144 PROCEEDINGS OF THE ACADEMY OF 



i 



near the summit, some of which measure as much as three times the diameter 

 of those of the corresponding pieces in the Canadian species of equal size. It 

 is true these are probably, to some degree, variable characters in this genus, 

 but not, we should think, to the extent exhibited between the Canadian species 

 and our specimens, in which latter they are constant. Again, where the 

 sutures of our species have been worn so as to expose the perforations, they 

 are seen to be less crowded, and not so numerous as in C. punctatus, while 

 none of the plates, even where apparently perfectly preserved, show any 

 traces of surface stria?. 



The deep concavity of the external surface of the plates in this genus, and 

 the sharply carinated character of the sutures between, together with the 

 irregularity in the size, firm and arrangement of the plates, give a very pecu- 

 liar appearance to the fossil, that might, at a first glance, cause it to be mis 

 taken for a coial. When only found in the condition of detached plates, they 

 present a singular appearance, well calculated to mislead even an experienced 

 palaeontologist who had not seen the entire fossil, or enough of the plates 

 united, to show their true characters. The fact that they are all deeply con- 

 cave, and when unworn, smooth on the outside, while the inner side is convex 

 and strongly rayed, would naturally lead to the conclusion that the outside is 

 the inner side, and vice versa. When a few of the plates are found united, 

 however, it is at once seen that the deep concavity is on the outside, and 

 the convexity and rajs within. These rays extend one from the prominent 

 middle of each plate to each of its sides, where they connect with those coming 

 from the middle of the adjacent plates. When three or four of the united 

 plates are placed with the inside upwards, the spaces between the rays are 

 seen to present the form of deep, triangular pyramidal cavities, the apex of 

 each cavity terminating at the meeting of the corners of each three of the con- 

 tiguous plates. The rays are as prominent as the convex centres of the plates, 

 and quite narrow or linear within, but widen rapidly towards the outside of 

 the plates. They are also each split longitudinally into parallel laminae by a 

 series of profound tdits extending nearly to the outer surface of the plates, 

 and it is these slits that are seen, like pores, at the prominent angular sutures, 

 where the edges of the plates at the latter have been worn partly away. It 

 is difficult to understand the use of these deep slits, or divisions of the internal 

 rays, since, as noticed by Mr. Billings, they seem never to pass entirely 

 through the plates, excepting where the prominent edges of the latter have 

 been worn away. 



Named in honor of Dr. B. F. Shumard, of St. Louis, whose labors in western 

 geology and palaeontology are well known. 



Locality and position. Cape Girardeau, Missouri. Trenton division of 

 Lower Silurian. 



Comakocystites Shumakdi, var. obconicds, M. & W. 



A single specimen in the collection from the same locality and position as 

 the species, just described, differs in being obconical instead of obovate, its 

 lower half tapering downwards gradually to the column. Its basal plates also 

 rise nearly vertically from the column, instead of extending out horizontally 

 as in the typical form of C. Shumardi. It has a part of the column attached, 

 showing it to be very nearly cylindrical, and composed of thin plates. In 

 form this specimen agrees nearly with Mr. Billings' figure 2, plate 5, decade 

 iii., Geol. Survey of Canada, from which it differs in having only five ranges 

 of plates above the base. It also agrees with the species we have just de- 

 scribed, ia having its plates above the middle proportionally larger, one of 

 these plates in a specimen only - 72 inch in height, measuring nearly a third 

 more in diameter than those of Mr. Billings' species near 1*50 inches in height. 



It is quite probable this form may belong to a distinct species, but as we 



[Aug. 



