AND OF PRODUCTS OF METABOLISM. 301 



ourable effect shows itself. Urea also acts favourably, 

 though not to the same extent as uric acid. If, there- 

 fore, these two simple bodies are capable of stimulating 

 the tissues to increased growth, it is possible that the 

 effects produced by animal excreta may be due to 

 minute quantities of other but more complex nitro- 

 genous bodies. That they are not due to simple urea 

 and uric acid was proved by the chemical analyses of the 

 fouled waters, for the amount of nitrogen found to be 

 present was never half sufficient, and as a rule was very 

 much less. As to the substances producing an ad- 

 verse influence on growth, no definite evidence was 

 obtained, but it seemed possible that they might be 

 derivatives of ammonia, perhaps amines or amido- 

 bodies. Thus ammonium salts themselves exert an ex- 

 ceedingly poisonous action, as may be gathered from 

 the following data: 



WEIGHT OF AMMONIUM 

 CHLORIDE PER LITRE. EFFECT PRODUCED. 



.0258 gin. Larvae diminished 7.3 per cent, in size. 



.0394 " " 19.0 per cent. " 



.1075 59 per cent, blastulse formed. Larvse lived 3 



days. 



.3745 37 per cent. " " No larvae. 



.7890 Most of the ova had disintegrated after 24 hours. 



That the effect produced by nitrogenous bodies depends 

 almost entirely upon the form in which the nitrogen is 

 combined, is shown by the fact that nitrates and nitrites 

 have no influence on larval growth unless the propor- 

 tions added be over 1 gram and .3 gram per litre re- 

 spectively. 



The products which every organism excretes prob- 

 ably consist, therefore, of various complex nitrogenous 



