AGRIUS CONVOLVULI. 339 



Ent., xxxii., p. 16). A ? taken in September, i860, at Yarmouth, is stated 

 to have laid a few eggs (Harrison, E. W.I., viii., p. 202) ; whilst an- 

 other, caught at Northfleet, 1870, laid 5 eggs (Button, Ent., v., p. 221). 

 A ? , caught at Christchurch, August 14th, 1885, laid four eggs (Druitt, 

 Ent., xviii., p. 259); another ?, captured October, 1887, at Starcross, 

 laid a number of fertile eggs that hatched in due course (Powley, Ent., 

 xx., p. 304), whilst a 2 was found October 7th, 1886, laying her eggs 

 on a paling at Weymouth in broad daylight, none of its foodplant being 

 near (Cambridge, Ent., xix., p. 280), another ? was found at Epsom, in 

 September, 1895, which also deposited a number of eggs (Morley, 

 Ent., xxviii., p. 281). 



Ovum. — Oval in shape, of a deep bluish-green colour * and 

 thus very unlike the yellower shade of most ova of the Sphingidae. 

 [Hellins describes the egg as pale green, and Poulton suggests that 

 Hellins- possibly only noted the colour when it had changed as 

 a result of the development of the larva.] The shape is also 

 different from that of the closely related species (S. lignstri), being 

 relatively broader. But the most peculiar point about the ovum is 

 its extremely small size when considered in relation to that of the 

 adult larva and the perfect insect. The two main axes of the 

 ovum measure respectively 1 '31x1111. and i'i5mm., while those of 

 the closely-allied, but much smaller, S. lignstri measure 175mm., and 

 i'5mm. (Poulton). I have not measurements of the egg, but I 

 noted that it is not more than two-fifths of the size of the egg of 

 S. lignstri, being decidedly small for so large a moth, and pale 

 green in colour (Hellins). Bartel also notes the egg as " green, 

 small in proportion to the size of the moth, scarcely as large 

 as the egg of Smerinthus ocellata" 



Habits of larva. — The newly-hatched larvae eat some of the 

 eggshell, but the quantity so eaten varies greatly. Of 14 ova 

 examined by Poulton (Trans. Ent. Soc. London, 1888, pp. 515 et sea.) 

 — five exhibited apertures of a size to admit of the emergence of the 

 larva but not larger, one had been nearly half-devoured, three had been 

 about three-fourths eaten, and five had been entirely devoured except 

 for the small part of the undersurface by which they were fixed to 

 the paper. Poulton states that, as soon as the newly-hatched 

 larvae are placed on the foodplant, they almost invariably rest on 

 the underside of a leaf, stretched along the midrib (as in other 

 Sphingids), the "Sphinx" attitude not being observed in this stage, 

 and, further, that but little variation in the position taken up is to be 

 noticed; a small number were observed stretched along the stem of the 

 foodplant or along a leafstalk, and very occasionally upon the upper- 

 side of a leaf (and probably then only because of repeated disturbance); 

 one larva was observed suspended by a silk thread, but the use of 

 silk is much less frequent than in larvae of S. lignstri. The following 

 tabulation shows that the first stadium is considerably longer in the 

 larvae which hatched first, thus tending to an equalisation of the 

 periods of development from the time at which the eggs were laid : 



August 29th, evening, 18 ova laid. September 5th, evening, first larva hatched 

 — 375mm. long. September 7th, morning, s econd larva hatched ; first larva 



* There is a remarkable note {Ent., vi., p. 545) by Mond, who states that a ? , 

 taken on September 12th, 1873, laid a good many eggs, which were, on October 3rd, 

 very dark brown. As the usual egg-period lasts normally about seven days and the 

 colour of the egg is green, one can only suppose that a mistake was made, or that 

 the eggs were infertile. 



