I have often been accused of slighting this quality in pictures; the fact being that I have avoided it only because I considered it too great and wonderful for me to deal with. The longer I thought, the more wonderful it always seemed; and it is, to myself personally, the quality, above all others, which gives me delight in pictures. Many others I admire, or respect; but this one I rejoice in. Expression, sentiment, truth to nature, are essential; but all these are not enough. I never care to look at a picture again, if it be ill composed; and if well composed I can hardly leave off looking at it. "Well composed." Does that mean according to rule? No. Precisely the contrary. Composed as only the man who did it could have done it; composed as no other picture is, or was, or ever can be again. Every great work stands alone. § 3. Yet there are certain elementary laws of arrangement traceable a little way; a few of these only I shall note, not caring to pursue the subject far in this work, so intricate it becomes even in its first elements: nor could it be treated with any approach to completeness, unless I were to give many and elaborate outlines of large pictures. I have a vague hope of entering on such a task, some future day. Meantime I shall only indicate the place which technical composition should hold in our scheme. And, first, let us understand what composition is, and how far it is required. § 4. Composition may be best defined as the help of everything in the picture by everything else. I wish the reader to dwell a little on this word "Help." It is a grave one. In substance which we call "inanimate," as of clouds, or stones, their atoms may cohere to each other, or consist with each other, but they do not help each other. The removal of one part does not injure the rest. But in a plant, the taking away of any one part does injure the rest. Hurt or remove any portion of the sap, bark, or pith, the rest is injured. If any part enters into a state in which it no more assists the rest, and has thus become "helpless," we call it also "dead." The power which causes the several portions of the plant to help each other, we call life. Much more is this so in an animal. We may take away the branch of a tree without much harm to it; but not the animal's limb. Thus, intensity of life is also intensity of helpfulness--completeness of depending of each part on all the rest. The ceasing of this help is what we call corruption; and in proportion to the perfectness of the help, is the dreadfulness of the loss. The more intense the life has been, the more terrible is its corruption.