The Barba^dos Sugar Cane Experiments. 357 



not followed by corresponding increases, but by 

 decreases in the yields. The latter results would appear 

 as though due to errors in the experiments if it were not 

 that they were almost universal during the experiments, 

 on two occasions only plots exceeded in yield those 

 given 75 lbs,, in one of which also the yield produced by 

 150 lbs. was higher than that by 75 lbs., but lower than 

 that produced by 1 1 2 lbs. These results can not be due 

 to the phosphoric anhydride contained in the manure, as 

 the sign of the amount of this substance necessary for the 

 plants being exceeded would have been diminished rate of 

 increase by increased manurings, and notdiminishedyields, 

 but were in my opinion due to the acidity of the superphos- 

 phate injuriously affefting the rootlets of the canes— 

 this, when the superphosphate is applied in large 

 quantities, not being sufficiently rapidly neutralised by 

 the aftive lime (calcium carbonate in the fine soil), 

 present in the soil of the plots. As the a6live lime 

 present in the soils of these fields amounted in one case 

 to '22 per cent., and in the other to '72 per cent,, the 

 results point to the necessity of great caution in the 

 application of superphosphates to our Demerara soils, in 

 the majority of which the amount of aftive lime present 

 is nil. 



The experiments with insoluble phosphates showed, 

 but poor results, their a6lion being scarcely appre- 

 ciable ; this was probably due to the Dodds' soils, 

 in common with the great majority of Barbados soils 

 and with many of our soils here, being very poor in 

 organic matters, so that the organic acids and carbonic 

 anhydride set free in the soil by oxidation were not 

 sufficient to influence the solubility of the phosphates. 



YY2 



