177 



tbat it is not adapted for stinging soft-bodied larvoe. The very curious 

 structural peculiarities of the abdomen, which we have just described, 

 and which are essential to permit the tip of the ovipositor to be pro- 

 jected against the trunk of the tree are also inconsistent with the motions 

 described by Professor Lintner. So, also, the labored force necessary to 

 bring the ovipositor in position, and iu the general act of ovipositiou in 

 Thalessa, does not agree with what is there described. While the rel- 

 ative length of ovipositor to body varies somewhat, the former generally 

 extends about five inches from the tip of the latter, and if brought 

 under the body would extend over three inches beyond the head. 

 Moreover there is no sharp lance at tii>, nor means of curving this last 

 so as to bring ic on the back of a caterpillar with the Ichneumon in the 

 position described by Lintner, 



The ovipositor of Thalessa is, in short, an elaborate boring and saw- 

 ing instrument. The simplest explanation of both Lintuer's and Wood- 

 ward's observations would be that, if the insect was Thalessa, she was 

 by chance boring a branch or trunk infested with Tremex at a place 

 where Dataua larvjc were massing, as they are known to congregate 

 for moulting purposes in masses upon the trunk. But, as will be seen, 

 Professor Lintner's statement is too explicit as to the alarm-jerk of the 

 stung Datana larvae to justify this first explanation of the riddle, and 

 the attitude assumed by Thalessa would not correspond to his descrip- 

 tion; so that upon careful consideration we are satisfied that the true 

 explanation is that some other large Ichneumonid was observed by both 

 and by both mistaken for Thalessa. Some of the large Ophionids of 

 the genera Thyreodon, Exochilum, or Heteropelma might be quite easily 

 mistaken therefor, especially at some little distance. 



Both Exochilum and Hetero])elma are parasitic upon Bombycid larvae, 

 which feed externally like Dataua, and in our breeding experience we 

 have found the commonest parasite of Datana ministra and .Datana in- 

 tegerrima to be a large undescribed Heteroiielma that might easily be 

 confounded with Thalessa litnator, unless one is quite careful in obser- 

 vation. It is true that the ovipositor iu these genera can not be ex- 

 tended to any great length, probably not more than half an inch ; but 

 the abdomen in oviposition is undoubtedly curved under the bod}' in 

 such way that the caterpillars are stung iu front of the parasite very 

 much as described by Mr. Lintner. The abdomen is long enough to 

 allow this, and it is the customary position with Ophionids when ovi- 

 positing. Another, black, species (H. flavicornis) resembles, in a similar 

 way, Thalessa atrata. 



The particular species of Heteropelma which we have bred from Da- 

 tana larva is undescribed, and at Mr. Cresson's request we add a de- 

 scription of it in this connection : 

 Heteropelma datanw sp. iiov. 



Feinale.—Average length 25™™; expanse SS"""'. Geueral color ferruginous-brown, 

 the abdomen verging to bronzy-black. Head: Antennte uniform yellowish-brown, a 

 ] 2357— No. C 2 



