This section is from the book "The Gardener V1", by William Thomson. Also available from Amazon: The New Organic Grower: A Master's Manual of Tools and Techniques for the Home and Market Gardener.
Perhaps this is as good a place as any to say a word on the best method of securing trees to walls. The old-fashioned way, which is still extensively practised, is to use shreds of strong tweed or other cloth, half-an-inch or more broad, and 2, 3, or 4 inches long, as need may dictate, and cast-iron wall-nails. So far as the trees are concerned the plan is a very good one, for the branches rest easily in the shreds without much chance of having the bark damaged; but it is a bad one for the walls, for the constant driving in and pulling out of the nails very soon loosens the mortar, thus letting in moisture, than which there is no greater destroyer of walls. The holes thus made also afford a safe retreat for many troublesome insects. A better way is to have the walls furnished with wire-work trellises. These are easily put up, and we need only refer to the illustrated advertisements, where the system is explained, while all the materials necessary are detailed. It may be as well to state, however, that the wires should be laid as closely against the walls as possible, for it is found that, when they are placed some distance from it, the trees do not receive the full benefit of the wall's shelter.
When the sun shines the walls are warmed by it, and they in turn raise the temperature of the air in close proximity to them, and the trees, which lie close to the walls, are benefited by the higher temperature. When, however, the trees are fixed to wires which are some inches from the wall, a current of air passes between the wall and the trees, and much of the benefit which the walls afford is thus neutralised.
There is another thing to which we might profitably refer here, and that is the distances at which branches should be trained. Hard and fast rules are generally laid down for this, but we unhesitatingly say that this is wrong. A recent writer in a contemporary says that the branches should be placed "so thinly that the leaves of one branch do not overlap those of another. If the spurs and leaves of a fruit-tree cover a space of 9 inches, the branches of that tree should not be less than 10 inches asunder. This is a safe rule to follow." We have no doubt that such is the practice of the writer of the sentences quoted, and we believe that the writer's practice is successful; but it makes some difference in the results sometimes when a practice, succcessful enough, perhaps, in the south of England, is applied four, five, or six hundred miles further north. The greatest mistake made by persons who have had no experience nor opportunities for observation is to overcrowd trees of all kinds. The sunshine ought to reach every part of a tree which bears fruit, and the more especially when the sun-rays are few and weak. This applies with double force to trees on walls. They are planted there partly for the sake of the shelter which the walls give, partly for the heat which the walls radiate.
If, however, the branches of the trees are so thickly trained that the wall is wholly covered with leaves, the wall can radiate little heat. To do this it must first be warmed, and for this purpose part of the wall should be purposely left hare. In the sunny south the bare portion might sometimes get damagingly hot; in the cold north a portion of bare wall is often necessary in order to raise the temperature a little.
Not far from where we write are two Easter Beurre Pear-trees and two Jefferson's Plums, planted at the same time and under precisely the same conditions every way, and all fan-trained; but the branches of one Pear and one Plum are 15 inches apart (which is generally considered ample) and the others are 2 feet. For what reason they were thus trained I know not, and may never know, for the gardener who planted and trained them has gone to his rest, and no one who was ever under him seems to consider it as anything but just one of his "freaks," as they call his many experiments; but this I know, that the thinly-trained trees seldom fail to produce a good crop of very fine fruit, while the others are neither so fruitful, nor yet is the fruit so fine in quality, and it is always some days later. We have again and again seen similar results from the same cause; and we are convinced that the cause is the larger amount of bare and therefore warming wall-surface. In trying to overreach others, it often happens that we only overreach ourselves; and it is exactly the same with plants.
We have said a good deal on this point, but not more than necessary, for it is a stumbling-block to many, and we don't remember having seen the same view expressed before.