3.--Dolland's "atmospheric recorder": 1, siphon and
float barometer; 2, balance (?) thermometer; 3, hygrometer; 4,
electrometer; 5, float rain gauge; 6, float evaporimeter; 7,
suspended-weight wind force indicator; 8, wind direction indicator; 9,
clock; 10, receivers for rain gauge and evaporimeter. (From _Official
... Catalogue of the Great Exhibition, 1851,_ London, 1851, pt. 2).]

The 17th century was not entirely unprepared for the idea of such a
self-registering instrument. Water clocks and other devices in which
natural forces governed a pointer were known in antiquity, as were
counters of the type of the odometer. A water clock described in Italy
in 1524 was essentially an inversion of one of Hooke's rain gauges, that
in which a bucket was balanced against a string of bullets.[15] The
mechanical clock also had a considerable history in the 17th century,
and had long since been applied to the operations of figures through
cams, as was almost certainly the case with the punches in Hooke's
clock. Still, the combination of an instrument-actuated pointer with a
clock-actuated time-scale and a means of obtaining a permanent record
represent a group of innovations which certainly ranks among the
greatest in the history of instrumentation. It appears that we owe these
innovations to Wren and Hooke.

Hooke's clock contributed nothing to the systematization of
meteorological observation, and the last record of it appears to have
been a note on its "re-fitting" in 1690. Its complexity is sufficient
reason for its ephemeral history, but complexity in machine design was
the fashion of the time and Hooke may have intended no more than a
mechanistic _tour de force_. On the other hand, he may have recognized
the desideratum to which later meteorologists frequently returned--the
need for simultaneous observations of several instruments on the same
register. In any case, no instrument so comprehensive seems to have been
attempted again until the middle of the 19th century, when George
Dolland exhibited one at the Great Exhibition in London (see fig. 3).
The weather elements recorded by Dolland's instrument were the same as
those recorded by Hooke's, except that atmospheric electricity (unknown
in Hooke's time) was recorded and sunshine was not recorded. Striking
hammers were used by Dolland for some of the instruments and "ever
pointed pencils" for the others. Dolland's barometer was a wheel
instrument controlling a hammer. His