y62 DIVISIONS OF THE GASTROPODA. 



the groups of which are based upon the structure of this organ. 

 All that will be attempted here, therefore, is to give a brief outline 

 of the characters and geological range of the principal families of 

 Prosobranchiates, without regard to the arrangement of these in 

 larger groups. In this outline the " holostomatous " families, as 

 the most ancient and the least highly organised, naturally take 

 precedence of the forms with a " siphonostomatous " shell. 



Family i. Patellid^e. — In this family the animal usually pos- 

 sesses a series of branchiae arranged marginally in a more or less 

 complete cycle round the foot. The shell is conical, with the apex 

 turned more or less clearly towards the front ; and the muscular 

 impression is horse-shoe-shaped and open in front, continuous or 

 broken up into separate cavities. The genus Acmcea, often regarded 

 as the type of a special family, differs from Patella proper in the fact 

 that there is always a cervical branchial plume, the marginal gills 

 being sometimes present, sometimes absent. As these types, how- 

 ever, cannot be distinguished from one another by their shells alone, 

 this distinction is palaeontologically inapplicable. 



The typical species of Patella, such as the common Limpet (P. 

 vulgata), have usually a radially -ribbed shell, with the apex sub- 

 central and turned forwards. Such types are clearly recognisable 

 in the Cretaceous and Tertiary rocks. In Helcion, again, the shell 

 is radially-ribbed, and the apex is shifted far forwards. Such types 

 abound in the Jurassic and Cretaceous rocks, and still survive. In 

 Acmcea, finally, the apex is usually subcentral and the surface is 



Fig. 644. — Tryblidium (Mctofitoi)ia) Xycteis, from the Ordovician of Canada, a, Side view; 

 b, View of the upper side. (Alter Billings.) 



generally smooth or feebly striated. This type cannot be clearly 

 separated from the preceding except by an examination of the 

 animal, but many of the fossil Limpets doubtless belong here. A 

 Limpet of this type has been described by Hall from the Upper 

 Cambrian (Potsdam Sandstone) of North America under the generic 

 name of Paloeacmcea. 



Most of the Palaeozoic Limpets belong to the genera Tryblidium and 

 Metoptoma, which agree with one another in the fact that the shell (fig. 

 644) is in the form of an obtuse or shallow cone, with the apex placed 

 very far forwards, so as to be submarginal. In Tryblidium, as shown by 



