236 Miss Haviland, Preliminary note on a 



exhaustion. The oviposition took from three to six minutes to com- 

 plete, which is not surprising when it is remembered that the ovi- 

 positor must be thrust through the chitin and body wall of the 

 aphid before probing for the Aphidiiis. Even then the organ 

 must possess an exquisite sense of touch, for the gut of the host 

 larva is so distended with food that the haemocoele is correspond- 

 ingly reduced ; and if the ovipositor were to be driven in a fraction 

 too far the egg might be deposited in the gut cavity, and be lost 

 at evacuation of the meconium. 



The egg is oval, with a pedicel or stalk at one end and a smooth 

 chorion. As development proceeds it becomes more spherical and 

 the stalk disappears. It hatches about two to three days after 

 oviposition. As in certain other internal Hymenoptera Parasitica, a 

 'pseudo-serosa,' or envelope of large deeply staining cells, is 

 developed round the embryo, and presumably fulfils a trophic 

 function. The remains of this embryonic membrane may be found 

 in the host when the larva has ruptured and emerged from it. 



The newly hatched larva is a remarkable form, armoured with 

 dark segmental plates of chitin, which render it easily visible 

 through the tissues of the host. There are a distinct head and 

 thirteen body segments, the last terminating in a long tail. The 

 mouth is produced into a proboscis, within which lie two simple 

 mandibles. The head is furnished with four pairs of chitinous 

 nodules, three on the ventral and one on the dorsal side. Each 

 bears a transparent spot at the summit, possibly sensory in func- 

 tion. The anus, which is dorsal to the cauda, is a large conspicuous 

 structure surrounded by a chitinous ring, and striae of chitin may 

 be seen radiating into the lumen. At this stage, as in other 

 Hymenoptera, there is no passage from the mid- to the hind-gut, 

 but the bulb-shaped cavity of the latter and the wide anus, suggest 

 that in this form it may serve some especial function in early larval 

 life. The duration of this stage is variable. In one observed case, 

 the chitin had been cast when the larva hatched, and was left 

 behind in the pseudo-serosa. In other instances, it lasted from two 

 to four days. Three or four of these larvae may be found in the 

 same host, but, so far as is known, only one reaches maturity. In 

 ecdysis the chitinous skin either splits down the ventral median 

 line or else transversely across the thorax. 



The second stage larva differs from the first chiefly in the 

 absence of the chitinous plates. It is transparent, and the gut 

 contents tinge it pale yellow. The mouth parts are less produced, 

 but the ventral papillae on the head are more conspicuous and the 

 first three segments bear ventral processes. 



In the third stage, the tail is greatly reduced, the thoracic 

 processes disappear, and the cephalic papillae are hardly visible. 

 It was not ascertained whether there was actually a moult between 



