54 
Proceedings of Boy al Society of Edinburgh, [jan. 17 , 
Before entering on the subject of the paper, I wish to make a few 
preliminary remarks on the cause of the difference in the readings 
given by different screens and by other ways of protecting the 
thermometers ; also to call attention to the interpretation we are 
entitled to put on curves of temperature drawn from readings taken 
at longer or shorter intervals of time. 
All the methods in use for taking the temperature of the air 
give different results. One cause of this difference is the more or 
less perfect way in which the thermometer is protected from the 
effects of radiation ; so that, while all tend to read too high during 
the day, some read higher than others. But in addition to this, 
there is another reason why the different arrangements give different 
results. This second disturbing element we will, for want of a 
special name, call the inertia of the apparatus. By the inertia of 
the apparatus is simply meant the resistance offered by the ther- 
mometer and its surroundings to change of temperature. The 
inertia may, therefore, be measured by the time taken by the ther- 
mometer to acquire the temperature of the air for a given amount 
of change of temperature. Bor example, suppose the temperature 
of the passing air to rise one degree, it would almost instantly 
heat up any small body, such as a cobweb, to its own temperature, 
but it would take a much longer time to heat up a larger body, 
though similarly exposed. The time required will depend on the 
mass and specific heat of the body, and on the shape and amount of 
surface it presents to the passing air. We see from this, that if the 
arrangement of apparatus we use to take the temperature of the air 
has a small inertia, it may, if the temperature is rising, indicate 
at first a higher temperature than an arrangement having a greater 
inertia ; and, if the temperature does not remain long enough 
at its highest point, the apparatus with small inertia will 
indicate a higher maximum temperature than the other. To 
illustrate this point, let us first consider a purely imaginary case. 
We know that the temperature of the air during the forenoon 
of a summer’s day is constantly changing. It does not rise regu- 
larly, but rises to a certain extent, then falls, then rises and falls 
again ; and though the general tendency may be upwards, there are 
many breaks in the curve representing the rise of temperature 
for the day. This results, as we shall see later on, from the manner 
