96 



the plots to which fertilizers were applied singly and two 

 by two, we find that the results in 1897 agree substantially 

 with those of 1890, 1891, 1892 and 1896 in ,showrng that all 

 thr^e of the usual fertilizer constituentis increase the yield 

 of cotton on this soil. 



The experiment at Jacksonville was conducted by Prof. 

 C. W. Daugette. The figures aif ord no euggestio-ns as to the 

 needs of this soil. Probably previous applications of ma- 

 nure, or previous methods of treatment, have rendered the 

 field unfit for experimental purposes. 



Experiment Made by Dr. Johx T. Gordon, Healing Springs 



Washington County. 



Gray, sandy soil, 12 inches deep ; sandy clay subsoil. 



The field is described as a gently rolling ridge between 

 two bra'nches, on which the original growth was long leaf 

 pine. It was in cotton in 1896, and for the three years pre- 

 ceding that time it was continuously in corn and cowpeae. 

 "There was no rust or other leaf disease. Leaves remained 

 green until the dry, hot winds came, about the last of Au- 

 gust and first of September, when the leaves seemed to 

 wither, at first in spots, afterwards pretty generally." 



Although the yields of the unfertilized plots point to uni- 

 formity in natural fertility, the results are perplexing. Ap- 

 parently .some undiscovered cause was more influential than 

 the fertilizers. This is the fifth test of fertilizers on this 

 soil. Previous results were either inconclusive or suggest- 

 ive of a deficiency of all three of the usual fertilizer ingre- 

 dients. 



Experiment JVIade by F. Funkey, 1^ IVIiles Soitth of Tus- 

 CUMBIA, Colbert County. 



Reddish soil and subsoil. 



This field had been in cultivation about fifty years. The 

 original forest growth was oak, blackjack oak and hickory. 

 Oats was the crop in 1894, corn in 1895 and 1896. The stand 



