178 HOW CROPS GROW. 



may be accidental, endeavored to approach a solution of 

 this question by comparing together the ashes of sam- 

 ples of the same plant, cultivated under the same circum- 

 stances in all respects, save that they were supplied with 

 unequal quantities of readily-available ash-ingredients. 

 The analyses of the ashes of buckwheat-stems, just 

 quoted, belong to this investigation. Wolff showed that, 

 by assuming the presence in each specimen of buckwheat- 

 straw of a certain excess of certain ingredients, and de- 

 ducting the same from the total ash, the residuary ingre- 

 dients closely approximated in their proportions to those 

 observed in the crop which grew in an unmanured soil. 

 The analyses just quoted (p. 163) are here "corrected" 

 in this manner, by the subtraction of a certain per cent 

 of those ingredients which in each case were furnished 

 to the plant by the fertilizer applied to it. The num- 

 bers of the analyses correspond with those on the previ- 

 ous page. 



123456 



20 p. c. ZOp.c. 25 p. c. 8.5 p. e. 16.6 p. c. 



Chloride Carbonate Carbonate Sulphate Carbonatet 



After deduction of of of of ofcalc'mand 

 of Nothing, pptas- potas- potas- magne- magne- 

 sium, sium. sium. slum. sium. 



Potash 31.7 27.0 32.5 33.5 30.6 28.0 



Chloride of potassium. 7.4 9.1 1.0 3.9 7.4 11.3 



Chloride of sodium... 4.6 3.8 4.0 4.7 3.7 1.9 



Lime 15.7 17.3 16.0 14.5 15.3 14.6 



Magnesia 1.7 2.4 4.1 1.7 2.3 2.9 



Sulphuric acid 4.7 3.5 3.4 5.4 2.1 4.1 



Phosphoric acid 10.3 11.7 8.1 11.2 11.8 11.7 



Carbonic acid 20.4 20.1 25.9 19.8 21.6 19.3 



Silica 3.6 5.2 5.2 5.3 5.2 6.1 



100.0 100.0 100.0 100.0 100.0 100.0 



The correspondence in the above analyses thus " cor- 

 rected," already tolerably close, might, as Wolff remarks 

 (loc. cit.), be made much more exact by a further correc- 

 tion, in which the quantities of the two most variable in- 

 gredients, viz., chlorine and sulphuric acid, should be 

 reduced to uniformity, and the analyses then be recalcu- 

 lated to per cent. 



