Mr. Rankin says the Peruvian Incas wore, as a distinction, two plumes on the front of the head, similar to those represented in the portraits of Tamerlane. (See _Conquest by the Mogols, &c._, p. 175.) I have seen, among the Wyandots of Sandusky, heads which one might suppose had been the originals of the portraits given in his plate: turban made of gaudy-coloured silk, with two short thick feathers stuck upright in front; the one red, the other white tipped with blue, the great desideratum being to have them of different colours, as strongly contrasted as possible.
The Kalmucs, when they celebrate any great festival, always wear coloured owls' feathers in their caps, &c. (See _Strahlenburg_, 4to., p. 434.) The Dacotas also wear owls' feathers. (See Long's _Expedition to Rocky Mountains_, vol. i. p. 161.) The Usbeck Tartar chiefs wore (perhaps _do_ wear) plumes of herons' feathers in their turbans; and the herons' plume of the Ottoman sultan is only a remnant of the costume in which their ancestors descended from Central Asia.
Your correspondent MR. G. T. HOARE is rather bold in describing the case he does as a "very common error;" and I cannot agree with him that the façade of Sennacherib's Palace (Layard's 2nd book on _Nineveh_) is an instance of the kind. The theory that horizontal lines in the plane of the picture should converge to a point on the horizontal line right and left of the visual ray, is by no means new; in truth, every line according to this view must form the segment of a circle more or less, according to circumstances. Apply this principle to the vertical lines of a tower or lofty building, and every such structure must be represented diminished at the top, the vertical lines converging to a vanishing point in the sky.
Some years since, this theory was brought forward by Mr. Parsey, and the subject fully discussed at scientific meetings. There was much ingenuity in the arguments employed, but the illustrations were so unsatisfactory that the system has never gained ground. The principles of perspective are most ably exemplified in many well-known works, as they set forth very satisfactory modes of delineation. The limits of your periodical prevent a fuller correspondence on this subject, or I think it would not be difficult to { 379} satisfy MR. HOARE that there are great difficulties attending his proposition.