FISHES OF OLD RED SANDSTONE. 367 



remained in doubt as to their true nature. 

 Agassiz says himself in his Preface : " I can 

 never forget the impression produced upon me 

 by the sight of these creatures, furnished with 

 appendages resembling wings, yet belonging, 

 as I had satisfied myself, to the class of fishes. 

 Here was a type entirely new to us, about to 

 reenter (for the first time since it had ceased 

 to exist) the series of beings ; nor could any- 

 thing, thus far revealed from extinct creations, 

 have led us to anticipate its existence. So 

 true is it that observation alone is a safe guide 

 to the laws of development of organized be- 

 ings, and that we must be on our guard against 

 all those systems of transformation of species 

 so lightly invented by the imagination." 



The author goes on to state that the discov- 

 ery of these fossils was mainly due to Hugh 

 Miller, and that his own work had been con- 

 fined to the identification of their character 

 and the determination of their relations to the 

 already known fossil fishes. This work, upon 

 a type so extraordinary, implied, however, in- 

 numerable and reiterated comparisons, and a 

 minute study of the least fragments of the re- 

 mains which could be procured. The materials 

 were chiefly obtained in Scotland ; but Sir 

 Roderick Murchison also contributed his own 



