daily rate, both the
curves would be straight lines and would coincide with the axis of x.
The curve C D, known as the curve of variation of daily rate, will
generally be found to follow changes of day and night, and of
temperature, and the fluctuations of the barometer and hygrometer; it is
the curve which reveals the true character of the clock. Hence in
testing a clock two things have to be determined: first, the daily rate
of error, and second, the average variations from that daily rate, in
other words the _irregularities_ of going. To test a clock well six
months' or a year's trial is needed, and it is desirable to have it
subjected to considerable changes of temperature.

The bibliography of horology is very extensive. Among modern works
Lord Grimthorpe's _Rudimentary Treatise on Clocks_, _Watches and
Bells_ (8th edition, London, 1903) is perhaps the most convenient.
Many references to older literature will be found in Thomas Reid's
Treatise on Clock and Watchmaking (1849). (G.; H. H. C.)

_Decorative Aspects._--In art the clock occupies a position of
considerable distinction, and antique examples are prized and collected
as much for the decorative qualities of their cases as for the
excellence of their time-keeping. French and English cabinet-makers have
especially excelled, although in entirely different ways, in the making
of clock cases. The one aimed at comely utility, often made actually
beautiful by fit proportion and the employment of finely grained woods;
the other sought a bold and dazzling splendour in which ornament overlay
material. It was not in either country until the latter part of the 17th
century that the cabinet-maker's opportunity came. The bracket or
chamber clock gave comparatively little scope to the worker in wood--in
its earlier period, indeed, it was almost invariably encased in brass or
other metal; and it was not until the introduction of the long pendulum
swinging in a small space that it became customary to encase clocks in
decorative woodwork. The long or "grandfather" clock dates from about
the fourth quarter of the 17th century--what is, perhaps, the earliest
surviving English dated specimen is inscribed with the date 1681.
Originally it was a development of the dome-shaped bracket clock, and
in the older examples the characteristic dome or canopy is preserved.
The first time-keepers of this type had oaken cases--indeed oak was
never entirely abandoned; but when w