=Daniell’s Pyrometer.=—In 1822 Daniell published an account of a pyrometer based on the expansion of a platinum rod enclosed in a plumbago tube. One end of the rod pressed against the end of the tube, whilst the other end was free to move, and was connected to a multiplying device which magnified the expansion, the increased movement being indicated by a pointer, moving over a dial. The scale on the dial was divided evenly into a suitable number of parts, it being assumed that the difference between the expansions of graphite and platinum was uniform at all temperatures. The scale was calibrated as far as possible by comparison with a mercury thermometer, the remainder being extrapolated. With this pyrometer Daniell obtained a value of 2233° F. for the melting point of silver, and 3479° F. for that of cast iron—results considerably higher than those now accepted, but much nearer than those obtained by Wedgwood. Daniell’s pyrometer was widely used, and its modern representatives are fairly common. Platinum, owing to its cost, is no longer used in these instruments, which are now generally made with a graphite rod encased in an iron tube, on the end of which the graduated dial is placed, as shown in fig. 69. Another form, commonly used in baker’s ovens, is constructed with an iron rod surrounded by a porcelain or fireclay tube. [Illustration: FIG. 69.—EXPANSION PYROMETER.] The defect of pyrometers of this type is that the coefficient of expansion of the materials alters with prolonged heating, causing the readings to become erroneous. Re-adjustment in boiling water or other substance does not compensate properly for this alteration, as both materials are not equally affected. Again, the readings will be too low unless the whole of the expanding parts are in the interior of the furnace, in which respect this pyrometer is inferior to a thermo-electric instrument, which may be inserted at any convenient depth, and may therefore be used for a greater variety of purposes. The chief recommendation is cheapness; but an expansion pyrometer should never be used for work of precision. A graphite rod in an iron enclosure gives more consistent results than other materials.