78 THE NAUTILUS. 



the more highly developed forms, belong entirely in the Northern 

 Hemisphere, would indicate this. I consider the radial beak sculp- 

 ture a character belonging to the older and simpler forms; the con- 

 centric beak sculpture belongs to the more recent and higher forms. 

 The zigzag radial sculpture of the group Rosanoramphus is a move 

 in the direction of concentric sculpture, hence of a higher order than 

 that which is strictly radial. 



The Tetragenae is a transition group. Its young are contained in 

 all four of its gills, filling them throughout, though they are more 

 numerous as a rule in the outer than in the inner, and this fact to- 

 gether with the general character of the shells and the beak sculp- 

 ture, which is generally more nearly concentric than radial, led me 

 to place these forms in the Unioninoe rather than to make a separate 

 sub-family for them. In every case where 1 have placed a form in this 

 group the shell has deep beak cavites. In the genus Pleurobema, 

 which seems to be the next step in the way of development, while 

 the shells are generally rather short, solid and inflated as in Quad- 

 rula, the beak cavities are invariably shallow, and in all cases that 

 I have examined the outer gills only contain embryos. Here we 

 have characters of the marsupia agreeing with those of the shells. 

 Care must be taken in the examination of the marsupia or appear- 

 ances may lead to wrong conclusions. According to von Ihering, 

 Sterki has found Quadrula heros with only the hinder part of the 

 outer gills filled with embryos. I have seen the same thing in other 

 Quadrulas. I have seen in some of the Unionidse the front part of 

 the gills rilled with embryos while all the rest was empty, and in a 

 number of cases a few ovisacs in the middle or in various parts of 

 the gills entirely empty, while the rest were full, or a few rilled while 

 all the rest were empty. In such cases the empty ones had simply 

 been discharged, the full ones had not. In all the forms which I 

 have mentioned so far the marsupia fill the entire gills and are pad- 

 like, that is, the ovisacs are not marked out separately by sulci. 



Advancing a little in the development of the family we find a 

 number of aberrant forms confined to the Mississippi and Gulf drain- 

 ages of the United States. Each group has certain characters of the 

 shells which we may call generic, though they are not striking. 

 But they are all very wonderful in the character of their marsupia. 

 In such groups as Ptychobranchus, Cyprogenia and Strophitus, the 

 marsupia are astonishing, and in all of them the ovasacs are dis- 

 tinctly marked. 



