EXPERIMENTAL STATIONS — TUENIP CROP 1882. 



373 



The amouDt of ash contained in the dry matter is greater 

 at Harelaw than at Pumpherston ; the turnips which contained 

 the larger proportion of water thus contained the larger pro- 

 portion of ash, the average over the whole plots of the stations 



being — 



Pumpherston, 

 Harelaw, . 



Water, 

 per cent. 



91-3 

 92-6 



Dry matter, 

 per cent. 



8-7 

 7-4 



Ash in dry matter, 

 per Cent. 



5-8 

 6-4 



The average amount of ash in the dry matter of the crop in 

 1879 was about 8-5 per cent., and now it is not so much as 

 6 '5 per cent. This result may be due to various causes, such as 

 the shorter time during which the crop was allowed to remain 

 on the land, and the very different condition in which the crop 

 was secured ; and perhaps the system of manuring to which the 

 land has been subjected may contribute towards the result, but 

 the method in which the ash was determined has no doubt much 

 to do with it. The amount of ash formed depends very much 

 on the method of incineration employed. 



For the purpose of this investigation, the absolute amouat of 

 ash contained in the roots is not of much importance, as it is the 

 relative proportions obtained by various modes of manuring that 

 are of chief interest. The essential thing to observe is that all 

 the samples be treated in the same way, and in the above analysis 

 the ash was determined in a very thorough manner. Dry 

 turnip powder is not easily incinerated ; it coheres into a mass 

 which prevents the access of air, and is found to be dark 

 coloured even after prolonged incineration, and the greater the 

 quantity employed the more imperfect is the ash. The method 

 adopted on this occasion was to take two grammes of the fine 

 dry turnip powder and mix with it one gramme of pure white 

 sand recently ignited and cooled in a drying chamber. This was 

 placed in a small platinum tray, and kept for an hour at a low 

 red heat in a muffle furnace. The somewhat coherent mass was 

 then gently pressed down with a glass plate exactly fitting the 

 tray, and then carefully stirred. The tray was then replaced in 

 the mullle and. kept therefor half an hour, when a perfectly 

 homogeneous white ash was obtained. The ashes were all done 

 in duplicate, and the duplicates were usually found to agree with 

 each other very closely. When larger quantities are used, and 

 no sand employed, the results are far from uniform, and always 

 higher than tliose obtained by the method here described. In 

 all cases where the duplicate samples didered from each other 



