TACHINIDAE OF ORIENTAL REGION 51 



encountered and seem to spend much of their time sitting motionless (and well 

 concealed) on the trunks of coconut and other trees not far from the ground, 

 and this secretive habit probably accounts for the rarity of specimens in museum 

 collections. The association with tree trunks is probably concerned with host- 

 seeking, for the known hosts include coleopterous and lepidopterous borer larvae 

 that tunnel in living wood: in the Oriental region Doleschalla elongata (Wulp) 

 attacks Sahyadrassus malabaricus Moore, a hepialid pest of teak and Eucalyptus 

 saplings (Beeson & Chatterjee, 1935), and in Bougainville and the Solomon Islands 

 a species of Doleschalla (not yet definitely named) attacks cerambycid beetle larvae. 



Mesnil (1975a) is the only recent author to have discussed the doleschallines in 

 any detail, and his account of the group appeared whilst the present work was in 

 preparation. As it differs in several important respects from my own view of the 

 group I present below some comments on the possible affinities and the generic 

 composition of the Doleschallini. 



Mesnil associates the Old World genus Torocca Walker and the New World 

 genera Cordyligaster Macquart and Eucordyligaster Townsend with Doleschalla 

 Walker (and immediate allies), so that together these forms constitute his subtribe 

 XXXVI (Doleschallina), and considers that the group so constituted is closely 

 related to the Thelairini. This interpretation of the tribal-group taxon seems to 

 me to be only very doubtfully warranted, for whilst Torocca appears in every way 

 to be phyletically close to Thelaira (as probably are the American Cordyligaster 

 and allies) this is not the case with Doleschalla. On the contrary, the entire adult 

 morphological facies (including male terminalia) of Doleschalla s.l. is essentially 

 that of the Proseninae (Dexiinae) and I can see no reason for excluding Doleschalla 

 from this subfamily. The host-relations, so far as they are known, do not contra- 

 indicate that Doleschalla should be regarded as a prosenine, for members of the 

 genus attack larval Coleoptera. One species is also known to attack a timber-boring 

 swift-moth in India, but the ecological similarity between timber-boring lepi- 

 dopterous larvae and timber-boring coleopterous larvae is so close that it seems 

 fair to suppose that Doleschalla - though a prosenine - has come to parasitize 

 Hepialidae because of their occupation of the same biological niche as timber-boring 

 beetles. In my view, therefore, Torocca and Doleschalla cannot legitimately be 

 regarded as contribal, and I treat the former as part of Thelairini and the latter 

 as constituting the monogeneric Old World tribe Doleschallini within the Proseninae 

 (Dexiinae). 



Four nominal genera belong in the Doleschallini, in the tribal sense here used, 

 but I recognize only the genus Doleschalla itself as valid. The other so-called 

 genera, namely Rhaphis Wulp, Doleschalhpsis Townsend and Macrosophia Town- 

 send, are untenable when the fauna is adequately studied, because of the essential 

 homogeneity of the group as a whole and the fact that intermediates exist between 

 the type-species on which the four generic names are based. Townsend (19366; 

 19396) recognized all four genera, in his usual splitting fashion, even though Wulp 

 (18966 : 139) had already established the synonymy of his own genus Rhaphis 

 with Doleschalla. 



Mesnil (1975a : 1348) has continued to accept Rhaphis as valid, despite Wulp's 



