OF CERTAIN CHALCIDID^ AND ICHNEUMONIDiE. 87 



posterior in the abdomen of the pupa. I put it by for future examination in a vessel of 

 water, having first made a drawing of it (fig. 2 d). The spiracles, on each side of its body, 

 are oval, corneous, and slightly project from the tegument, and are situated one at the 

 anterior part of each segment, a little above the longitudinal trachea, and immediately 

 anterior to the trachea that supplies the dorsal surface of each segment. The whole of 

 the tracheal vessels are distinct and distended with air. On looking at it this morning I 

 found it dead." So that although the parasite may reside for many months bathed by 

 the fluids of the Sphinx, it perishes when a change occurs in the degree of activity of its 

 respiratory functions. The length of time which it remains in the nymph-state is about 

 a month or six weeks at the utmost, as most of the specimens I have bred from the pupa 

 have appeared in June. The perfect insect makes its way out of the dead pupa of the 

 Sphinx by perforating the case with its mandibles, on the dorsal surface, and sometimes, 

 as in the preparation now exhibited, it becomes fixed in the orifice and unable to escape 

 (fig. 4). 



The body of the larva (fig. 2 a to d) is composed of fourteen segments, or, if the pedal 

 process of the last segment be reckoned, of fifteen. It is elongated, somewhat tapering, 

 and curved in its earlier stages of growth; but is thick, fat, and pointed at its anal 

 extremity, when mature. The pedal or terminal portion of the last segment is pointed and 

 projecting, and is opposable to a process from the inferior margin of the thirteenth seg- 

 ment, with which it forms a kind of forceps, or prehensile organ by which the larva may 

 affix itself, and change its position in the body of the Sphinx The lateral margins of all 

 the segments are thinned and project as tubercles. These are well-marked in the pro-, 

 meso-, and metathoracic and pre-abdominal segments, but are most distinctly tubercular 

 from the fifth to the eleventh inclusive. These latter segments have also distinct tubercles 

 or segmental appendages on each side of the ventral surface in the shape of mammae, and 

 in the position of the false feet of the Terebrantiate Hymenoptera to which they may be 

 regarded as analogous, and as subservient to the movements of the larva within the Sphinx. 

 The lateral tubercles of the Ichneumon-larva have already been noticed by naturalists, but 

 I believe this is the first time that ventral tubercles also have been discovered. In the 

 very young larva they are situated nearer to the side of the body than in the full-grown, 

 and become more and more approximated to the median line, as the growth of the larva 

 proceeds, by the greater extent of growth and development of the dorsal than of the ventral 

 portion of the segments ; thus beautifully illustrating the corresponding process of growth 

 of the segments, and the approximation of the limbs to the median line of the body, in the 

 Myriapoda. But although pedal tubercles exist along the ventral surface of the abdominal 

 segments, the future true legs of the perfect insect are indicated only by six white points 

 on the ventral surface of the thoracic segments, in the precise situation, however, of the 

 tubercles on the abdominal. 



The head (fig. 5) and mandibles (d) of the larva are strong, corneous and of a yellow 

 colour, with the margins and apices of the mandibles black, curved and sharp-pointed, 

 fitted only for piercing and suction, and not for manducation. The maxillae (e) are three- 

 jointed, with the terminal joint broad, triangular, soft and membranous, the second joint 

 very short, and the basal joint strong and elongated. The labium (/) is triangular, with a 



N 2 



