186 Agricultural Experiment Station, Ithaca, N. Y. 



(c) The rest of the combustible matter, foriuin^ but a small por- 



tion of it, contains nitrogen, and is called nitrogenous matter 

 orproteids. Though small in proportion it is very important. 

 The nitrogen for it must come from the soil. 



(d) The production of starch, sugar, cell-walls and acids is the 



chief work accomplished in the growth of the vine and its 

 fruit. 



(e) The quality of the grape for eating depends largely on the 



relative proportions of sugar and acid. The ripening of the 

 grape consists largely in changes in these proportions. 



(f) Can the grape grower modify these proportions by fertilizing 



or other treatment of the plant ? 



(g) The fermenting of the grape juice depends on its sugar, the 



access of air and the assistance of micro-organisms. 



Hoio can the Farmer tell ivhat Fertilizer his Soil Needs f 



(Given at the Jamestown School.) 



(By G. C. Caldwell.) 



Nitrogen, phos})horic acid and potash, are the only plant foods 



that need any special looking after in the management of manuring 



the soil. 



The old-time management compared with more modern methods. 

 In some respects the modern method better, in others not. Neither 

 method is based on any real knowledge of what any particular crop 

 growing on any particular soil requires. No royal road to this 

 knowledge. The chemical analysis of the soil will not give a reliable 

 answer to the question in hand, because the soil of a field cannot be 

 fairly sampled. A recent striking illustration of the difficulty. 



The question must be put to the soil and crops together in each 

 case, by plot experiment. The difficulties of this method discussed. 

 How the experiment should be conducted, — 

 a : Selection and preparation of the tield. 

 b : Size, shape and arrangement of the plots. 

 c: Fertilizers to be used, and when and how to be applied, 

 d : Cultivation of the crop. 

 e : Harvesting and measurement of the crop, 

 f : Corrections that may be made to even np the results. 

 The use of the results as an answer to the question put. Con- 

 cerning the repetition of the experiment. 



