VISIBLE AND NEAR-VISIBLE RADIATION 203 



the photochemical effectiveness in producing vitamin D should coincide 

 with the absorption curves. In fact, there is considerable evidence that 

 the long-wave-length portion of the absorption curves contributes to the 

 development of antirachitic material, while the shorter wave-length 

 region may be destructive to the biologically active component, so that 

 it may be desirable in evaluating the dosage from a particular ultra-violet 

 source to determine the relative distribution m at least two portions of the 

 region independently. Furthermore, there seems to be little basis for 

 concluding that the therapeutic effectiveness in the application of ultra- 

 violet to a patient necessarily coincides with the production of the bio- 

 logically active material in vitro. It is still more difficult to see why the 

 erythemal effectiveness should be used for determining the therapeutic 

 value. In fact, one might be definitely detrimental to the other. In such 

 a case, the region of 0.280 m might be preferable. 



Turning again to Fig. 30, the erythemal curve has been reproduced, 

 dotted curve E, for direct comparison with the relative-sensitivity curves 

 of the two methods of ultra-violet measurement in most common use, 

 shown by the heavy curves, Q-BF and ZnS-2. It is immediately 

 evident that these weighting curves have little in common with the bio- 

 logical effectiveness of any of the phenomena as they are known. The 

 value of such methods lies simply in that they enable an operator to 

 reproduce a given dosage for a given type of source. They will tell us 

 nothing, however, as to the relative effectiveness of two different types 

 of sources. In the case of the sun, the relative energy effective in ery- 

 thema has been plotted in the lower section, curve E, Fig. 30, for com- 

 parison with the relative-response curves of the standard and zinc 

 sulfide methods. These points are brought out not as a criticism of the 

 methods which have been developed, which undoubtedly serve a useful 

 purpose in dosage control, but rather to indicate the definite limitations 

 of such methods and the dangers of too wide generalization. 



In the lower section of Fig. 34 the relative-sensitivity curves of two 

 photocells have been plotted which may be of especial value in working 

 with phenomena occurring in this region. The wave-length threshold 

 of the titanium cell, occurring at about 3160 A, is very convenient as to 

 both the erythemal threshold and the probable absorption of ergosterol. 

 If the titanium cell is used in thin corex A (dotted curve), its sensitivity 

 continues well beyond the region where any effective radiation is likely 

 to be available from a quartz-housed source. When combined with a suit- 

 able glass filter, the short-wave-length limit may be adjusted as de^sired. 

 Silver, on the other hand, has a threshold in the vicinity of 2750 A, the 

 short-wave-length limit depending again upon the type of window. By a 

 combination of these two cells, the relative intensities in different 

 portions of the region from 2400 to 3200 A can be fairly well evaluated. 



