154 



The dimimition in amonnt of solublo materials in tlie old field soil is very 

 striking;, especially in the important i)lant foods, potassium and calcium. The 

 very ai)parent ditferences of solubility lietwoen the top soil and subsoil emphasize 

 the well-established fact that the plant food of the subsoil is of relatively low 

 availal)ility. 



IMPORTANCE OF THE METHOD OF SAMPLING. 



If we bear in mind that the samples of top soil for this investigation were not 

 taken to uniform de])th. Imt were taken to the point where a marked difference 

 was noticeable between top and subsoil, not exceeding 10 inches, meaning an 

 irregular depth of from 4 to 10 inches, and also that the samples were not 

 averages in the true sense of the term, but were obtained by digging in one spot, 

 the regularity of the above results is even more remarkaltle. Uuiiuestionably, 

 for a comparison of this kind, the samples of top soil should be taken to a uni- 

 form de])th, not exceeding that ordinarily turned by the plow, and it is also 

 important to obtain an average by mixing portions taken from many places. No 

 assayer would consider liis work well done if he took for his sample a shovelful 

 of ore from one pl.ice in a carload : neither should the agricultural chemist be 

 content with his shovelful of earth from one spot in seven or eight hundred tons 

 comjwsing the first C) inches on an acre of land. The writer believes the old 

 directions for collecting soil samples are unfortunate and affected by theoretical 

 considerations that have no determinalile relation to the practice of the farmer. 

 If one sample be taken to the depth of 10 inches because the soil show^s no 

 difference of color above that depth, and another is taken to 7 inches because of 

 a change of ajipearance there, while the farmer plows all to 7 inches, we have 

 put ourselves out of relation with his work and have obscured existing relations 

 in the one case by mixing into our sample an uncertain amount of material 

 admittedly poorer in available plant food than that which comes into use, while 

 in the other case we get only that which is actually used. From the chemical 

 standpoint, as well as from the standpoint of the farmer, it is the character of 

 the first few inches of top soil that determines whether the crop will thrive or 

 fail. If the first roots of the young plant do not find within easy i-each enough 

 available food to support a healthy, vigorous growth they will never be able to 

 penetrate to the subsoil for the necessary supply of moisture and additional 

 food, and the crop will be stunted or a failure. The writer does not desire to 

 discourage the analysis of sulisoils in connection with top soils, but advocates 

 sampling to uniform depths, the first, representing the top soil, being well within 

 the depth of ordinary plowing. In the work of the Kentucky station 6 inches 

 has been selected. The samples are taken preferably by means of a tube made 

 of 1 5-inch steel bicycle tul)ing provided with a tempered steel cutting edge, so 

 constructed that the core of soil cut out is a little less in diameter than the 

 interior of the tube, and reenforced at the top by a heavy steel ring. With such 

 a tube of proper length samples can be taken to a considerable depth without 

 disturbing the surrounding soil and with very little effort. A coi'e from the first 

 G inches weighs about lOo grams, so that ten or fifteen cores will give enough 

 .soil for the usual processes of analysis, and the sample so taken will represent 

 a much better average tlian if taken from only one spot. Difficulties may be 

 encountered with these tubes in a soil filled with small stones, and it may be 

 necessary in such cases to have recourse to the spade or other large tool, but the 

 tube will be found nuich the best in most situations. 



The method with carl)onated water, like the method of Dyer with citric acid, 

 is an attempt to imitate natural solvents, but in using water more strongly 

 charged with carbon dioxid than is the natural soil water the i)rinciple is 

 recognized that the solvents used in the ialioi-atory must be stronger than the 

 natural ones if we would i)r()duce in a siiort time effects conunensurate with 

 those resulting from the long-continued action of natural solvents in the field 

 aided liy the continuous removal of certain constituents from the solution by 

 the plant roots. This selective actiop, of the roots of plants is a process very 

 difficult to imitate in the laboratory, although it must be a very important factor 

 in determining the result in the field. But even admitting that carbonated 

 water may be satisfactory, luider proper conditions, as a solvent for calcium 

 carbonate and that a determination of the ])otash dissolved sinuiltaneously 

 may give valuable information, the very natiu-e of the solvent renders it diffi- 

 cult to ajtply in known and constant strength. And its inefiiciency in dissolv- 

 ing ])hosiihates renders it us(»less as a means of detei'mining the availability of 

 the phosphorous compounds. There ai)pears to be no practical or theoretical 

 reason why a dilute solution of one of the more active acids, such as hydro- 



