Averaging the results given in the above table, the following comparisons can be made as to the effects of P2O5 on quality.

The above schedule of prices is based on pool ratings by Mr. Walter Edwards and 1923 Pool prices. Average prices per pound of the tobacco grown on the different plots were computed on the same basis as the nitrogen plots. This is given in Table XII.

CONCLUSIONS FROM THREE YEARS DATA ON THE PHOSPHORIC ACID PLOTS

The most striking result of these experiments is the bad effect of increasing the phosphoric acid to 306 lbs. per acre. This is evidenced first by the reduced yield, the average being 31 pounds less than where no mineral phosphoric acid was used. In quality it was rated as "poor" in 1922 as compared with "excellent" for the Pl plot. The sorting data for 1923 show little difference in the tobacco taken from the different plots. In 1924, however, the quality was so poor that it was pooled at an average price of 4 cents per pound less than where no mineral phosphoric acid was added. The net return after deducting the cost of the fertilizer was $77.36 per acre less than where no mineral phosphoric acid was used.

These results are not in accord with those presented by Jenkins as discussed above. No explanation of the contradictory results will be attempted until the tests have been carried further.

The highest average yield for the three years was obtained where 225 lbs. of phosphoric acid were supplied by addition of both acid phosphate and precipitated bone to the organic carriers (P1) but the best quality was obtained on P3 where the acid phosphate was omitted but the P2Os brought up to 191 lbs. by addition of precipitated bone. The latter also gave the highest net return per acre of all the phosphoric acid plots in 1924. The effect of omitting all carriers of mineral phosphoric acid was noticeable only by a slight reduction in quality.

POTASH SERIES.

This series of experiments was begun in 1923 and continued through 1924 with three treatments in duplicate on the same field as the nitrogen and phosphoric acid series. The object of the tests was to compare the effect of supplying one half or all of the potash in double sulfate of potash magnesia instead of supplying it all in high grade sulfate of potash. The occurrence of "sand drown" during 1922 in the other plots previously mentioned suggested the advisability of supplying magnesia in the fertilizer ration. Since the cheapest and most convenient carrier of magnesia is the double manure salt, it has been most frequently used as the source of this element in fertilizer mixtures. It seemed advisable therefore to find out what effect its use would have on the yield and quality of the leaf.

Goessman, in the experiments in Massachusetts previously referred to, after three years of testing, sums up his conclusions in regard to it: "Our results with potash magnesia sulfate as main potash sources of a tobacco fertilizer are not encouraging". In rating the ten fertilizer formulas which were tried he places the two which contained double manure salt at the foot.

Jenkins, in the five year experiment at Poquonock, found that the plot treated with double manure salt gave a higher yield than the plots treated with any other carrier of potash. The quality of leaf, however, was not so good and the fire-holding capacity was less than all the rest except high grade sulfate which stood at the foot of the list although the yield was good. Thus he ranks double sulfate above high grade sulfate while Goessman does just the

reverse.

In the face of these contradictory experimental results there appeared to be need of further tests. It was decided to compare plots where double manure salt was the only source with those in which high grade sulfate was the source and also with plots where the ration contained a mixture of the two carriers. The fertilizer ration of each of the plots is as follows: