46 R. W. CROSSKEY 



from the ICZN, and it has therefore been unavoidable to make a choice for purposes 

 of the present work as to whether to adopt the usage of Dexia, anticipating a 

 Commission case, or whether to use the currently correct nomenclature under the 

 Code. From contacts with fellow specialists it appears unlikely that an application 

 to the Commission to set aside Westwood's type-designation and to have rustica 

 fixed as type-species of Dexia would be opposed, and it has therefore been decided 

 here to adopt the 'nomenclature of usage' for Dexia and Phyllomya, pending a 

 Commission ruling when the case is put. 



It follows from this that Dexilla Westwood, a name employed in an earlier paper 

 dealing with Oriental Prosenini (Crosskey, 1967c), is here treated as an absolute 

 synonym of Dexia Meigen, since both will have rustica as type-species. In this 

 earlier paper I emphasized that many so-called genera from the Oriental area that 

 had been proposed by Townsend could not be satisfactorily separated from Dexia 

 (as Dexilla) when the whole Old World fauna was carefully considered. While 

 preparing the present revisionary work I have again considered these Townsendian 

 genera, and maintain my earlier view that none of them can be justified in the 

 light of the characters that were supposed to differentiate them or in the 

 light of any other combinations of characters. There is no doubt that the Prosenini 

 (Dexiini) have been badly and unnecessarily 'split' on the basis of characters that 

 are completely unstable in the Z)c;«a-complex (such as the presence or absence of 

 a small pra seta, the number of stpl setae, and the number of post ia setae) and 

 that it is necessary to sink many names as synonyms in order to revert to a broader 

 and more definable concept of Dexia. Accordingly I treat Townsend's names 

 Phasiodexia, Eoptilodexia, Eomyocera, Sumatrodexia, Calotheresia, Eomyoceropsis, 

 Asbellopsis and Barydexia as synonyms of Dexia. Also treated as straight synonyms 

 are the subgeneric names Dexillina and Dexillosa proposed by Kolomiets, as (taking 

 Dexia s.l. as a whole) I am unable to find any really convincing way in which a 

 useful subgeneric classification within Dexia can be formulated. 



The principal characteristics that define Dexia and differentiate it from other 

 genera of Dexiini are as follows. 



Dexia Meigen. Head always with a strong (usually subfusiform) facial carina. Para- 

 frontals bare or virtually so. Palpi well developed. Proboscis shorter than head height. 

 Propleuron bare (one species with some minute hairs medially). Humeral callus with two 

 strong setae, sometimes a weak third differentiated. Pteropleural seta present (cf. the 

 Australian genus Senostoma). Pleural hairing almost always all dark. Usually three 

 post dc setae. Cell R$ open. Bend of vein M abrupt, often appendiculate, near to 

 wing margin. Second costal sector haired ventrally. Forms with rather slender facies and 

 usually with extremely elongate legs. 



It will be noted that the character of haired ventral surface to the second costal 

 sector has been italicized in the foregoing characterization. This has been done 

 to draw special attention to what appears to be a character of considerable sig- 

 nificance in defining Dexia. The occurrence of hairing along the lower surface 

 of the second costal sector is unusual in the Proseninae as a whole, and the sector 

 is bare below in all Rutiliini and Doleschallini and in almost all Prosenini other 

 than Dexia. The character of haired second costal sector therefore appears to be 



