390 BK1TISH BUTTERFLIES. 



energy to feed on the 27th, on the 28th stood quite still all day, with- 

 out capacity for eating, and on the 29th was still standing in the same 

 place, dead. 



Habitats. — Eough slopes at the foot of the chalk and limestone 

 hills of the southern counties of England, sparsely clad chalk or lime- 

 stone downs, rough uncultivated meadows on chalk, and similar 

 localities where Hippocrepis comosa grows, are the favoured haunts of 

 this excessively local species in Britain. Hence it occasionally visits 

 gardens in the neighbourhood, as noted by Webb at Dover, in 1896, 

 and at Reigate, between 1896 and 1900, by Prideaux. It lives in 

 " the Warren " at Folkestone, quite on the face of the cliffs, as well 

 as on the sweep of downs above the town; it is found on the cliffs near 

 Beachy Head, stretching back inland on the chalk range, of which 

 this is the termination, abounds on the slopes near Reigate, and is 

 found, in fact, in most suitable places on the downs of Kent, Surrey, 

 Sussex, Berks, Hants, and the Isle of Wight, extending into the 

 counties of Hertfordshire, Bedfordshire, Bucks, Wilts, Oxford- 

 shire and Gloucestershire, on similar ground, where, however, the 

 number of its localities is exceedingly small ; it is locally common 

 on the chalk and limestone hills and downs in Dorset, on the 

 limestone cliffs near Babbacombe in Devonshire, in a small rough 

 field near the railway at Buckland Dinham, near Frome, in 

 Somerset, on the oolitic limestone above Box Tunnel near Bath, and 

 occurs in Gloucestershire on the Cots wolds, near Painswick, etc., 

 but, except in the counties mentioned, does not now occur elsewhere in 

 the British Isles. Even in some of these localities, it is very local ; 

 on the Ranmore slope, in Surrey, it is found commonly on a piece of 

 ground 100 yards by 30 yards, but it is not to be seen anywhere else. 

 Prout observes (Ent. Rec, iv., p. 278) that, in July, 1893, the second- 

 brood of this species reappeared in some of its old haunts near 

 Sandown and Brading, after an absence of some years. The history 

 of its taking possession, and losing hold, of ground in Bucks and 

 Oxford, is related at length by various observers ; thus Clarke uotes 

 (Ent. Rec, xiv., p. 24) that, on a hill near Marlow, where the butterfly 

 had hitherto been quite unknown, it appeared for the first time in 

 1899, continued in 1900 (a year in which the species was very abundant 

 in Britain), appeared to have quite settled itself in 1901, but three 

 years later, 1904, had entirely disappeared, due probably to the spread 

 of tall coarse grasses that had replaced the few patches of Hippocrepis 

 (<>p. cit., xviii., p. 23). Similarly, in 1899, the species occurred for 

 the first time on the chalkdowns to the south-east and east of Oxford ; 

 an odd specimen had been captured in 1894 or 1895 at Streatley, but, 

 in 1899, on the downs between Streatley and Blewberry, it was 

 observed in abundance in various places, where it had not been noticed, 

 and it was suggested by Holland (Ent. Rec, xiv., p. 50) that it had 

 reached Marlow (see supra) by way of the Berkshire downs ; Bell also 

 noted it for the first time in early September, 1899, on Pyrton Hill, 

 in the Oxfordshire Chilterns, where also it was found in 1900 and 

 1901 (op. cit., p. 51), but this sudden recrudescence and extension 

 of the haunts of the species must have been due to generally favour- 

 able local conditions, as the species was well-known to Henderson, on 

 the Streatley downs, in 1869 and following years (Ent. Rec, xiv., 

 pp. 136-7), and it possibly retained a precarious hold of the ground 



