Page images
PDF
EPUB

.od of variable length, turning on a pivot, must be stretched all round from time to time upon a moveable centre, rising as the work proceeds, in order to regulate the internal out line. Such is the strength of this form that the highest kilns are seldom built more tha one brick thick, although this dimension would be altogether insufficient for a common wal of the same height. It is, probably, this principle which has conduced to the existence o the Round Towers of Ireland. That of Kilkenny, for example, 100 ft. in height, wa built on, or very near, the surface, for at 2 ft. below it, wood coffins with skeletons wer found partly under the walls, thus affording an unstable foundation.

POINTED ARCH VAULTING.

1499 v. We now proceed to enter into a view of the general forms of groining in pointe architecture, observing, by the way, that the groins at the arrises, up to the twelfth century were seldom moulded with more than a simple torus or some fillets. In the twelft century, however, the torus is doubled, and th doubling parted by a fillet. Towards the end o the twelfth century, three tori often occur; and at the beginning of the thirteenth, the moulder arrises become similar to the moulded archivolt of the arches, both in their form and arrangement In France, until the middle of the fifteenth cen tury, the arrises of the groins only were moulded but in this country the practice took place much earlier, for, instead of simple groining, the introduction of a number of subdivisions in the soffit of arches had become common. In fig. 590e i given a plan of the soffit of a vault of this kind in which A is an arc doubleau (by which is un derstood an arc supposited below another at cer tain intervals, and concentric with the latter); I is an upper arch, called by the French antiqua ries formeret; C, the wall arch, or formeret di mur; D is a diagonal rib, or croisée d'ogire; E intermediate rib or tierceron; FF, summit rib or liernes; G, the key or boss, clef de voute. Willis has used the French terms here given, and as we have no simple terms to expres them in English, it may be convenient to adopt the practice. 1499w. The ribs formed by the intersections of the groins perform the office of supporting

Fig. 590e.

Mr

the vaulting which lies upon them, they in their tri being borne by the pillars. Thus, in the simpl groin (fig. 590f), the arches AA, and diagona rib C, carry the vaulting BB, a rebate being formed at the lower part of the ribs on which the vaulting lies. This figure exhibits the simples form of groining in any species of vaulting, th intersecting arches being of equal height Th contrivance in its earliest state was ingenious, and the study attractive, and we cannot be surprise at Dr. Robison observing, in respect of the artist of the thirteenth and two following centuries, tha "an art so multifarious, and so much out of th road of ordinary thought, could not but becom an object of fond study to the architects mos eminent for ingenuity and invention: becoming thus the dupes of their own ingenuity, they wer fond of displaying it where not necessary." Thi observation would be fully verified had we room fo showing the reader the infinite number of device that ingenuity has created: he will, however, fron the few elementary ones that we do give, be enabled to see the germs of countless others 1499r. Ware, in his Tracts on l'aults and Bridges, 1822-a work which, notwithstand ing the quaint method in which the subject is treated, contains extremely valuable matte -has made some remarks which we must introduce at length, or justice would not b done to them. "In the vaulting," he says, "of the aisles of Durham and Canterbur cathedrals are to be observed the arcs doubleaux and groined ribs in round-headed vault: In the naves of the same buildings is the same character of vaulting, except that the arc of the vault is pointed. Some vaults of this kind are to be distinguished from others by th

Fig. 590.

positing of the stones of the vault between the ribs, which, instead of being parallel to each side of the plan, as in Roman groined vaults, take a mean direction between the groined rib and the ribs of the arches over the sides; whence they meet at the vertex at an acute angle, and are received by stones running along the vertex, cut in the form of a ratchet. The advantage of this method consists in requiring less centering, and originates in the position of the ribs at the springing." "From these beginnings vaulting began to assume those practical advantages which the joint adaptation of the pointed arch and ribs was calculated to produce." The second step differed from the first, inasmuch as at the vertex of the vault a continued keystone or ridge projects below the surface of the vault, and forms a feature similar to the ribs. But here it was necessary that the ridge should be a stone ot great length, or having artificially that property, because its suspension by a thinner vault than itself would be unsafe, unless assisted by the rib arches over the diagonals and side, a distance equal to half the width of the vault. To obviate this objection, other ribs were introduced at intervals, which may be conceived to be groined ribs over various oblongs, one side continually decreasing. This practice had a further advantage, as the panels or vaults between the ribs might become proportionally thinner as the principal supports increased. It is now that the apparent magic hardiness of pointed vaulting and the high embowered roof began to display itself; from slender columns to stretch shades as broad as those of the oak's thick branches, and, in the levity of the panel to the rib, to imitate that of the leaf to the branch." "On comparing rib-pointed vaulting with Roman vaulting, it will be invariably found that the rib itself is thinner than the uniform thickness of the Roman vault under similar circumstances; and that the panel, which is the principal part of the vault in superficial quantity, sometimes does not exceed one ninth part of the rib in thickness. The Gothic architects, it has been expressively said, have given to stone an apparent flexibility equal to the most ductile metals, and have made it forget its nature, weaning it from its fondness to descend to the centre."

[blocks in formation]

1499y. In the second example (fig. 590g.), another rib, a b, is introduced, which on plan produces the form of a star of four points. The forms of these thus inserted ribs result from curves of the lines on the plan in the space to be vaulted.

As many radii are drawn

[graphic][subsumed][subsumed]

from the angles of the plan as there are ribs intended, until they mutually intersect each other. The curvatures of the ribs will be elongated as they recede from the primitive arch, till they reach the centre on the place where the groins cross, and where of course the elongated curve is a maximum. The ribs thus form, when they are of the same curvature portions of an inverted conoid.

14992. In the next example (fig. 590h.), the primitive arches are unequal in height, the arch A being higher than B. The plan remains the same as in that immediately preceding: but from the inequality of height, a d, c b, must be joined by curved lines, determined or one side by the point a, where e a intersects the longer arch. A curved summit rib, as well longitudinally as transversely, may occur with equal or unequal heights of primitive arches (as in fig. 590i.); but the stellar form on the plan still remains, though differently modified, with the same, or a less or greater, number of ribs on the plan (fig. 590k.). By truncating, as it were, the summit ribs, level or otherwise, with the tops of the primitive arches, and introducing on the plan a polygon or a circle touching quadrants inscribed in the square, we obtain, by means of the rising conoidal quadrants, figures which perform the office of a keystone. In this, as we have above observed, the construction of the work is totally different from rib vaulting, inasmuch as each course, in rising, supports the next, after the manner of a dome, and is not dependent on ribs for carrying the filling-in pieces. Hence the distinction between fanwork and radiating rib work so judiciously made by Mr. Willis.

Fig. 5901.

1499aa. The sixth example (fig. 5904) has primitive arches of different heights, forming an irregular star on plan, that is to say, the points are of different angles. The figure will scarcely need explanation after what has been already said in relation to the subject.

1499bb. A polygonal space may be vaulted in three different ways. First, by a central column serving for the reception of the ribs of the vault, the column or pillar performing in such case the office of a wall, as in the chapter-houses of Worcester, Salisbury, Wells, and Lincoln. This mode evidently admits of the largest space being covered, on account of the subdivision of the whole area by means of the central pillar. The second mode is by a pendent for the reception of the arches, as in the Lady Chapel at Caudebec, (given in the section MASONRY). This mode is necessarily restricted in practice to small spans, on account of the limits attached to the power of materials; albeit in theory its range is as extensive as the former. The last method is by at

[blocks in formation]

once vaulting the space from wall to wall, as in fig. 590m., like the vaulting to the kitchen of the monastery of Durham Cathedral, or. fig. 590n., similar to the chapter-house at York of which, the upper part being of wood, Ware quaintly observes, "The people of Yorkshire fondly admire and justly boast of their cathedral and chapter-house. The principle of vaulting at the chapter-house may be admired and imagined in stone; not so the vault o the nave; it is manifestly one of those sham productions which cheat where there is ne merit in deceiving." The principle, as Ware justly observes, is perfectly masonic, and might be easily carried out with stone ribs and panel stones, it being nothing more thar an extension of that exhibited in the third example of simple groining (fig. 590f.) above given; and the same remark applies to the Durham kitchen.

1499cc. We propose to offer explanations of the nature of the vaulting at King's College Chapel at Cambridge, and the silly story related by Walpole of Sir Christopher Wren Saying, "that if any man would show him where to place the first stone he would

engage to build another" (vault like it). divided into oblong severies, whose shorter It must be evident that the curves of the inverted quadrants must intersect each other previous to the whole quadrant of the circle being completed. Hence these intersections form a curved summit line lowest against the windows or smaller sides of the oblong. This summit line of the vaulting of the building in the direction of its length forms a series of curves, though from the angle under which it is seen it is scarcely perceptible. Mr. Ware says, "It is observable, in the construction of this vault, that the principle of using freestone for the ribs, and tufa for the panels, has not been followed; but the whole vault has been got out of the same description of stone, and with an uniform face, and the panels worked afterwards, and 10 duced to a tenuity hardly credible except from measurement. The artists of this building might be trusted in the decoration of a vault with what is now called tracery; they knew how to render it the chief support, and what was the superfluous stone to be taken away: every part has a place, not only proper, but necessary; and in the ribs which adorn the vault we may in vain look for false positions. This is the ocular music which affords universal pleasure."

The vault of the chapel in question is sides are placed longitudinally (fig. 5900.)

Fig. 5900

1499dd. We now return to the consideration of two more modes of simple vaulting. In England, the summit ribs of the vault are almost always found running longitudinally and transversely in the various examples. In Germany the summit ribs are more frequently omitted than introduced. Thus in the example fig. 5904, the scheme is merely a square diagonally placed within the severy, subdivided into four parts and connected with the basepoints of the groins by ribs not parallel to the alternate sides of the inserted square. This, however, sometimes occurs in English buildings, as in the monument of Archbishop Stratford, at Canterbury Cathedral; though in that the central portion is not domical. It is to be remarked that the intersecting arches are not of equal height, otherwise the arrangement could not occur.

1499ee. In the example fig. 590p, the arrangement completely assumes what Mr. Willis calls the stellar form. Here in the soffit a star of six points is the figure on which the projection depends, the points radiating from the angles of an hexagon, and thus forming a cluster of lozenges whose middle longitudinal sides produce another longitudinal lozenge to connect the centres of the pattern. The longitudinal arches are, as in the preceding figure, lower than the transverse arches. Mr. Willis says, "the principal distinction between these and our own fanvaulting is the substitution of lozenge-headed compartments in the fans, for the English horizontal transom rib. We have also lozenge-headed compartments in our early vaulting, but they are never so symmetrically arranged in stars throughout."

Fig. 590p.

1499ff. From the simple lines or principles above given, it is easy to perceive through what numberless ramifications of form they may be carried. Another form is that called hexpartite vaulting, where the ribs spring from the angles, and two others from a shaft placed in the middle of each long side, thus making sir divisions. This is a step beyond the quadripartite groining shown in fig. 590f. Examples of hexpartite vaulting are scarce in England, but it may be seen in the chapel of St. Blaise in Westminster Abbey, the choir of Canterbury Cathedral, and in many parts of Lincoln Minster.

1499gg. It would be difficult to find a system of vaulting more unlike any English example than that in Anjou generally, of which the Hospital at Angers is a fair specimen. It is always excessively domical in its sections, both longitudinal and transverse; and has eight ribs, the cells being filled in with stones exactly parallel with the centre or ridge of each cell: the ribs are edge-roll mouldings.

1499hh. Besides the books named above, Prof. Willis On Vaulting, and by T. Eagles, 1874, both read at the Royal Institute of British Architects, the Dictionnaire by Violletle-Duc, the Lectures by Sir G. G. Scott, R.A., and the paper by W. H. Wood, in Builder for 1883, xliv., 55, should be referred to. A very complete outline of the subject has been printed by Prof. Babcock, of the Cornell University, Ithaca, New York, for his

courses of lectures.

SECT. IX.

WALLS AND PIERS.

1500. The thickness which is to be assigned to walls and points of support, that their stability may be insured, depends on the weight they have to sustain, and on their forma tion with proper materials; still more on the proportion which their bases bear to their heights. The crushing of stone and brick, by mere superimposed weight, is of extremely rare occurrence in practice, even with soft stone and with bad bricks. The result of some few experiments that have been made as to the resistance of some of our bricks and stones. to a crushing force, by George Rennie, in 1818, are here subjoined. Some later experiments made by the Commissioners mentioned in Book II. chap. ii., and appended to their Report on Stone, &c., in 1839; with a few others; as well as some important trials made in 1864 by a committee of the Institute of British Architects, given in Transactions, 1863-64, are likewise added.

TABLE OF CRUSHING FORCE OF MATERIALS, BY GEORGE RENNIE (Phil. Trans. 1818).

[blocks in formation]

1501. The above experiments lose much of their practical value from our knowledge that the interior particles in granulated substances are protected from yielding by the lateral resistance of the exterior ones; but to what extent it is impossible to estimate, because so much depends on the internal structure of the body. We are, however, thus far informed, that, taking into account the weight with which a point of support is loaded, its thickness ought to be regulated in an inverse ratio to the crushing weight of the material employed. In Gothic structures we often see, for instance, in chapter houses

« PreviousContinue »