kept within a much smaller compass than by any

other method.

This mode of training is best adapted for dwarf trees, or such as do not form coarse thick branches, nor range extensively, for this purpose. Apple trees should be such as are grafted on paradise stocks; and to keep them with greater certainty within prescribed bounds, and at the same time healthy and prolific, a bed or border should be made of a light and dry porous soil, not too rich, of the depth of nine inches or a foot, on a substratum impervious both to the roots and to water, on the principles recommended for wall trees.

The plants intended for this mode of training should have from four to six branches of equal strength, growing from a stem of from three to six inches from the ground, and may be planted from eight feet to any distance apart.

When the branches are from four to six feet in length, let three long stakes be driven into the earth, from one foot and a half to two feet from the stem of the tree, at equal distances, so as to form a triangle; the stakes may be from four feet to any height; then let two strong hoops, of a diameter to fill the space between the stakes, be fixed horizontally, one about one foot and a half from the earth, and the other two feet above it; between the stakes, fix to the hoops two small laths or stakes at equal distances, the branches must then be brought down to an angle of from 40 to 60 degrees, and fixed to the laths or stakes at equal distances, and

each carried round the circle, rising in the same de gree like so many cork-screws entwining one into the other. Small circular iron bars not only form much neater fixtures, but more durable; and the best method of fixing is to drive small oak plugs into the earth, with a hole in the centre to receive the bars. Three bars of three-quarter inch, placed in the earth triangularly, and three of quarter or half inch, at equal distances between, fixed to the hoops, will form a proper support.

As it regards the trees, a circle of stakes fixed in the earth would answer as good a purpose as hoops; but the hoops admit of more room for working the earth round the stem; and if large wire be used, and painted, it will produce a neater and more elegant effect.

To elucidate this description, I have given sketches in plate 3. figures A. 1, 2, & 3. Suppose the centre dot in figure A. to be the plant, with its branches as first fastened, the three large dots to represent the stakes, and the six smaller the laths. Figure 1. shews the manner in which the hoops and laths are fixed; and figure 2. represents a plant when first fixed; and figure 3. the second or third year after.

It will be seen that this mode is in every respect conformable to the principles upon which the system of training wall trees and common Espaliers is founded; and consequently trees trained in this manner will be, in every respect, productive of the same effect.

The branches being continued in the same elevation, and parallel with each other, they will always form their strongest shoots at the point buds, and may be carried to any height without running into confusion, or crossing.

If any strong shoots are formed near the base, they may either be fixed and carried on parallel with the stems, or cut clean out.

The bearers, which will be natural studs or spurs, may remain as they are formed, within or without the circle, or on the upper or lower sides; and if they grow out far, or long, they may be tied in.

This method is extremely well calculated to train vines in the open ground, and to ripen grapes, as it will easily admit of being covered with glasses...

It is also well calculated to train fruit trees in pots for occasional or constant forcing in the hothouse.

When Espaliers or dwarf trees are found to grow too luxuriant, and to exceed their bounds, the better way to check their growth is to open the earth two or three feet round the stem, and cut through one or more of those roots that grow the strongest, and that run deepest into the soil; by these means, the form and regularity of the tree will be altered, as it would be by shortening and cutting out the branches.

If trees are found to throw out too much wood towards the stem or base, and the point buds or leading shoots do not grow sufficiently strong to

carry on the horizontal branches, these must be raised to a greater elevation.

The growth of these trees is determined by the elevation or depression of the branches, the same as that of wall trees.

ON STANDARD FRUIT TREES.

THE

HE system in general practice, of raising and training standard trees, is as imperfect and deficient, particularly in shaping or forming their heads and its consequences, as that of wall trees, and is as capable of being improved.

Hitt has recommended the shape best adapted to every desirable purpose, but why his method has not been adopted, I am at a loss to guess, unless it is from his not having described a more simple mode of commencing the formation of his trees, or from the same cause that his wall trees failed, viz. his not having sufficiently attended to original and natural principles, in his directions.

Many of the objects described by different authors, as desirable, but difficult of attainment, are brought easily within the reach of every one, by the mode I shall explain.

Hitt recommends that the shape or figure should be conical, like the natural growth of the fir tree; and in this manner, almost all fruit trees are naturally inclined to grow to a certain height; but Hitt depresses the horizontal branches too much,