110 BULLETIN 699, U. S. DEPARTMENT OF AGRICULTURE. 



single year, it appears that their relative natural fertility varied 

 greatly, yet check plot No. 23, which yielded 3 less bushels of corn 

 than check No. 12 in 1907, exceeded the yield of this latter plot in 

 1908, by nearly 11 bushels. These same inconsistencies are noticeable 

 throughout the table. The plot receiving a fair application of lime 

 gave better yields than any of the phosphate treated plots, yet the 

 plots treated with basic slag, which is richer in lime than any of the 

 other phosphates, gave poorer yields than either of the " floats " plots 

 and two out of three of the acid phosphate plots. While the raw 

 rock phosphate plots produced on the whole better results than those 

 treated with either of the other forms of phosphoric acid, the average 

 yield of com on these plots for the entire five-year period was only 

 1 bushel greater than the average on all the check plots. 



In view of the fact that the applications of raw rock were entirely 

 too light to prove effective and moreover were only from one-sixth 

 to one-half as great as those of acid phosphate, no comparison be- 

 tween the two forms of phosphoric acid would be justified even had 

 the soil been responsive to phosphate treatment. Practically the 

 only conclusion to be drawn from these results is that all forms of 

 phosphoric acid were ineffective. 



EUett and Hill ^ also published the results of four pot experiments 

 with various phosphates on different but important Virginia soil 

 types. The tests were made in the greenhouse and were similar in 

 all respects. Twenty -five pounds of each soil type was taken for each 

 pot, the fertilizer added, and the whole thoroughly mixed. The 

 moisture content was maintained in each pot throughout the experi- 

 ment at about one-half the water holding capacity of each soil. The 

 seed planted, the number of plants per pot, and the fertilizer treat- 

 ments were the same for each series of pots. Every treatment was 

 run in quadruplicate. The various soil types employed were as fol- 

 lows: (1) A typical Coastal Plain sandy soil, low in phosphoric 

 acid ; (2) a Piedmont " Red Lands " clay soil well supplied with phos- 

 phoric acid and potash but low in nitrogen; (3) a Piedmont choco- 

 late-colored clay loam suitable for general agriculture, with rather 

 high contents of potash and phosphoric acid, but low in nitrogen; 

 (4) a Piedmont gray sandy soil lower in its content of phosphoric 

 acid and potash than the chocolate clay loam and very poor in 

 nitrogen. 



The results of the pot experiments with these soils are given in 

 Table LXVI. 



iVa. Agr. Expt. Sta. (Blacksburg). Bui. No. 200, p. 24 (1912). 



