=70. Rutherford's Maximum Thermometer.=--The maximum thermometer, invented by Dr. John Rutherford, differs from an ordinary thermometer in having a small cylinder of steel, porcelain, or aluminium, moving freely in the tube beyond the mercury, so as to form an index. The stem of the thermometer is fixed horizontally on the frame, which must be suspended in the same position, as represented in fig. 53. The instrument is set by holding it bulb downward, so as to allow the index to fall by its own gravity into contact with the mercury. Increase of heat produces expansion of the mercury, which consequently pushes forward the index. When the temperature decreases, the mercury recedes from the index, leaving it so that the extremity which was in contact with the mercury indicates upon the scale the highest temperature since the instrument was last set. [Illustration: Fig. 53.] As it is easily constructed and is comparatively cheap, it is still employed for ordinary purposes. Its disadvantages are, firstly, its liability of soon getting out of order by the index becoming embedded in the mercury, or fixed by oxidation, thus rendering it altogether useless; secondly, the ease with which the index can be displaced by the wind moving the instrument, or other accidental disturbance, so as to cause it to give erroneous indications occasionally; and thirdly, its consequent total unfitness for use at sea. In the part of the tube beyond the mercury, a small quantity of air is enclosed for the purpose of preventing the metal flowing freely in the tube. This necessitates the construction of a larger bulb, which renders the thermometer less sensitive. Moreover, as it frequently happens that some mercury passes the index, particles of air insinuate themselves in the metal, and cause separations in the column, which very often can be removed only by a maker. To facilitate this re-adjustment, a small chamber is left at the end of the tube, and the mercury being expanded into it by heat until the index and air bubbles are forced into it, if possible, upon the cooling down again, by a little management, the mercury will contract, leaving the air and index behind. Yet sometimes the index cannot be moved in the least from its place of fixture, so that the instrument must be virtually reconstructed. =71. Phillip's Maximum Thermometer.=--A maximum thermometer, better perhaps in its action than Rutherford's, has been suggested by Professor John Phillips, of Oxford. A small portion of air is introduced into an ordinary thermometer, so as to cut off about half an inch of the mercurial thread near its end in the tube. This forms a maximum thermometer, when the stem is arranged horizontally. The isolated portion is pushed forward by expansion, and is left in this position when the mercury contracts. The end remote from the bulb shows on the scale the maximum temperature.