188 OUTLINES OF PHYSICAL CHEMISTRY
directly or indirectly by comparison with an air-thermo- meter. The reason for this is to be found in the peculiarities of the relative expansions of mercury and glass. Whatever be the method of control adopted, it is necessary, in order to read the true temperature, to take into consideration the following remarks :
1. When a thermometer has been heated to a rather high temperature (for example, to 100°), the glass does not immediately regain its initial volume. An expanded state is retained such that in melting ice the mercury descends below zero. This residual expansion disappears .only slowly (after days or months), so that, if a thermo- meter is to be used constantly at a high temperature, it is advisable to correct the graduation of the instrument after having maintained it at the temperature in question for about half an hour. (This residual expansion is reduced to a minimum in thermometers made of Jena glass.)
2. The same thermometer in the same medium denotes a slightly different temperature according as the mercury column has come to rest after rising or after falling. This difference tends to disappear if the instrument be gently tapped. To make comparable observations it is necessary always to proceed in the same way, and best to arrange that the mercury reaches its point of equilibrium in rising.
8. To make an exact reading the eye must be brought to the same level as the mercury surface. The so-called parallax error is thus avoided.
The Calorimeter
Thermo-chemical determinations are generally carried out in a water calorimeter by the method of mixture.
The calorimeter consists of a cylindrical vessel of at least 500 c.c. capacity. For smaller dimensions, the surface area becomes relatively very large compared with the volume, and the temperature of the calorimeter is too strongly influenced by that of the surrounding medium.
�� �
