1018 Agricultural Gazette of N.S.W. [Dec. 2, 1908, 



samples taken as speedily as possible for the various nitrogen determinations. 

 But no matter how quickly we worked we could not altogether avoid loss ; 

 our fisrures for total nitro<;en and free ammonia are therefore low. Still, the 

 loss is much less than if the samples had been dried before the nitrogen was 

 determined, for the nitrogen found in the dried material is always le.ss than 

 one expects by calculating from the amount in the original wet silage. It 

 will be observed that in three cases the dirt'erem-e is approximately equal to- 

 the amount of ammonia present in wet silage, but in the other cases it is 

 much greatei 



Samples 1 to 4 were drawn from the 1905 silage, sample 5 was taken in 

 1906, and sample 6 comes from a silo on another farm. 



Another sample of silage was quickly drawn for the determination of 

 the volatile acid. It was distilled in steam, and the distillate titrated with 

 standard alkali. The results are only approximate, the higher acids come 

 over so slowly that the distillate never really becomes neutral. The non- 

 volatile acid was estimated by titrating the residue in the flask, but we soon 

 gave up this detei'mination because of the difficulty of getting a sharp end 

 reaction. 



Table 5. 

 Percentage of nitrogen in the dry matter of silage. 



A third sample was dried, ground, and used for the estimation of ether 

 extract, fibre, ash, and nitrogen-free extract. The ether extract results are 

 only approximate ; samples continue losing weight for days in the extraction 

 apparatus. Some of the acids are only slightly soluble in ether ; the colouring 

 matter also dissolves with difficulty. 



The bulk of the sample was then used for qualitativeexamination. — (Journal 

 of Agricultural Science.) 



