244 COSMIC PHILOSOPHY. ft. i. 



logical state, and set in motion a whole series of phenomena 

 which obscure those which we wish to observe. It is instruc- 

 tive, and often amusing, to read some treatise on experimental 

 physiology, like those of Magendie and Claude Bernard, and 

 see how easy it is for equally careful investigators to arrive 

 at totally irreconcilable results. It is not to be denied that 

 experiment is of vast importance in biology, and has already 

 achieved wonders. Nevertheless, the practical study of 

 experimentation should never be begun in biology, but in 

 chemistry or physics, where the conditions are simpler. 

 Having learned from these sciences the general theory of 

 sound experimenting, we may afterward safely proceed to 

 apply the same method to vital phenomena. 



The additional implement possessed by the organic sciences 

 is comparison, to which corresponds the Method of Concomi- 

 tant Variations, already described. It is true we can also 

 employ this method to a large extent in the simpler sciences, 

 but it is in biology that it attains its maximum efficiency. 

 Here we have a series of instances already prepared for us 

 by Nature, in which certain antecedents and consequents 

 vary together. We have a vast hierarchy of organisms, each 

 exhibiting some organ and the corresponding function more or 

 less developed than it is in the others. To trace the functions 

 of the nervous system, or to follow the process of digestion, 

 in its increasing complication, from the star-fish up to man, 

 is to employ the logical method of comparison. And if any 

 one wishes to realize the immense power of this method, let 

 him reflect upon the revolution which was wrought in the 

 science of biology when Lamarck and Cuvier began the work 

 of comparison upon a large scale. 



Hence, it is that biology is eminently the science of classi- 

 fication ; and if skill in the use of this powerful auxiliary of 

 thought is to be acquired, it must be sought in the compara- 

 tive study of the vegetable and animal kingdoms. Theoretical 

 lo^ic may divide and subdivide as much as it likes ; but 



