Notices respectijig New Books. 61 



the same care as the Choisy-le-Roi flint-glass. At the same time, on 

 comparing together the results obtained on thermometers of com- 

 mon glass, the same conclusion is arrived at as for the thermome- 

 ters made of the flint-glass of Choisy-le-Roi, namely, mercurial ther- 

 mometers constructed of the different varieties of common glass., which 

 are at present used in the manufacture of chemical instruments., do 

 7iot proceed in strict accordance heyond the fixed points which have 

 been used to regulate their scales ; but the differences are so small 

 that t/iey mat/ be neglected in most experiments, especially/ if care be 

 taken to reject glasses which contain a sensible quantity of lead, which 

 may readily be detected when they are worked in the lamp. 



The thermometers made of green glass, similar to that which is 

 used in Paris for organic analyses, and of a Swedish glass, remark- 

 able for its infusibility, gave results sufiiciently analogous to those 

 which had been obtained from the more numerous observations 

 made with the other thermometers. Only, the coefficient of dilata- 

 tion of these two kinds of glass, which, besides, were of a different 

 chemical composition, was not the same as that of the other glasses. 



After having demonstrated that thermo-electrical currents cannot 

 be used for the measurement of temperatures, in a portion of his 

 work of which we shall not speak at present, since we shall copy it 

 verbatim, M. Regnault terminates his memoir with some general 

 conclusions, of which the following is a summary. 



The air-thermometer is the only instrument of measurement which 

 can be applied with confidence to the determination of elevated tem- 

 peratures ; this is the only one M. Regnault will employ when the 

 temperatures exceed 100°. 



The air-thermometer should be founded on the measurement of 

 the changes of elastic force which a given volume of air undergoes 

 when it is l^rought to different temperatures. It is requisite, as far 

 as possible, to arrange the air-thermometer so as to determine di- 

 rectly by experiment the elastic forces between 0° and 100°, the 

 reservoir being plunged into melting ice or kept in the vapour of 

 boiling water. But if, from the arrangement of the apparatus, the 

 direct determination of the two fixed points of the thermometric 

 scale is impossible, as sometimes happens, it becomes necessary to 

 take the point of departure of the air-thermometer at the tempera- 

 ture of the surrounding medium, ascertained by a mercurial ther- 

 mometer, and then to deduce, by calculation, the elements proper 

 to the apparatus for the temperature of melting ice. 



When the thermometer contains air having an elastic force of 

 760 millims. at 0°, and if it does not exceed the temperature of 350°, 

 the elastic force of the air within will not become greater than 

 1720 millims.; there will be no fear, therefore, of a permanent de- 

 formity of the envelope being produced. But at higher tempera- 

 tures there may be a fear of this alteration of form ; — 1, because the 

 internal pressure becomes considerable ; 2, because the glass may 

 undergo a sensible softening. It is consequently advisable to intro- 

 duce into the thermometer air having a weaker elastic force when 



