152 



not only that the diminution of tlio gross amounts of plant food caused by crop-" 

 \m\'^ could ho denionstr; ted by clioniical analysis, hut that the amount of plant 

 food caiJaijlc of Ixmhu exti-actcd liy a weak solvent and j)resuniahly available to 

 plants was dnninishcd also as the soil bccanio less productive. This investifia- 

 tion had for its fundamental ])roj)osition the comparison of rich soils against 

 the same after they had been im])Overished by cultivation, with the expressed 

 object to ascertain what constituents had been diminished in quantity and solu- 

 bility and to furnish a guide for the agriculturist in rei)leinshing the same. 

 Such was the work planned in 1S.'')4 by Dr. Robert Peter, chemist, and David 

 Dale Owen, chief geologist, of the Kentucky Geological and Agricultural Survey ; 

 and because the results are hidden in a series of State reports long out of i)rint 

 is the writer's excuse for ])resenting a smumary of them here. 



In those days there were no experimental fields under careful and continuous 

 observation, from which the chemist might draw samples for comparative an- 

 alysis: but the large areas of virgin land in Kentucky, more especially in the 

 Bluegrass region, offered standards of great fertility, while in the cultivated 

 fields examples were to be found of the depletion of available plant food result- 

 ing from continuous cr(ip])ing under an impi-ovident husbandry. To get mate- 

 rial for comparative analysis it was planned to collect samples of soils in sets 

 of three: First, the virgin soil: .second, soil of the same type from an adjacent 

 old field ten to fifty years in cultivation; third, the subsoil of tht> old field. 

 These samples were sulimitted to analysis by digestion in hydrochloric acid of 

 1.1 specific gravity to deterndne the acid-soluble mineral plant food, and also 

 by digestion in water saturated with carbon dioxid under a pressure of two 

 atuiosidieres. While the first process or " general analysis " gave very interest- 

 ing and instructive results, showing in a large majority of instances a material 

 diminution of mineral plant food by cultivation, it is with the latter that we 

 are now concerned. And the writer has tabulated about a hundred analyses, 

 taken at random, but using only such as exhibited the complete set of virgin 

 soil, old field soil, and subsoil of old field. 



The digestion in carbonated water was performed by putting 30 grams of air- 

 dried soil into a strong S-ounce 1)ottle. filling quickly with the charged water, 

 thus using approximately eight parts of solvent to one of soil, and corking 

 securely. The bottle was then jnit in a warm place, where the temperature 

 never exceeded 100° F., and allowed to remain about a month, shaking wel] 

 each day. The liciuid was theu filtered out. evaporated to dryness, and the resi- 

 due weighed, burned, and weighed again to deterndne " organic and volatile 

 matter." The nuneral part was then dissolved in hydrochloric acid, and the 

 solution analyzed in the same way as for the " gener.-d analysis." It was soon 

 found that the amount of ])hosphorus contained in the extract was exceedingly 

 small, even when the soil was very rich in this substance. So the deternunation 

 was onntted. The extracts often contained very noticeable (juantities of man 

 ganese, especially in the most fertile soils: but this deternunation was rarely 

 made. Therefore neither of these deternunations is tabulated. The tleter- 

 minations of soda and silica have also been omitted from the tabulation, because 

 they showed no regularit.v and seemed to have no possible significance. These 

 irregularities are easily exi)lained by the fact that the soda determination is a 

 difference, subject to relatively l;irge errors, and by the difficulty in obtaining 

 the soil solution perfectly .clear by filtration. The table gives a comparison of 

 the total extract, organic and volatile matter in the same, i)recipitate by am- 

 monia, calcium carbonate, magnesia, sulphur trioxid, and potassium oxid in 

 thirty-three sets of three samples each, from twenty-three Kentucky coxmties. 

 representing five great geological divisions. ;is compiled from analyses by Dr. 

 Robert Peter, State geologist, in 1 S,">4-1 SHO. The full-faced type is used when the 

 quantity is less in the virgin soil than in the cultivated soil or greater in the 

 subsoil than in the top soil. 



