TENTHREDINID^E . 



213 



family, the second segment is shorter than half the length of 

 the abdomen, being much longer and less high and compressed 

 than in the Cynipides, and the ovipositor is retracted within 

 the abdomen. These insects are true internal parasites, re- 

 sembling the Chalcids. Ibalia is a parasite on a wood-beetle. 

 This genus has, by Walsh, been placed in the Cynipides. 

 Figites has feather-like antennae in the male ; it is a parasite 

 on the larvae of Sarcophaga. The genus Allotria is a para- 

 site on Aphis. 



Walsh states thaj two genera, which he has identified as 

 Kleidotoma and Eucoila are true Figitidw, and "have the 

 wings fringed like a Mymar, and the former has them emargi- 

 nate at tip with the radial area in my species distinctly open, 

 and the latter simple at tip with the radial area in my species 

 marginally closed by a coarse brown vein." Eucoila is sup- 

 posed to be parasitic on some insect attacking the turnip. 



TENTHREDINID^E Leach. The Saw-flies connect the Hymen- 

 optera with the Lepidoptera. In the perfect state they con- 

 form to the Hymeiiop- 

 terous type, but as b 



larvae they would often 



be mistaken for Lepi- rf 



dopterous larvae, and |;7.- : 

 in their habits closely K 

 resemble many cater- 

 pillars. The three 

 divisions of the body, 

 usually so trenchantly 

 marked 'in the higher 

 Hymenoptera, are here 

 less distinct, since the abdomen is sessile, its basal ring being 

 broad and applied closely to the thorax, while the succeeding 

 rings are very equal in size. The head is broad and the thorax 

 wide, closely resembling that of the Lepidoptera. The wings 

 (Fig. 144, fore- wing) are larger in proportion to the rest 

 of the body than usual ; they are more net- veined, the cells 

 being more numerous and extending to the outer margin.* 



*In treating of this family we avail ourselves largely of the important Avork on 

 the American species, publishing at the time of writing, by Mr. E. Norton, in the 

 Transactions of the American Entomological Society, vols. 1, 2. We therefore 



144> 



