136 I^VER TEBEA TE FA UXA OF DA KG TA FORMA TION— WHITE, vol. xvii. 



Concluding remarTcs. — It is true that the collection of invertebrate fos- 

 sils described in this article does not add materially to our knowledge of 

 biological forms, but in several respects it possesses unusual interest. 

 It is, as has already been mentioned, one of only three collections of 

 in vertebrateremaius thathave been made from the Dakota formation, the 

 strata of which we have abundant evidence to believe originally occu- 

 pied many thousands of square miles. It indicates more distinctly 

 than any previously discovered facts have done, the noumarine char- 

 acter of that formation. It embraces four genera which have never 

 before been recognized in collections from its strata. Although that 

 formation lies at the base of the Upjier Cretaceous series, a majority 

 of the species which this collection contains belong to genera repre- 

 sentatives of which are among the characteristic members of the mol- 

 luscan fauna now living in the waters of the Mississippi drainage 

 system. 



Of the few species that were discovered by Dr. Hayden a part belong 

 to genera which are generally regarded as indicating a marine habitat, 

 a part are such forms as usually inhabit estuarine or other brackish 

 waters, and one Avas referred by Mr. Meek, who described these fossils,* 

 to the genus Margaritano. I have no reason to doubt that this species 

 belongs to the Uuionid?e, but tlu^. ty^e sjjecimens do not satisfactorily 

 show the hinge structure and other features upon the modification of 

 which the different genera of that family are established. 



Dr. Hayden did not find the forms which have just been mentioned 

 as indicating a marine habitat in immediate association with the shell 

 which Mr. Meek referred to MargarUana^ its only associate having been 

 a form which he referred to Gyrcna. According to our present knowl- 

 edge of the habitat of the different molluscau genera the association of 

 Margaritana and Cyrena is incongruous, because the former genus is 

 never found living in saline waters, and the latter never in fresh. I 

 think, however, that the shell referred to Cyrena by Mr. Meek may be 

 properly referred to Corhicida, the shell characteristics of which genus 

 are so nearly like those of Cyrena that it is often difficult or impossible 

 to diagnose them as different in the fossil condition. Species of Corbi- 

 cula are not unfrecpiently found living in fresh waters, and we have 

 abundant evidence that fossil, if not living, forms of Unio and Corbictila 

 lived and thrived together. I therefore regard it as reasonable to infer 

 that the Dakota strata in which the two species referred to were dis- 

 covered were deposited in fresh, or at most, in brackish waters. The 

 discovery of remains of a couple of species»of characteristic Dakota 

 plants commingled with these fossil molluscan forms leaves little or no 

 room for question as to the Dakota age of those strata. I have also 

 little doubt that the layers from which came the other molluscan forms 

 discovered by Dr. Hayden in the district just mentioned are near the 



*For descriptions and iioures of these species, see Vol, ix, U. S. Geol. Surv. Terr., 

 pi. 1, pp. 92, 114, 159, 206, 251. 



