.588 SCIENTIFIC RECORD FOR 1883. 



shut off by the rise of the mercury as the temperature rises. (Phil. 

 Mag., May, 1883, V, XV, 339.) 



2. Expansion and Change of State. 



Kapoustine Las devised a simple method of showing the expansion 

 of a solid bar. It is supported at its two ends, one end resting against 

 a fixed point, and the other upon a sewing needle placed upon a hori- 

 zontal plate of ground glass. When the bar expands by heat the 

 needle rolls upon the glass, and a light wooden index fixed upon its 

 point renders this rotation visible. Greater precision can be given by 

 the use of a mirror. (J. Soc. Phys. Ghent. Russe, XIV, 64; J. Phys , 

 December, 1883, II, n, 57C.) 



In order to throw some light on the question whether a given bar 

 can have differing lengths at the same temperature, Woodward, 

 Wheeler, Flint, and Voigt have made a series of experiments with bars 

 of various metals, measuring them at the temperature of melting ice. 

 Seven bars were used — two meter bars of steel, two of glass, one of 

 zinc, one of copper, and one of brass. The comparisons w r ere made by 

 means of micrometer-microscopes, magnifying about thirty diameters 

 and reading to 92.1 and 95.3 microns, respectively, for each turn of the 

 screw. They were mounted on an oak beam one meter apart, the 

 whole being supported on stone piers. The bar to be compared is 

 placed in a wooden box 1.1 by 0.1 by 0.1 meter, supported at one-fourth 

 and three-fourths the length of the bar from either end. The box is 

 filled with finely pounded ice, spaces being left near the graduations. 

 Taking one of the steel bars as standard, several sets of comparisons 

 are made with another bar. This latter is placed in water, which is 

 gradually raised to boiling. It is then cooled gradually to the melting 

 point of ice and a second set of comparisons made. The same bar is 

 then cooled to —6° or —8° F., returned to the melting point of ice, and 

 again compared. The zinc bar, raised to 208° F. and then cooled to 

 the temperature of melting ice, was found to be 139 microns longer 

 than before. Kept in melting ice for four days, it shortened 39 microns, 

 leaving the final length still 90 microns shorter than the initial length. 

 It was then cooled to —8° F. iu the open air and again brought to the 

 temperature of melting ice. Its length had now diminished by 30 mi- 

 crons below that originally given. Kept in melting ice for a day and a 

 half produced no change ; but exposure to the temperature of the com- 

 paring room for one day increased its length permanently by 15 mi- 

 crons, leaving it still 15 microns too short at the temperature of melt- 

 ing ice. Exposed to an air temperature of 70° F. for four hours in- 

 creased its length when cooled to the temperature of melting ice 26 

 microns, leaving its final length 11 microns greater than the initial. 

 The zinc bar then having been subjected to a total range of 216° F. 

 varied in its length at the temperature of melting ice 169 microns. No 

 similar set was observed in the copper bar as compared with the steel, 



