364 THE FERTILITY OF SOILS IN THE TROPICS 



the soil in excess of that which has been added in the 

 manure. The only satisfactory explanation of this result 

 appears to lie in the assumption that it is occasioned by 

 the operations of organisms of the Azotobacter type, 

 the existence of which has been demonstrated in the 

 soils in question. 



Results in connection with the soils of these same plots 

 are also available to show that in the cases where 

 moderately heavy dressings of organic manures have 

 been applied the rate of decay of organic matter is more 

 rapid than when such manures have been withheld, afford- 

 ing additional evidence that liberal applications of organic 

 manure result in an increase in numbers and vigour of 

 the bacterial flora. 



In relation to nitrifying power, all of the soils 

 examined showed normal properties; it is, however, again 

 significant that those soils which have been manured with 

 organic manures gave higher values for this property as 

 a general rule. A point of interest in this connection is 

 that though many of the soils in question are naturally 

 deficient in lime they still showed marked formation of 

 nitrates under laboratory conditions; the suggestion has 

 been made that perhaps under these circumstances 

 ammonia formed in the course of ammonification may 

 function as a base for the neutralization of nitric acid, 

 though no experimental evidence is as yet available in 

 favour of this view. 



A considerable range of manurial experiments in the 

 Leeward Islands over a period of fifteen years' duration 

 has fully confirmed the view that the most satisfactory 

 way of maintaining fertility lies in the application of 

 organic manures. A result such as this sheds important 

 confirmatory evidence on the views outlined above. 



It is submitted that the above results taken with the 

 indications of manurial experiments, both in the Leeward 

 Islands and elsewhere, point definitely to the importance 

 of maintaining the supply of organic matter in tropical 

 soils; if this is withheld diminution in fertility takes place 

 owing to the effect on the biological relationships which 

 ensue. 



There would further appear reason to believe that the 



