48 



BIOLOGICAL EFFECTS OF RADIATION 



tions corresponding to the particular voltage wave form used, there can 

 be no direct or sunple relationship between the qualities of X-ray beams 

 of very different voltage origin. On the other hand, when constant voltage 

 is applied to a tube, the spectral distribution remains the same and is char- 

 acteristic of that particular voltage. In such cases, it would be sufficient 

 for fixing the quality to state only the target material, voltage, and filter. 

 It is obvious from the foregoing discussions that peak voltages in particu- 

 lar may be very misleading. Under most practical conditions (52, 50) 

 the effective (root mean square) voltage will give a much more nearly 

 consistent measure of the quaUty, insofar as voltage measurements alone 



are concerned. 



Direct determination of the spectral-energy distribution is possible 

 only by means of the ionization spectrometer (23, 24). This process is, 



^cVc-td 



Fig. 3.- 



'c 

 Ti me 

 -Typical X-ray voltage wave forms. 



however, so complicated and laborious as to be precluded in practice. 

 Consequently, indirect methods of quality determination are usually 

 resorted to, most of them based upon the laws of X-ray absorption in 

 copper and ahuninum. 



Com,plete Absorption Curve.— It has been shown theoretically by 

 Silberstein (39) that a given X-ray spectral distribution will yield a 

 discrete absorption curve (in, say, copper) ; and conversely, that from a 

 given absorption curve can be derived the spectral distribution with 

 sufficient accuracy for most practical purposes. Therefore, a statement 

 of the complete absorption curve in copper or aluminum is taken as the 

 quality of the radiation (55, 51). 



To obtain the complete absorption curve, filters of the absorbing 

 material (copper or aluminum) are successively placed in the X-ray beam 

 and the transmitted intensities read with an ionization chamber (dis- 

 cussed in detail later on). If h is the intensity of radiation, incident 



10 



100^ 



on a filter /i, and / the intensity transmitted through /i, then log 



is the logarithm of the percentage transmission for filter /i. Plotting the 



values of logio 100^ against a number of values of /then gives a complete 



io 



