I think that any cultivated observer who may dwell ever so briefly on the tree scenery of Great Britain, will admit that the contrasts of colour, weak and little varied as they generally are, present to him one of its most pleasing features. If, then, the slight existing variation of colour, restricted principally to the contrast between light and dark green, is admitted to be an element of beauty, may we not justly infer that we should gain something if we varied and increased the contrasts by the use of stronger and more distinct colours? I think, then, that I may fairly assume that, on a free and full consideration of this subject, it will be generally admitted that a greater variety of colour would be an improvement in the tree scenery of our English gardens, parks, and landscapes.

I have next to show that the object I seek is attainable. The arrangement of the colours of flowers in the flower-garden has of late years been worked out with wonderful skill and effect. What were our flower-gardens in this respect thirty years ago? I remember that results predicted then were considered impossible by the many, although they have been accomplished, and more than accomplished, long ago. Now, as far as I am aware, no one has yet applied the same principles in the arrangement of trees and shrubs with coloured leaves. I have been told that it cannot be done. But after a long study of the question and numerous experiments, I have come to a different conclusion, which I submit with all deference to those who think otherwise. I believe that here, as in the flower-garden, there only needs a beginning, and progress will be rapid and success certain.

Many years ago I formed a collection of pictorial trees and shrubs, and planted them closely together, with the view of watching their development and eradicating those kinds which might prove undesirable on more mature acquaintance. This plan I vigorously pursued, and now find myself in possession of a select list, which I believe is sufficient to carry out all that I shall advance.

In addition to the ordinary or prevailing green, I find that I have five colours, or tones of colour, with which to work - 1. Light green; 2. Dark green; 3. Reddish purple; 4. Yellow or golden; 5. White or silvery; - and these may be combined in a variety of ways. Dark-bluish green has a good effect when placed in contrast with light-yellowish green; white with dark green; reddish purple with light green; reddish purple with yellow; yellow with dark green. And these contrasts by no means exhaust our resources. I merely quote them from among a number of experiments which I have actually tried and found agreeable to my taste. I have, indeed, no intention here of laying down any precise or definite rules for the application of these principles; I aim at no more than to show that the materials in colour exist, leaving their combination to be dealt with by the ingenuity and industry of a cultivated taste. It would be chimerical to suppose, unfair to expect, that any person taking this subject in hand without previous study, or without the fullest acquaintance with the materials which exist - some of them newly introduced would at once realise any great measure of success.

To such an individual the scheme would probably appear Utopian. He might, by a momentary effort, call up in his mind the short list of old and familiar trees with purple, white, and yellow leaves - the purple Beech, the white Poplar, the variegated Turkey Oak, and some few others still among the most valuable - but so few in number that he would dismiss the subject as impracticable. But unless familiar with the black and yellow Oaks, the yellow Elm, Acacia, and Alder, the white-leaved Acer Ne-gundo, and the many beautiful Maples recently introduced from Japan - the host of richly variegated trees only now becoming plentiful, - in a word, unless familiarly acquainted with the latest introductions of this kind, he would, I submit, be drawing his conclusions from incomplete information.

In order to bring my views to a practical test, I have here a diagram, in which I have merely sought to obtain the identical colours existing among trees and shrubs, and must refer you to the specimens exhibited to show that these colours really exist. This diagram, hastily executed, will perhaps also give some idea of the effect of the arrangement of the colours which we possess. The light green here is supposed to represent the Larch, the dark green the Yew, the reddish purple the purple Beech, the yellow the golden Oak and the white variegated Acer Negundo. There is also introduced here the ordinary green of nature, which may fairly be taken as the groundwork of our operations.

Now I am well aware, and would not ignore the fact, that the colours of the leaves of trees are influenced in some degree by cultivation and soil; but this does not affect our argument, as in the majority of cases they still bear the same relation to each other.

I have now to offer a few general remarks, with brief examples in support of the views which I have advanced. Let us remember that we are working with pictorial trees for pictorial effect. We may have spring pictures, summer pictures, autumn pictures, and permanent pictures. Summer and permanent pictures are the most valuable because of their greater durability. Specimens of these are before you, and a list of their names will be given at the end of this paper. The materials for spring and autumn pictures can only be shown in spring and autumn. The varying tints of the unfolding leaves of some trees in spring, and the glowing colours of the leaves of other trees in autumn, must be familiar to all observers; and these trees are beautiful in their seasons, whether regarded individually or in combination. But they are transitory. The varied and telling colours of spring, ordinarily, quickly subside into the universal green; and the bright leaves of autumn fall speedily before the frost and gales of that season. Yet both are desirable. The warm red and yellow tints of the unfolding leaves are peculiarly cheering in the colds days of early spring, and should be introduced freely when planting.