tem of improvement, and your illuftration of its excellence, I will next confider your defence of the detail of his practice. If, as you fay, no man of taste can hesitate between the natural group of trees compofed of various growths, and a formal patch of firs (and, I will venture to add, of any other trees) which, as you well obferve, "too often disfigure a lawn under "the name of a clump"-why not strive to imitate those natural groups, by attending to the principle on which they please? The strong argument against Mr. Brown, and that which I ftated in my Effay,* is, that in the course of a long practice, and therefore with many opportunities of seeing their effects, he never made a clump like a natural group, though he did make many natural groups like clumps; I therefore may fairly conclude that he preferred

*Effay on the Picturefque, page 359.

the

the latter and as he never (as far as I have obferved,) connected one group with another, but always detached them as much as poffible, I may also infer that he studied diftinctness, not connection.

Now, unless I am totally wrong in all my notions, CONNECTION is the leading principle of your art, and it is the principle that has been, of all others, the most flagrantly and fyftematically violated. It is by means of this fyftem of making every thing diftinct and feparate, that Mr. Brown has been enabled to do fuch rapid and extenfive mifchief; and thence it is that he is fo much more an object of the painter's indignation than his strait-lined predeceffors. He was a mere gardener, but he chose to be a landscape-gardener, without knowing the first principles of a landscape: the confequences have been fuch as might be expected; for as nothing is fo eafily, fo quickly deftroyed as connection,

nection, so nothing is restored with greater difficulty, or by a more tedious process.

Two of the principal defects in the compofition of landscapes, whether real or painted, are the oppofite extremes of objects being too crouded, or too fcattered: your cenfure, therefore, of fingle trees dotted over the whole furface of a park, or any other ground, is perfectly juft. Such scattered trees are rendered much more disgusting by heavy cradle fences, and, unlefs in very good foils, they alfo (as you obferve) are generally starving. I can speak very strongly as to the bad confequence of this practice in every point of view, from its having been in too great a degree my own; and it is by no means the only infance in which I could offer my own former practice (for I do not perfevere in what I think wrong) as a warning to others.

There cannot be a doubt, that the most

certain

certain expedient for producing future beauty, is to prepare and fence the ground, and to fet more plants than are meant to remain; for the young plants must neither be stunted, browfed, nor ftarved. But where those maffes (as is ufually the case) are formed of trees of equal growths, and left close together in one thick lump, the variety they give to any ground scarcely deferves that name. The remedy I propofed* (after stating the defects of the ufual method) was to mix a large proportion of the lower growths in every plantation; this, in my opinion, would not only prevent their flat, heavy, uniform appearance, but would alfo furnish means for varying and foftening the abrupt lines of their outside boundaries, and correcting that folitary, infulated look which they fstill would have. The method of doing it which I fhould recommend, would be to

*Effay on the Picturefque, page 309.

take

As

take trees, both of the larger and fmaller growths, from the plantation itself, (after they are grown ftrong enough to resist animals) and to transplant them on the outfide of the fence; where a stiff formal outline is apt to remain, even when the fence itself has been taken away. thefe plants would be to be carried fo short a way, though large, they might be removed with safety; and would want no fence, but merely to be staked till they had taken root. Their effect would alfo be immediate; they would at once break, vary, and soften the hard line of the clump by partially concealing it, which trees alone would not effect; but by such a mixture of thorns, hollies, &c. with foreft trees, the most painter-like groups and thickets might be formed.

This feems to me the true ufe of planting trees and bushes detached from the larger maffes; and thus much it may be fufficient