[BARNES] TEMPERATURE RECORD OF NOCTURNAL RADIATION 131 



earth's radiation through water and clear ice. To do this two exactly 

 similar platinum thermometers were constructed from flat mica plates 

 10 X 4 cm. in size. The wire, 50 ohms resistance, was wound on 

 and soldered to copper leads threaded in the mica. Each thermometer 

 was placed in a box, blackened on the inside, and provided with a 

 hinged cover. The thermometers were connected differentially to a 

 Callendar electric recorder which was kindly lent me by Dr. Porter, 

 Director of the Mining Department at McGill University. 



Fig. 2 shows the thermometers in their respective boxes. 



The boxes were placed side by side on the roof of the porch over 

 the entrance to the Physics Building, and wires were taken to the 

 laboratory^ across the front of the building and in at the window, where 

 thcv ■were attached to the recorder. The recorder itself was placed 



Fig. 2. Radiation Thermometers. 

 on a firm shelf, and was very conveniently located for examining the 

 character of the records produced. Wlien both of the covers of the 

 boxes containing the thermometers, which, we shall designate P and C,i 

 were closed down the thermometers indicated equal temperatures and 

 were sul)ject only to atmospheric disturbances. If either P or C were 

 opened to the sky at night, the exposed thermometer at once became 

 cooled by its own radiation below the closed thermometer situated 

 alongside of it. 



To obtain the screening of any material it was merely necessary 

 to place it over the exposed thermometer, when the trace on the 

 recorder indicated the effect. A I30X with a glass bottom was made in 

 which water or a saturated solution of salf, when the air temperature 

 was below the freezing point, could be placed. 



Glass covers were made for the boxes which could be slipped in 

 place under the hinged wooden covers. 



