48 Serial Classifications to be renounced. 



diversity, the fossil species of this formation are in nothing 

 inferior to that of a very extensive fauna belonging to the 

 actual creation. 



From the whole of the facts above noticed, it appears to 

 me to follow, that not only do the fishes of the old red sand- 

 stone constitute a distinct fauna, independent of those belong- 

 ing to otlier formations, but that they also present, in their 

 organization, the most remarkable analogy to the earliest 

 phases of the embryonic development of the osseous fishes of 

 our own epoch, and a not less obvious parallelism with the 

 lower degrees of certain types of the class, as they now exist 

 on the surface of the globe. What is most curious in these 

 connections is, that it is not with the corresponding types of 

 the actual creation that these ancient fishes can be con- 

 sidered as parallel ; for example, the osseous fishes of that 

 period had nothing in common with the osseous fishes of this 

 period, nor did the Placoides of the most ancient formations 

 in general resemble those of the present creation. Neither 

 do the Ganoides exhibit more than remote resemblances to 

 the existing Ganoides ; but these same Ganoides approach, in 

 a multitude of characters, the Placoides of our own period, 

 and even the inferior types of this order. And yet, along 

 with this, they have also certain relations to reptiles, al- 

 though this class of animals did not actually appear till 

 later. These relations I would call prospective analogies, so 

 frequent is it to meet with prophetic resemblances, in the 

 series of formations, among types succeeding each other, and 

 which, after having for a long time presented the combined 

 characters of many groups, do not become distinct till a later 

 period. These facts appear to me deserving of our most se- 

 rious attention ; for they shew us, always with increasing ur- 

 gency, the necessity of renouncing serial classifications, in 

 order to express the real relations of living beings. If, in effect, 

 the most ancient fossil fishes of the order Ganoides shew strik- 

 ing resemblances to the Cyclostomes and Plagiostomes of 

 our era — if these same Ganoides have, besides, certain analo- 

 gies to reptiles, and, in particular, to the Labyrinthodontes 

 — if these relations disappear in more recent eras — if these 

 families themselves become progressively extinct and are 



