122 MASS. EXPERIMENT STATION BULLETIN 176. 



A second series of percolation experiments was tried in which but 1 

 kilogram of soU was used, and proportionately smaller amounts of water 

 were percolated through it, until the total percolate amounted to 1 liter. 

 The percolate was filtered through porcelain and subsequently^ yielded no 

 precipitate with ammonium hydroxide. 



Iron and manganese were both found and determined by the colori- 

 metric methods. Both limed and unlimed soils from plots 1, 6, 7 and 8 

 were used in this series. All the extracts yielded colorimetric tests for 

 iron, but only those from the unlimed soils showed any manganese. The 

 results on the unlimed soils are given in Table III. 



Table III. — Milligrams Manganese Oxide (Mn-Oi) and Iron Oxide 

 {Fe^Ozj removed in Water from 100 Grams of Unlimed Soil. 



The amounts of manganese from the soils of plots 1, 6 and 8 are closely 

 like those obtained in the previous series with 8 kilograms of soil. 



The iron obtained is about one-half the amount of aluminium oxide 

 tabulated in the previous series. 



There were in the laboratory samples of soil from plots 5 and 6 which 

 were collected four years before, in 1912. Plot 5 had received the same 

 amount of sulfate of ammonia that had been applied to plot 6. Both 

 samples were from the unlimed halves of the plots. One kilogram of each 

 was treated as in the previous experiment. The exi-racts showed the 

 presence of aluminium and iron, but were most striking in the tests for 

 manganese. Plot 5 yielded 2.36 mg. Mn304, and plot 6 yielded 3.18 

 mg. Mn304, from 100 grams of soil. This shows that the formation 

 of salts of aluminium, iron and manganese by ammonium sulfate was as 

 marked four years ago as in 1916. 



All these experiments showed that ammonium suKate persistently 

 formed soluble salts of aluminium, iron and manganese in the soil of 

 Field A. 



It was next decided to secure samples of soils from other fields that had 

 received ammonium sulfate as a fertilizer over a considerable period of 

 time. The desired soils were obtained from the agricultural experiment 

 stations of Ohio and Rhode Island by the kindly co-operation of Director 

 Thorne and Director Hartwell. 



The soil of the Ohio experiment field is a rather heavy clay loam. The 

 samples were taken from Section C of the continuous five-year rotation 

 experiment described in Circular No. 144 of the Ohio Agricultural Experi- 

 ment Station. The plots selected for our purpose were Nos. 8 and 24. 



