differential thermometer, with some of its applications. 131 
the limits of error ) as the squares of the distances of the 
centre of the balls from the end of the instrument. 
This property is demonstrated by Lambert to hold with 
regard to light. Photometria, page 56. Theor. VI. 
Experiment 4. 
Place two heated balls on one side of the instrument, and 
one on the other, (the whole being of the same temperature) 
and move the instrument till the fluid remains at zero, and 
the distances of the centres of the balls will be as 1 to the 
square root of 2. 
Since the effects of radiant heat from the heated balls 
diminish as the squares of the distances, how are we to 
account for the striking deviation from this law, when the 
large canister was used ? Professor Leslie says “ such a 
striking deviation from the properties of rectilineal emanations 
must originate somehow, either wholly, or in part, from an 
imperfect reflection” As the same thing holds without reflectors, 
it appears that the cause assigned by Mr. Leslie cannot be 
the true one. 
