THE COASTS OP SAINTONCE. 291 



nearly allied genera of this group. Although it 

 is abnormal in regard to the lymphatic respiration, 

 the Branchellion in all other respects is merely an 

 ordinary leech. Classification, which is not science, 

 but which ought as far as possible to be the ex- 

 pression of it, has here a double fact to record, and 

 it will consequently be necessary again to deviate 

 from some of those rules which have been deduced 

 from too exclusive a study of the higher animals. 



By introducing into zoology the great principle 

 of the subordination of characters discovered by 

 Laurent de Jussieu*, Cuvier rendered an immense 

 service to science. From that moment the value 

 and not the number of characters determined the 

 division of the animal kingdom, like that of the 

 vegetable kingdom, into naturally sub-divided groups. 

 But in arriving at the appreciation of this value the 

 two men of genius whose names we have quoted 

 proceeded in a very different manner, for while 

 Jussieu consulted only the evidences of observation 

 and experimental research, Cuvier, as he himself de- 

 clares, attached a pre-eminent and almost exclusive 

 value to reasoning.! He judged of the importance 

 of the organs by the importance of the functions, 

 and then proceeded to the characters yielded by 

 those organs. 



Nothing can be more apparently rational and lo- 

 gical than this, but unfortunately nature often seems 

 to take a malicious pleasure in sporting with our 



* [A sketch of the scientific services of the different members of 

 the family of De Jussieu is given in the Appendix, Note III.] 

 f Ht'yne animal, 2 C edition, Introduction. 

 u2 



