M. Agassiz on Fossil Fishes. 335 



a description of tho skeleton of a living species, in order to 

 facilitate and complete the knowledge of the fossil species, 

 which, not being usually preserved entire, are on that account 

 more difficult to determine. If, on the contrary, he treats of 

 families wholly extinct, the author has endeavoured to afford 

 a similar advantage by restored figures, representing the fish 

 with the form and ornaments which may be supposed to have 

 belonged to it, judging from the preserved remains. 



2. Fossil Fishes important in an anatomical point of view. 



M. Agassiz's work is not a less important contribution to 

 anatomy than it is to zoology and geology. Obliged to study 

 minutely, not only the external form, but also all the parts of 

 the skeleton in living fishes, in order to determine the detached 

 analogous pieces met with in the strata of the earth, the author 

 was under the necessity of tracing, with the greatest care, the 

 numerous modifications which these same bones undergo in the 

 diflferent families of the class of fishes. There can be no doubt 

 that this is the most difficult part of the work ; for in no other 

 class of the vertebrata is the osseous frame-work so variable. 

 We have only once to examine the head of a fish in order to 

 perceive the difficulty of referring all the pieces to a constant 

 type ; for not only are the bones of the head more numerous 

 among fishes than among the other vertebrata, but they are 

 combined in so many different manners, that it is very difficult 

 to detect their true relations. On the other hand, age induces 

 very considerable changes, not only in the form and dimen- 

 sions of the different bones, but even in their structure, and to 

 such a degree that the same bone often cannot be recognised 

 in the different stages of life. Hence the necessity of studying 

 the development of all the parts of the skeleton, in order to 

 be in a condition to distinguish with certainty the essential 

 characters from the secondary characters — what is constant 

 from what is transitory. Considered in this light, the re- 

 searches which the author has undertaken, in connection with 

 M. Vogt, on the embryology of the Salmonidse, must have afford- 

 ed him great assistance ; they have, above all, given him the 

 means of appreciating the relative value of the different organs, 



VOL. XXXYII. NO. LXXIV, OCTOBER. 1844. Y 



