The four staple colors for men's wear, are black, blue, brown, and olive. Other colors, such as drab, grey, mixed, etc., being so far as the principal garments go, what are termed "fancy colors," should be very cautiously used.

As was remarked above, black has the effect of diminishing size, but it has another more important effect, which is to test, in the severest way, the wearer's claims to a distinguished appearance. It is a very high compliment to any man to tell him that black becomes him, and it is probably owing to this property that black is chosen, par excellence, for evening or ball dress. Men, therefore, of average or ordinary pretensions to stylish contour, should bear this in mind, and, when such color is not indispensable, should be careful how far they depend on their own intrinsic dignity.

Blue, of almost any shade, becomes a light complexion, besides being an admirable set-off to black velvet, which can, in almost all cases, be judiciously used in the collar, in which case, a lighter shade of blue (also becoming such a complexion) can be worn without killing (as it is technically termed), the darker shade of the coat—the velvet harmonizing both.

Brown being what is termed a warm color, is eminently adapted for fall and winter wear—olive and dark green, for summer.

When Beau Brummel was asked what constituted a well-dressed man, he replied, "Good linen—plenty of it, and country washing." This, perhaps, is rather too primitive. The almost equally short opinion of the French critic is decidedly more comprehensive—"un homme bien coiffé, et bien chaussé, peut se présenter partout." Under any circumstances, however, it may be laid down as immutable, that the extremities are most important parts, when considered as objects for dress, and that a well appointed hat, faultlessly-fitting gloves, and immaculate boots, are three essentials to a well-dressed man, without which the otherwise best constituted dress will appear unfinished.

Besides the necessity for the greatest care required in the selection of colors, with regard to their harmonizing with each other, and their general adaptation to the complexion or contour of the wearer, there is another matter of the first importance, and this is, the cut. Of course, everything should be sacrificed to perfect ease, as any garment which pinches, or incommodes the wearer, will strongly militate against the easy deportment of even the most graceful, and tend to give a contracted and constrained appearance. Every garment, therefore, should leave the wearer perfectly free and uncontrolled in every motion; and, having set out with this proviso, the artiste may proceed to invest his work with all the minute and seemingly immaterial graces and touches, which, although scarcely to be remarked, still impart an air or character, which is unmistakable, and is expressed in the French word chique.

Wadding, or stuffing, should be avoided as much as possible. A little may be judiciously used to round off the more salient points of an angular figure, but when it is used for the purpose of creating an egregiously false impression of superior form, it is simply snobbish. Some one has called hypocrisy "the homage which vice pays to virtue." Wadding is the homage which snobbishness pays to symmetry!

A well-dressed man will never be the first to set a new fashion; he will allow others to hazard the innovation, and decline the questionable honor of being the first to advertise a novelty. Two lines of Pope (I believe), admirably illustrate the middle course:—

"Be not the first by whom the new is tried,
Nor yet the last by whom 'tis set aside."

Besides which he will find it far easier to become a critic than an author; and as there is sure to be a vast number of men who "greatly daring" dress, he will merely be at the trouble of discriminating which is worthy of selection or rejection; he will thus verify the old saw, that "fools make feasts and wise men eat thereof," and avoid, by means of his own knowledge of the becoming, the solecisms which are pretty certain to occur in a number of experiments.