REPORT OF THE DIRECTOR I9IO 175 



THE FAUNA OF COFFIN ISLAND 

 Nodosinella clarkei nov. 



Shells long, slender, branching, nodose, usually nearly straight. 

 Test thick, imperforate so far as can be determined. Nodes well 

 defined, quite as wide as long in all specimens sufficiently exposed 

 to show full diameter. Diameter i mm, 6 or 7 

 nodes in 5 mm. Shells apparently monothala- 

 mous. In sections cut deep enough to avoid the 

 sharp, keel-like edge of the constriction between 

 chambers no septa are distinctly shown. Xo 

 indubitable septa seen. 



Remarks. This species appears to be related 

 to N. (Dentalina) priscilla Dawson,^ 

 but differs in being i mm in diameter instead of ^ 

 a fortieth of an inch, and the modes are wider ^5^ 



than lone instead of beins^ considerably long^er ^ 



., . ,., -,1 , Nodosinella clarkei 



than wide, and the test is thicker. In all these nov. x 1.5. 



. Loose at Grand Entry 



respects it agrees closely with N. d 1 g 1 1 a t a 

 Brady. Our specimens differ from material of either species 

 as described and figured, in branching rather liberally. It would 

 seem to be impossible that the great masses of specimens mentioned 

 by Dawson as occurring in the Windsor limestone, even though they 

 were fragmentary, did not include branching forms. None of our 

 specimens show the plane base indicated by Brady for the British 

 species. The shells are i mm in diameter while Brady's specimens 

 varied from i to 2 mm in diameter. 



These tests are uniformly about a millimeter in diameter, except 

 where a segment is enlarged to give oflf two or more branches, 

 while Brady's species frequently reach a diameter of 2 mm. The 

 tests in thin section under the microscope appear to be nearly homo- 

 geneous with a little rusty coloring. No indication of foramina is 

 visible. Considering the very fine calcareous character of the matrix 

 their presence in the tests originally seems improbable. From the 

 state of preservation it seems questionable if they could be referred 

 to the Lagenidae as suggested by SpandeP and others. While a 

 larger amount of material may demonstrate the presence of the 

 characters of this family, I am inclined to leave the specimens in 

 Brady's genus since the ramose character of the species seems in- 

 compatible with such shells as Nodosaria. 



1 op. cit. p. 285. 



^ Die Foramen, des Deutschen Zechs. etc. p. 6, 1898, 



