170 K Merritt — Light from Incandescent Lamps. 



were drawn after each test showing the variations of T„ T„ 

 and of the lamp current C. The changes in T 2 were found to 

 be slight and to correspond almost exactly with changes in the 

 lamp current. 



In calculating the rise of temperature it was necessary to allow 

 for the time occupied by a particle of water in passing through 

 the calorimeter. Since the rise in temperature of each parti- 

 cle is the difference between its temperature when entering 

 and its temperature when leaving, T\ must be observed before 

 T 2 . The exact difference in time depends on the rate of flow 

 and the capacity of the calorimeter. The correction to be ap- 

 plied to Tj was usually small. 



The rate of flow, F, was kept constant by an arrangement 

 similar to the " Mariotte's bottle " used in illustrating the 

 laws of hydrostatics. D (fig. 1) is a large carboy filled with 

 distilled water, and closed by an air-tight stopper. Through 

 this stopper two glass tubes A and B pass down into the water, 

 reaching nearly to the bottom of the carboy. The tube A is 

 open to the air at the top, while B is connected by means of a 

 rubber tube with the entrance tube, E. The carboy being air- 

 tight except for the tube A, the rate of flow must depend on 

 the difference in level between the surface of the water in the 

 calorimeter and the lower end of A. The flow will therefore 

 remain constant until the level of the water in the carboy has 

 fallen below the bottom of the tube A. As a matter of fact F 

 never varied more that 0*1 per cent in an hour or more. F 

 was measured by weighing the water discharged in a known 

 time. 



The correction for radiation was complicated by the fact that 

 the water in the calorimeter did not come quite to the top, but 

 left a layer of air two or three centimeters thick between the 

 lid and the surface of the water. For this reason the radiation 

 coefficient for the lid was less than for the sides and bottom. 

 The two coefficients were determined as follows : On a warm 

 day, when the water in the carboy was several degrees cooler 

 than the air, the water was allowed to flow through the calori- 

 meter for two hours, and readings were taken of T 1 and T 2 . 

 The rate of flow was also determined in the usual way. Since 

 the lamp was not running, the rise in temperature must have 

 been due to absorption of heat from the air. This rise was 

 only about 0°*3, while the difference between the temperature 

 of the air and the mean temperature of the calorimeter was 

 almost 5°. Hence it was not far wrong to say that the surface 

 temperature of the calorimeter was the same at all points, and 

 equal to the mean temperature of the water. This experiment 

 then gave the sum of the two coefficients R x and R 2 . The 

 coefficient, R 1? for the lid was then determined separately. A 



