122 
MESSRS. W. R. BOUSFIELD AND C. ELSPETH BOUSFIELD 
former paper were the use of a cylindrical Dewar vessel immersed in a water bath 
as the calorimeter, and of a new “ mercury resistance thermometer ” as an electrical 
heater. The open end of the Dewar vessel was closed by an obturator consisting of 
a well-fitting platinum box containing water, which was kept at a certain temperature 
above the contents of the calorimeter in order to prevent condensation of vapour. 
A stirrer and thermometers and the leads of the heater were passed through suitable 
channels in the obturator, and through these orifices and around the obturator there 
was a small unavoidable escape of vapour from the calorimeter. The heat loss due 
to this was the subject of a correction which, in the neighbourhood of 80° C., became 
sufficiently uncertain to deprive the results of the great accuracy which was sought. 
The experiments were therefore not carried beyond 80° C. The experiments described 
in the present paper are only carried to 40° C., within which range corrections can be 
applied of the same order of accuracy as that of the electrical measurements. In 
order to render the apparatus more easily workable by two observers only, certain 
changes of method have now been made which must be noted. 
In the former experiments it was the business of one observer to keep the heating 
current as nearly as possible constant with the help of an ampere balance in the 
circuit and an auxiliary mercury thermometer resistance, the ends of which were 
shunted through a battery of standard cadmium cells and a galvanometer. It was 
the business of another observer, by means of electrical heaters and gas burners, to 
keep the temperature of the bath exactly equal to the rising temperature within the 
calorimeter so that there was no heat transfer through the Dewar vessel, and also to 
keep the temperature of the obturator exactly 10° C. above that of the calorimeter. 
At the same time constant notes of the temperatures had to be made. This process 
involved a good deal of strain and an amount of skill which could only be obtained 
by long practice. The following modifications were therefore made :— 
(а) The ampere balance and standard cells have been abandoned and the heating 
regulated by observation of a watt balance in the circuit. Thus the electrical energy 
is now kept constant instead of the current. This involves some loss of accuracy, 
but gives some compensating advantages. 
(б) Instead of endeavouring to keep the temperature of the bath exactly equal to 
that of the contents of the calorimeter, the bath heating is only adjusted every 
two or three minutes, so as to keep the temperature within ±0°'5 C. of that of the 
calorimeter contents. The bath temperatures are noted every minute and plotted 
(see fig. l) and a suitable correction applied for the temperature differences. 
(c) The obturator is at the beginning of every run filled with water at such a 
temperature that at the end of the run (which lasts about 15 minutes) its tempera¬ 
ture will have sunk to one or two degrees above that of the calorimeter contents. The 
obturator temperatures are also noted every minute, and subsequently plotted in 
order to arrive at the proper correction. 
With the above changes the observations can be easily carried out and the 
