138 
PROFESSOR HUGH L. CALLENDAR OX 
The later work of Ludix (Zurich, 1895), under the direction of Professor Pernet, 
extending from 0° to 100° C., requires further notice on account of the wider range of 
the experiments, and the great attention paid to the thermometry. He employed 
mercurial thermometers of the Paris type, with all the usual precautions. He adopted 
the method of mixture, adjusting the quantity of hot water in each case to give the 
same rise of temperature, from 11° to 18°, in the calorimeter. The discrepancies of 
individual measurements at any one point do not exceed '3 per cent., but he did not 
vary the conditions of experiment materially, and it may be questioned whether the 
well-known constant errors of the method could have been eliminated by the devices 
which he adopted. His results (see Table XIV., p. 144) give a minimum at 25° and 
a maximum at 87° C., the values being ‘9935 and P0075 respectively, in terms of 
the mean specific heat between 0° and 100°. The rapid rise from 25° to 75° may 
possibly have been due to radiation error from the hot water supply; and the 
subsequent fall between 90° and 100° to the inevitable loss of heat by evaporation of 
the nearly boiling water on its way to the calorimeter. His values, reduced to a unit 
at 20° C., are given for comparison in Table XIV. The agreement with our values is 
remarkably close at the lower temperatures ; but the difference at 80° amounts to 
nearly 1 per cent. In addition to the fact that his curve cannot easily be reconciled 
with Regnault, there is this theoretical difficulty in accepting his values at higher 
temperatures. The quantity which he actually observed was the mean specific heat 
between the higher temperature and the final temperature of his calorimeter. His 
values of the specific heat itself were obtained by differentiating the curve, and really 
depend on small differences at the higher points between observations which are 
themselves difficult and uncertain. His values for the mean specific heat differ much 
less from ours. The peculiar advantage of the method we adopted is that the specific 
heat itself is determined over a range of 8° to 10° at each point by adding accurately 
measured quantities of energy to the water at the desired temperature. There is no 
possibility of evaporation or heat-loss in transference as in the method of mixture, and 
the protection from external radiation is much more perfect. 
(44.) The Work of Miculescu. 
The work of Miculescu (‘Ann. Chim. Rhys.,’ XXVII., p. 202, 1892), though not 
directly affecting the question of the variation of the specific heat, is interesting as an 
example of a steady-flow method of calorimetry. The apparatus consisted of a Joule 
calorimeter mounted horizontally, with the paddles directly driven by an electric 
motor, the torque being observed by the “ cradle-method ” of Despretz and Brackett. 
The heat generated was measured by passing a steady current of water round the 
calorimeter, and observing with a platinum-iron thermocouple the rise of temperature 
of the stream. The rate of heating was about 50 calories per second, the rise of 
temperature 2°, and the capacity of the calorimeter about 3 litres. The work has 
