130 THE ANTARCTIC MANUAL. 
The rule * for finding the term of cooling referred to the shortest 
possible intervals of time, and the smallest logarithmic differences, 
from observed values of d@ and di is: Divide the modulus of the 
system of logarithms, 0°434295 by the logarithmic difference dl, and 
multiply the quotient by the interval of time d@. The product ts the 
term of cooling expressed in the same wnits of time that have been 
used in expressing the time interval d @. 
For the above case we have for the true term of cooling— 
0°434295 
0669 * 10= 64°917 seconds. 
R= 
In the application of this rule we may make the interval any- 
thing we please. We may, therefore, choose it so that the loss of heat 
during it is expressed by a simple fraction, such as one-half. The 
convenience of this method was first pointed out by Leslie, and it 
affords by far the best practical method. It is sufficient, for instance, 
to heat the thermometer to 10° or 11° above the temperature of the 
atmosphere, take time when it is exactly 8° warmer than the air, and 
again when its temperature has fallen to 4° above that of the air. 
The excess heat present at the beginning of the interval is 1, and 
that at the end of the interval is 3. Then we have— 
dl = log. 1 — log. } = 0°30103; 
and the term of cooling is 
0434295 
R= Q30103 7 93 
or, very approximately, R = oe d 0. 
Hence the rule to find the term of cooling when the time in 
which one-half of the heat excess is lost is: Multiply that interval 
d@ by 101 and divide it by 70. The quotient is the term of cooling. 
It is evident that there is no difficulty in making this observation 
in a shop, and the time in which the instrument loses half its heat 
is sufficient without further computation to give a good idea of what 
has been called its thermal nimbleness, The want of a term commonly 
used to express this important property shows how much the 
property itself has been neglected. 
Table XIII. gives an example of the use of the method of the 
“half-fall’’ in the case of a thermometer, first with its bulb plain, 
and secondly with its bulb silvered. 
“ «Experimental Inquiry into the Nature of Heat,’ by John Leslie. Edinburgh, 
1804. Page 265. 
