AND ANATOMY OF CERTAIN HYMENOPTERA PARASITICA. 391 



eggs inside the body o£ a very bristly caterpillar, at least its own size, and 

 often bigger. According to Martelli the Apanteles quite often disturbs its 

 prospective victim. The latter makes violent movements, signifying that it 

 feels that "there is something afoot ^'; the parasite then draws back and 

 waits till the caterpillar's agitation passes off ; the parasite then moves up 

 again, and if it is successful it approaches quite near, folds its antennae back 

 over the dorsal region of its abdomen, curves the latter little by little 

 between its legs, after having raised itself on them. When the extremity of 

 its body gently touches the lateral side of the victim's body, it suddenly 

 darts its ovipositor into the latter and then clings on to the squirming 

 caterpillar. The latter turns and twists, vainly trying to throw off its 

 tormentor ; the Apanteles hangs on, suspended in the air by its front and 

 middle pair of legs. For ten minutes the parasite clings on, and during 

 this time it lays in the hfemocoel of the Pieris larva from sixteen to thirty 

 eggs. It then leaves go, and flies off. The elongate Apanteles eggs give 

 rise by monembryony to the same number of larvse as there were eggs. No 

 account has ever been given of any stages of their development. The larvae 

 of Apanteles and Microgaster are much alike, and several observers have 

 given more or less valuable descriptions of them. 



Ratzeberg (14) has given a very short description of a Microgaster larva in 

 his work on the " Forest Ichneumonidse.^^ He mentions the presence of 

 tracheal tubes, but does not give any detailed attention to the anatomy of 

 the larva. Seurat (12), in a paper on the structure of the entomophagous 

 hymenopterous ])arasites, has given a short description of Apanteles 

 glomeratus and Microplitis seurati (Marsh.). Seurat describes the larva as 

 being formed of thirteen segments plus the head. The last segment consists 

 of an enormous vesicle. There are no tracheal tubes in Seurat's figure, and 

 as the skin is very thin this observer thinks that at this time (when the larva 

 is half-grown) respiration is purely cutaneous. The gut consists of a tube 

 constricted so much at the proctodseum as to be blind, free communication 

 not existing between stomach and rectum. The rectum in its posterior 

 region gives insertion to two Malpighian tubes which run forwards towards 

 the thoracic region. As will be shown below (p. 394), there are no tubes of 

 any kind attached to the hinder region of the gut. 



The jaws in the live specimens may be seen to be continually in motion, 

 and serve to break up the fat-body of the host caterpillar. 



The heart contracts from the back, forwards, and one can see the blood 

 liquid from the vesicle passing into the openings in this region to go 

 forwards. 



There is nothing peculiar about the nervous system except that the 

 abdominal chain does not enter the vesicle. 



When the larvse of Apanteles are a certain size cutaneous respiration does 

 not suffice ; it is at this moment that the tracheal tubes fill themselves with 



