EXPERIMENTAL WORK WITH RAW ROCK PHOSPHATE. 63 



The results on corn given in Table XXXII are rather inconclusive, 

 even though they represent the crops of six years. The average pro- 

 duction of the no-phosphate plots was in many instances greater than 

 that of the plots receiving phosphate treatments. Only in the cases 

 of the plots treated with reverted phosphate of iron and aluminum 

 were there fairly consistent increases obtained over the check plots. 

 In 1903 the average yield of the check plots exceeded that of any of 

 the others except the Florida soft phosphate plots and equaled the 

 average yield of these. 



The yields of the various soluble phosphate plots were very incon- 

 sistent. For instance, the plot treated with double acid phosphate 

 gave on the average an increase in yield over and above the no- 

 phosphate plots, yet the plot treated with dissolved South Carolina 

 rock (ordinary acid phosphate), which was applied in such quan- 

 tities as to furnish the same amount of phosphoric acid, gave a yield 

 considerably below the average of the check plots. 



The results obtained on the hay crop (Table XXXIII), however, 

 were more consistent. With the exception of the first crop of the 

 dissolved South Carolina rock plot, the yields from the less soluble 

 phosphate plots were greater than from the water soluble phosphate 

 plots, and in nearly every instance the yields of the treated plots 

 were considerably greater than those of the checks. Here, again, 

 the reverted phosphate of iron and aluminum gave the highest yields. 



Although the conditions of this experiment were apparently more 

 favorable to the soluble phosphates the results indicate that for corn 

 and hay the relatively insoluble phosphates of iron and aluminum 

 in a very finely divided or precipitated condition, were more effective 

 than the former. 



In the case of the two crops of wheat grown on these same plots 

 the opposite results were obtained (Table XXXIII). The yields 

 of the soluble phosphate plots in most instances exceeded the yields 

 of the relatively insoluble phosphate plots quite appreciably, and all 

 the treated plots gave considerable increases in yield over the check 

 plots. 



Regarding the influence of decaying organic matter in the soil, the 

 turning under of rye seemed to increase somewhat the effectiveness 

 of the insoluble phosphates, but the plots on which clover was grown 

 and subsequently turned gave on the average smaller yields than 

 similar plots receiving no organic matter. 



In studying the results of these experiments two questions natu- 

 rally arise. First, how uniform was the soil of the various plots, 

 and, second, what would have been the effect of applications of lime ? 



Another five-year experiment undertaken by the Maryland station 

 some time after the one just described and discussed, gives consider- 

 afeion to these points. In this later experiment an attempt was 



