548 L. H. G R A Y 



value of the stopping power of the soHd relative to the gas for all 

 groups, the energj'^ lost by the ionizing particles per unit volume of 

 solid is equal to p times the energy lost per unit volume of gas. The 

 energy lost in the gas is equal to the number of ions formed multiplied 

 b}^ the average energy, TFgas, expended in the production of a pair of 

 ions in the gas. We have, therefore: 



(£'r) solid = P X TFgas X (J.) gas (15) 



where E^, is the energy lost by the ionizing particles per unit volume 

 of solid (JtOgas is the number of ions formed per unit volume of gas 

 and p is the linear stopping power of the solid relative to the gas. Al- 

 ternatively : 



(Em) solid = Pw X TFgas X (-/to) gas (16) 



where the energy and ionization are expressed per unit mass and p^ 

 is the mass stopping power. 



If the solid and gas have the same atomic composition, then it is 

 generally assumed (23)* that, irrespective of the state of chemical 

 combination of the atoms, pn is unity, so that : 



Em = TFgas X J,n (17) 



The ideal method of estimating E^ is therefore to measure Jm in a 

 small ionization chamber of which both walls and gas have the same 

 atomic composition as the tissue to be irradiated. The dose in 

 equivalent roentgens is by definition E^ divided by Eo, the X-ray 

 energy absorbed per unit mass of air exposed to 1 roentgen of X raj'^s. 

 The X-raj^ energy absorbed in 0.00129 g. of air exposed to 1 roentgen 

 generates, hj definition, ions of either sign carrying 1 electrostatic 

 unit of charge. If, therefore, ionization is expressed throughout in 

 electrostatic units of charge collected: 



Eo = (1/0.00129) X 32.5 = 32.5/d. (e.v./g.) 



where d. = density of air at 0°C. and 760 mm. of mercury. There- 

 fore the dose received by any material in equivalent roentgens follows 

 from equation (17); thus: 



-l^m " gas X i* TO '' gas X »/ » 



1 — in 



dose = -— = 



Eo Eo density of gas X Eo 



and inserting the value of £"0 : 



* Recent experiments by Appleyard (105), however, reopen the question. 



