We must feel that a column, no matter how beautiful, is supporting something. A floor, always a plane surface, must not be tiled or decorated in any way to express relief. This would apparently destroy the essential constructive quality of a floor, viz., flatness. For the same reason, all shams, such as painted arches, pillars, etc., are not legitimate. As long as they do not actually exist, they are evidently not necessary to the construction, and have no purpose save an imaginary decorative one, and in the words of Owen Jones, construction must be decorated - not decoration constructed.

III. Scope of Ornament. - The scope of ornamental art is almost boundless. It is applied to objects large and small, adapted to the most various uses, constructed of the most different materials. As the ornamentation is always to be subordinate to the object, considerations regarding size, use, position, material, etc., must govern it. An ornament that would be admirable applied to one object, might be detestable if applied to another. A design cannot be made without reference to its future application.

First: The material must be considered. Heavy and hard materials, such as wood and stone, will not admit of as delicate curves and lines as textile fabrics, such as cotton and woolen goods, laces, etc.

Second: The manner in which the article is to be made, whether by weaving, cutting, carving, casting, etc.

Third: The position the object is to occupy. If elevated or otherwise remote from the eye, elaborate finish and minute detail are useless. Ornamental art, from time immemorial, has attained its greatest excellence and exercised its greatest influence in connection with architecture.

In fact, the study of ornament is inseparable from that of architecture. It is upon architectural forms that the greatest artists have in all ages expended their greatest efforts and skill, and in a treatise on historic ornament they are decidedly the most interesting and important object of study.

IV. Material of Ornament. - The two great sources of ornament are geometry and nature. The latter includes the former; for not only must natural forms, in order to be available as material for ornament, be first conventionalized, or reduced to regular, symmetrical, geometric outlines, but any and all designs, whether the unit of repetition be geometric or conventional, must be founded upon geometric construction. This refers to the regularity, repetition, and distribution of parts; so that every good design, if reduced to its principal lines of construction, would exhibit but a few geometric lines and inclosing spaces. Many designs are not only geometric in their basis or plan, but make use of geometric figures as the units or materials of design. Such designs, however, rank lower than those in which natural forms conventionalized are taken as the subjects of repetition; and as the ornament rises in the scale toward perfection, even the geometric basis becomes less and less apparent, and sinks into a decidedly subordinate position; so that in many of the most perfect specimens it can be traced only in a few leading lines of the composition.

Its presence, however, is necessary, and is the foundation, if not the most important element, of beauty in the design.

Relation Between Nature And Ornamental Art

While the natural world, including leaves, flowers, animals, etc., is the greatest source of ornament, it is generally the opinion of the best authorities, derived from the study of the best styles and by a consideration of the principles of fitness and propriety which underlie the entire physical and moral world, that natural forms in ornamental and decorative art should not be literally copied or imitated. That is the aim of painting, sculpture, and the other representative arts, where the object is to present something to the eye which will suggest at once the actual presence of the object. To produce that effect, the object, whether animal or vegetable, is represented as much as possible in the actual circumstances of its existence, surrounded by the necessary conditions of its well-being and growth. A frame is placed around it, to shut it off as much as possible from other surroundings, and thus help us delude ourselves that we are in the presence of the real thing, either as it would impress us through our senses or our imagination.

But in ornamental art the case is entirely different. As it is to be applied and consequently subordinated to something, and does not exist for itself, it would be impossible, except in very rare instances, to introduce in a design a natural object in a realistic manner and not violate some important law of its growth or the conditions of its well-being. For instance, to exactly repeat a certain rose, with all the accidents of its growth, many times in a carpet is not natural. Nature never repeats herself. Moreover, to tread on that which is supposed to suggest to us real roses is barbarous. It would really be outraging and distorting nature while pretending to be her faithful disciple and imitator.

We not only derive from nature the most important materials for our designs, but also the various modes of arranging this material. Various modes of repetition - radical, bilateral, etc. - were all probably suggested by some natural arrangement observed in flowers, leaves, etc. Of these different arrangements it is curious to note that the bilateral is more characteristic of the higher forms of nature and the radiating of the lower. The leading principles of ornament - symmetry, proportion, rhythm, contrast, unity, variety, repose, etc. - are all exemplified in natural forms. The latter have also suggested many of the most important architectural forms. The Gothic cathedral, with its clustered columns branching and forming pointed arches overhead, was probably suggested by a grove of trees with overarching branches and boughs. The idea of the column was derived from the papyrus plant, a species of reed growing in the river Nile. The bud or flower suggested the capital of the column; the stalk, the shaft; and the bulbous root, the pedestal.