= 
we 
TRANSACTIONS OF SECTION A, 625 
heat of a liquid in terms of the international electrical units, and gave a few results 
which had been obtained in the cases of water and mercury. The whole apparatus 
was also exhibited in action to several members of Section A, on the occasion of 
their visit to McGill College. One of the main objects of the work was the 
determination of the mode of variation of the specific heat of water over the range 
0° to 100° C., for which the method was peculiarly suited. 
The progress of the investigation has been somewhat delayed by the removal 
of Professor Callendar to London in May 1898. Since that time the work has 
been in the sole charge of Mr, Barnes, who has now succeeded in obtaining satis- 
factory results over the greater part of the range to be covered. 
The general principle of the method, and the construction of the apparatus, 
will be readily understood by reference to the diagram of the Steady-Flow Electric 
Calorimeter given in fig. 1. A steady current of water flowing through a fine 
Fic, 1.—Diagram of Steady-Flow Electric Calorimeter. 
INFLOW GLASS VACUUM JACKET 
tube, AB, is heated by a steady electric current through a central conductor of 
platinum. The steady difference of temperature between the inflowing and out- 
flowing water is observed by means of a differential pair of platinum thermometers 
at either end. The bulbs of these thermometers are surrounded by thick copper 
tubes, which by their conductivity serve at once to equalise the temperature, and 
to prevent the generation of heat by the current in the immediate neighbourhood 
of the bulbs of the thermometers. The leads CC serve for the introduction of the 
current, and the leads PP, which are carefully insulated, for the measurement of 
the difference of potential on the central conductor. The flow tube is constructed 
of glass, and is sealed at either end, at some distance beyond the bulbs of the 
thermometers, into a glass vacuum jacket, the function of which is to diminish as 
much as possible the external loss of heat. The whole is enclosed in an external 
copper jacket (not shown in the figure), containing water in rapid circulation at a 
constant temperature maintained by means of a very delicate electric regulator. 
Neglecting small corrections, the general equation of the method may be stated 
n the following form :— 
ECt=JMd6+H. 
The difference of potential E on tke central conductor is measured in terms of 
the Clark cell by means of a very accurately calibrated potentiometer, which 
serves also to measure the current C by the observation of the difference of potential 
on a standard resistance R included in the circuit. 
The Clark cells chiefly employed in this work were of the hermetically sealed 
type described by the authors in the ‘ Proc. Roy. Soc.’ October 1897. They 
were kept immersed in a regulated water bath at 15° C., and have maintained 
their relative differences constant to one or two parts in 100,000 for the last 
two years, 
The standard resistance R consists of four bare platinum silver wires in parallel 
wound on mica frames and immersed in oil at a constant temperature. The coils 
were annealed at a red heat after winding on the mica, and are not appreciably 
heated by the passage of the currents employed in the work. 
The time of flow ¢ of the mass of water, M, was generally about fifteen to 
twenty minutes, and was recorded automatically on an electric chronograph 
reading to ‘01 second, on which the seconds were marked by a standard clock. 
The letter J stands for the number of joules in one calorie at a temperature 
which is the mean of the range, d6, through which the water is heated. 
The mass of water, M, was generally a quantity of the order of 500 grammes. 
1899, ss 
