STATE EXPERIMENT STATION 231 



As to soil analyses, I am so blind as not to be able to see 

 much benefit in them as usually made. If you can find a 

 half dozen soils, more or less, which evidence a graded pro- 

 ductiveness, and can show any positive relations between com- 

 position and fertility, that would [be] a valuable result. My 

 friend Dr. E. W. Hilgard's methods (1st Census Report on 

 Cotton Production of U. S.) appear to be most promising. 

 Dr. G. E. Moore's plan of mechanical analysis (see 10th Cen- 

 sus Report on Tobacco culture) is worthy of investigation. 

 I have long desired to do something in soil analysis but have 

 not yet been able to undertake any systematic and extended 

 work. What to do and where to stop in soil analysis are very 

 perplexing questions. Yours very truly, S. W. Johnson. 



(S. W. J. TO A. D. G.) 



March 27th, 1886. 



Dear Sir, In answer to yours of yesterday, I believe it 

 would be a great gain were : All our States to adopt one uni- 

 form Fertilizer Law; were all samples for official analysis 

 taken by honest experts; were all analyses made by uniform 

 (if good) methods. I also believe that Experiment Stations 

 and Departments of Agriculture are more properly supported 

 by a general tax than by a special tax on manufacture of 

 fertilizers. 



There are, however, two sides to all these questions. To 

 illustrate : 



1. The laws that now appear to work well in Georgia and 

 North Carolina would not apply in Connecticut. 



2. The detriment caused by the imperfect sampling of 

 honest men or by the "cooked" samples of an occasional 

 rogue would be offset by danger of systematic fraud by 

 corrupt experts. 



3. Uniformity in methods of analysis is no remedy for 

 bad methods or unskilful use of good ones. 



