80 SHELLS AND SHELL-FISH. PART I. 



certain species which exhibited so many intermediate 

 characters, that, but for the pecuharly small and incurved 

 channel of Cancellaria, we should have been at a loss 

 where to place them : one, we more particularly remem- 

 ber, had an exceedingly wide umbilicus, nearly as 

 large as the Scolymus umbilicm^is* , with a very simi- 

 lar shaped shell; and it is not a little remarkable that, 

 in the figure of this very Scolymus, the base of the 

 pillar, instead of being straight, as in all the other 

 species, is represented as curved inwards, and with an 

 exceedingly small notch, — two characters which are 

 absolutely peculiar, in this group, to Cancellaria. 



(JO.) The fourth genus, Rhinedomus, is typically 

 represented by the Cancellaria senticosus of Lamarck, 

 but of which there are several species. It retains suf- 

 ficient of the characters of CancellaricE to show its true 

 afiinity to them, for the exterior surface is longitudinally 

 ribbed, and rough with little points, and there are two 

 distinct folds, or rather one deep groove, at the base of 

 the pillar : but in all essential respects the true cha- 

 racters of Rhinedomus are remarkably different : the 

 basal canal is fully developed, being deeply notched ; 

 and, what is very unusual, there is a distinct, though 

 not a very deep, sinus close to the base of the outer 

 lip. The animal, made known to us by the valuable 

 researches of MM. Quoy and Gaimard, is most remark- 

 able ; but, until we are acquainted with that of Cancel- 

 laria, it furnishes no comparison with them. The other 

 species of Rhinedomus have the spire less elongated, 

 and thus open a passage to our new genus Polytropa, 

 examples of which will be found in two well-known 

 British shells, the Purpura lapillus and imbricatus of 

 Lamarck : this genus agrees, indeed, with Purpura, in 

 common with many other of its prototypes, in having 

 the inner lip flattened j but the base of the shell, in- 

 stead of being truncate, forms a short straight channel; 

 the spire is always as long (generally much longer) as 



* Ency. Method, pi. 431. bk, fig. 1. 



