446 ANNUAL REPORT OF THE Off. Doc. 



or high-grade siili)hate, depending upon which of the formulas is de- 

 sired, or in 100 or 200 pounds of low-grade sulphtae. Upon the rela- 

 tive values of these latter carriers, we have very little evidences as 



yet. 



The above amounts are intended as annual aj)i)lications for bearing 

 trees of medium age where most of the ground is to be covered. For 

 younger trees, they may profitably be reduced, approximately in 

 proportion to the amount of soil to be covered. On older trees or in 

 special conditions, the combination is expected to be varied as later 

 results direct. For example, in some instances, especially in con- 

 nection with sod, we have found that the amount of nitrogen is 

 apparently a little too low, while with tillage and leguminous cover- 

 crops it is likely to be somewhat higher than necessary. 



As to where fertilization is needed, this is more difficult to define, 

 and probably the only certain method of determining it is by actual 

 trial. These trials are very simple. Merely leave a typical portion of 

 the orchard untreated, lor three or more seasons, as a check on the 

 value of treatment, and carefullv mark and record the trees in at 

 least one of the groups. There are a number of points, however, that 

 will aid one in determining the relative size to make these groups, or 

 in other words, aid in deciding whether to leave most of the orchard 

 in the check or in the treated portion. 



In general, for two years at least, the check should be much the 

 larger in most young orchards or in any orchard that is doing well 

 through late August and September. The fact that the trees are 

 well loaded in a given year, however, is no sufficient reason for 

 omitting the fertilizer that year. In fact, that is one of the best 

 reasons and limes for applying a proper fertilizer rather liberally, in 

 order to prevent the total absence of a crop the following year and in 

 the long run tend to steady the annual production. 



In case of the reverse conditions — old orchards or those not re- 

 taining a thrifty look throughout the season or not growing and 

 bearing regularly and satisfactorily — it is best to reverse the pro- 

 cedure and fertilize most of the orchard, leaving only a small 

 block as a check. In all cases, however, we strongly advise the use 

 of a check until the real value of the treatment is thoroughly es- 

 tablished. Even then it is not desirable to omit either the check 

 or the treatment entirely. The less valuable one may be reduced 

 to a small space, but it should not be omitted entirely if the grower 

 cares much for his orchard. One of our experiments, for example, 

 showed practically no response until the fifth year, and then, when 

 the cropping strain began to appear, quite marked differences arose 

 in favor of the properly fertilized plots. 



The time of ap]»lication we also consider important, especially in 

 the case of the nitrates. While our evidence is by no means com- 

 plete on this point, yet we have some indications that nitrates applied 

 too early in the season may be wholly lost to the trees. Other evi- 

 dence leads us to believe that distinct harm may be done by making 

 nitrate applications too near the fruit-setting time, especially in 

 the case of peaches. We feel, therefore, that nitrates should be 

 applied not earlier than petal-fall in apples, and probably not later 

 than the first of July, though we have had some very good results 

 from applications made as late as July 8. Making the applications 



