10] 6 Agricultural Gazette of N.S.TF. [Dec. 2, 1908. 



The pi'utciii surtcrs cousidci'uhlv, though not in i-culity ;is inucli us the 

 nitrogen-free extract. Hyth-olytic decomposition, complicated hv bacterial 

 action, takes phice, and altnough there is Tiot iiimli, if any, loss of free- 

 iiiti'oi;('ii, the iii'w nit I'diiciioiis coniiMHinds air li'ss \'aluable as food than the 

 protein. 



Tahlk 4. 



Mean losses in the sih), l'J()4 :Ind 1905. 



Dry 

 matter. 



Ether 

 extract. 



Nitrogen- 

 extract. 



Fibre. 



Total I'l-otoiii 

 nitrogen, nitrojjeii. 



Noii- 



protein 



nitro'^en. 



Ash. 



Fur- 

 fiirol. 



The figures show an al)solute loss of nitrogen, but we are satisfied that a 

 certain amount of annnonia is given off during .sampling, and we have no 

 evidence to show how or to what extent losses of nitrogen occur in the silo. 

 In laboratory experiments on silage made in bottles we onlv observed losses 

 of nitrogen when air was admitted, and this did not happen in the silo. 



The figures given for the loss in ether extract are quite accidental, since 

 the value obtained in any pai-ticulai' l)ag depends on its position in the silo ; 

 the bags in the top half showed a considerable loss — 32 per cent, on the 

 average — those in the loioer half showed an average gain of G pei' cent. 

 The ether exti'act of silage contains a number of soluble acids which 

 obviously wash downwards. Further, since these acids are derived horn the 

 nitiogen-free extract or protein of the maize it follows that no comparison is 

 iji any case possible. 



A similar downwash occurs with the soluble ash constituents, the upper 

 bags losing on an average 17 per cent, and the lower bags gaining '1 per 

 cent., hence the average figure given in Table 4 is of no value. The in- 

 soluble ash c(mstituents are of cour.se not liable to this movement ; but their 

 amount is so small and variable — depending partly on the presence of stray 

 soil — that the experimental error becomes too large to give the calculations 

 any value. 



The general nature of the losses outlined above is probably the same in 

 all silos, though the actual amount vai-ies : our losses are highei- than those 

 observed by American investigators, but there is considerable ditl'erence 

 between American and Knglish maize at the tinie of cutting; ours is much 

 less mature, and contains a lower proportion of nitrogen-free extract. 



When a farmer makes silage on a large .scale he often wants to know what 

 loss takes place in his silo. The most comincing way of demonstrating this 

 is, of course, to bury weighed bags of material as we have done, but a 

 sufficiently accurate result can be obtained by assuming that the fibre under- 

 goes no change, or indy a 5 per cent, loss, and calculating on this basis the 



