Page:Elementary Text-book of Physics (Anthony, 1897).djvu/204

190 reached by the end of the mercury column is marked on the stem, as before. The space between these two marks is then divided into a number of equal parts.

While all makers of thermometers have adopted the same standard temperatures for the fixed points of the scale, they differ as to the number of divisions between these points. The thermometers used for scientific purposes, and in general use in France, have the space between the fixed points divided into a hundred equal parts or degrees. The melting-point of ice is marked 0°, and the boiling-point 100°. This scale is called the Centigrade or Celsius scale.

The Réaumur scale, in use in Germany, has eighty degrees between the melting- and boiling-points, and the boiling-point is marked 80°.

The Fahrenheit scale, in general use in England and America, has a hundred and eighty degrees between the melting- and boiling-points. The former is marked 32°, and the latter 212°.

The divisions in all these cases are extended below the zero point, and are numbered from zero downward. Temperatures below zero must, therefore, be read and treated as negative quantities.

A few points in the process of construction of a thermometer deserve notice. It is found that glass, after it has been heated to a high temperature, and again cooled, does not for some time return to its original volume. The bulb of a thermometer must be heated in the process of filling with mercury, and it will not return to its normal volume for some months. The construction of the scale should not be proceeded with until the reservoir has ceased to contract. For bhe same reason, if the thermometer be used for high temperatures, even the temperature of boiling water, time must be given for the reservoir to return to its original volume before it is used for the measurement of low temperatures.

It is essential that the diameter of the tube should be nearly uniform throughout, and that the divisions of the scale should represent equal capacities in the tube. To test the tube a thread of mercury about 50 millimetres long is introduced, and its length is