is not sufficiently distinct; because these other arts require to be first cleared away. Let us proceed, then, by regular steps:--There are causal or principal, and co-operative or subordinate arts. To the causal class belong the arts of washing and mending, of carding and spinning the threads, and the other arts of working in wool; these are chiefly of two kinds, falling under the two great categories of composition and division. Carding is of the latter sort. But our concern is chiefly with that part of the art of wool-working which composes, and of which one kind twists and the other interlaces the threads, whether the firmer texture of the warp or the looser texture of the woof. These are adapted to each other, and the orderly composition of them forms a woollen garment. And the art which presides over these operations is the art of weaving. But why did we go through this circuitous process, instead of saying at once that weaving is the art of entwining the warp and the woof? In order that our labour may not seem to be lost, I must explain the whole nature of excess and defect. There are two arts of measuring--one is concerned with relative size, and the other has reference to a mean or standard of what is meet. The difference between good and evil is the difference between a mean or measure and excess or defect. All things require to be compared, not only with one another, but with the mean, without which there would be no beauty and no art, whether the art of the statesman or the art of weaving or any other; for all the arts guard against excess or defect, which are real evils. This we must endeavour to show, if the arts are to exist; and the proof of this will be a harder piece of work than the demonstration of the existence of not-being which we proved in our discussion about the Sophist. At present I am content with the indirect proof that the existence of such a standard is necessary to the existence of the arts. The standard or measure, which we are now only applying to the arts, may be some day required with a view to the demonstration of absolute truth. We may now divide this art of measurement into two parts; placing in the one part all the arts which measure the relative size or number of objects, and in the other all those which depend upon a mean or standard. Many accomplished men say that the art of measurement has to do with all things, but these persons, although in this notion of theirs they may very likely be right, are apt to fail in seeing the differences of classes--they jumble together in one the 'more' and the 'too much,' which are very different things. Whereas the right way is to find the differences of classes, and to comprehend the things which have any affinity under the same class.