23 



to its height ; the edge entire, thin and sharp in most of the specimens, 

 but in others with apparently one, two or three small notches. The area, 

 owing to the abrupt curvature of the valve at the beak, is inverted, or 

 overhangs the hinge line, as shown in fig. 2a, (which represents, however, 

 an extreme form.) Sometimes it is nearly in the plane of the lateral 

 margin, but in all the specimens examined, in which it can be seen, it 

 overhangs the hinge-line more or less. 



Dorsal valve concave, the curvature conforming to that of the ventral 

 valve. Area small, and at nearly a right angle to the ventral area. 



In the interior of the ventral valve the muscular impressions occupy a 

 large, ovate or flabelhform space, extending one-half or three-fourths the 

 whole length of the shell. The divaricator scars are divided longitudinally 

 into a variable number of lobes, usually ten or twelve, but specimens occur 

 with as many as fifteen. In the one represented in pi. 2, fig. 2, there are 

 five lobes distinctly marked out, but it can be seen that each of these is 

 divided by an obscure ridge (not shown in the figure) into two, making 

 ten in all. In others, from the same locality, there are ten or more lobes, 

 well-defined, especially towards the front. The two divaricator scars are 

 usually divided throughout their whole length by a medial septum which 

 sometimes extends nearly to the beak (w^ood cuts No. 10, 11.) The im- 

 pressions of the occlusors occupy a small ovate space, partly above and 

 partly between the divaricators. The remainder of the interior of the 

 shell is densely tuberculated. In most specimens it is somewhat smooth 

 round the front, or marked with obscure radiating stride. 



Nearly all the space beneath the area is filled with solid shell, so that 

 the hollow of the umbo is reduced to a small conical cavity, which some- 

 times reaches nearly to the beak, but often scarcely penetrates beyond 

 the hinge line (wood cuts, No. 10, 11, c.) There is no rostral septum. 

 The bilobed process on the inner surface of the deltidium is, in some spe- 

 cimens, obscurely represented by a small tubercle with a groove along the 

 middle, as in S. filitexta. In others it appears to be absent altogether. 



The interior of the dorsal valve is not shown in any of the specimens 

 from Gaspe. The structure of the ventral valve, however, clearly proves 

 that, when the two valves are in connection, the divaricator processes can- 

 not project beneath the area, as they do in S. demissa. On each side of 

 the mouth of the small conical cavity, beneath the deltidium, there is a 

 shallow ovate pit excavated in the substance of the shell. The free 

 extremities of the divaricator processes occupied these pits. The struc- 

 ture shows that the divaricator processes are short, and their extremities 

 distant from each other about one fine, in a specimen two inches in width. 



This is well own in the specimens from which the wood cuts, figs. 12 



