692 
Proceedings of the Poycd Society 
the screens shown in figs. 2 and 3. A third method is to reduce 
the absorbing power of the thermometer bulb, and expose it freely 
to the air, so that it may absorb as little radiant heat as possible, 
and make the bulb small, so that the heat received may be carried 
away by the air as quickly as possible. This plan has been adopted 
in the thermometer with silvered bulb. 
In all these different methods of working, the enemy we have 
been contending with is radiation, and though we may have 
succeeded in greatly reducing its influence, yet it must be admitted 
that it has still some power. Seeing then we have been unable 
thoroughly to vanquish our enemy, perhaps the wisest course now 
open to us is to see if we cannot make an ally of it, and enlist it 
in our service. 
We know that large bodies are more highly heated by radiation 
than small ones ; we also know that different coloured bodies are 
heated to different degrees by radiant heat. Here then we have 
the foundation of a method of estimating the temperature of the air, 
by observing the difference in temperature, either of different sized 
or of different coloured bodies. For the obvious reason that the 
changes of temperature in different sized bodies do not take place at 
the same time, they evidently are not suitable for our purpose ; but, 
on the other hand, different coloured bodies are. If we can And out 
the relation between the absorbing powers of two substances, then 
we can, by simply coating the bulbs of two thermometers with these 
and noting their different temperatures, tell what the temperature 
of the air is. Suppose, for instance, one substance heated the bulb 
of the thermometer twice as much above the temperature of the air 
as the other, then we should only require to take the readings of 
the two thermometers, and subtract the difference between them 
from the lowest, to get the true temperature of the air during the 
day, and add the difference to the highest, to get the temperature at 
night. 
In practically carrying out this plan and selecting the substances 
most suitable for coating the bulb, one great object evidently is 
to get substances which will not change, and which can be easily 
kept in order. How in these respects no substance is more suitable 
than glass, and one thermometer ought therefore to be kept with 
its bulb clean. As for the other bull), something is required 
