THE EVOLUTION OF THE THERMOMETER. 417 



as the Swedish and afterwards as the centigrade thermo- 

 meter, and was not used to any extent in France until after 

 the Revolution, was divided centesimally. The boiling 

 point of water was at first marked as o° and the freezing 

 point as ioo°, but, at the suggestion of Linnaeus, this 

 graduation was reversed, the temperature of melting ice 

 being taken as zero and that of boiling water as ioo°. 



Towards the middle of the seventeenth century, atten- 

 tion began to be directed to methods for measuring higher 

 temperatures than can be shown by ordinary thermometers, 

 and a paper in the Philosophical Transactions for 1747 

 (vol. xl., pp. 397 to 408) stated that Dr. Muschenbroek of 

 Utrecht had invented a " metalline" thermometer, in which 

 the expansion or contraction of rods of metal was caused to 

 show changes of temperature. The instrument, as improved 

 by Dr. Desaguliers, has been described in his Experimental 

 Philosophy (vol. i., p. 421, 1734). 



The next important improvement in this branch of ther- 

 mometry was the introduction of the Wedgwood pyrometer 

 [Phil. Trans., vol. lxxii., p. 305, 1782 ; and vol. Ixxiv., p. 

 358, 1784), in which the high temperatures of a furnace 

 were measured by the diminution in bulk of a block pre- 

 pared from a pure fire-clay under constant conditions as„ to 

 the amount of moisture used, and the pressure applied in 

 moulding. 



Thermometers for registering the highest and lowest 

 temperatures attained during the absence of the operator, 

 were introduced as soon as the ordinary thermometer at- 

 tained to the dignity of an instrument of precision, and 

 although want of space forbids a description of them, the 

 following brief account, with the sources from which it has 

 been derived, may be of use to those who wish to gain in- 

 formation as to the earlier history of this form of thermo- 

 meter. J. H. van Swinden {Dissert, sur la Comp. der 

 Thermom. (?), p. 253, 1770) described, about 1770, maxi- 

 mum and minimum thermometers made on a plan first 

 contributed by Bernouilli to Leibnitz, and Lord Cavendish, 

 in 1757, described somewhat similar instruments of his own 

 invention {Phil. Trans., vol. 1., p. 300). In 1782 the well- 



