142 



In flosins, I wish to ompliasizo the opinion of myself, and surely of yon who 

 have had lonirer experience, that no man is justilied in drawinj; eondnsions 

 from one y«>ar"s worlv with pot .cultures or Held e.\i)eriments ; and to call yoiu' 

 attention to the fact that the results shown in this tahle are the actual yields 

 ohtained without any corri-ctions ; that they are from only oue of (mr soil- 

 oxperiment tields. of which we have more than twenty in various parts of the 

 State, from near the Wisconsin line almost to the Ohio Uiver ; and that the 

 results which I have j^iven in detail, while they furnish some of our most 

 decisive data, are yet in harmony with the jreneral average of all results on 

 similar soils, and not only with other i)lat exi)eriments and tield trials, but also 

 in harmony with pot-culture experiments and with the chemieal composition of 

 the soil. 



H. J. Waters, of Missouri. Have you been able to see the gain the first year 

 from the application of tinely ground rockV 



C. G. IIoPKixs. A good many of our experiments with the ground rock began 

 this year. Last year, as an average of IG tests where we applied ground rock 

 without any organic matter, we made a gain of 4.G bushels per acre ; where we 

 applied it with stable manure we made a gain of 8.6 bushels above that made 

 by the rock phosphate without the organic matter. 



H. J. Waters. I would like to ask Doctor Armsliy if he has comimted the 

 results of the experiments that have been made with unacidulatcxl rock at the 

 Pennsylvania station. 



H. P. Arm SHY, of Pennsylvania. Those experiments ran three or four rota- 

 tions and were then tliscoutinued. They showed no material superiority of 

 the soluble over the insoluble phosphates. 



A Member. Why is it that the nitrogen gave increasingly larger results when 

 used year after year with the phosphorus? 



C. G. Hopkins. Where we were using phosphorus without nitrogen the nitro- 

 gen would run down in the soil, because no legume crops had been used. It 

 is very necessary to make annual applications of nitrogen in order to keep up 

 the nitrogen in those soils. It is an element the supply of which will be ex- 

 hausted quickly if we do not take some measures to preserve and maintain it. 

 Phosphorus, however, is the controlling element, as you will see from the aver- 

 ages. We can not draw any conclusions from a single yield on a single plat, 

 but the averages from four different tests show that the first year nitrogen 

 produced an increase, each year the increase growing larger, until in the fifth 

 year it was 6 bushels of wheat. 



Potassium produced the first year an average of ('> bushels of corn to the 

 acre, and if we needed any furtluM" evidence that would add a little more 

 strength to the statenient that we nuist not draw any final conclusion from one 

 year's work, it is quite evident to me that the action of this potassiinn salt 

 was a corrosive action. We know that common table salt will increase the 

 yield of many lands for several yeai-s. but the result is not permanent. And 

 here the average of the tests fnr the four years shows that more potassium was 

 not needed in this case. 



I would say that potassium frequently produces an injurious effect: that its 

 use as often results in a decrease as in an increase. 



R. J. KEDniNG. of Georgia. I discovered fifteen or sixteen years ago in my 

 first experiment on our soils that potash in any form containing chlorin was 

 oftener injurious than beneficial, and we try now to be very particular never 

 to ai)ply an excess of such salt. 



C. (I. Hopkins. I might say that where our soils are very low in available 

 potassium we ai)ply potassium as we did in this soil and get marked increases. 

 On <mr jieaty and swamji lands we ai)ply not less than 100 i)i)unds or perhaps 2(X) 

 poumls per acre of itotassium suljihate or muriate. We have increased the 

 yield from 4 or 5 bushels per acre up to 70 bushels. Those lauds are exceed- 



