172 PROCEEDINGS OF THE NATIONAL MUSEUM vol. ni 



from the tropicopolitau brevUabiatus (Newport) . For instance Attems 

 regarded them as distinct (1929, p. 112), as did Kraus recently (1957, 

 p. 368). On the other hand Biicherl, who has evidently seen many 

 specimens referable to one or the other from Brazil, considers brasUi- 

 anus a junior synonym of tlie senior Newport name (1941, p. 354). 



Family Schendyiidae 

 Subfamily Ballopliilinae 



Ballophilus riveroi Chamberlin?. One adult male, with 55 pedal 

 segments, Sea Cow Bay, Tortola, March 26, 1958, from Berlese sif tings. 

 The Tortola specimen agrees closely with Professor Chamberlin's 

 original description of riveroi (1950, p. 157), but this description 

 seems insufficiently detailed to permit a positive identification with 

 my single specimen. If the tw^o are actually conspecific and in turn 

 congeneric with the Liberian-type species, davicornis Cook, then the 

 West Indian riveroi, represented in Puerto Rico and Tortola, and the 

 Peruvian peruanus Verhoeff (1941, p. 70) represent the only species 

 of the genus known in the Western Hemisphere at the present time, 

 theu- congeners having been recorded from Africa and the Indo- 

 Australian region. 



CarltohiiUe.\\ new genus 



What little we know now^ of the Ballophilinae does not justify our 

 speaking very confidently of intergeneric relationships. Indeed we 

 may not even be dealing here with a single evolutionary, i.e., mono- 

 phyletic, unit. Too little is understood of generic characterization 

 and content, and too little is known of inter- and intra-specific varia- 

 bility. Certain so-called key or diagnostic characters may reasonably 

 be suspected of having undergone evolutionary convergency. In this 

 connection one might cite the ventral pore fields and coxopleural 

 glands, perhaps even the prosternal sclerotic lines as well. Until a 

 great deal more is known and miderstood, hovrever, these possibilities, 

 as well as the existing ballophiline system, should be regarded as 

 conjectural and provisional. 



It is unrealistic to speak here of relationships, though we can speak 

 of resemblances, a concept that is quite different because it does not 

 necessarily imply anything about descent and evolutionary affinity. 

 The underlying key to genera is intended to accomplish no more than 

 single out the various ballopliiline genera as we now know them. Its 

 groupings do not necessarily imply anything about community of 

 descent. I am certain that its ultility will be short-lived. For the 

 time being, however, it does synthesize what we believe may be 

 meaningful. 



