24 BULLETIN 712, U. S. DEPAETMENT OF AGEICULTURE. 



became infected many times as severely as those which received in 

 addition to the dormant spray the proper foliage spraying through 

 the active growing season, as shown in Tables I, II, and III. The 

 reason for this lack of effect of the dormant spray is to be found in its 

 inability to reach the overwintering mycelium in the buds. If the 

 perithecia alone were concerned there might be some beneficial effect 

 from such sprays, but since these are known to be unnecessary in 

 carrying over and disseminating the fungus little is to be gained by 

 their destruction. 



PRUNING EXPERIMENTS. 



Many growers have endeavored to control mildew by pruning out 

 the infected parts, both in the winter and while the infections are 

 active during the growing season. By the removal of infected twigs 

 the immediate effect secured is to eliminate a source of infection for 

 the remainder of the season; the permanent effect is to prevent the 

 overwintering of the fungus and the spread of the disease the follow- 

 ing year. Hence, if all mildewed shoots and infected buds could be 

 removed, the disease would doubtless be eradicated, and the j)ractice 

 of such pruning could be depended upon to hold the disease in check. 

 However, in cases of severe infection, eradication by this method 

 would involve not only the serious mutilation of the trees and often 

 the destruction of practically all the new growth, but also a great 

 amount of labor. In order to demonstrate the practicability and 

 efficiencj^ of this method the following experiments were undertaken. 



Six Jonathan trees were very carefully gone over before they came 

 into leaf, and every sign of mildew that could be detected was re- 

 moved. At this time the terminals infected the previous year and 

 bearing infected buds could be detected by their silvery gray color 

 and glistening appearance in the sunlight, their stunted growth, and 

 the reddish color and slender, elongated shajDC of their lateral buds, 

 which were also less advanced than normal healthy buds. 



After the active growing season began it was found that in spite 

 of all the care in pruning, these trees were liberally covered with in- 

 fected leaves coming from widely distributed buds on older branches 

 that had escaped detection earlier in the year. To have distinguished 

 these with certainty as infected at that time would have involved an 

 inspection of every single bud on the trees, a task manifestly imprac- 

 ticable. A'^'lien this result became apparent the experiment was con- 

 tinued to demonstrate the benefits to be derived from dormant prun- 

 ing as a supplemental aid to foliage spraying during the growing 

 season. The trees were sprayed at intervals of four weeks until the 

 middle of July, receiving in all four applications of lime-sulphur 

 solution diluted 1| to 50, the same as plat 3 in 1915. For comparison, 

 six unpruned Jonathan trees in the immediate vicinity of the pruned 



