Dracaenas require strong heat and abundance of moisture. Various methods are in use in propagating them. Old plants frequently become too tall to be entirely satisfactory, as their habit of growth produces a long, straight stem, destitute of leaves except near the top. It may then be desirable to get the tops rooted in small pots, making lower plants, which are more attractive and more easily handled. One method is to saw a pot in halves, then bring the halves together, enclosing the stem a little below the base of the lower leaves. Bind the pot together with wire. It may rest on a platform, supported by stakes set in the soil of the larger pot, or be held in any convenient manner which will keep it securely in place. The pot is then filled with soil, placing a little moss in the bottom in the ordinary manner, and is kept well watered to induce roots to form. Slits may be made in the bark or an incision made part way through, and lengthwise of the stem, as a means of facilitating root growth.

On January 13 a plant of Dracma Goldieana, layered in this way October 21. was examined, and also one of D. terminalis, which was layered December 4 by bending on a piece of tin in the shape of a funnel and fastening it around the stem with wires and stakes. Neither of these gave any indication of rooting, while two cuttings of the latter made at the same time, December 4. were rooted nicely. These cuttings were made by cutting off the tops just far enough below the leaves for a sufficient stem, then set in 3-inch pots and placed in the propagating case. The older stems of the plants were cut into pieces from one to two inches in length, placed in a 6-inch pot of light soil and covered with sand. The shoots from these were also to be seen at this date.

December 11, two top cuttings were put in 3½-inch pots. One showed roots January 15, and both were nicely rooted January 27, with strong, healthy roots circling around the bottom of the pots. Several of the inner leaves of the plant of D. terminalis layered in the tin funnel from December 4 to January 13 had par-tially shriveled and died. At the latter date these leaves were cut away, the funnel removed, and the top cut off and simply set in the moss in the propagating case, without any soil whatever. January 27, only twelve days afterward, it was breaking for roots, and January 31, there were roots one-fourth to one-half an inch long. The same day, January 31, a cutting of D. Youngi was put in the moss in a part of the propagating case which receives somewhat stronger bottom heat. The bark at the base of this cutting was slit with a single cut of the knife on two sides, and on two other sides a narrow strip was taken out. February 7, just one week afterward, two roots had formed well up the side of the stem.

The bark at the base of the stem was breaking for roots between the cuts, but showed little or no signs of rooting directly from them, indicating that these were rather a disadvantage than otherwise.

February 12, a plant of D. braziliensis, which had been layered two months in a pot sawed in halves, was examined and showed no signs of rooting. One plant layered in this manner has produced roots, as can be seen by digging into the soil.

From these trials it would seem that cuttings are preferable to stem-rooting by layers, if a suitable place for propagating is at hand, it being much quicker and equally certain - not a plant or even a leaf having been lost from any of these cuttings, except from the one first layered in tin. Probably rooting would be more certain with an incision half way through the stem diminishing the flow of sap somewhat; this was not done in most of these cases. Partially girdling the stem just below the pot might produce the same effect. Keeping the soil about the old roots dry and that in the upper pot moist would also be likely to aid in the formation of roots there. By this treatment we have the top of our plant as handsome as before, not a leaf being lost in the operation, with the additional advantage of a low plant in a small pot, with young, vigorous and healthy roots.

The propagating case used in rooting such plants is a very useful and very simple affair. It consists of a box about 4 feet long, 3 feet wide and 2 feet deep, with tight board sides and two small sash of two lights each, like the ones often used for cellar windows, hinged on top. This is placed just over the pipes to get good bottom heat. Common roofing slates are used lor the bottom, and may be held in place by slats. The slate is covered with three or four inches of moist sphagnum- moss Such a case could be readily made by any one accustomed to using tools. In it the most fastidious tropical plants take root readily, either directly in the moss or in pots plunged in it. A little air is given under the sash. It is surprising to lift some of these cuttings with a handful of moss and find it filled with roots.

Cuttings of Nepenthes are considered among the most difficult plants to root, but in this case it is done readily. The cutting is stuck through the drainage hole in a small pot from the bottom, and the pot inverted in the moss. The fine black roots, which look something like coarse horse hairs form from the end of the cutting, inside the pot. When the plant is well rooted, the pot is broken to free the plant, which is then potted off in very coarse material. Pennsylvania. Fred. W. Card.