OSTRAOODA. 671 



Drepanella bllateralis.] , 



group. With these we may include, without materially altering our conception of 

 the genus, the new D. bilateralis, although in this species the characteristic sickle- 

 shaped marginal ridge is wanting posteriorly. But the other Minnesota species, D. 

 higeneris, is certainly a remarkable form. In size aud general appearance it agrees 

 very well with D. crassinoda and D. ampla having the sickle-shaped ridge well 

 developed, and two large centro-dorsal nodes, separated by a depression, as in the 

 latter species. But the peculiar feature is that these nodes are prolonged below and 

 united by a slender connection, giving us precisely the horseshoe ridge of a Bollia. 

 The question arising at once is, why should the species not be viewed as a Bollia, 

 rather than a Drepanella. 



I have decided for Drepanella on what I believe to be good genealogical grounds. 

 In the first place, aside from the ventral connection of the nodes, all the characters 

 of the species are those of Drepanella. The marginal ridge, it is true, runs farther 

 up on the anterior end than on any of the other species, yet its extremity is thin and 

 the mere fact that it is a trifle longer than usual cannot be of much consequence. 

 But the most important evidence on the question is furnished by D. ampla var. elon- 

 gate, of which a copy of the original figure is given above. In this variety, namely, 

 there is a well defined depression between the nodes precisely as in D. higeneris, and 

 all that is required to produce the loop of the latter, is a slight raising of the nodes, 

 together with the lower border of the depression. This is not, I believe, supposing 

 too much, for a ventral coalescence of the anterior and posterior lobes or nodes is 

 not by any means restricted to Bollia. Indeed it occurred under one form or another, 

 among many types of Beyrichiidce. That this is true, a glance at plate xliv may 

 suffice to prove. One form is shown in fig. 4, another, and widely different, in fig. 6, 

 while 8, 10,' 12, 15, 17, 20, 23, 26, and 27 illustrate other types of the same condition. 



Drepanella bilateralis, n. sp. 



PLATE XLVI, FIGS. 35-38. 



Size.— Greatest length 2.7 mm.; length of hinge 2.15 mm.; greatest hight 1.64 mm.; greatest thick- 

 ness, about 1.3 mm.; thickness, not including nodes and ridge, about 0.6 mm. 



Valves suboval or oblong-subquadrate, compressed; dorsal margin straight; 

 distinctly angular at the extremities; anterior end a trifie narrower, and the outline 

 less convex than the posterior; ventral margin nearly straight centrally. Running 

 parallel with and close to the ventral margin a strong ridge, somewhat thickened at 

 each end, but not continuing up the posterior end as in the other species. Above 

 this two irregularly triangular and very prominent large nodes extend to the dorsal 

 edge, beyond which their pointed extremities occasionally project. The last is true 

 also of a small central tubercle. 



