united about the last of March, when the plants should be taken from the sphagnum and set in the body of the house to finish their growth." All who have had any experience in the propagation of trees by grafting in spring, are well aware of the flight of time, in the hurry of work that must be done in a few days or not at all. It is true that the season for grafting may be prolonged or extended a little by cutting the cions in winter and storing them in a cool, moist place, where they remain dormant after vegetation has started in the open air; but this does not affect the stocks, and these may come on slowly or rapidly, varying with the seasons, and the grafter must not only watch for opportune moments, but take his chances of striking the right time and conditions, in order to be successful. With such hard wood trees as the hickories it is better to be a little ahead of time than a few days too late, for frosts, and even quite a severe freeze, will not injure a dormant cion, and under the most favorable conditions the union between stock and cion is a rather slow process. For this reason I advise giving as much time as possible, and while I do not claim to having had any personal experience as a grafter, in the South, still I am inclined to think that grafting in the fall, and not later than December, would be preferable to later in winter or spring. By giving the cion and stock two or three months in which to form granulations and cohesion, there would be more certainty of success. Of course, I now refer to what is called crown grafting on the root below the surface of the ground, and when the cion is fixed in place with the usual ligatures of waxed paper or cloth, the soil is drawn back into place and the cion entirely covered with it, but very lightly over the terminal bud. [Illustration: FIG. 68. CROWN GRAFTING ON ROOTS OF THE HICKORY.] Where small stocks are not at hand, the roots of large trees may be severed and the end partly lifted towards the surface, as shown in Fig. 68, and when grafted, allowed to remain in position until the following season, and then taken up entire or with roots enough to insure future growth. The same or a similar process may be practiced to propagate a choice variety of the hickory, and a mere severing of the roots will insure the production of suckers from near the severed end, as shown in Fig. 69. [Illustration: FIG. 69. SPROUTS FROM SEVERED HICKORY ROOTS.]