294 THEORIES OF FERTILISER ACTION [chap. 



stearic acid, picoline-carboxylic acid, etc., which when 

 introduced into water cultures are toxic to seedling 

 plants. The compounds isolated are, however, all of 

 them products of the oxidation and decay of proteins, 

 fats, and other compounds contained in plant residues ; 

 there is no evidence to show that they are specific 

 excretions from particular plants or that they are more 

 abundant in soil impoverished by the continuous 

 growth of a crop than in soil which would be usually 

 termed rich. Again, it has not been demonstrated that 

 such substances, although harmful to young plants in 

 water culture, are toxic under soil conditions ; it is well 

 known how exceedingly sensitive are plants in water 

 culture, where growth, for example, is inhibited by 

 traces of copper not to be detected by ordinary methods 

 of analysis. A body like ammonia, itself a product of 

 protein decay and present in the soil, is exceedingly 

 toxic to water cultures, yet when applied to the soil it 

 increases the growth of the plant. Turning to the 

 fertiliser side of the theory, evidence is yet lacking to 

 show that fertilisers in such dilute solutions as they 

 form in the soil water can exert any precipitating or 

 destructive action on such toxic substances as have 

 been extracted from the soil ; particularly the specific 

 action of fertilisers is difficult to explain. Why should 

 substances so dissimilar as nitrate of soda and 

 sulphate of ammonia exert the same sort of action on 

 the same toxin ? Why should phosphates cause all 

 classes of plants to develop in one direction, or why 

 should they be appropriate to the toxins of all plants 

 on one particular type of soil, whereas potash answers 

 on another soil type? 



Lastly, there is a lack of evidence for the funda- 

 mental thesis that the rotation will take the place 

 pf fertilisers and that the yield only falls off when a 



