THE OSTRACIN^j OB TRUNK-FISHES. 



195 



above mentioned_, being either very minute, or the skin 

 only granulated. These genera have been already 

 named by Cuvier, Alutera, Triacanthus, and Monocan- 

 thus. In this latter genus^ some of the forms, as that 

 of Mon. Ufilamentosus Less. {Jig. 2Q.), are singularly 



grotesque ; but 

 the sub- genera 

 have not been 

 investigated, and 

 much remains to 

 be done in deter- 

 mining their lo- 

 , cation: some will, 

 V doubtless, enter 

 as aberrant types 

 in the other ge- 

 nera ; nor is it at all probable that the genus Tria- 

 canthus should contain only one typical example. If 

 the ichthyologist wishes to study the relations of all 

 these new divisions, he will find they follow each other 

 in the same series as that in which we have noticed the 

 primary famihes of the entire order. The analogies, 

 indeed, of the whole of this family, to that of the Chce- 

 todonidcB, with which so many writers have incidentally 

 compared them (one of the best proofs of the analogy 

 being natural), are most particularly beautiful. But 

 we have no space for this inviting subject. 



(172.) The sub-family OstracincB is composed of the 

 trunk or tortoise fish ( O. m'gus^\i^'^,,fig. 30.) ; so called 



from, their bodies being often quadrangular like a trunk or 



o 2 



