The Measurement of Radiation 181 



media would be roughly parallel for all qualities. If the average 

 atomic mmibers of two media are fairly close, then the relative 

 importance of any one type of energy-conversion process (Comp- 

 ton, photoelectric, etc.) will be similar in the two media, and 

 so the variation of the gross energy conversion with quality will 

 be similar for the two media. 



Thus Villard in 1908 first suggested a unit based on air ioniza- 

 tion : that quantity of radiation which, by ionization, liberates one 

 electrostatic unit of electricity per cubic centimeter of air under 

 normal conditions of temperature and pressure. Much work re- 

 mained to be done, however, before a satisfactory realization of 

 the idea underlying this proposal was possible. Much of the 

 difficulty lay in the phenomenon of the ''wall effect" of the ioniza- 

 tion chamber. The radiation causes the emission of secondary 

 electrons from the walls of the chamber, so that the observed 

 ionization in the air of the chamber, instead of depending uniquely 

 on the radiation itself, is determined by a complex set of factors 

 such as the nature of the walls and the size of the chamber. The 

 surmounting of these difficulties and the development of the 

 theory of the ionization chamber will be referred to later. 



The necessity for general agreement on a satisfactory unit 

 became ever more pressing, and in 1923, the first steps were 

 taken by the British Rontgen and Physical Societies. Discus- 

 sions followed with the first international congress of radiology 

 in 1925, and finally matured at the second international congress 

 in 1928. The unit of X-ray quantity, or dose, called the "ront- 

 gen" (symbol, r) was defined as "the quantity of X-radiation 

 which, when the secondary electrons are fully utilized, and the 

 wall effect of the chamber is avoided, produces in 1 cubic centi- 

 meter of atmospheric air at 0° C and 760 millimeters mercury 

 pressure, such a degree of conductivity that one electrostatic unit 

 of charge is measured at saturation current." 



The ''Free-air" Chajnber 



In order to make measurements in accordance with this 

 definition, a rather special technique is necessary, namely, the 



