No. 4.] GERMAN EXPERIMENT STATIONS. 403 



As my final word, I wish to present some general deduc- 

 tions, drawn by Professor Wagner from the results of his 

 many experiments. 



1. By a liberal manuring with potash and superphos- 

 phate, we must make the nitrogen-gathering plants nitrogen- 

 hungry^ in order that they will take from that nitrogen mag- 

 azine, the air, the largest possible quantity. 



2. For the nourishment of the nitrogen-consuming plants 

 — the grains and hoed crops — we nnist take the greatest 

 amount of nitrogen from the air, as the very chea})est source. 

 This end can be best attained by a rich manuring of mead- 

 ows, by devothig an increased area to leguminous crops, 

 and especially by planting leguminous crops to be ploughed 

 in as green manure, and by manuring them lil)erally with 

 potash and superphosphate. 



3. In so far as the nitro2:en taken from the air or obtained 

 from stable manure and ^reen manuring is not sufficient for 

 the growing crops, we must make up for it hy applying as 

 much sulphate of ammonia or nitrate of soda as the crops 

 can use to advantage, considering the condition of soil and 

 climate. 



4. The manure obtained from animals fed with legumi- 

 nous crops is naturally rich in nitrogen, and the loss of this 

 nitrogen can be prevented by frequently scattering super- 

 phosphate or sulphate of potash and miignesia into the stall 

 and over the manure. 



5. We must try to make the crops appropriate all the 

 nitrogen given to the S(>il in the form of stal)le manure, 

 green manure. Chili saltpetre or sulphate of annnonia. To 

 attain this end we should manure lil^erally with potash and 

 phosphoric acid. 



G. When the soil is found lacking in lime, we must 

 spread marl or quicklime over it, so that the phmt may not 

 be prevented from taking up the nitrogen, potash and phos- 

 phoric acid. 



The above rules must not in every case be taken as an 

 infallible ffuide, but rather looked at in the liijht of a oreneral 

 statement, and as such are productive of satisfactory results. 



