Thermal Conductivities of Marble and Slate. 44] 
but we can find no evidence of change of conductivity with 
the temperature. 
In order to form an estimate of the difference in tempera- 
ture between the adjacent faces of two slabs separated by one 
of the interspaces, we made a series of experiments by increas- 
ing the width of one of the spaces by various amounts. We 
found that this difference increased rapidly with the width 
of the space, being about 10° C. for a space 1™™ wide and 
nearly 30°. for a space 1°65™" wide. Our observations seemed 
to show that the temperatures on two sides of a crack 0°2™™ 
wide did not differ by so much as 2° C. and that the thermal 
junction gave either temperature within 1° C. 
Two sets of experiments were made with the second form 
of apparatus upon slabs cut parallel with the cleavage from the 
same block of slate. Although further measurements, which 
we hope to make soon by other methods, are desirable, our 
results seem to indicate an increase of conductivity of about 
30 per cent between 70° C. and 300° C. They are as follows: 
Distance in millimeters 
from the Cold Face. Temperature in Case I. Temperature in Case II, 
0-0 3 49°5 455 
9°9 91°6 96°1 
19°6 128°8 140°5 
29°4 161°0 182°5 
39°2 196°0 219°9 
48°9 229°6 258°9 
58°8 260°5 295°0 
68°9 . 2Ou«2 d00'l 
The range of temperature in the second series was increased 
by decreasing the thickness of the sheet of asbestos between 
the hottest slab and the iron plate next it. 
Our acknowledgments are due to the American Academy of 
Arts and Sciences, who have made an appropriation from the 
Rumford Fund in aid of our work. 
_ The Jefferson Physical Laboratory, Cambridge. 
