23 



that the amount of nitrates lost in the drainage water was 

 greater with increased dressings of ammonia salts. 



As soon as this loss was recognized, an experiment was 

 begun to ascertain whether the effect of the loss of nitrogen 

 by drainage during the winter made an appreciable difference 

 to the wheat crop. The following table shows the corn pro- 

 duced, in bushels, the weight of straw, and the weight of the 

 total produce, in each case per acre, obtained by the use of 

 ammonia salts, autumn sown and spring sown, for the years 

 1873-9 inclusive: * 



TABLE X. 



Autumn sowing gave the best result in only one year 

 1874 an exceptionally dry season ; while in four years the 

 result was as decidedly in favour of spring sowing, and in 

 two years there was no material difference. 



These experiments established the fact that ammonia 

 salts should not be applied to the wheat crop in autumn ; 

 so in 1878 and following years they were put on to the 

 experimental wheat crop in the spring. In 1884, a further 

 change was adopted, one quarter of the ammonia salts being 

 applied in autumn and the remainder in spring ; and this 

 practice has been continued since. A small quantity of 

 nitrogen is thus at the disposal of the plant for its autumn 

 growth ; but it is not till more active vegetation has com- 

 menced in spring that the full quantity is put on. The 

 next table [p. 24] gives the average produce of wheat in 

 bushels per acre per annum, obtained at Rothamsted by con- 

 tinuously manuring with minerals only, and with minerals 

 and ammonia salts, for each of the three periods 1852-77 

 inclusive, when the ammonia salts were applied in the 

 autumn; 1878-83 inclusive, when they were put on in 

 the spring; and 1884-94 inclusive, when one-quarter of 

 the ammonia salts were applied in the autumn and the rest 

 in the spring. 



Lawes and Gilbert, " Journal of the Royal Agricultural Society of 

 England," Vol. XYL, S.b. 



