Abstract

The Libros del Saber de Astronomia is a compilation of various Arabic astronomical works translated into Castilian in the second half of the thirteenth century, under the direction of King Alfonso X of Spain. A section describing a mercury clock has been suggested to be of particular significance in view of the likely invention of the mechanical clock around this period, so a new translation into modern technical English has been prepared. The clock is shown to consist essentially of an annulus of mercury-filled chambers separated by narrow partitions, each partition being perforated by a very small hole. This annulus is attached concentrically to a pulley wheel around which a cord is wrapped, and a fixed weight hung upon the cord. The torque raises the mercury on the opposite side, but as the resulting hydrostatic head then causes it to percolate through the holes connecting the chambers the wheel slowly rotates to maintain its equilibrium. The weight therefore gradually descends in a controlled manner. It is claimed that, by suitable adjustment of the driving weight, the wheel could be made to turn once in 4 hours. A 6:1 reduction gear therefore enabled the rete of an astrolabe to be rotated once in 24 hours, modelling the real sky so that the time could be read off directly against the plate.A working model has been constructed on these principles, but utilizing modern materials to contain the mercury. It does work, but the range of rotation periods is quite limited and the reproducibility poor . It is shown that this is a consequence of the compounding of errors in what amounts to a ring of ‘mercury glasses’ joined end-to-end, for a single ‘mercury glass’ has a standard deviation of ±1.7 per cent and five or six are in use at any one time. Errors and omissions in the Castilian account suggest that the translator had never seen the instrument they attempt to describe.The mercury clock of the Libros is not isochronous, and its constant unidirectional rotation contrasts with the oscillators that are the hallmark of all advanced timekeepers. It is concluded that, although an interesting experiment, it had no influence on the evolution of clockwork in the western world