116 DEVONIAN FAUNA. 



The present species differs from P. vasiforme by its less arched dorsal outline, its 

 more regularly and slowly tapering form, its more flattened ventral side, and its more 

 elliptic apical section. By these points it seems pretty clearly distinguishable, but 

 the finding of more numerous and better-preserved specimens may possibly modify 

 this conclusion. It is only by the examination of a large series of specimens that two 

 forms differing considerably on certain points can with safety be placed together as 

 belonging to a single very variable species. And though to describe forms as distinct 

 that may ultimately be found to belong together leads to the undue multiplication 

 of specific names, this evil, great though it be, appears to be less in the end than that 

 of the confusion caused by grouping two or three species under a single name ; the 

 consequence of which is that the information respecting them becomes hopelessly 

 confused, and when in the future they are found to be distinct, erroneous notions are 

 much more likely to be produced concerning them. Moreover, in the case of a fossil 

 occurring in different formations, it is much easier to arrive at truth by having to 

 identify two differently named and described fossils as the same, than by having to 

 separate two species that bear one name in both formations. It would be, I think, a 

 most useful and interesting task to compare carefully the kindred species of different 

 formations, and the same species when occurring in several formations, with a 

 view to discover how far their palgeontological relations as at present received are 

 accurate. The result of such a course of comparative palaeontology would probably 

 lead to many modifications in our present views. 



Returning to the species und er notice, I do not think that there is any reason to 

 suppose that P. vasiforme will have to be connected with it. From P. Marri it is 

 separated by its shape and by the absence of any signs of longitudinal ornamentation, 

 and from the other similar Devonshire species by the position of its siphuncle. 



0. suhpjrif ormis, d' Arch, and de Vern., is clearly different from the shell described 

 under that name by Miinster,^ and on the other hand it appears to me identical 

 with Phillips's shell. The siphuncle is on the shorter diameter, and the contours 

 and septa are similar. More of the shape of the aperture is shown in the French 

 authors' figure, and this, unless it be a mere restoration, may help to determine 

 the genus. 



Affinities. — Phragmoceras Brateri, Miinster,^ is a more compressed and recurved 

 shell and the septa seem rather more distant. Its body-chamber has a more 

 arched outline, at least laterally, if our and Phillips's specimens are at all perfect in 

 that portion. 



Ph. ventricosum, Sowerby,^ of the Upper Silurian, is much more recurved than 

 either the present species or P. Marri. 



1 1840, Miinster, ' Beitr.,' pt. 3, p. 103, pi. xx, fig. 10. 



^ 1840, Miinst., ' Beitr.' pt. 3, p. 105, pi. i, figs. 10 a, b, and c. 



3 1838, Sowerby, ' Murch. Sil. Syst.,' p. 621, pi. x, figs. 4—6. 



