270 CLASSICS OF MODERN SCIENCE 

 tion are formed, they must combine according to certain laws so as to 

 form a systematic whole, similar to an organism. Both these points 

 must be clearly proved, in order to establish the truth of the foregoing 

 view. But it is otherwise if this view be adduced merely as an hypoth- 

 esis, which may serve as a guide for new investigations. In such 

 case the inferences are sufficiently probable to justify such an hypoth- 

 esis, if only the two points just mentioned can be shown to accord 

 with it. 



With reference to the first of these points, it would certainly be 

 impossible, in our ignorance as to the cause of chemical phenomena in 

 general, to prove that a crystal capable of imbibition must produce 

 chemical changes in substances surrounding it ; but then we could not 

 infer, from the manner in which spongy platinum is formed, that it 

 would act so peculiarly upon oxygen and hydrogen. But in order to 

 render this view tenable as a possible hypothesis, it is only necessary to 

 see that it may be a consequence. It cannot be denied that it may: 

 there are several reasons for it, though they certainly are but weak. 

 For instance, since all cells possess this metabolic power, it is more 

 likely to depend on a certain position of the molecules, which in all 

 probability is essentially the same in all cells, than on the chemical 

 combination of the molecules, which is very different in different cells. 

 The presence, too, of different substances on the inner and outer sur- 

 face of the cell-membrane in some measure implies that a certain 

 direction of the axes of the atoms may be essential to the metabolic 

 phenomena of the cells. I think, therefore, that the cause of the 

 metabolic phenomena resides in that definite mode of arrangement 

 of the molecules which occurs in crystals, combined with the capac- 

 ity which the solution has to penetrate between these regularly 

 deposited molecules (by means of which, presuming the molecules 

 to possess polarity, a sort of galvanic pile will be formed), and 

 that the same phenomena would be observed in an ordinary crystal, 

 if it could be rendered capable of imbibition. And then perhaps the 

 differences of quality in the metabolic phenomena depend upon their 

 chemical composition. 



In order to render tenable the hypothesis contained in the second 

 point, it is merely necessary to show that crystals capable of imbibition 

 can unite with one another according to certain laws. If at their first 

 formation all crystals were isolated, if they held no relation whatever 



