proportion will be agreeable in any two parts of a building. I add, that concordant notes are produced by wind-inftruments, which, as to proportion, appear not to have even the flightest refemblance to a building.

With refpect to the other notion, inftituting a comparison between proportion, in numbers and proportion in quantity, I urge, that number and quantity are so distinct from each other, as to afford no probability of any natural relation between them. Quantity is a real quality of every body; number is not a real quality, but merely an idea that arifes upon viewing a plurality of things in fucceffion. An arithmetical pro portion is agreeable in numbers; but have we from this any reafon to conclude that it must also be agreeable in quantity? At this rate, a geometrical proportion, and many others, ought alfo to be agreeable in both. A certain proportion may coincide in both; and among an endless variety of proportions, it would be wonderful, if there never fhould be a coincidence: one example is given of this coincidence, in the numbers 16, 24, and 36; but to be convinced that it is merely accidental, we need but reflect, that the fame proportions are not applicable to the external figure of a house, and far lefs to a column.

That we are framed by nature to relish propor tion as well as regularity, is indifputable; but that agreeable proportion, like concord in founds,

is confined to certain precife measures, is not warranted by experience: on the contrary, we learn from experience, that various proportions are equally agreeable; that proportion admits more and lefs; and that we are not fenfible of difproportion till the difference between the quantities compared become the most striking circumstance. Columns evidently admit different proportions, equally agreeable; and fo do houses, rooms, and other parts of a building. This opens an interefting reflection; which is, That the foregoing difference between concord and proportion, is an additional inftance of that admirable harmony which fubfifts among the feveral branches of the human frame: the ear is an accurate judge of founds, and of their smallest differences; and that concord in founds fhould be regulated by accurate measures, is perfectly well fuited to this accuracy of perception: the eye is more uncertain about the fize of a large object, than of one that is fmall; and in different fituations the fame object appears of different fizes. Delicacy of feeling therefore with respect to proportion in quantities, would be an useless quality; and it is much better ordered, that there fhould be fuch a latitude with respect to agreeable proportions, as to correfpond to the uncertainty of the eye with refpect to quantity..

But all the beauties of this fcene are not yet difplay'd; and it is too interefting to be paffed over in a curfory view. I proceed to obferve,

that

that to make the eye as delicate with refpect to proportion as the ear is with respect to concord, would not only be an ufelefs quality, but be the fource of continual pain and uneafinefs. 1 need go no farther for a proof than the very room I poffefs at prefent; for every step I take, varies to me, in appearance, the proportion of the length and breadth: at that rate, I fhould not be happy but in one precife fpot, where the proportion appears agreeable. Let me further obferve, that it would be fingular indeed, to find in the nature of man, any two principles in perpetual oppofition to each other: which would precisely be the cafe, if proportion were circumfcribed like concord; for it would exclude all but one of thofe proportions that utility requires in different buildings, and in difierent parts of the fame building.

1

It is ludicrous to obferve writers acknowledging the neceffity of accurate proportions, and yet differing widely about them. Laying afide reasoning and philofophy, one fact univerfally agreed on ought to have undeceived them, that the fame proportions which please in a model are not agreeable in a large building: a room 48 feet in length and 24 in breadth and height, is well proportioned; but a room 12 feet wide and high and 24 long, approaches to a gallery.

Perrault, in his comparison of the ancients and moderns *, is the only author who runs to the

* p. 94.

oppofite

opposite extreme; maintaining, that the different proportions affigned to each order of columns are arbitrary, and that the beauty of these proportions is entirely the effect of cuftom. This bewrays ignorance of human nature, which evidently delights in proportion, as well as in regularity, order, and propriety. But without any acquaintance with human nature, a single reflection might have convinced him of his error; that if these proportions had not originally been agreeable, they could not have been established by cuftom: if a thing be univerfal, it must be natural.

To illuftrate the present point, I shall add a few examples of the agreeableness of different proportions. In a fumptuous edifice, the capital rooms ought to be large, for otherwife they will not be proportioned to the size of the building: and for the fame reafon, a very large room is improper in a small house. But in things thus related, the mind requires not a precife or fingle proportion, rejecting all others; on the contrary, many different proportions are made equally welcome. It is only when a proportion becomes loose and diftant, that the agreeableness abates, and at last vanisheth. In all buildings accordingly, we find rooms of different proportions equally agreeable, even where the proportion is not influenced by utility. With refpect to the height of a room, the proportion it ought to bear to the length and breadth, is extremely arbitrary; and

it cannot be otherwife, confidering the uncertainty of the eye as to the height of a room, when it exceeds 17 or 18 feet. In columns again, even architects must confefs, that the proportion of height and thicknefs varies betwixt 8 diameters and 10, and that every proportion between these two extremes is agreeable. But this is not all. There must certainly be a further variation of proportion, depending on the fize of the column: a row of columns 10 feet high, and a row twice that height, require different proportions: the intercolumniations must alfo differ in proportion according to the height of the row.

Proportion of parts is not only itself a beauty, but is infeparably connected with a beauty of the highest relifh, that of concord or harmony; which will be plain from what follows. A room of which the parts are all finely adjusted to each other, ftrikes us with the beauty of proportion. It strikes us at the fame time with a pleasure far fuperior: the length, the breadth, the height, the windows, raife each of them feparately an emotion: these emotions are fimilar; and though faint when felt separately, they produce in conjunction the emotion of concord or harmony, which is extremely pleasant *. On the other hand, where the length of a room far exceeds the breadth, the mind comparing together parts fo intimately connected, immediately perceives

* Chap. 2. part 4.

a