C. GENERAL PHYSICS. 153 



the upper part of the tube which was not exposed to the 

 heat of the operation. When the portion condensed in the 

 upper end of the tube was united, by cautious tapping, with 

 the main column, the instrument again registered correctly. 

 The observer of this phenomenon determined thereupon by 

 experiment that a quantity of mercury corresponding to from 

 1 to 15 will be volatilized in the manner described if the 

 mercury column of a thermometer is exposed for several days 

 to a temperature of from 60 to 100 Centigrade. This obser- 

 vation is worthy of the special attention of experimenters, 

 since the small quantity of mercury thus condensed in the up- 

 per end of the thermometer tube may be readily overlooked, 

 and thus give rise to serious errors of observation. Frese- 

 nius's Zeitsch. filr Analyt. Chem., XIV., 81. 



RELIABILITY OF SIEMEXS's PYROMETER. 



The Siemens pyrometer has been subject to a careful in- 

 vestigation by a committee of the British Association, in 

 order to decide whether or not the resistance is altered after 

 exposure to high temperatures. Four instruments were ex- 

 amined, three of which were found to be considerably altered 

 after having been exposed to a high temperature ; the fourth 

 gave results showing it to be sufficient for industrial ap- 

 plication, if not for strictly scientific observations. 12 A, 

 X., 373. 



A NEW MERCURIAL MINIMUM AXD MAXIMUM THERMOMETER. 



Mr. Denton describes a maximum and minimum thermom- 

 eter combined in one, by which both registrations of temper- 

 ature are obtained from one mercurial bulb, both indices are 

 moved by the mercury pressing on their ends, and, inde- 

 pendent of the self-registering feature, the actual tempera- 

 ture is shown, at any moment, by two separate columns of 

 mercury. In the construction of the instrument the tube 

 of the maximum thermometer is bent at the top and turned 

 downward, and dips into an hermetically sealed chamber, 

 which is itself more than half filled with mercury. An in- 

 crease of temperature raises the index of the maximum ther- 

 mometer and pushes down the mercury in the other leg of 

 the tube. A diminution of temperature leaves the maximum 

 index in its place, and allows the mercury in the other leg 



G 2 



