166 THE INFLUENCE OF INANIMATE SURKOUNDINGS. 



which regularly ensued showed that the salts present and dis- 

 coverable by chemical tests had really no influence that could be 

 detected. 



Even injurious gases which might be formed, in a certain 

 proportion to the body of water and the number of animals, 

 from the faeces and other decomposing organic matter, cannot be 

 regarded as causing the effects referable to the volume of water. 

 Suppose we take two vessels containing equal volumes of water ? 

 but place only one animal in one and twenty in the other, these 

 last of course will disengage the larger amount of injurious gases, 

 and consequently, in the first instance, a certain retardation in 

 their growth might be caused. But since the one isolated 

 individual grows immensely faster than those that live together, 

 this one will very soon yield as much, and at last more, faecal 

 matter than the twenty ; so that growth must cease with it at 

 least as soon as with the others. But since this is precisely 

 not the case for the curve of time remains the same for each, 

 while it is only the size attained within a determined period 

 which varies it is, it seems to me, thereby proved that the 

 effects of the injurious gases produced by the animals them- 

 selves cannot be the cause of the effects of volume. 



What, then, is the real cause ? I regret to say that I cannot 

 give any answer to this question. With the assistance of my 

 friend Hilger I have long been trying, but altogether in vain, 

 to find anything present whatever, even in the smallest quantity, 

 to which this effect of volume could be ascribed. I think, 

 however, that we are justified in the following hypothesis. It 

 would seem to follow from my experiments that some substance 

 as yet unknown must be present in the water, probably in 

 a very minute quantity, which, by its relations to the water that 

 holds it in solution and by its osmotic affinity to the skin of 

 the animal, can be absorbed only in a determined and extremely 

 small quantity, and also within a definite period and in a definite 

 amount of water. 66 Now, if this substance were simply a stimu- 

 lant, and thus, without actually contributing to growth, were 

 nevertheless essential to it like the oil to the steam-engine 

 it must be absorbed in the quantity which is most favourable 

 if the normal growth is to be accomplished within a fixed time. 



