342 BRITISH LEPIDOPT£RA. 



off. In a state of nature, feeding as they do on a plant so near the 

 ground, no doubt even a heavy caterpillar might not, by dropping, 

 suffer injury. That they avoid sunshine and hide during the day, Sturt 

 was convinced is not the case, for they seemed to prefer feeding 

 in the sunshine if placed there, and would feed freely by 

 gaslight. Poulton adds (Trans. Ent. Soc. Lond., 1888, pp. 548- 

 549) that the larvae are extremely irritable when touched, ejecting 

 from the mouth, on the slightest provocation, large quantities ot a green 

 fluid, containing fragments of leaves. This habit, he says, is quite 

 unknown in Smerinthus, Manduca, or Sphinx liguslri, except when 

 the larvae are excessively irritated ; it is, however, well-known to 

 occur in Hyles euphorbiae ; the habit did not take place in 

 the earlier larval stadia of A. convolvuli. When irritated, the mature 

 larvae curled up and remained in this position for a very long time. 

 They were fond of drinking the drops of water which were some- 

 times introduced on the foodplant. They also freely ate the brown 

 and withered leaves which were occasionally introduced when the 

 plant became frost-bitten in late autumn. The antennae were in a 

 state of continual and rapid vibration. The " Sphinx-like" attitude was 

 never observed in large larvae ; this fact is, doubtless, due to the usual 

 horizontal position of the larva upon its foodplant, which chiefly creeps 

 along the ground. Bartel notes {Pal. Gross-SchmetL, ii., p. 39) that, 

 by day, the larva is extraordinarily skilful in concealing itself, and 

 is found almost exclusively by country people who gather the bind- 

 weed for fodder ; it would appear that, in Egypt, its mode of life is 

 different, for Schneider has met with it repeatedly, usually by day, 

 sitting exposed on the foodplant. Powell writes (in litt.) : " At Hyeres 

 the larvae are usually abundant in September and October ; generally 

 they are of the brown variety and are found feeding on a low-growing 

 Convolvulus, but I have found the greenish form of larva on cultivated 

 Convolvulus in gardens. One may frequently see a dozen or more 

 larvae in a morning when out collecting; I took several on October 

 12th, 1892, rushing wildly along the road." Mathew notes ( E.M.M., 

 xviii., p. 96) that, on September 12th, 1878, he found a single 

 larva in a vineyard, on a plant of Persicaria, at Gallipoli, which 

 turned to a pupa in a few days and, on October 16th, produced 

 an imago ; he further records (En/., xxxi., p. 115) that, on Sep- 

 tember 17th, 1897, at Pirano on the Gulf of Trieste, while 

 passing a fence by the side of a small patch of Indian corn, he 

 noticed a brown-looking object near the ground upon one of the 

 palings, partially hidden among some sprays of common bindweed 

 which was climbing up the fence, and found it to be a large full-grown 

 larva of A. convolvuli^ almost the exact colour of the piece of rail upon 

 which it was resting. Search at the time failed to produce more, 

 but, the next day, on gathering some of the plant for food at the 

 same place, a small larva about a week old was found. On 

 September 20th, three more small larvae were found at Corfu on 

 bindweed brought in for food. On two other occasions food was 

 obtained at Vido, and each time eggs and small larvae were 

 obtained. The larvae fed up very rapidly — one found that had just hatch- 

 ed on September 27th had buried by October 18th, but an accident upset 

 the earliest formed pupa", although four were quite healthy on February 

 15th when the ship was at Alexandria. The same observer states (loc. 



