CORNELL UNIVERSITY EXPERIMENT STATION BULLETINS. 267 



tion through two winters. It is apparent, therefore, that any fertilizer 

 which is not at once available to the plant, bnt which gives up its materials 

 comparatively late in the season, will maintain a vigorous growth and 

 probably delay fruitfulness. Coarse stable manures belong to this class. 

 It is some time before they become thoroughly decomposed and incorpor- 

 ated with the soil, and if applied heavily it is probable that they will give 

 unsatisfactory results. If the season were long enough to allow the plant 

 to live out its natural lifetime it is conceivable that the materials would 

 be gradually used and that the total productiveness of the plant would be 

 as great, if, in fact, not greater, than it would have been under a treatment 

 which caused it to bear heavily at an earlier period. Thoroughly decom- 

 posed manures ought to give quicker and therefore more satisfactory 

 results than coarse ones, and proper concentrated fertilizers might give 

 better results than either. It would be interesting to know if stable 

 manure applied in the fall — and which therefore becomes thoroughly 

 incorporated with the soil before spring — will give earlier fruitage than 

 similar manure applied in the spring. We tried this experiment the past 

 season, but the soil in the plots proved to be so heterogenous in character 

 that we have no confidence in the results. We made a very suggestive 

 test in this direction last year with nitrate of soda. Upon a certain area 

 the material was applied all at once early in the season (June 25), and 

 upon another equal area the same amount of nitrate of soda was applied 

 in four applications from June 25 to August 28. Up to about the first 

 of October, the yield from the first or single treatment area was 20 per 

 cent, greater than from the other; but when the last picking (October 5) 

 was added, the single treatment area fell some 8 per cent, behind. " This 

 means," as the bulletin states, "that the intermittent application of fertil- 

 izer in lot two was beginning to be felt late in the season, while the single 

 early application of the same amount of fertilizer gave quicker results. 

 Frost held ofp until the second week in October, so that it happened that 

 the intermittent fertilizing gave us the better result, but had frost come 

 the last of September, as it frequently does at Ithaca, it would have given 

 us the poorer result." This year the frost did come early (October 1) 

 and the intermittent fertilizing gave us the poorer result. The figures 

 which follow afford a most striking confirmation of the foregoing 

 remarks, and they will bear careful study. Four plots, containing 

 15 plants each, were under experiment. In the first plot, 8 lbs. 

 of nitrate of soda were applied at one time early in the season (June 

 20); in the second, the same amount was applied in four applications 

 from June 20 to July 27; in the third, this amount was applied at four 

 different times from June 20 to August 26, thereby extending the growing 

 season very greatly; the fourth plot had no fertilizer. The plants were set 

 in the field June 1, and they were all Ignotum. Table I gives the yields 

 until frost (October 1). Table II gives the total yield found by adding to 

 Table I all the fruits which remained on the plants after they had been 

 killed by the first frost. 



