This operation finished, the operator has years of anxiety before him ere the fruits of his labours will prove themselves good or bad. Notwithstanding, he will watch every stage from the present time till then as a faithful mother would her tender offspring. If, through the course of growth, any of these fruits should appear deformed in any way, remove them, as in all probability the seeds may not prove so good as those of their larger or better brethren. And further, I would recommend every one of them to receive support from the time they take their last swelling until they arrive at maturity, as I find from experience that all fruits, no matter of what sorts, swell from one-fourth to one-half larger when supported than if left to hang from the branch; and I am further convinced that the seeds also must share in the benefit, and consequently give stronger and healthier seedlings.

The fruit being ripe and gathered, remove the seeds with care. Various ways and times for sowing have been recommended, and each and all, I believe, has been followed with less or more success. Some have recommended the placing of the seeds in a phial amongst sand for one or two months, afterwards taking them out into paper and allowing them to remain there till the following autumn, when they should be sown - M. de Jonghe giving as his reason for this course that, "after remaining in this situation, . . . the skins of the seeds will acquire a consistence, and the kernel a firmness, that will enable it to vegetate with greater vigour when committed to the soil" By this means, however, a year is lost, and a considerable amount of labour also, compared to the mode practised by that eminent pomolo-nist, Mr Rivers, and with what success it is not necessary for me to say. His method will be found in extenso in the ' Journal of the Horticultural Society,' vol ix. beginning at page 292: I will, however, give the substance of it here. When the Pear-eating season begins he prepares 9-inch pots, filling them with two-thirds of loam to one-third of rotten manure, with a little sand; these are placed upon tiles, away in an open space not likely to be the haunt of slugs.

The Pear with the desired seed being ready for use, he eats it, removes the pips one by one, and presses them down about an inch into the pot. Thereafter he writes a label bearing the parent names and date of sowing, covers the pot with a slate to exclude vermin, and the operation is finished. Fifteen pips he considers sufficient for a 9-inch pot. In this way he proceeds until the pots for receiving the seed out of doors become frozen, when he has a few in the greenhouse for a like purpose, which he continues to sow in whenever the fruit is ripe. He has no particular season for sowing, but does so from October till May or June, as the case may be; but those seeds sown after March seldom vegetate till the following year. As soon as the seedlings appear on the surface of the soil he removes the slates from the top of the pot. Seedlings thus sown may remain in the pots for the season, when they ought to be planted out into rows, say 2 feet apart, as soon as they are ripe in the wood, which ought to be about the beginning of October. The course to be followed with regard to pruning is similar to that of a young tree of any sort, details for which shall follow hereafter. In all probability it will be five or six years before these will bear fruit.

A year or two may be saved by removing a scion, if strong enough, the first year, and grafting it upon a stock of some age. Mr Rivers recommends them to be rind-grafted "upon old dwarf Pears without names, which may be bought at a cheap rate from the nursery".

Having thus treated of the raising of seedlings, I now come to speak of the stocks which are in use for grafting, which I have already named. The nature of the soil and climate has much to do with the stock which ought to be used. M. Du Breuil, in his extensive experiments with the chalky soils about Rouen, found that the Wild Pear suited best for grafting, and that the Quince was almost or altogether useless. Dr Lindley has said that for light and loamy soils the Quince was best, but for chalky soils the Wild Pear. I am further of opinion that in wet cold localities, such as we have in the west of Scotland, the Quince, although shorter-lived, is the best stock for the Pear as a standard, or I should rather say for pyramids. I have invariably noticed in such localities as I have indicated that standard Pears, when grafted upon the Wild Pear, in the course of time permeated the soil to such a depth that the roots got imbedded into the cold wet subsoil to such an extent that the result was canker, the dwarfing and cracking of the fruit, and latterly partial or entire failure of crops. By using the Quince many of these things are obviated.

It not being of such a rank habit of growth as a stock, the roots run and keep more on the surface, and therefore are not nearly so liable to canker and decay. And further, these bear fruit much earlier, are earlier ripe in the season, and the tree is handsomer and more ornamental. The fruit, no doubt, is scarcely so large, but what it wants in this is made up by flavour, colour, and appearance. The Quince of Portugal is the best variety for a stock, as it is the strongest and most robust of the family. The easiest and perhaps the best method to obtain stocks of the Quince is to layer the young shoots in winter, and allow no more than one bud to appear out of the soil. In all probability this bud will form a shoot about 3 feet the first season, which ought to be cut down in winter to about 6 inches, having one or two inches more removed before the graft is inserted. It is advisable to have as little of this stock above the ground as possible; and the reason for this is, that the Pear being of quicker growth than the Quince, in the course of a few years the scion would be much thicker than the stock, which, besides being unsightly, might be liable to be broken over by a storm. By having the union as near the soil as possible, both of these contingencies are guarded against.