REPORT ON UPPER PALEOZOIC FOSSILS FROM CHINA. 309 



Of the ventral valve, unfortunately, our collections contain only very imperfect 

 examples. The marginal outline was of course the same as that of the other valve. 

 The general shape was apparently conical, strongly sloping backward, and with the 

 apex bent in the same direction. The interior of the valve was furnished with two 

 strong dental plates. 



The details of the surface are unknown, but whatever markings were present 

 must have been delicate, as no traces are to be found upon exfoliated specimens. 



For the Carboniferous this is a singular type, and unfortunately it is imper- 

 fectly known. Among Carboniferous species it has no close allies so far as I have 

 discovered. It is in fact more suggestive of the Cyrtiae of the Silurian and Devonian, 

 such as C. cyrtiniiformis and C. exporrecta. Too little is known, however, with regard 

 to the areal structures to say definitely whether it is or is not a Cyrtia. In some 

 respects it suggests a diminutive representative of the disjiinctus group of Spiriferi, 

 but the imperfect definition of the fold and sinus and the equality of the radii 

 negative a close affinity with those forms. In some respects also it is of the general 

 type of our Devonian species Spirifer hungerfordi. 



Of the Chinese forms figured by Kayser that from the Silurian of Tshau-tien 

 identified as Spirifer radiatns possesses the most resemblance. His figures, however, 

 appear to represent a form without lateral plications, and therefore one which can 

 not be identical with typical 5. radiatus. The resemblance between his figures and 

 my specimens (except in the matter of size) is rather striking, but both forms are 

 so imperfectly known that it may be entirely without significance. 



Locality and Horizon. Pre-Pennsylvanian (?); near Ta-miau-ss'i, East Ss'i- 

 ch'uan (stations 6, 8, and 9). 



Rhynchonella ? sp. 



Plate 29, Figure 8. 



The little shells included under this title are associated with those referred to 

 as Dalmanella ? sp., and although they are tentatively assigned to widely different 

 genera, it is not certain that their relationship may not be rather close. Few if 

 any examples included under the present title attain a length in excess of 6 mm., 

 and the average is less. The width is always greater than the length. 



The ventral valve is rather strongly convex in the posterior portion, and some- 

 times in the anterior. The beak is prominent, pointed, and incurved. There is 

 no well-marked sinus, though sometimes the median portion appears to be depressed. 

 The surface is traversed by from 25 to 30 rigid angular ribs. 



The dorsal valve is transversely subelliptical ; the beak small and depressed. 

 Perhaps by deeper exfoliation, the posterior portion sometimes has a mesial depres- 

 sion, which, however, soon dies out. Some specimens have an indistinct fold, and 

 the ribs are in number and character as in the ventral valve. 



The most typical examples of this division differ from Dalmanella ' sp. in lack- 

 ing the mesial angulation of the ventral valve, instead of which they are regularly 

 rounded transversely, or even in some cases with indication toward the front of a 

 mesial depression. The ribs are coarser, more distinct and angular, and they rarely 

 if ever bifurcate. In the other valve there is little difference in configuration, and 

 a discrimination rests almost wholly upon the character of the ribs, though in some 

 cases the convexity is greater, and there is occasional evidence of a fold, instead of 

 a depression, along the mesial line. In many instances, even in ventral valves, and 

 still more so in dorsals, it is impossible to refer a specimen confidently to one group 

 rather than the other; yet some of the ventrals of the division under consideration 

 have a distinctly rhynchonelloid expression, while in those referred to Dalmanella ? 



