INSTITUTE OF SCIENCE, PHILADELPHIA. 493 



all the characteristics of Pelecypods were in a formative stage, and among these 

 ancient forms variety ran riot. Nearly all the modern types of hinge are fore- 

 shadowed in the Silurian, together with singular forms which have been weeded 

 out by natural selection. Successful development seems to have taken place 

 in three principal lines, or, rather, as the evolution has not been linear, three 

 principal groups, more or less parallel among themselves, yet divergent each 

 as regards the other two. These were regarded by me as Ordinal divisions, 

 and separated in 1889 under the names oi Prionodcsmacea, Teleodcsmacca, and 

 Auomalodesniacea, respectively. The characteristics of these groups will be 

 discussed later; suffice it at present to say that they were based not on the 

 forms of hinge-teeth as such, which would have brought together many incon- 

 grous forms and separated near relatives, but on the development of a general 

 type in each case, to which, in spite of present superficial incongruities, the 

 pedigree of existing genera could be referred. So far as the hinges are con- 

 cerned, the Prionodcsmacea are the product of evolution applied to the devel- 

 opment of (among other things) teeth transverse to the hinge-margin, of crural 

 or of amorphous teeth; the Teleodesiiiacea,o'i se.^x&%^X^ox^. of teeth into different 

 classes on one hinge-plate, including, especially, the utilization of lateral teeth 

 parallel to the hinge-margin; and the Anomalodesinacca. of such experiments 

 in hinge-building as tended to supersede teeth by developing the chondro- 

 phore and resilium, or to retain archaic edentulism by specialties of habitat. 



A number of types of teeth have been recognized by various writers. For 

 the serial alternating teeth of Area and its analogues Neumayr proposed the 

 name of Taxodont, or rather classified as Taxodonta the mollusks which possess 

 them. To a portion o^ the Tcleodesmacea he gave the name o^ Heterudonta on 

 account of the alternation of the cardinals, failing to fully grasp the dynamic 

 principles involved which show that all efficient hinges must be Heterodont. 

 For the bivalves of the Silurian which, in their first stages of dental develop- 

 ment, throw immense light on the evolution of the hinge, and yet are hardly 

 rangeable with modern types, he proposed the name of Cryptodonta. His Desino- 

 donta, founded on the misapplication of a true and fertile conception, may be 

 mostly ignored. 



For hinges like those of Schizodiis, Myophoria, and Trigonia, Steinmann 

 proposed the term Schisodont. To these Neumayr correctly adds the Naiades, 

 while it seems probable that Lyrodesvia, Avicula and its allies should also be 

 included. The more carefully these hinges are studied the more obvious it is 

 that their amorphous character and facile variation distinguish this type of 

 tooth from any other in the modern faunas, though in the Paleozoic beds, in 

 this as in all other cases where sufficiently full data are available, the most dis- 

 tinct types converge. Bittner, in his criticism on Neumayr's classification, was 

 ill-advised in his attempt to harmonize the hinge of Trigonia with that of the 

 Teleodonts. Both are heterodont, as has already been pointed out, like all 

 teeth in the class Pelecypoda ; but the former are certainly more nearly re- 



