364 BRITISH LEPIDOPTERA. 



says (Ent. Rec, vii., p. 226) that the larvae buried to a depth ot 

 about 8 inches, though the pot was a large one and they could 

 have gone down at least another 4ms. They formed particularly 

 large earthen cavities, in which were the pupae — remarkably active, 

 very soft and delicate-looking and of a pale shining reddish- 

 brown colour. Proudlock states that the puparium consists of 

 a large smooth cavity made in the earth from a depth of 7ms. 

 to i2ins. Mathew says (in litt.) ; "The fullfed larva buries 

 itself 3 or 4 inches below the surface of the ground, and there 

 constructs a large chamber or cocoon similar to that of Sphinx 

 ligustri or Manduca atropos. The sides are coated with a silky or 

 glutinous secretion which, when dry, becomes hard and brittle. One 

 moth was found to have emerged from the pupa and had died in 

 the puparium, having been unable to force its way through, possibly 

 from the earth in the breeding-cage having become too dry." Bell- 

 Marley notes (Ent., xxxi., p. 68) two puparia, made on December 12th, 

 1897, one so thin that it fell to pieces when touched, the other 

 very* thick * like that of Manduca atropos. 



Pupa. — [My pupae of A. convolvuli are all dead and only 

 one is a fairly good one and that has the segments rather 

 contracted, so that I cannot go into many details of the pupa with 

 much confidence as to accuracy.] The pupa is about 60mm. in 

 length, flattened in front in the thoracic region, but without any 

 S curve. The labrum is quite to the front, though the frons just 

 dorsal to it by something less than a millimetre projects as a rounded 

 eminence, but with a shallow dip between the two. The frons, 

 labrum, and first part of the maxillae are in a line, which is not 

 transverse by a dip ventrally of 17 ; this line is about 5mm. long 

 of which 3 , 5mm. are maxillae; the maxillae then curve ventrally and 

 proceed directly backwards until 15mm. from their origin when they 

 make a curl towards the pupa and then forwards for 7 -5mm. touching 

 the pupa and usually being soldered to it for 4mm. of this, ending 

 somewhat bulbously. The trunk or horn thus formed is rather more 

 than 2mm. wide, rather less in thickness, the round smooth bulbous end 

 being about 2'5mm. in diameter; basally it is also slightly thick. 

 The anterior margin being set back to the labrum, there is about 

 3mm. free as it were anteriorly, whilst the posterior surface is still 

 masked in the origin of the maxillae between the cheeks, which run 

 down beside it parallel to each other for about 4' 5mm. ; the margins 

 then turn directly outwards, curving under the eyes for 5mm. 

 to reach the antennae, the margin of the maxillae following the cheeks 

 for nearly 4mm., the rest being occupied by 1st leg. There is a 

 very narrow spindle of 1st femur against maxilla at from 9mm. 

 to 14mm. from front of pupa. The other measurements given in 

 table below as well as these, are from the only fair specimen I have, 

 which may or may not be an average one. It is, I fancy, rather 

 below average in size, and has the free segments contracted. 

 These measurements show the pupa to have little or no flattening ; 

 it is not, however, cylindrical. An axial line would come to the 

 surface in front near the base of the antennae, and 2*5111111. dorsal 

 to the most prominent front of the pupa, the dorsal line from 



* We suspect from this description that these remarks are based on insufficient 

 data ; the puparium of M. atropos is most thin and brittle (see _postea) . 



