8 Bulletin American Museum of Natural History. [Vol. XXIII, 



acute mandibles; the body consists of a number of sharply marked segments, 

 of which the anterior are longer and broader, the posteriof smaller and often 

 telescoped into one another so that it is difficult to ascertain their exact 

 number. There are probably three thoracic and ten abdominal segments. 

 The terminal segment bears a pair of hair-like cerci. Were it not for the 

 absence of legs, these larvje might be regarded as campodeiform and likened 

 to the youngest stages of such parasites as the Stylopidae and Mcloidae. 

 Several of my preparations show these larvee attached to the necks of worker 

 semipupte or pupse, as represented in Plate II, Fig. 13. Twice I have seen 

 a pair of these larvae attached symmetrically on the sides of the same pupa. 

 In other cases they were found on the nuchal or sternal surfaces. 



So different are these minute, sharply segmented and dark brown larvfe 

 from those of other Chalcidid larvae, of which I have seen descriptions or 

 figures, that I should never have regarded them as belonging to the life- 

 cycle of 0. viridis, had I not seen stages like those represented in Plate II, 

 Figs. 14 and 15. These figures represent semipupse of instahilis soldiers 

 with undoubted Orasema larvae .3 mm. in length attached, in the one case 

 to the sternal surface between the pro- and mesothoracic segments, in the 

 other to the nuchal surface. In Plate II, Fig. 14, the larva has its long axis 

 at right angles to that of its host; in Fig. 15 the parasite and host are simi- 

 larly oriented. The dark brown segments of the younger larva are repre- 

 sented in both cases by dark bands on a yellowish white background. The 

 cerci have disappeared. The larvae have plunged their mandibles into 

 their host and have begun to absorb its juices, and this has led to a separa- 

 tion of the more heavily chitinized sclerites and great expansion of the 

 intervening membranes. 



Succeeding stages in the growth of these larvae are shown in Plate II, 

 Figs. 16, 17 and Plate V, Fig. 66. All of these represent female semipupae 

 of instabilis, and in each the Orasema larva, which is attached as in Fig. 14, 

 has attained a length of .7-. 9 mm. In Plate II, Fig. 17, which is drawn 

 from a stained and mounted specimen, the parasite is somewhat shrunken 

 through dehydration and clearing, but in Fig. 16, from an alcoholic speci- 

 men, the skin of the larva is smooth and tense. The fact that all traces of 

 the dark bands have disappeared is probably due to the intervention of an 

 ecdysis between this stage and the one represented in Figs. 14 and 15. It 

 was stages like Figs. 16 and 17, which Avere first seen in my artificial nests, 

 and led me to an erroneous interpretation. The parasite in this stage was- 

 yellowish and semitransparent, while the semipupa to which it was attached 

 was opaque, waxy white and more or less shriveled. The larva, moreover, 

 seemed to make its appearance very suddenly, and this, coupled with the fact 

 that the ants kept licking it till its surface glistened with saliva, led me to 



