CHEMICAL AND PHYSICAL NOTES, 137 
volumes of mercury and of glass respectively. The sum of these, J, 
is the water value of the bulb as a whole. If the values of 2 be com- 
pared with those of f in the first part of the table, the agreement will 
be found to be very close. With regard to the mercury, there is not 
room for much discrepancy, because it is an elementary body, and it 
can be used in the construction of thermometers only when it is in 
a state of purity. It is otherwise with the glass. Neither Berthelot 
nor the makers give its composition, its density, or its specific heat. 
Although the composition of the glass is of great importance, a know- 
ledge of it is not necessary for thermometric or calorimetric purposes ; 
on the other hand, a knowledge of both the density and the specific 
heat of the glass is essential. Yet it is very rarely furnished. The 
values used in the table are commonly occurring ones; and the agree- 
ment in the water values of the glass arrived at in the first and 
second parts of the table, lines e and #, shows that they apply to the 
glass used. Having obtained the total volume of the bulb, m, and its 
: l : : 
water value /, we obtain at once a the specific heat per unit 
volume of the bulb considered as a whole, or in other words it is the 
water value of 1 c.c. of bulb. The values of m agree very closely, 
the extremes being 0°484 and 0°472, and the mean 0°476. The 
factor used in the rule, page 134, is 0°475, which is a more con- 
venient number than 0°476, and it has now been shown how it is 
arrived at. 
The third part of the table gives the centesimal composition by 
volume of the bulbs of the four thermometers. The thermometers 
are by different makers, and the quantity of mercury in each shows 
how different they must be in pattern ; yet there is great resemblance 
in their composition by volume. The mean composition is 61 
volumes of mercury to 39 volumes of glass, and we shall never be far 
wrong if we take the volumes of the mercury and glass in a thermo- 
meter bulb to be in the proportion 3 : 2. 
Method of Determining the External Volume or Displacement of the 
Bulb.—We have shown how the factor 0°475 is arrived at; it now 
remains to show how the volume or displacement of the bulb is deter- 
mined. It can be roughly ascertained by actual displacement of 
water in a graduated vessel, but this method is not sufficiently 
delicate. When the thermometer has a spherical bulb, its diameter 
must be measured with calipers. The result is not usually satisfac- 
tory, because the length to be measured is very short, and it is never 
certain that the bulb is truly spherical. Fortunately it is usual 
nowadays to make thermometers with cylindrical bulbs, and their 
