MICHIGAN EXPERIMENT STATION BULLETINS. 



l'ii-4 



III examiniDg the yields m detail, attention is called to the fact that each 

 row contains a nothing: plot. Observe that the yields of these nothing 

 plots decrease gradually from the tirst until the last. The soil in each row 

 seems to be very uniform, and the nothing plot of that row should be used 

 as a base of comparison for the plots of that row, and to compare the plots 

 of difPerent rows it should be done by comparing the gains of the plote 

 over the nothing plots of those rows. The following table shows the gain 

 of the fertilized plots over the nothing plots. 



Table No. 12. — Gain per acre in bushels of fertilizer plots over nothing plots. 



The action of the mulch was not as beneficial this season as in 1891. It 

 held the moisture when the ground was too wet already. The plots treated 

 with manure presented a very thrifty appearance during the whole season. 

 They seemed to rid themselves of the surplus water in June, and they 

 stood the drought well in August. The yield of plot XIII was not so large 

 as the appearance of the vines would indicate. 



The difference between the treatment of plots III and IV from that of 

 plots VII and VIII is that the former contain muriate of potash and 

 the latter sulphate of potash. The latter two average 2 4 bu. per acre 

 more than the former, which is not enough difference to form a definite 

 conclusion, but as the sulphate is cheaper than the muriate, its use would 

 be more economical. Plots VII and VIII differ from plots XI and XII 

 in regard to treatment, in that the former contained sulphate of ammonia, 

 and the latter nitrate of soda. The gain of the former two over the latter 

 two is 21.1 bu. per acre, which tends to prove the superiority of 

 sulphate of ammonia over nitrate of soda. Plot XX has the nitrate of 

 soda combined with the muriate of potash instead of the sulphate of 

 potash, as in plot XI (plot XX has the fertilizer applied under the seed 

 hence it can be compared only with plots that have the fertilizer applied 

 likewise). It has been previously shown that muriate and sulphate of 

 potash produce nearly similar results. If this is true plots XI and XX 

 should show the same gain, but there is 52.7 bu. per acre difference in 



25 



