Febeuaky 26, 1904.] 



SCIENCE. 



345 



nitrogen or phosphorous or both were applied; 

 yielded about the same, 36 bushels when 

 potassium, 40 bushels when potassium and 

 nitrogen or 38 bushels when potassium and 

 phosphorus were applied. But when potas- 

 sium, nitrogen and phosphorus were all ap- 

 plied, the indications of the analysis were 

 flatly contradicted by a yield of 60 bushels. 



In an 'Added Note' to the circular it is 

 stated : " In connection with the discussion 

 which followed the reading of this and several 

 other addresses relating to this general sub- 

 ject at the convention in Washington, the 

 fact was clearly developed that some of the 

 new analytical methods devised by the Bureau 

 of Soils and used in the work reported in Bul- 

 letin No. 22, instead of being ' very accurate 

 methods of analysis,' are absolutely untrust- 

 worthy." This statement is not in accord 

 with the facts. The only method mentioned 

 in the discussion was that for determining 

 phosphates. The validity of the method itself 

 was not questioned and the discussion was con- 

 fined to the discrepancy in the solubility of the 

 phosphates in the Eothamsted soils, as shown 

 by the results reported in the bulletin, and 

 those reported on the same soil samples in an- 

 other publication.* During the public dis- 

 cussion referred to it was distinctly and ex- 

 plicitly stated that the authors of Bulletin No. 

 22 were aware of the discrepancy between 

 their results and those in the publication cited, , 

 that they believed they knew the reasons there- 

 for through work which was being done upon 

 the solubility of phosphates, in the laboratory 

 of the bureaii, and that they had satisfied 

 themselves that the results given were sub- 

 stantially correct. 



Nevertheless, in the ' Added Note ' it is 

 stated that the absolute untrustworthiness of 

 the methods used ' is further established by an 

 examination of the data which are given in the 

 publications referred to,' and a table is sub- 

 mitted in which there is a comparison of the 

 number of pounds of phosphorus per acre, to 

 a depth of seven inches, in the Eothamsted 

 soils, as calculated from the data in the two 

 publications. In this table results are stated, 

 ' reported ' by Bureau of Soils, three minutes' 



* Jour. A m. Ohem. Soc, 24, 79, 1902. 



extraction with distilled water, whereas the 

 method actually employed and described in de- 

 tail was to stir the soil in water vigorously 

 for three minutes, then allow to stand 

 twenty minutes before decanting and filtering, 

 and the work of King was cited to show the 

 significance of the time element. Equally 

 inaccurate is the heading to the other column 

 of figures which are stated as ' obtained ' after 

 fifteen hours' extraction with dilute acid. As 

 a matter of fact, according to the statement 

 in the paper from which the data were taken, 

 the soils were digested for five hours in a 

 hydrochloric acid solution, which contained 

 enough hydrochloric acid to be a iV/200 solu- 

 tion when the carbonates of the alkaline earths, 

 etc., were nevitralized, and here also the im- 

 portance of the time element was emphasized 

 by the author of the method. Beyond the in- 

 excusable carelessness of misquoting results 

 and statements in a controversial paper, these 

 inaccuracies are objectionable because pur- 

 posely stated in such a way as to infer in- 

 vidious and quite inaccurate comparisons. 

 Moreover, it is not at all clear why the phos- 

 phorus as determined in the two investigations 

 should be compared on the basis of an acre 

 surface with a depth of seven inches, for it is 

 inconceivable that any one at 'this day, and in 

 view of the well-known work of Darwin and 

 others, would suppose that the same identical 

 seven iiiches of soil would remain at the sur- 

 face for any considerable period of time. 



Following the table, the statement is made 

 that the author of the Journal article cited 

 " determined the phosphorus by the absolute 

 gravimetric method of the Association of 

 Official Agricultural Chemists, and there is no 

 reason to doubt the accuracy of the results thus 

 obtained. The Bureau of Soils used a newlj 

 devised colorimetric method which evidently 

 gives results about a thousand per cent, above 

 the truth." These statements are incorrect. 

 The procedure of the Association of Official 

 Agricultural Chemists was not followed; but 

 an entirely different one, which is not absolute, 

 bvit indirect; is not a gravimetric, but a 

 volumetric one; and the accuracy of the pro 

 eedure which was actually used has not been 

 established by any published work upon it 



