40 PROGRESS OF PALAEONTOLOGY n 



hand and with the ruminants on the other, differed 

 from both to such an extent that it could find a 

 place in neither group. In fact, it held, in some 

 respects, an intermediate position, tending to 

 bridge over the interval between these two groups, 

 which in the existing fauna are so distinct. In 

 the same way, the Pa Iceotherium tended to connect 

 forms so different as the tapir, the rhinoceros, and 

 the horse. Subsequent investigations have brought 

 to light a variety of facts of the same order, the 

 most curious and striking of which are those which 

 prove the existence, in the mesozoic epoch, of a 

 series of forms intermediate between birds and 

 reptiles two classes of vertebrate animals which 

 at present appear to be more widely separated 

 than any others. Yet the interval between them 

 is completely filled, in the mesozoic fauna, by 

 birds which have reptilian characters, on the one 

 side, and reptiles which have ornithic characters, on 

 the other. So again, while the group of fishes, 

 termed ganoids, is, at the present time, so distinct 

 from that of the dipnoi, or mudfishes, that they 

 have been reckoned as distinct orders, the 

 Devonian strata present us with forms of which 

 it is impossible to say with certainty whether they 

 are dipnoi or whether they are ganoids. 



Agassiz's long and elaborate researches upon 

 fossil fishes, published between 1833 and 1842, 

 led him to suggest the existence of another kind 

 of relation between ancient and modern forms of 



