low down as possible in the tower, and rising so high that the mouths of the bells, when at rest, are a few feet above the window sills; this is good both for the tower and the bells. For the tower, because it brings the weight of the bells low down, and, if the frame is properly made, reduces the vibration of the tower consequent on the ringing of the bells to a minimum. For the bells themselves because it places them where their sound has the best chance of getting out. A frame of this sort should only touch the tower at the corbels from which it springs, and should, if room permit, narrow considerably as it goes up, in order to prevent stupid people from steadying it by wedging it tight to the walls, and thereby seriously endangering the stability of the tower. The lead floor for the protection of the lower part of the tower from rain and melting snow, ought to be independent of the frame. It may be below it, or part-way up it, the timbers of the frame passing through properly contrived holes in the floor; but the floor ought never to be close under the mouths of the bells, or it will tend to deaden their sound.

There ought to be at least two floors between the bell chamber and the ringing chamber, or if there can be only one, it ought to be well packed.

Where a sanctus bell is wanted, the question will arise whether the architect ought to make any special provision for it, and if so, what? It is tolerably certain, that, in the fifteenth and sixteenth centuries, the sanctus bell was a small bell set apart for its special purpose, and, generally hanging away from the other bells; very often not in the steeple at all, but in a turret or gable over the

chancel arch. This arrangement seems still to be very suitable to country churches. But in towns the custom which exists in some places of using the largest bell for the sanctus, seems to be more in accordance with the spirit of Archbishop Pecham's constitution, in commenting upon which Lyndwode remarks, "Hæc pulsatio fieri debet de campanis illis quæ longius possent audiri.” Where this is to be done the architect must provide, by a squint, if necessary, that the sexton, when handling the bell rope, shall be able to see the altar.2

I

Musical chimes are an ornamental addition to the belongings of a church, to which no objection can be made. But I cannot agree with those who would have us adopt the Flemish carillons. Apart from the constant jarring of chords, which do not work exactly together, the feeble tinkling of a lot of little bells is far inferior in effect to the old English chime on eight or ten bells of reasonable size. The carillon at Boston is as satisfactory as any of the Belgian ones which I have heard, but I should be sorry to see any more such set up in England. The proposal to place one in St Paul's Cathedral is almost comic, seeing that the five ton bell on which the clock strikes, is often scarcely audible in the churchyard.

1 Quoted by Mr Edward Peacock in the Proceedings of the Society of Antiquaries of London. 2d series, vol. v. p. 27.

2 If the sackering bell is also used, this is scarcely necessary.

SECTION XXIII.

OF CLOCKS, SAND-GLASSES, AND SUN-DIALS.

THE custom of providing churches with some apparatus for measuring time, dates back far beyond the use of the clock, as we understand the word. It is indeed probable that the first clocks were produced in monasteries, for in them mechanism was most likely to be studied, and also there was need of some means of measuring time, not only by day, when the sun-dial might be used, but also at night, in order that the canonical hours might be properly kept. But, whatever be its origin, it is certain that the clock of wheel-work early became common in churches. In England until quite recent times almost every church had its clock. But at the present moment it is the exception for a new church to be provided with one; indeed, an architect is now not unfrequently called

1 Durandus (Lib. I., Cap. I., Sec. 35), writing in the latter half of the thirteenth century, speaks of the horologium as if it were part of the usual furniture of a church. But it is possible that it was some sort of waterclock or rather alarum, like the horologium from which the monks of St Edmundsbury fetched water to put out a fire in their church in 1198. (See Jocelin of Brakelond, Camden Society's edition, p. 78.) De Hcnecort in his sketch book (c. 1245), has drawn "li masons don orologe," which his editors believe to be intended for a wheel-clock, though there is nothing which shows it with certainty. Many large churches are known to have had wheel-clocks in the fourteenth century.

upon to build a whole church, for about the price of a good clock and bells, without which, little more than half a century ago, none would have been considered complete.

Church clocks are of two sorts, external and internal. We have to consider of each of these, whether it should now be introduced, and if so, what form should be given to it. The external clock has really nothing to do with the church or its services, but is set up for the benefit of the neighbourhood, in the steeple, as being the most convenient and conspicuous position which can be found for it, and also that the church bells may be used for it to strike upon. I suppose that at the present time there are few houses in England which do not possess a clock of some sort, and consequently the necessity for public clocks is not now so great as it once was. But they still have a use, if it be only to afford a common standard by which the private clocks may be regulated. I think, therefore, that in all villages and small towns, where there is no other public clock, one may very properly be provided in the steeple; but in large towns they are rather a nuisance than otherwise, from their frequently not keeping exact time. A little inaccuracy in one clock, which is the standard of a whole neighbourhood, is of little importance; but when there are twenty

1 I have myself heard the clocks of London striking midnight for more than five minutes. The clock of the Houses of Parliament has, however, done away with the practical inconvenience, for, being audible everywhere, it has superseded all others. Nevertheless, close to it are the clock of the abbey, and that of St Margaret's church, and, only a little further off, that of the Horse Guards with its horrible "ting tang."

standards the case is different. Town churches are therefore best without external clocks.

A clock being decided upon, we have next to determine the number, position, size, and architectural treatment of its dials; first, however, noting that there are cases in which the dial may be altogether omitted. I do not know that I should ever deliberately design a tower intended to contain a clock, without making some provision for a dial, though it is really of very little use. It is the striking which is of real value, and I cannot help thinking that the slight advantage of a dial is a poor compensation for the disfigurement often caused to an old tower by its introduction.

If there are dials, they must be placed where they are most likely to be visible to the largest number of persons. This will depend on the position of the church; if, as is often the case in villages, it stands a little apart from the houses, one dial facing them will be sufficient. If the church stands more amongst the other buildings, there may be two, three, or four dials, as the situation dictates. The best position for a dial is the usual one; namely, the stage immediately below the belfry. It is there sufficiently raised to be well visible. If placed higher it will appear insignificant, or else it must be made so large as to subordinate the tower to it. But a clock in a church steeple is, it should be remembered, only an accidental addition, not the main cause of its erection, as is often the case in the tower of a civil building. There is a very good formula used by clockmakers to determine the proper size of a dial. which is,