

 on August 6th, 1900, feeding on the bedstraw growing on the landward 

 side of the seawall ; they were mostly in the final instar, and many quite 

 fullgrown. Referring to their protective habits, he states that they were 

 far less conspicuous than one would have considered possible ; the 

 weight of the larva bends the bedstraw downwards, and the larva 

 feeds with its ventral area uppermost, and, in this position, the 

 colour corresponds well with the sage-green tinted leaves of the 

 small reeds and coarse grasses among which the bedstraw grows. He 

 says that the larvae were so crowded that one could count two or three 

 dozen without moving, probably because the foodplant only grows 

 on a small part of the seawall, about 100 yards in length, and no 

 trace of Galium verum could be found elsewhere in the neighbourhood. 

 Hay ward observes that the larvae were so common in 1865, at Devonport, 

 that above 100 were taken in various stages of growth during the first 

 three weeks of August, every clump of Galium sometimes containing 

 several. In 1888, the larvae were exceedingly abundant on the 

 southeast coast of Kent throughout the whole of September. At 

 Walmer and Kingsdown, on the small isolated patches of Galium 

 growing on the shingle, the larvae of this species (and those of 

 Celerio gallii) were in unusual abundance, those of S. stellatarum 

 hiding low down during the day. The larvae pupated in due course, 

 and either emerged (or died failing to do so) during late October 

 and November. In July, 1894, high up on the French side of 

 the pass leading over the Little St. Bernard, a magnificent plant of 

 Galium verum, growing on a rocky wall by the roadside, tempted one 

 to examine it, with the result that a very large fullfed larva of 

 S. stellatarum was discovered clinging to the underside of one .of the 

 main flowering branches, but not before its presence was made 

 known by its coming in contact with the hand ; it was then feeding, 

 and the time about 7.30 p.m. (Tutt). Larvae were found feeding by 

 night at Painswick (Watkins) ; in nature, nocturnal feeders, and 

 found with difficulty, as a rule between 6 p.m. and dusk. Wherever 

 the foodplant shows signs of being much eaten, search should be 

 made (Ransom) ; appear to prefer the plants of Galium mollugo 

 growing on old walls exposed to the sun (Chaumette) ; appear to 

 feed exposed on G. verum and G. mollugo growing on sunny banks 



