682 
Proceedings of the Pogal Soeiety 
mined, and whether this addition will work equally well in all 
weathers or not remains to be seen. 
Radiation Thermometers. 
Some observations have been made with the radiation ther- 
mometer already described. In addition to this instrument, another 
has been prepared. In this second instrument a small circular 
chamber is made in the centre of the black surface, and covered 
with thin glass let in level with that surface. The thermometer 
with its bujb blackened is exposed in the middle of this chamber. 
This instrument always reads higher than the other one in which 
the bulb forms part of the surface. For this reason, it is perhaps 
not so satisfactory as the other, as it possesses something of the 
so-called “bottling up” powers of the black bulb in vacuo. These 
radiation thermometers were generally exposed in sunshine while 
the trials with the screens were going on. Their indications, of 
course, varied from hour to hour and from day to day. The 
highest temperatures observed were on the afternoon of the 7th 
August, when the thermometer fixed in the black surface rose to 
a temperature of 144°, and the one in small chamber in the black 
surface indicated a temperature of 154°. On the 8th, and on some 
other days, the temperature was nearly as high. When the 
observations were made, the instruments were generally placed 
near the ground with the black surface perpendicular to the sun's 
rays, and freely exposed to any wind that might be blowing. On 
dull and sunless days these thermometers often read 10° or 15° 
higher than the temperature of the air. 
Thermometers with Protected Bulbs. 
A short time ago, I received one of the thermometers specially 
constructed for the experiments with small bulbs, to which refer- 
ence has been made in the former part of this paper. The 
directions for the construction of this thermometer were to make 
the bulb as long as might be necessary to give the desired capacity, 
but its diameter as small as possible. The bulb of the instrument 
sent is, however, not by any means so small as desired, nor so small 
as it might easily have been made. Owing, however, to an unex- 
