t4 CLASSIFICATION. 



* * Varices numerous. 



Subgenus Vitulakia. Shell oblong ; spire short ; body whorl long ; 

 canal very short, wide ; outer lip thickened and dentate within. 

 Varices nearly obsolete. 



Subgenus Ocinebra. Spire elevated ; canal more or less closed ; 

 varices foliated, sometimes spinose. 



UROSALPINX. Fusiform. No proper varices, which are replaced by 

 longitudinal ribs. 



EUPLEURA. Ranelliform, with a pair of lateral varices, one on either 

 side, and intermediate smaller varices ; aperture dentate within. 



TYPHIS. Ovate or oblong, with projecting hollow tubes between the 

 three spinose varices ; aperture sub-orbicular, prolonged in front into a 

 closed siphonal canal. 



TROPHON. Varices numerous, lamelliform or laciniated ; spire promi- 

 nent ; aperture ovate ; canal open, usually turned to the left ; shell 

 white, often dark-colored within the aperture ; typically, Arctic and 

 Antarctic. 



Sub-Family PURPUKIN.E. 



As already stated, whilst the Muricida) naturally subdivide into two 

 groups, one of them (Murices) distinguished by varices on the shell oper- 

 culum with temiinal initial point, whilst the other (Purpuras) has nodules 

 but no varices, patulous columella, short canal or mere basal notch, oper- 

 culum with lateral nucleus ; yet on the confines of these two groups occur 

 forms which partake of the characters of either, and the classification of 

 which is entirely arbitrary. Ocinebra, species of Trophon, Urosalpinx 

 and Eupleura, have undoubted relationships with Purpura, yet are 

 classed with Murex— partly because the species have usually been con- 

 sidered or were described as Murices ; on the other hand. Purpura erispata 

 and its allies possess the variceal features of Murex. Kobelt has, on this 

 account, included them in his catalogue of the genus Murex ; but on 

 account of the extreme variability of the species (some specimens being 

 without varices) and the number of connecting forms between the smoother 

 varieties and typical Purpui-ae, I prefer to retain them in the group to 

 which they have usually been referred. 



If the difficulty of defining these two sub-families is great, still greater 

 does it become when we descend to the genera and subgenera of either of 

 them. Various authors have attempted it, fiom the "groups" of Kiener's 

 monograph to the genera and subgenera of H. & A. xVdams. I adopt the 

 latter as a mere convenience, premising that nature presents her specific 

 forms here (as freciuently elsewhere) in such continuous series, that no 



