252 LOUIS AGASSIZ. 



circumstance, other consideratious sliould not be neglected wliicli may 

 complete our insight into the general plan of creation." * 



In this point of view Agassiz had already perceived through his study 

 of fossil fishes the importance of embryology. He had for the first time 

 propounded the theory that the animals of our period in their embry- 

 onic condition resemble the ancient representatives of the same tyi)e 

 who lived in anterior geological ages. 



Classification, based upon these considerations, will no longer be a 

 system invented by such or such a naturahst, but will result strictly from 

 facts observed in nature. It will be proved, for example, that an insect 

 in various degrees of its development resembles successively types of 

 various classes of the branches of the Articulata, and that it assumes the 

 character of a perfect insect only after it has achieved its metamorpho- 

 sis. " When we study the gradual development of the insect * * * 

 we have a simple, natui'al scale by which to estimate the comparative 

 rank of these animals. Since we cannot suppose that there is a retro- 

 grade movement in the development of any animal, we must believe 

 tliat the insect stands highest (compared with Crustacea and other ar- 

 ticulates), and our classification in this instance is dictated by nature 

 herself." t 



" It may therefore be considered as a general fact, very likely to be 

 more fully illustrated as investigations cover (i wider ground, that the 

 phases of embryonic development of all hving animals correspond to the 

 order of succession of their extinct representatives in past geological 

 times. As far as this goes the oldest representatives of every class may 

 then be considered as embryonic types of their resi)ective orders or fami- 

 Ues among the living." | 



The class of Echinoderms furnishes a remarkable example. It is thus 

 that the embryonic phases of the European Comatula corresi)ond with 

 the principal forms of the Crinoids which characterize the successive 

 geological periods ; Cistoids of paleozoic rocks ; Platycrinoids of the car- 

 boniferous period ; Pentacrinoids of the freestone and the oolite. Anal- 

 ogous facts are found in the families of the Asteroids and the Echinoids. 

 The Trilobites are the embryonic type of the Entomostracans ; the Deca- 

 pods of the oohte that of our crabs ; the heterocercal Ganoids that of 

 the Lepidistes. 



By the side of these emhryonic types Agassiz recognised prophetic 

 types. "Embryonic types exemplify only the peculiarities of develop- 

 ment of the higher representatives of thek' own types ; while prophetic 

 types exemi)lify structural combinations observed at a later period in 

 two or several distinct types." One of the most striking examples he 

 cites is that of the sauroid fishes. "These fishes, which have preceded 

 the appearance of reptiles, present a combination of ichthjac and reptilian 



*Au Essay on Classiticatiou. 8vo., Loudon., 1859. Chap, i, sec. 32, pj). 205, 20G. 

 t A jouruey in Brazil. Boston, 18G8, chap, i, p. 21. 

 X Essay ou Classihcatiou, chap, i, sec. 25, p. 174. 



