Bombyliidae, 101 



the ventral side there are similar hairs and they are tuft-like placed 

 at the lateral margins. The last segment terminates in two com- 

 pressed, three-pointed spines, the upper point of which is the largest 

 and curved somewhat upwards. The sheaths of the hind metatarsi 

 stretch out beyond the wing-sheaths and are divergent at the tips. 

 There are prothoracic and seven pairs of abdominal spiracles. 



The biology of the larva is rather curious as it lives as parasite 

 on parasitic Hymenoptera and also on parasitic Diplera; the parasites 

 which the Hemipenthes larva infests seem all to live in Lepidopterous 

 larvæ. The above described pupa was found in the cocoon of an 

 Ophion v^hich it left before the imago escaped. Reaumur (Méni. VI, 

 PL 27, fig, 13) mentions H. morio "d'un nid creusé dans le bois" ; 

 the author thought that it was stored as food by a fossorial wasp, 

 but it is most probable that it had come from some Ichneumonid ; 

 V, Roser bred it from a pupa of Banchus? (Wiirtemb, Gorrespon- 

 denzbl. I, 1840, 52); Portchinshy (Les parasites des criquets nuisibles 

 en Russie, St. Petersb. 1895, 14) bred it from cocoons of Ophion and 

 Banchus which infested Panolis piniperda, and Wassiliew (Zeitschr. f. 

 wiss. Insektenbiol. I, 1905, 1874) bred it from the pupa of Masicera 

 silvatica, parasitic on Dendrolinus pini. The larva hibernates in the 

 cocoon, and the pupation and development of the imago take place 

 in the following summer. It does not seem to be known in what a 

 way the Hemipenthes larva gets on to the parasitical Hymenopterous 

 or Dipterous larva. 



The species o{ Hemipenthes occur on dry and sandy piaces, flying 

 in bright sunshine. Of the genus only one species is known from the 

 palæarctic region, and it also occurs in Denmark (see below). 



Remarks. Low founded the genus Hemipenthes in 1869 and based 

 it on the presence of pulvilli; he knew only one European species, 

 H. morio. Osten Sacken (Biol. Centrali-Americana, 1887, Dipt. 129) 

 declares that besides morio., A. velutinus Hoffm. (as already stated by 

 Schiner) also has pulvilli and that small pulvilli are also found in 

 A.maurus L. and in some American species; he thinks therefore that 

 Hemipenthes cannot stand as a genus. Also Bezzi (Zeitschr. f. syst. 

 Hymenopterol. und Dipterol. VIII, 1908, 34) declares that Hemipenthes 

 cannot stand, as a generic distinctlon from the pulvilli is not possible. 

 (Bezzi uses the name Hemipenthes for Anthrax auct. becauses he uses 

 Anthrax for Aryyramoeha Schin., see above under Argyramoeha.) 

 Perhaps the genus will prove not to be valid, but for the present I 

 retain it here, because it seems to me, that the parasitic habit on 

 Hymenoptera and Diptera may be of some importance in contrast 

 to that of Anthrax. Now it is interesting to see that the species 



