FERTILIZERS. 979 



a«ter and determined (1) the effect on tbe solubility of the phosphoric 

 aeid of adding known quantities of the oxid of iron or alumina, and (2) 

 the solubility of the latter at different periods. 



In a superphosphate containing 25.97 per cent of soluble phosphoric 

 acid and 3,21 per cent of insoluble phosphoric acid the addition of an 

 amount of sulphate of iron containing Fe203 equal to 1.36 per cent of 

 the superphosphate resulted in the immediate reversion of 1.86 per 

 cent of acid, 2.39 per cent in one year, and 2.42 per cent in 15 months. 

 Aluminum sulphate was added to the same superphosjihate in an amount 

 furnishing AI2O3 equal to 1.1 per cent of the whole mixture, but the 

 results were much less marked than in the case of the addition of iron. 

 The immediate reversion was 0.17 per cent, at the end of a year 0.83 

 per cent, and at the end of 15 months 1.02 per cent. An examination 

 of other samples of superphosphate showed that reversion was due 

 principally to the oxid of iron present. 



A superphosphate which showed when made 36.5 per cent of soluble 

 phosphoric acid contained at the end of a year 38.19 per cent, and it 

 is noted that in case of high percentage superphosphates there is fre- 

 quently an increase rather than a decrease of soluble phosphoric acid 

 on keeping. 



"A review of the facts already obtained makes it pretty certain that there is a 

 wide difference in the behavior of oxid of iron and alumina in the manufacture of 

 superphosphate. The first, for practical purposes, seeuis to revert phosphate in what 

 may be called theoretical proportions, viz, 160:310, or as nearly as possible 2 parts 

 of phosphate to 1 of oxid of iron, while the alumina, which, if acting in the same 

 manner as oxid of iron, would revert three times its weight of phosphate, at most 

 seems to cause a retrogression of its own weight. It has long been known that 

 superphosphate manufactured from Florida phosphate increases in the percentage 

 of soluble phosphate instead of retrograding, and as this phosphate contains 

 siarcely any oxid of iron, but 2 to 3 per cent of alumina, it was difficult to under- 

 stand this behavior. The foregoing results, however, explain the matter. 



'■Under these circumstances, and especially as many of the phosphates now com- 

 ing into the market contain alumina in larger proportions than oxid of iron, it appears 

 to me that there is pressing need for a revision of the terms of the contracts on which 

 phosphates are usually sold, and for a differentiation of oxid of iron and alumina in 

 the analysis upon which the sales are made." 



In the discussion following the reading of this paper its essential 

 conclusions were confirmed by several other analysts. 



Farmyard manure (Agl. Jour. Cape Colony, 8 (1895), No. 5, p. 111'). — In trials of 

 covered and uncovered manure on potatoes and wheat during 2 years the results 

 strongly favored the covered manure. 



Utilization of the refuse bones of the farm, L. Grandeau {Jour. Agr. Prat., 69 

 {1895), No. 15, pp. 527, 528). 



Nitrate deposits in South Africa {L'Engrai,9, 10 {1895), No. 15, p. 55^).— The dis- 

 covery of nitrate deposits in caves near Prieska at an elevation of 3,000 to 4,000 ft. 

 is noted. The material is composed generally of nitrate of lime, but sometimes con- 

 tains 95 per cent of finely crystallized nitrate of potash. The nitrates are believed 

 to be formed from the accumulated excrement of rabbits, which breed in numbers in 

 that locality. An English company has been organized for the exploitation of the 

 ■deposits. 



