5 o R. W. CROSSKEY 



able aberrant form now being described, when it was said that it would not satis- 

 factorily fit into any of the described genera of Simuliidae. At that time only the 

 quite unique larva and the exceptional form of pupa had been studied, the larva in 

 particular being found to have characters apparently allying it to Prosimuliine 

 black-flies. For the detailed study and description now given, both sexes of the 

 adult have been dissected from mature pupae (reared or wild-caught adult flies are 

 not yet available) and from the characters of these adults I have been forced to the 

 conclusion that, in spite of the very atypical immature stages, the new species 

 atlanticum must be placed in the genus Simulium sensu lato, where for a balanced 

 classification it should form a new subgenus of its own (Dexomyia sgen. n.). 



It has all the following adult characters of Simulium s.l. : Costa and R\ with 

 spinules as well as hairs ; vein Cu% sinuous ; Rs simple ; no evident basal cell ; 

 calcipala and pedisulcus present ; antennae u-segmented ; last segment of female 

 maxillary palp very long and slender ; male style with one apical spinule ; katepi- 

 sternum delimited by deep almost complete mesepisternal sulcus. In addition it 

 shows in the pupa no long strong terminal hooks, no contorted or anchor-like apico- 

 lateral spines, and the pupal abdominal cuticle in the form of a thin transparent 

 pellicle without brown sclerotization as in Simulium s.l. (i.e. without Prosimulium- 

 like or Cnephia-like features) ; furthermore, the cocoon is a strongly-woven shoe- 

 shaped pocket. 



The main clue to the probable phyletic affinities of Dexomyia within the genus 

 Simulium lies in the adult male, which (except for hairing of the pleural membrane) 

 is virtually indistinguishable from that of Eusimulium. The male hypopygium 

 (Text-fig. 311) with its keeled lamellate ventral plate and single very strong out- 

 wardly-directed parameral hook on each side is exactly of the type found in the 

 ruficorne-group of Old World Eusimulium, and generally similar to that of Pomeroy- 

 ellum from the African mainland. This type of hypopygium does not seem to occur 

 in any South American segregate of Simulium, or in the North American fauna 

 (although that of the latipes-group of Eusimulium there is similar) and the conclusion 

 appears justified that S.(D.) atlanticum sp. n. has derived from Eusimulium-like 

 forms similar to the ruficorne-gronp and reached St. Helena from the Eur- African 

 side of the Atlantic. But although the male hypopygium is so identical with the 

 ruficorne-group, perhaps indicating monophyly, this alone would not in my view 

 justify including atlanticum in the subgenus Eusimulium, to which it will not fit on 

 its whole constellation of characters (though there are other resemblances to 

 Eusimulium in the haired base to the radius, the simple female gonapophyses, the 

 slender fore tarsi, the very large female claw-tooth, and the four-filamented pupal 

 gill). The subgenus Dexomyia is thus seen as having a closer relationship with 

 Eusimulium than any other world subgenus of Simulium, at least more characters 

 in common whatever the phyletic significance of this may be. 



Both sexes in Dexomyia have the scape and pedicel segments of the antenna rather 

 large and the flagellar segments rather loosely articulated in the manner of more 

 ' primitive ' Prosimuliine forms, and the head form of the female is also exceptional 

 among Simulium species for the eyes are relatively reduced in size, so that in profile 



