TABLE VI. SUMMARY OF RESULTS FOR 1924 ON NITROGEN PLOTS.

SUMMARY OF THREE YEARS EXPERIMENTS ON NITROGEN RATION.

In summarizing the results of the first three years of these tests, it must be kept in mind that the experiments are not yet complete and any tentative decisions drawn may be reversed by results of the experiments of 1925 or subsequent years. Also it must be remembered that only one of the three years was a normal tobacco year, viz., 1923. The season of 1922 was unusually wet while 1924 was abnormally dry. It is conceivable that the results in a normal year might be different.

1 Ave. yield for 3 years.

*-Through a mistake which was not discovered until after the conclusion of these experiments plots N4 and N5 were treated with a low grade nitrate of potash (18% NH and 12.76% K2O) instead of a high grade (15% NH3 and 43.4% K2O) as was considered in working out the formula. The total amount of ammonia added to N4 was 275.6 lbs. and the potash was reduced to 106.5 lbs. per acre. Plot N5 likewise had the ammonia increased to 277.4 lbs. and the K2O reduced to 70.4. It is quite certain this mistake also occurred in 1923. For this reason, the results obtained on these two plots in 1923 and 1924 are of doubtful significance.

Nevertheless with the preceding reservations in mind, let us see what tentative answer can be made to the questions proposed on page 4.

In

1. Can all the nitrogen be supplied from mineral sources? answering this question it is best to disregard, for the reason previously stated, the plots on which nitrate of potash was used. Thus the question really is: Can all of the nitrogen be supplied from a mixture of nitrate of soda and sulfate of ammonia? This may be answered by comparing plots N1 (basal ration) with N3 (all nitrogen in sulfate of ammonia and nitrate of soda). The average yield of the N1 plots for the three years was 1,520 lbs. and for N3 was 1,573 lbs., showing thus a small increase in weight of cured leaves. In 1922 the quality of tobacco raised on N3 was rated poor as compared with N1 rated excellent. The sorting records for 1923 do not show that the N3 tobacco was inferior to N1. In 1924 the average price of the N1 tobacco was 2c. per pound higher than N3. When, however, this is balanced against the reduced cost of the fertilizer used on N3, the net return to the grower was only 90c. per acre less than where the more expensive fertilizer was used. However, since the quality was undoubtedly inferior during two out of the three years it would seem best for the grower to avoid the entire substitution of mineral sources of nitrogen and keep up the quality of the tobacco even though the immediate net return is no larger.

2. Can one half of the nitrogen be supplied from sulfate of ammonia and nitrate of soda? This question may be answered by comparing the records of the N1 plots with the N2 plots. The average yield on the N2 plots was 70 lbs. per acre less than the N1 plots for the three years. In 1922 the N2 tobacco was rated as "fair in quality" compared with N1 rated "excellent". On the other hand the sorting records of 1923 show somewhat better quality in N2 than in N1. The average price for the 1924 crop was practically the same for N1 and N2. The sorting records indicate about the same quality. The net return for the N2 tobacco was $5.56 per acre greater than for the N1 tobacco in 1924 because of the reduced price of the fertilizer. In answer to the question then we may say that the results during 3 years of tests indicate that mineral carriers of nitrogen may be used to advantage to supply one half of the ammonia.

3. Can one half of the nitrogen be supplied to advantage in dry ground fish? This may be answered by comparison of the records of the N1 and the N6 plots. The average yield of the N6 plots for the three years was 1,571 lbs. as compared with 1,520 lbs. for the N1 plots. In 1922 the quality of the two was rated the same. In 1923, N6 had more light wrappers, more seconds, fewer fillers and brokes and much longer fire-holding capacity than N1. In 1924 the average price of the N6 tobacco was about 1 cent less

than N1 and the net return per acre after deducting the cost of the fertilizer was $10.87 less for the fish plots than for the basal ration plots. Since the favorable results of the first two years are contradicted by the third year's results it would seem best to delay a decision until after further tests. At least we can say that the fish ration compared very favorably with the basal and it was superior during two of the three years.

4. Can one half of the nitrogen be supplied in tankage? The answer is obtained by comparing the records of plots N7 with N1 during the three years. The average annual yield per acre of the N7 plots was 1,543 as compared with 1,520 for the N1 plots. In quality N7 was rated "good" in 1922 as compared with "excellent" for the N1 plots. The sorting records of 1923 show a higher percentage of light wrappers and seconds, a smaller percentage of fillers and brokes and a better fire-holding capacity on the N7 tobacco than the N1 plots. The situation is just reversed in 1924 when the average price per pound for the tankage tobacco was 2.2 cents less than N1 tobacco and the net return per acre was $35.55 less than where the basal ration was used. Here again, the results of the three years are contradictory and final decision must be reserved until after further trials have been made. The case does not seem quite as favorable for tankage, however, as for the substitution of fish.

*

PHOSPHORIC ACID SERIES.

This series was begun in 1922 with the object of determining how much phosphoric acid should be used on tobacco. There is considerable difference of opinion on this point among tobacco growers. The only experiments dealing with this point on Connecticut Valley tobacco are those recorded by Jenkins which were conducted in co-operation with the United States Department of Agriculture. In experiments on shade tobacco they found that by increasing the acre application of phosphoric acid from 210 to 310 pounds, the yield was increased by 25 pounds when precipitated bone was used, 55 pounds with double superphosphate and 88 pounds with acid phosphate. The burn and quality was good on all but best where precipitated bone was used to supply the extra hundred pounds of phosphoric acid. In another test, on broadleaf, they increased the phosphoric acid from 176 to 276 pounds by addition on successive plots of (1) acid phosphate (2) Thomas slag (3) double superphosphate and (4) precipitated bone. The addition of acid phosphate did not increase the yield, slag increased it by 108 lbs. per acre, double superphosphate by 144 lbs. and precipitated bone by 408 lbs. The burn was freer on the plots without added phosphoric acid but the leaf grown on added double super

* Jenkins, E. H., Studies on the tobacco crop of Connecticut. Conn. Agric. Expt. Station, Bulletin 180:28-30. 1914.

phosphate and on precipitated bone had better quality than the others. From these data one would expect that there would be a distinct advantage in raising the amount of phosphoric acid to about 300 lbs. per acre.

At the Windsor station during 1922, '23 and '24 four plots in triplicate received varying amounts of phosphoric acid according to the following formulas. The amount of ammonia and potash was the same in all. The plots were on the same field as the nitrogen series during these years and all treatment throughout the season after the application of the fertilizer was the same for all plots.

PLOT P1. BASAL RATION 225 LBS. P2O5. SAME AS N1.

PLOT P2. BASAL RATION BUT WITHOUT ACID PHOSPHATE OR PRECIPITATED BONE. 75 LBS. P2O.

PLOT P3. 190.5 LBS. PO, IN PRECIPITATED BONE, COTTONSEED MEAL

AND CASTOR POMACE.

PLOT P4. 306 LBS. PO IN PRECIPITATED BONE, COTTONSEED MEAL

AND CASTOR POMACE.

During 1922 the plots were carefully watched for any signs of earlier maturity. The triplicate plots, P4, almost from the start showed a marked tendency to early ripening and the buds appeared on the plots of this series fully a week before the others. This plot, P4, had received a very high application, 306 lbs., of phosphoric acid per acre. On none of the other plots of this experiment (on which phosphoric acid was applied in amounts ranging from 75-225 lbs. to the acre) was there noted any such tendency to

early maturity. All the tobacco was harvested at the same time in spite of the early maturity of the plants in plots P4. In the following table will be found the amounts of phosphoric acid applied per acre and the resulting yields with Chapman's notes as to quality.

TABLE VII. EFFECTS OF DIFFERENT AMOUNTS OF PHOSPHORIC ACID ON THE YIELD AND QUALITY OF PRIMED HAVANA-1922.

Within the limits of this experiment, varying the amounts of phosphoric acid had little effect on the yield. Only one treatment, P4 (with an excessively high application of phosphoric acid) showed any appreciable effect and this was on the quality rather than on the yield. The tobacco from P4 was inferior and showed many double colors, probably due to the fact that it was harvested over-ripe. The colors of P2 were inclined to be more green than the other plots but the tobacco was of fine quality otherwise.

In 1923 tests of plots P1-4 and replicates were conducted just as in 1922, the location of the plots and treatment being the same. The unfavorable effect of high application of P2O, was apparently the same as in 1922 but the notes are scanty. The tobacco was primed. Records were taken from sample hands of leaves as in the case of the nitrogen series. Burn tests were made. Table VIII includes all records taken.

These data do not indicate that there has been a reduction in yield from omission of all mineral carriers of P2O5. The fireholding capacity is strikingly best on the plot which received the least P2O5. In view of the fact that acid phosphate is believed to be detrimental to the burn of the leaf, it is puzzling to find that P1 which received acid phosphate had a longer fire-holding capacity than P3 or P4 which received no acid phosphate, and that P1 also had the highest yield.

During 1924 no wide differences in growth and maturity were observed. Plot P2 might possibly have been slower and Plot P4 more rapid in developing buds and flowers than the other plots of this series.

The tobacco was taken down, stripped, sorted and weighed at the same time as the nitrogen plots. The sorting results are given in Table IX.