54 JOURNAL OF THE 



taken as in method, and only 1 grm. as well for the comparison and 

 the determinations were carefully duplicated : 



Acid Phosphate. ^ol. PoOr^. Average. Difference. 



2 grms. taken, < i o 35 ( 13.33 



1 grm, taken, 5 12 82 i^ 12.72 



0.39 



The method was followed exactly in each cage and the final wash- 

 ings did not react acid ; and yet by using 1 grm. a result was ob- 

 tained 0.39 per cent, higher than that found with 2 grms. as in the 

 Association method. What causes this difference ? 



Presumably because a larger surface proportion of the 1 grm. is 

 exposed while on the filter to the action of the water than is the 

 case with the 2 grms., and for this' reason is more completely 

 washed and more phosphoric acid is consequently dissolved. 



H. B. BATTLE. 

 Laboratory of the JY. C. Ag. Exp. Station, 

 Raleigh, May Qth, 1886. 



ON THE DETERMINATION OF MOISTURE IN 

 COMMERCIAL FERTILIZERS. 



H. B. BATTLE. 



Superphosphates manufactured from phosphate rock by treat- 

 ment with sulphuric acid contain not only hygroscopic water from 

 the dilute acid, but also water combined with calcium sulphate 

 formed by the chemical action. And when organic matter is added 

 in the preparation of ammoniated phosphates, moisture is again 

 brought into the mixture, and this is more securely retained than 

 hygroscopic water in the plain superphosphates. The difficulty met 

 with in driving off the different forms of moisture in its estimation 

 is two-fold : (1) the impossibility of expelling all combined water at 

 lOC'C, and (2), the likelihood of the oxidation of the organic mat- 

 ter if a higher temperature is reached, and so impairing the result. 



