582 THE PALEONTOLOGY OF MINNESOTA.. 



LNucahdae. 



above the top of the Trenton, ' Group III is by far the largest section of the genus 

 both in its specific and individual development. It may justly be called the nuculoid 

 section, since .not only the general expression of the shell is decidedly like Nucula, 

 but its internal characters likewise approach those of that remarkably persistent 

 type more closely than is the case with any of the other groups here defined, 1 

 think that the evidence indicates very strongly that Nmula was developed from this 

 stock. As is well known, that genus is distinguished from Ctenodonta chiefly in 

 having a small but well defined internal cartilage pit immediately beneath the 

 beaks. Now, although in the Trentou forms of Group III the hinge denticles form 

 a perfectly continuous series, this cannot be said of the Hudson River species. In 

 many, if not all, of these, namely, the series of teeth are more or less distinctly 

 interrupted beneath the beaks by the incipient development of an at least similar 

 pit. So far as it is possible to say, true species of Nucula occur in the Devonian, so 

 it is but natural to assume that the missing links between them and the Ctenodonta 

 levata group of species are to be found in the intervening Upper Silurian deposits. 

 But here we meet with an obstacle in the JEact that none of the Upper Silurian shells 

 that have been referred to Ctenodonta (TelUnomya) and Nucula*, with the possible 

 exception of TelUnomya, curta Hall, of the Clinton group, belong to the C. levata 

 section. It does not, however, follow that such species did not exist, though we 

 must admit that it is a strange, if not a significant fact that they have not yet been 

 found. Still, the significance of their absence is lessened when we consider that the 

 Upper Silurian deposits throughout are relatively poor in remains of Lamellibran- 

 chiata. It is also to be remarked that the forms which have occurred belong chiefly 

 to families widely diflferent from the Nuculidce. It is possible that the Devonian 

 genera Palceoneilo and Nuculites also came from this stock, such a development being 

 faintly indicated by C. fecunda and C. nuculiformis; but taking all the characters 

 into consideration, and the direction of the variation that may be followed into the 

 lower divisions of the Upper Silurian, Clidophorus seems to me a more likely ancestor 

 for those genera. 



Group IV may be a departure from the C. recurva group, but, as it seems to me 

 to be a more primitive type, I would rather consider the relation as reversed. The 

 only objection to the latter arrangement may be removed at any time, since it is 

 nothing more than that C. compressa, a typical species of the recurva group, has been 

 found somewhat lower in the Trenton formation than the earliest known member 

 of the pectunculoides group. 



*Very little is known of the hinge of the Up. Sll. species that have been referred to TeUinomya by Hall and others, 

 so that we are justified In doubting that they really belong to the genus. Those Icnown to possess a denticulated hinge 

 are much jnore lilce PalcBoneilo than Oterwdonta. 



