When an unimaginative painter is about to draw a tree, (and we will suppose him, for better illustration of the point in question, to have good feeling and correct knowledge of the nature of trees,) he probably lays on his paper such a general form as he knows to be characteristic of the tree to be drawn, and such as he believes will fall in agreeably with the other masses of his picture, which we will suppose partly prepared. When this form is set down, he assuredly finds it has done something he did not intend it to do. It has mimicked some prominent line, or overpowered some necessary mass. He begins pruning and changing, and after several experiments, succeeds in obtaining a form which does no material mischief to any other. To this form he proceeds to attach a trunk, and having probably a received notion or rule (for the unimaginative painter never works without a principle) that tree trunks ought to lean first one way and then the other as they go up, and ought not to stand under the middle of the tree, he sketches a serpentine form of requisite propriety; when it has gone up far enough, that is till it begins to look disagreeably long, he will begin to ramify it, and if there be another tree in the picture with two large branches, he knows that this, by all laws of composition, ought to have three or four, or some different number; one because he knows that if three or four branches start from the same point they will look formal, therefore he makes them start from points one above another, and because equal distances are improper, therefore they shall start at unequal distances. When they are fairly started, he knows they must undulate or go backwards and forwards, which accordingly he makes them do at random; and because he knows that all forms ought to be contrasted, therefore he makes one bend down while the other three go up. The three that go up he knows must not go up without interfering with each other, and so he makes two of them cross. He thinks it also proper that there should be variety of character in them, so he makes the one that bends down graceful and flexible, and of the two that cross, he splinters one and makes a stump of it. He repeats the process among the more complicated minor boughs, until coming to the smallest, he thinks farther care unnecessary, but draws them freely, and by chance. Having to put on the foliage, he will make it flow properly in the direction of the tree's growth, he will make all the extremities graceful, but will be grievously plagued by finding them come all alike, and at last will be obliged to spoil a number of them altogether, in order to obtain opposition. They will not, however, be united in this their spoliation, but will remain uncomfortably separate and individually ill-tempered. He consoles himself by the reflection that it is unnatural for all of them to be equally perfect. Sec. 12. Laws of art, the safeguard of the unimaginative.