1887.] Mr John Aitken on Thermometer Screens. 
59 
It will he observed that while these curves show fairly well the 
effect of inertia on the rate of the heating and cooling of the 
thermometers, yet they do not show that the arrangement with 
smallest inertia gives the highest readings, as we concluded from a 
consideration of the curves in fig. 1. The reason for this is that 
the curves in fig. 2 are complicated by radiation effects. The large 
bulb and one Stevenson screen read higher than the fine-bulbed 
thermometer. This was caused by those arrangements being 
more affected by radiation than the fine bulb. If we take readings 
given by the screen C and the fine bulb — the two which are least 
affected by radiation — it will be observed that the fine bulb rises 
above and falls below C, somewhat in the manner indicated in the 
imaginary curves fig. 1, only it rises too high, owing to its being 
more heated than C by radiation. 
It may be as well to note here that the two thermometers exposed 
under the sunshade did not give more correct temperatures than 
the screens, hut it must he remembered that they were not coated 
with silver when these readings were taken. It may be interesting 
to note that the fine bulb gave lower readings than the large 
bulb, though similarly exposed ; this will be seen from an examina- 
tion of the curves FB and LB. All the curves in this figure 
follow each other more closely than they would under many con- 
ditions. The reason for their comparative closeness on this occasion 
was doubtless the amount of wind that w r as blowing at the time 
the readings were taken. It is evident that a quick circulation of 
air will have great influence in reducing the time necessary to heat 
or cool the screens, and will have very much less effect on the 
exposed bulbs. I much regret I was unable to return to this 
investigation till late in the season, by which time the radiation 
effects were greatly reduced, and, owing to the amount of had 
weather, very few observations were made. The curves shown in 
%• 2 are the result of the only one-minute observations I have 
been able to make. If these readings had been taken in spring 
or summer, and when there was sunshine and little wind, the 
fluctuations would have been much more marked, and the relative 
steepness of the different curves better brought out. 
Turning now to the consideration of the question as to what 
interpretation we are entitled to put on curves of temperature made 
