ing out the plant, they should be placed in a cold frame and still kept close. After being a week in this situation they may be potted into larger pots. This is a very interesting method of propagation, and the plants made in this manner form very pretty bushes of compact growth; it is applicable to all roses: even Moss Roses will strike root if treated as above: they require more patience, as they are longer in forming their roots than many, as are also the Provence. Care must be taken that the shoots before being formed into cuttings, are perfectly ripe: an invariable sign of their maturity is when the terminal bud is formed at the end of the shoot; this shows that they have made their first growth; to hasten this, the plants should be placed in the most sunny situation, so as to mature their shoots as early as possible.

Cuttings of Hybrid China Roses, Hybrid Bourbons, and of all the climbing roses, may be raised with facility by planting them in a shady border in September. They may be made about ten inches in length, two thirds of which should be planted in the soil: in fact, they can scarcely be planted too deep: one, or at most two, buds above the surface will be enough; on these buds the leaves must be left untouched. These will be fit for planting out the following autumn.

PROPAGATION BY BUDDING.

The

A

This seems at present, owing to the strong wish manifested by the present generation to do every thing quickly, to be the favourite mode of propagation. A summer rose from a cutting requires at least two seasons to form a blooming plant. A layer is occasionally very capricious, and very loth to make roots; indeed, of some varieties, particularly of Rosa alba, they will not by any means be induced to form roots when layered, and are very difficult even to be propagated by cuttings from the forcing-house; but these become perfectly docile and manageable when budded, in one season only, forming large and handsome plants. operation of budding is difficult to describe. longitudinal cut, not so deep as to cut into the wood but merely through the bark should be made in the clear part of the shoot; thus making the diagonal cut at the top of the incision. I differ from most of those who have given directions for budding, as they make the incision thus, T: my practice has arisen from the frequent inconvenience sustained by shoots from standard stocks being broken off by the wind, when the cut is made at right angles: with the diagonal incision an accident rarely happens: the bark on both sides this incision must be opened with the flat handle peculiar to the budding-knife, and the bud

inserted: the slice of bark taken off the shoot with the bud in the centre should not be more than an half-an-inch in length; the incision being made of the same length: this is the length used by experienced budders, who pride themselves upon performing the operation in the neatest manner possible. When the bud is inserted, cut off with your knife (which should be very sharp) a piece from the upper part of the plate, i. e. the piece of bark with the bud attached, so that it fits closely to the diagonal cut at top; then bind it up firmly with cotton twist, such as the tallow-chandlers use for the wicks of candles; the finest quality is best: this is the most eligible binding known and far preferable to matting or worsted. Many writers recommend the wood to be left in the plate: in cases where the bud is unripe this may be very well; but, as a general rule, always remove it. Take buds that are mature, and by placing the thumb-nail at the top of the plate, peel cleanly the wood from the bark: if a remnant of wood is left on or near the eye of the bud, let it remain; it will do no harm; but if attempted to be removed, the eye is liable to be bruised and injured. Budding may be commenced in June, and performed as late as the second week in September; if done in June, the only shoots fit to take buds from are those that have shed their bloom: on these alone the buds are mature. I have occasionally known them to suc

ceed in October.

After August it is at the best uncertain, as the success of the operation entirely depends upon the state of the weather. In taking the wood from the bark, it will seem occasionally as if the eye or root of the bud is dragged out; it will then appear hollow: this only appears so, and is not of the least consequence, at least with roses, as those apparently hollow buds take as readily as those with the eye prominent.

PROPAGATION BY GRAFTING.

This may be performed in the forcing-house in January, and in the open air in February and March. There are many modes of grafting: those most eligible for roses are the common "whip-grafting," using clay as a covering, and "cleft-grafting," using wax or pitch: the former is generally the most successful; and if the stocks are potted a year before being used, strong blooming plants of the perpetual roses may be made in three months.

A neighbouring amateur has been very fortunate in grafting roses, merely gathering his stocks from the hedges in January and February, and immediately grafting and potting them after the operation; in doing so covering the union of the graft firmly with mould, using no clay, so as to leave only three or four buds above the surface,

and placing them in a gentle hotbed, in a common garden-frame, keeping them very close. In this simple method of operating I have seen eighteen out of twenty grafts grow, but, owing to the stocks not being established in pots a year, as they ought to have been, these plants have not made strong and luxuriant shoots the first season. Stocks may be potted in October, if none can be had established in pots: these may be used in January or February with much success.

In whip-grafting of roses in pots it will be as well to omit the usual tongue by which in open air the graft is, as it were, hung on the stock; this tongueing weakens rose-grafts too much; as their shoots are generally pithy, a slice of bark with a very small portion of wood about 12 inch in length, taken from one side of the stock where the bark is clear and free from knots, is all that is required; then take part of a shoot about six inches in length, and pare its lower end down quite thin till it fits accurately on the place in length and breadth, from whence the slice of bark and wood from the stock was taken; bind it firmly with strong bass, which has been soaked in water, and then place clay over it, so as to leave no crack for the admission of air: presuming this graft to be in a pot, it may be plunged in sawdust or old tan, leaving two buds of the graft above the surface in a gentle hotbed, and kept close till it has put forth its shoots: when these are three inches in