1895. PROCEEDINGS OF THE XJTIONAL MUSEUM. 299 



JJnio and Castalia, the cardinals being somewhat elongated and broken; 

 and these, with the laterals, are more or less corrugated, and show traces 

 of vertical striation. It was named Castalia sulcata by Kiauss, but 

 was placed in Unio by Lea, and as its specific name was preoccupied in 

 the latter genus, he changed it to l:rai<.s.si. 



dastaUa duprei, Lea, shows characters in the teeth which approach 

 Hyria. It is a smooth, light yellowish green shell of thin texture, trian- 

 gular in outline, and much intlated, with an excessively high, sharp 

 keel running from the beaks to the posterior basal margiu. The cardi- 

 nals are much elongated and sometimes brokeu, as in Hyria. The 

 arch of the hinge plate under the beaks is high and sharp. There is no 

 radiating sculpture, and there appears to be none of any kind on the 

 beaks. I agree with von Ihering that this should quite probably be 

 placed in a new genus. 



Hyria, on the other hand, seems to be equally connected with Unio. 

 In U. stevensi, Lea, from northern South America, the form, sculpture, 

 and external appearance are decidedly like that of Hyria corruyata, it 

 being furnished with quite a distinct anterior dorsal wing and a slight 

 hint at one posteriorly. This species of Hyria is sometimes destitute 

 of a wing behind, and this part of the shell occasionally ends in a some- 

 what obtuse angle. The hinge teeth of Unio stevensi partake, to some 

 extent, of the characters of both genera, though they are more U nion- 

 oid than Hyrioid. The si^ecies should probably, however, be placed in 

 Hyria. 



Unio ortoni, Lea, of which the type — a single left valve, and the 

 only specimen I have seen — is in the Museum collection (No. 25430, 

 U. S. X. M.), approaches the form of U7iio somewhat, but its sculpture 

 is very much like that of Hyria, and its cardinals are multifid. It is 

 very doubtful which genus should receive it, and it quite probably 

 should have a new generic name. 



I think there can be little doubt that the relation between these four 

 genera, Unio, Hyria, Castalia and Gastalina, is a close one anatomically 

 and conchologically, and that they must all be placed in one family in 

 any natural arrangement. Yet in a classification based upon the devel- 

 opment or wantof develoi^meut of the siphons, the former has been made 

 the type of one family, the Unionidiie, and the other three have been 

 placed in another, the Mutelidte. Glabaris, which, as I have shown, 

 may have either perfect siphons or an open mantle, has generally been 

 placed in the genus Anodonta, in the Unionidte, though some authors 

 give it a place in the other family. 2Iycetopus, which has an open 

 mantle, has generally been put in the Unionidje, but it is, as I expect 

 to show farther on, more likely a member of the Mutelidie. 



So far as I am aware, nothing is known of the larval state of any 

 of the African Naiades, so that the character of the embryo, on which 

 von Ihering bases his classificaiuion, can not yet be used in determining 

 the relationships of the peculiarly African genera. 



