194 FARMYARD MANURE [chap. 



the deep pit from 6th November 1896 to 21st February 

 1897, when the dung made was cleared out, sampled, 

 and analysed. The experiment was then resumed until 

 2 1st May, after which the dung was left in the box for 

 another month, until 17th to i8th June, without any 

 beasts to keep it trodden down, the weather being 

 meantime very hot. The results appear under items 4 

 and 5 in Table LIV. It will be seen from this that the 

 loss of nitrogen was much greater during the second 

 series, which only differed from the first in the fact that 

 the dung lay without trampling for a month during the 

 summer. 



Taking these results as a whole, it is seen that, even 

 with the most careful management, the loss in making 

 the dung amounts to 1 3 per cent, of the total nitrogen 

 supplied in the food, in addition to 6 per cent, or so 

 which the animals retain. This loss increases with 

 great rapidity if the conditions are less favourable; the 

 minimum is only attained if the dung be kept trampled 

 beneath the animals in a deep box, for if it be left to 

 itself for a time, or if it be made in a shallow stall and 

 thrown out daily into a heap, as is often the practice, 

 the loss rises to between 30 and 40 per cent. 



In connection with the first-mentioned experiment, 

 Maercker and Schneidewind made determinations of 

 the state in which the nitrogen exists in the dung, 

 whether it was soluble and therefore active, or msolubic 

 and comparatively inactive. From the known digesti- 

 bility of the foods consumed, it was possible to calculate 

 what proportion of the nitrogen in each food left the 

 body in a digested condition as urea and kindred 

 bodies dissolved in the urine, and what proportion 

 consisted of undigested and insoluble compounds in 

 the faeces. Maercker and Schneidewind found not only 

 that the loss had fallen upon the active nitrogen — i.e., 



