840 Agassiz’s Contributions to the 
ordinal relations, and then structural type would have been the 
sole reliance in classification, as it is in a great number of in- 
stances, 
runs downs into numerous monotypic genera, it appears to us to 
be failing of one great purpose, which is, to exhibit groups of 
species, in their true relations. 
The application of the views brought forward by Prof. Ag: 
we suspect will give the greatest occasion for diversity of judg- 
ment. He has not sketched out the system in zoology in order 
to exemplify his principles, and has only mentioned hypothetically 
the names of the Classes of Vertebrates. Instead of the sare 
number, Mammals, Birds, Reptiles, and Fishes, there are the fol 
lowing: Mammals, Birds, Reptiles, Amphibians, Selachians, Ga- 
noids, Fishes proper, and Myzontes,—the last four being as 
memberments of the group of fishes, and the preceding tw®, of 
Reptiles. The subdivision of the Reptiles had been before sug- 
red; but the Fishes are here for the first time divided. On 
this subject, he observes, p. 186: 
“The number and limits of the classes of this branch (the Vertebrate} 
are not yet satisfactorily ascertained. At least, naturalists. do not : 
agree about them. For my part, I believe that the Marsupialia canno 
be separated from the Placental Mammalia, as a distinct class, since We 
observe, within the limits of another type of Vertebrata, the Selachians, 
which cannot be subdivided into classes, similar differences in the m 
ars 
be sef 
decided 
sf gated, 
e@ case 
ould con- 
