THE 
AMERICAN JOURNAL OF SCIENCE 
[THIRD SERIES.] 
Oe 
Art. XLVIL—The Temperature Variation of the Thermal 
Conductivities of Marble and Slate; by B. O. PEtrcE and 
R. W. WILison. 
Iy the course of an investigation on the absolute thermal con- 
ductivities of certain poor conductors, which has occupied us 
for the last two years, we have obtained some preliminary meas- 
ures of the temperature variation of conductivity, which we are 
permitted by the Rumford Committee of the American Acad- 
emy of Arts and Sciences to present here in advance of the 
publication of the complete paper. 
Our experiments were made by the so-called “ Wall Method ” 
upon square prisms. In one form of apparatus which we have 
used the prism to be tested is enclosed between two planed iron 
plates, one of which forms the top of a jacketed chamber through 
which steam may be passed for some hours at a time, while the 
other is the bottom of a vessel containing a little water and a 
large quantity of ice, with suitable stirrers and scrapers to insure 
the constant removal of the film of water immediately against 
the surface. These plates are fastened firmly together by bolts 
around their edges to insure close contact with the body under 
experiment. It is thus possible to keep the boundary planes 
at temperatures nearly a hundred centigrade degrees apart ; the 
continuous observation of the interior temperatures renders it 
easy to see when the stationary state is reached and to insure ~ 
that it is maintained. 
The actual measurement of the interior temperatures is a mat- 
ter of more difficulty. Inasmuch as the prism must be thin, it 
is necessary to determine with considerable accuracy the posi- 
Am. Jour. Sct.—TaHirp Smerizs, Vou. L, No. 300.—DECEMBER, 1895. 
29 ‘ 
