68 Applied Biophysics 



here that this accuracy is probably better than is needed in 

 clinical work. What is needed is an instrument which, under 

 varied conditions of use, will always indicate within a few per 

 cent the total incident radiation. A vacuum thermocouple is 

 more accurate in the sense that it measures a certain quantity 

 very precisely, but as we shall see, under certain not unusual 

 clinical conditions, the quantity which it does measure is very 

 different from the quantity which the clinician needs for con- 

 trolling his treatment. Our instrument must, of course, be cali- 

 brated against known radiation sources, or by other methods, 

 but when this is done it is found to have a linear response, and 

 the millivoltmeter or other instrument used to measure the 

 thermoelectric current may then be calibrated with a linear scale 

 of pyrons. Typical examples of determinations of total flux 

 under various clinical radiation sources are shown in the figures 

 on pages 69 and 70. 



When this instrument is used to measure the incident flux 

 under various types of clinical radiation source, it is usually 

 found that the flux increases with time. This is due to the 

 fact that the glass envelopes of the electric lamps, metal re- 

 flectors, and other parts of the source, become heated in course 

 of time, and these in their turn become sources of radiant energy. 

 The temperatures attained by these parts of the source are low, 

 but in many cases they are of considerable area, so that they 

 may eventually come to provide the major part of the flux 

 received by the patient. On the other hand, the radiation which 

 they do provide is all low-temperature, long-wave radiation 

 which is absorbed by glass. Therefore, a glass-enclosed instru- 

 ment will show little or no increase even under circumstances 

 when, in the course of an hour, this instrument will show a 

 three-fold increase of flux. We have pointed this out in a dis- 

 cussion of radiant-heat cradles,*'^ where we found that the flux 

 at the center of the cradle increased from 0.4 pyrons to 1.2 

 pyrons in an hour. The patient, of course, will respond to this 

 change in ways which may be unpleasant, but for the reasons 

 given, a glass-enclosed thermocouple will not respond to it. This 

 time factor is thus of peculiar importance in estimating radiant- 



