12 ANNALS NEW YORK ACADEMY OF SCIENCES 



the introduction of ribs, and they often become extremely numerous. 

 "While the order of introduction of the types of surface ornamentation is 

 fairly constant, the precise volution at which a new feature will appear 

 is dependent on the degree of acceleration which the shell has attained 

 and often shows considerable variation within the same Linnsean species. 

 In highly accelerated forms, both spirals and ribs may appear together 

 at the close of the protoconch stage. This is seen in Cerithium adan- 

 soni, plate iii, fig. 3. 



The amount of embracing of the whorls in shells of this genus varies 

 within considerable limits, but as a rule the body volution is shorter than 

 the remainder of the spire. 



The aperture is oval, with an oblique, more or less widely open an- 

 terior canal. Primitive forms have no posterior tooth, but in some 

 Eocenic and later forms a projection of the callus of the inner lip forms 

 a distinct tooth which defines a short canal between itself and the outer 

 lip. 



In the earlier literature of the subject, the name Cerithium was applied 

 to all forms having in common the characters of a high spire, a short 

 body whorl and a short anterior canal, although Bruguiere himself recog- 

 nized the heterogeneous character of the genus, for he separated it into 

 three groups which are now properly recognized as distinct genera. Fur- 

 ther subdivision has been found necessary from time to time, until in 

 M. Cossmann's Essais de Paleoconchologie, volume 7, after the separa- 

 tion of several groups of family rank, we find seventeen genera, thirty- 

 one subgenera and forty-five sections. 



Choosing Cerithium nodulosum Bruguiere as the genotype, M. Coss- 

 mann lays especial stress upon the projecting tooth on the anterior por- 

 tion of the outer lip, and he restricts the genus so as to include those 

 forms in which this tooth crosses the opening of the canal. This ena- 

 bles him to place only two species within the genus, the genotype and 

 Cerithium erythrcBonense {=C. tuberosum), but a close study of the 

 anterior tooth seems to indicate that its importance as a basis for classifi- 

 cation has been greatly overestimated. It appears on the last portion of 

 the body whorl, a position in which the highest degree of variation in the 

 shell is to be expected, and features appearing at this stage should serve 

 as reasons for separating end members of evolutionary series, but not for 

 uniting species as genera, if the classification is to be a natural one — 

 that is, based on community of descent. The tooth is formed by the 

 more rapid growth of the anterior part of the outer lip, and when a 

 spiral is present on this part of the shell, it often determines the point 

 at which an extra amount of calcareous material is deposited. The de- 



