“If the wood is tender, adventitious roots spring without difficulty from the interred part, and the methods already described suffice for the success of the layering; but woods of dense structure are more or less obstinate about taking root, and might remain in the ground indefinitely without yielding. In such cases our art must intervene, based on the plant’s manner of living. Let us recall the effect of a band drawn tightly about a stalk or trunk. Above this line of strangulation the descending sap accumulates more and more, since it can no longer continue on its course between the wood and the bark, this latter being compressed by the ligature. It accumulates and produces a ring-shaped swelling where the plant tries to discharge on the outside the superabundance of matter arrested in its passage. Let this protuberance be heaped about with fresh earth, and adventitious roots will speedily be developed to allow the sap to continue its descent. A tiny streamlet, running free, follows its channel without effort and without any undermining of obstacles. But if we obstruct its passage the accumulating body of water will gain power to open new vents for itself through the dam. Sap does likewise. Circulating freely in its natural channel, it is not diverted from its course by any allurements on its way; and unless the conditions present in wood and bark favor the growth of new roots, no sap will be expended for this purpose. But if its usual passage is barred, the sap devotes its energies to the formation of adventitious roots in order that it may, through them, resume its interrupted course. A like result follows if a ring of bark is removed from the buried part of the branch or shoot that we wish to take root. The arrested sap produces a ring-shaped swelling on the upper edge of the wound, and from this swelling spring roots. “Now let us apply these theoretical principles. If the wood is compact and for that reason rebels against the laws of simple layering, we will take a piece of wire and strangle (that is the word) the branch we are operating upon; that is to say, we will bind it tight, but without breaking the bark. The compression should be made just below a bud or eye, and about midway of the part that is to be underground. This process is called layering by strangulation. “Or again, still midway of the part to be bedded in the earth, and immediately under a bud, we cut the bark all around the branch without injuring the wood; a second incision is made a centimeter and a half lower down; then tearing off the strip of bark between the two circumcisions, we remove it all in one piece. This method is known as annular incision from the ring of bark thus taken away.