THE TRENTON LIMKSTONE OF MANITOBA. 79 



tious. Some medium sized ones, a little less or a little more than two inuhes in thick- 

 ness, have as few as three or four annulations to the inch and others as many as six. The 

 annulations, although always rounded at the summit, are by no means always " low," as 

 described by Professor Whitfield, but are often so prominent as to give a strongly ribbed 

 appearance to the shell, and the concave spaces between them are not infrequently 

 broader than the annulations themselves. 



Detached siphuncles of this species are by no means rare in Manitoba, the largest 

 known to the writer (from Big Sturgeon Island) being fifteen inches and a half in length, 

 nearly one inch and a quarter thick at its smaller end, and two inches and an eighth at 

 its larger. The very large and apparently single siphuncular sheath is elongated, conical 

 and rather thin walled, the test of the wall being about half a millimetre thick. The 

 endosiphon has not been observed. 



Endoceras crassisiphonatum. (Sp. Nov.) 



Plates VI, figs. 1-4, and VII, fig. 1. 



Siphuucle (the only part of the shell known) very long and thick, attaining appar- 

 ently to a length of considerably more than four feet, circular in transverse section, nearly 

 cylindrical, but alternately slightly swollen and as slightly constricted at distant but 

 regular intervals, the constrictions, which cross the siphumle somewhat obliquely, being 

 probably caused by the overlapping of the posterior portion of the necks of the septa ; 

 increase in thickness very slow but regular, at the rate, so far as known, of three-tenths 

 of an iuch per foot ; septa unknown, though the distances apart of the annular siphuncular 

 constrictions and their obliquity seem to indicate that the septa also were widely distant, 

 and the siphuucle itself either marginal or submarginal. Endosiphon narrow and nearly 

 cylindrical posteriorly, but widening irregularly and gradually anteriorly. At the anterior 

 end of the thickest specimen collected (which is represented in outline on Plates VI, fig. 

 4, and VII, fig. 1) the diameter of the endosiphon is a little more than half that of the 

 siphuucle. In another specimen (the original of figure 3 on Plate VI) the interior of the 

 narrow posterior end of the siplruncle appears to be portioned ofi by a few transverse con- 

 cave dissepiments. 



Collected at Lower Fort G-arry by Mr. Donald Gunn in 1858 ; and at East Selkirk by 

 Dr. R. Bell in 1880, by Messrs. McCharles and Weston in 1884 and by Mr. Lambe 

 in 1890. 



The most perfect specimen in the Survey collection, which it will be convenient to 

 designate as No. 1, and which is represented in outline, one-fourth of the natural size, on 

 Plate VI, fig. 1, was collected at East Selkirk by Mr. McCharles. Its actual length is 

 three feet all but au inch, and it is obviously imperfect at both ends. It is the only 

 specimen known to the writer in which the increase in thickness is v.ny obvious. At 

 the smaller end its maximum thickness is an iuch and a tenth, and at the larger end just 

 two inches. Its rate of increase, therefore, as already remarked, is three-tenths of an inch 

 per foot. 



Another large fragment, collected by Mr. Lambe at East Selkirk, which is represented 

 in outline, of natural size, on Plate VII, fig. 1, and which may be indicated as specimen 



