592 
DR. J. T. ROTTOMLET ON THERMAL 
have made some progress, particularly in the determination of radiation from incan¬ 
descent filaments of carbon. The loss of heat from a heated body, however, depends 
to some extent on the form and dimensions of the body, and it seemed very important 
to experiment on the emission of heat from bodies differing in form from the wires 
I had been using, and larger in dimensions. I desire now to give an account of some 
experiments in this direction.'^ 
In the autumn of 1889 I commenced to experiment on loss of heat from the copper 
globes used in the experiments of Mr. D. Macfarlane, ‘Roy. Soc. Proc.,’ 1872, 
p. 93. These globes are two exactly equal solid copper spheres, 4 centims. in 
diameter ; one has a smooth copper surface, the other is now thoroughly well electro¬ 
plated with silver. A considerable number of preliminary experiments were tried, 
using apparatus similar to that which Macfarlane employed, and in particular the 
same enclosure; and then, as my object was to experiment first with full air pressure 
and afterwards with vacuum more or less complete up to the highest attainable, new 
and suitable arrangements were made for this purpose. 
It is unnecessary here to describe in detail the apparatus used and the mode of 
experimentiug in these preliminary investigations.! It was similar in every respect to 
that of Macfarlane. The thermo-junctions were new, however, and were standard¬ 
ized for the purpose of these experiments ; the galvanometer arrangements were also 
fresh and were very carefully made to secure sensitiveness and permanence of the 
thermometric value of the deflections. The globes had their surfaces prepared in 
the waj^s shown in the columns of the table on p. 593 ; and the metallic enclosure 
used by Macfarlane was employed. The headings of the table will, it is hoped, 
make it self-explanatory. 
* The experiments described in what follows were concluded some time ago. I have delayed the 
giving an account of them in the hopes of finding time and opportunity for obtaining further results; 
but I have been unable to do so up to the present, and judge it best to publish now the results which I 
have obtained. 
f In connection with these preliminaiy experiments I Avish to acknoAvledge Amry cordially the great 
assistance I received from Mr, James H. Gray, M.A., B.Sc., at the time of the experiments a young 
student in the Laboratory, noAv the holder of the first 1851 Exhibition Scholarship for Glasgow Uni¬ 
versity. To his care and assiduity I am greatly indebted. 
