METAMORPHOSES OF ANIMALS. 127 



one type compared with that of other types, as well as 

 between the ultimate histological differences which all 

 exhibit within certain limits. Though important frag- 

 ments have been contributed upon these different points, 

 I know how much remains to be done, from the little I 

 have thus far been able to gather myself by systematic 

 research in this direction. 



I satisfied myself long ago that Embryology furnishes 

 the most trustworthy standard to determine the relative 

 rank among animals. A careful comparison of the suc- 

 cessive stages of development of the higher Batrachians 

 furnishes, perhaps, the most striking example of the 

 importance of such investigations. The earlier stages of 

 the Tadpole exemplify the structure and form of these 

 Ichthyoids which have either no legs or very imperfect 

 legs, with and without external gills ; next it assumes a 

 shape reminding us more of the Tritons and Salamanders, 

 and ends with the structure of the Frog or Toad. 1 

 A comparison between the two latter families proves 

 further that the Toads are higher than the Frogs, not 

 only on account of their more terrestrial habits (see 

 Section 16), but because the embryonic web, which, to 

 some extent, still unites the fingers in the Frogs, dis- 

 appears entirely in the Toads, and, possibly also, because 

 glands are developed in their skin, which do not exist in 

 Frogs. A similar comparison of the successive changes 

 of a new species of Comatula discovered by Prof. Holmes 

 in the harbour of Charleston, hi South Carolina, has shown 

 me in what relation the different types of Crinoids of past 

 ages stand to these changes, and has furnished a standard 

 to determine their relative rank ; as it cannot be doubted 

 that the earlier stages of growth of an animal exhibit a 



1 AGASSIZ (L.), Twelve Lectures, etc., p. 8. 



