NO. 6 BODY RADIATION ALDRICH 1 7 



element used on the subject. This suggestion seemed especially feasi- 

 ble since our results indicate that cloth radiates nearly as a black body. 

 Accordingly, brown canton-flannel cloth was hung forming a cur- 

 tained room 2^ meters high and li by 2 meters in area, enclosing the 

 calorimeter and with the melikeron mounting projecting through the 

 curtain. The same cloth also formed the ceiling and floor. For part 

 of the tests a current of air of known velocity from an electric fan 

 outside the curtain was admitted through a hole in the cloth. The air 

 velocity at the calorimeter was measured with a Katathermometer, 

 an instrument invented by Prof. Leonard Hill, of England (see The 

 Science of Ventilation and Open Air Treatment, British Govt. Report, 

 1919). and serving admirably for this purpose. The motor of the 

 electric fan was run on storage batteries to insure a more constant 

 air current. The Katathermometer was kindly furnished by Mr. 

 Dufifield, of the New York Commission. 



The results of these tests are found in table D. Table F is a con- 

 densed summary of both tables C and D. From these tables a num- 

 ber of conclusions can be drawn : 



(i) The amount radiated from the horizontal cylindrical calori- 

 meter is about 7% less than from the vertical cylindrical calorimeter. 



(2) The estimated wall temperatures in the preliminary calori- 

 meter experiments are too low. From this cause the amount radiated 

 should probably be lowered at least 10%. Much more weight can be 

 placed in the measured wall temperatures of the cloth walls. 



(3) For air motions greater than 75 feet per minute, the melikeron 

 is unsatisfactory for use. An irregular drift of the galvanometer zero 

 due probably to small fluctuating convection currents makes the in- 

 strument unreliable. 



(4) In the ]>reliminai-y experiments the melikeron gives appre- 

 ciably higher results than the thermoelement, and in the second set 

 of experiments this discrepancy disappears. The cause of this is 

 explained in the following section. 



RELATIONSHIP BETWEEN MELIKERON AND THERMO- 

 ELEMENT RESULTS 



In the preceding experiments, recorded in tables B, C, D and E, 

 we have 265 comparisons of temperatures determined directly by 

 thermoelement and computed from the radiation as measured with 

 the melikeron, and including skin, clothing, hair, shoes, wall, and 

 cloth-covered calorimeter temperatures. By a study of these com- 

 parisons we can determine their relationship, with a view to using 

 only the thermoelement in a new series of experiments. The thermo- 



