ANTHEOCERA FILIPENDULAE. 527 



in June. Henderson found it in abundance in Lundy Island, from the 

 sea level to the highest point of the west coast, near the lighthouse ; 

 the imagines swarmed, whilst the pale straw-coloured cocoons, and the 

 black protruding pupa-cases, were attached to grass culms, heather 

 stems, rushes, and even chalk blocks. B. Adkin records it as common 

 in the Scilly Isles, and R. Adkin from Sutherland, up to 2,000 feet. 

 At Folkestone, Dover, Kingsdown, Freshwater, Sandown, Shanklinand 

 Ventnor, it abounds on the cliffs near the sea, and at Cuxton and 

 Reigate on the chalk downs, inland. So varied indeed are its localities, 

 that one might mention probably most of the possible natural conditions 

 in this country without exhausting them. The banks of a railway 

 cutting at Madeley (Daltry), railway banks at Owston (Dixon), railway 

 banks and rough grass fields nr. Enfield, also the railway banks in the 

 New Forest (Edelsten), on the sandhills nr. Burghead (Gordon), on the 

 coast of Kincardine and Forfar (Reid), coast districts of Kent and 

 Essex (James), more plentiful on the coast of Norfolk (Atmore), on 

 the sandhills between Troon and Ayr, also in the Cumbraes (Dunsmore), 

 on the downs at Eastbourne (Edelsten), and the sides of the limestone 

 hillsides at Clevedon (Mason), a rough grassy undercliff near Brooke, 

 Isle of Wight (Hodges), in fields bordering Epping Forest (Bayne), a 

 grass slope by the sea at Swanage (Alderson), are a few of the reported 

 localities. Harker says that the imago emerges generally from 10.30 a.m. 

 to noon. On the continent it extends from Sicily, Italy, and southern 

 Europe generally, to Finmark and the shores of the White Sea ; it is 

 recorded from the Canary Islands, but not from continental Africa, and 

 the recorded Asiatic localities (excluding Asia Minor), are of the most 

 uncertain character. We have found it high up the mountains in 

 Piedmont (to 5,000 ft.), on the wooded slopes of Lac Bourget, near 

 Aix-les-Bains, in the Forest of Fontainebleau, on heathy ground, and 

 other widely differing places. There are described Arctic and Alpine 

 races, as well as southern and eastern ones. It is rather rare in the 

 Channel Islands, on the cliffs, although it occurs abundantly in the Scilly 

 Isles, etc. Selys notes that in his garden, at Longchamps, there is a 

 gently sloping field, the lower end of which is marshy and abuts on 

 the Geer. In this marshy part he finds A. trifolii, in the upper drier 

 part, A. filipendulae, only occasionally one sees a stray A. trifolii 

 in the latter part, although the field is continuous. This localisation 

 is possibly due to a real specialisation of food-plants, of which, how- 

 ever, little is yet known. Selys has never observed any cross-pairing 

 between the species on this ground, nor noticed any examples that he 

 would consider as possible hybrids. 



Time of appearance. — The insect generally known in Britain as 

 the " early" A. filipendulae, occurring in late May and early June is 

 referred to A. hippocrepidis, Stephs. The true A. -filipendulae, how- 

 ever, sometimes occurs in late June, and continues to do so in different 

 districts until the end of September, the latest appearing, generally, 

 in the marshy districts of our southern coasts. Martorell says that 

 at Barcelona it occurs in May and lasts a month, whilst, in the 

 meadows of Spanish Galicia, Velado notes it as appearing in May 

 and June (are these hippocrepidis, St. ?). In Roumania it does not 

 occur until the beginning of July, and continues until the end of 

 August (Caradja) ; in France, in Ille-et-Vilaine it emerges in July 

 and August, abounding at St, Malo, on the dunes, in early August, 



