8 



est percentage of nitrogen, when no nitrogen was applied 

 to it. With this exception, the highest percentage of nitro- 

 gen is found in plot 3 which has nitrogen fertilization, and 

 the lowest percentage where nitrogen was used, is higher 

 than the average of those where no nitrogen was added, 

 even when the high percentage of plot 8 is included in 

 the estimate. As has already been noted, the variation in 

 phosphoric acid seems to obey no rule, the percentages in 

 the two soils being practically the same. 



In the beginning of this discussion it was stated that the 

 garden soil was in a high state of cultivation to begin with, 

 and it was to be expected, that the influence of fertilizers 

 here, both on the composition of the plant and on the yield 

 of seed cotton, would be less strongly marked than in the 

 poorer soil. While this is the case, it is, also, true that by 

 fertilization with potash and nitrogen the percentages of 

 these constituents even here are increased. This is notably 

 true in plot 9, where all three fertilizers were applied and 

 where are found the highest percentages of these ingredi- 

 ents. 



The average effect of fertilization on the percentages of 

 the fertilizing constituents in the plants may best be seen 

 by reference to Table II. By the term "fertilization" in 

 this table is to be understood the use of the particular in- 

 gredient in question, without reference to the other in- 

 gredients. Thus when percentages of potash are considered, 

 fertilization with potash without reference to phosphoric 

 acid or nitrogen is solely considered. 



