were as well covered as could be done, but some of the parts were difficult of access.

The work was continued in 1894, and after one application had been made all the knots were cut out. This was done April twenty-fifth, and the knots from each portion were counted so that the relative number of branches in each portion could be estimated and a basis for comparison might be made when the knots should again be cut. They had been allowed to remain on the tree during the preceding year, so that the chances for further infection should be as favorable as possible. At the time of taeir removal no marked difference could be seen in favor of either portion as regards the relative abundance of the knots, athough on the sprayed portion they may not have been quite so thick.

The Bordeaux mixture was again used in 1894. It was applied to the same portion treated in 1893 and upon the following dates: April ninth and twenty-fifth, May twenty-first, June sixth and twentyseventh, July tenth and twentieth and August first. This number of treatments gave the limbs and foliage a blue color, which they retained until the leaves fell to the ground, and at this writing, December third, the branches still show a decided coloration.

Figure 5 represents the thicket as it appeared November twentysixth, before the knots were cut out after the summer's growth. The illustration does not represent plainly the number of knots present, but it will be noticed that there are many more on the unsprayed than upon the sprayed part. The line between the two plots passes across the page from A to B. The knots were all cut out after the photograph was taken. The appearance of the thicket after the operation is shown in Figure 6. It will be noticed what a large proportion of small branches has been removed from the unsprayed part. When the value of these branches is considered, the seriousness of this loss can be realized. The sprayed portion lost but a very small number of such twigs and is still in very good condition. The table below shows. the numbers obtained in the spring and those removed in the fall. the term "new" knots are meant those which appeared to have been produced by a new infection; "old" knots are those which apparently grew from the mycelium which had not been removed when the knots were cut in the spring.

By

April 25, 1894, unsprayed portions, knots removed, 2,002; sprayed portions, knots removed, 1,155. November 26, 1894, unsprayed portion, total number removed, 3,529; new knots, 3,466; sprayed portion, total number removed, 240; new knots, 165.

The above figures are full of encouragement. They show, not that the black knot fungus has been held under absolute control (and what fungus disease is), but that it is susceptible to treatment. There was a large increase, nearly 57 per cent., in the number of knots cut in the fall from the unsprayed portion over those cut in the spring. Had the same proportionate increase taken place in the sprayed part, a supposition which must be allowed had no applications been made, the number of knots produced would have been within a fraction of 2,000. Consequently, the 165 new knots should be compared with 2,000, and not with 1,155, as shown by the table. It must also be considered that this reduction took place in the face of circumstances which had been made as favorable for the spreading of disease as possible. The sprayed portion had been covered with knots during the first year's treatments. It was practically in contact with a large number of diseased plants. And finally, we are still in the dark as to the best time of making the application, for we do not know with certainty when the infections take place, nor when the knot appears after infection has occurred. If this were known the matter would be considerably simplified, and still better results obtained. It would appear from the above that an orchard may be kept practically free from the knot if the trees are sprayed at the proper time. One or two treatments might be given before the blossoms open, and later applications probably need not be made oftener than is required to control the leaf-blight and the rotting of plums. If, in addition to this, the comparatively few knots which will undoubtedly appear be removed as soon as discovered, no grower need have much fear of this disease, even though his trees be considerably exposed to infection from neighboring trees or thickets. The treatments which the above* experiment seems to justify me in recommending are as follows:

First. During the latter part of March or early in April.
Second. When the buds are beginning to swell.

Third. During the latter part of May, or as soon as the new knots begin to show their velvety coating.

Fourth. About two weeks after the third.

Fifth and sixth. At intervals of two or three weeks after the fourth, but these two will probably not always be necessary.

*It is unfortunate that other experiments, begun at the same time and carried on in a manner similar to the above, have failed to give results, from the fact that no new knots formed upon either the sprayed or the unsprayed portions. The plum orchard of Geo. T. Powell, Ghent, N. Y., has been under treatment during two years, as well as other trees in the vicinity of Ithaca. All have failed to develop new knots. But the work will be continued, and it is hoped to make other reports in the future.