162 RADIATION BIOLOGY 



In particular, whenever the neutron nucleus interaction consists of elastic 

 collisions* only, as in the case of moderately fast neutrons interacting with 

 most light nuclei of biological interest, the dose can be calculated as 



D = Xj^iF^a'^ni^ (12) 



where Fj is the energy flux. To a first approximation, for fast neutrons 

 on tissue, D can be considered the result of ionization due to recoil nuclei 

 of hydrogen, carbon, oxygen, and nitrogen, for which it may be assumed 

 that E^'/Ej = 2Ai/{l + Ai)^, where Ai is the atomic weight of atoms of 

 type i. Since o-^' values for these elements are given in the literature, it is 

 possible to estimate roughly the dose when neutron beams of known 

 composition are used on small biological systems, for it may then be 

 assumed that any one neutron does not experience more than one collision. 



This computational approach is of limited usefulness for many pur- 

 poses, however, and the problem then arises as to how to best measure a 

 neutron dose. In principle, this can be done by the cavity method, where 

 the fundamental relation D = pm ' J ' W is valid whenever the ionization 

 in the cavity is produced essentially by particles originating in the walls 

 (Gray, 1944; Rossi and Failla, 1950). 



Difficulties of a practical nature, however, occur in selecting a volume 

 of linear dimensions small compared to the range of the recoil nuclei since 

 this generally will be very small. The situation can be circumvented by 

 using a gas of atomic composition identical to that of the wall, in which 

 case the only limitation on its size is that the neutron intensity be sensibly 

 constant throughout its volume. Rossi and Failla have simulated the 

 composition of tissue (C5H40O18N) by a mixture consisting of gelatin 20.15 

 per cent, glycerin 5.18 per cent, water 66.23 per cent, and sucrose 8.43 per 

 cent, which will harden into a resolvent gel if about 0.5 per cent formal- 

 dehyde is added, and by a gas mixture having the following composition 

 (partial pressure in centimeters of mercury): CH4, 29.22; H2, 50.12; O2, 

 48.96; and air, 3.53. The gas mixture, however, is explosive. Such a 

 chamber can be calibrated by exposing it to 7 rays and by determining the 

 reading R corresponding to 1 r. This corresponds very closely to 93 ergs 

 per gram of tissue. Since for a tissue-equivalent chamber pm = 1, irre- 

 spective of the type of ionizing particle, the energy absorbed when the 

 reading is equal to L can be calculated as 



^ R ^^ We 



where the subscripts p and e refer to protons and electrons, respectively. 

 To a first approximation Wp/We may be assumed to be equal for all 



* Where the sum of the kinetic energies of the neutron and the nucleus is identical 

 before and after impact. 



