530 



L. H. GRAY 



roentgens per minute. The difficulty of preventing the evaporation 

 of the Hthium under intense ion bombardment is a factor that has 

 generally weighed in favor of beryllium targets at the highest ener- 

 gies. The distribution of initial energy among the recoil protons 

 generated by 900 e.kv. deuterium ions bombarding lithium and for 

 8 m.e.v. deuterium ions bombarding beryllium has been calculated 

 by Lea {1, p. 21). Though the energy spectrum extends to 15 and 



1000 



100 



10 - 



0.1 0.2 0.5 1.0 2.0 5.0 10 20 

 ACCELERATING VOLTAGE IN MeV. 



Fig. 14. Curves giving neutron output in equivalent roentgens per minute at 

 5 cm. from the target for the (D-D), (Li-D), (Be-D), and (C-D) reactions over 

 the voltage range 300 kv. to 5 m.e.v. Data from Reddemann {96) for (D-D) 

 neutrons and from Amaldi, Hafstad, and Tuve {97) for (Li-D), (Be-D), and 

 (C-D) neutrons. The 500 ^lamp. curves are for 10% D+, and 90% D+a; the 1 



Mamp. curve is for 100% D 



+. * 



Ordinate: dose rate in r.e.p./min. at 5 cm. 



to 12 m.e.v., respectively, there is so great a preponderance of low 

 energy neutrons in each case that the mean values are about 4 and 6 

 m.e.v. compared to about 2 m.e.v. for (D-D) neutrons. The mean 

 initial energies of the particles by which a given cell is ionized are 

 about half these figures. The (D-Be) neutrons are accompanied by 

 7 radiation to the extent of one or more quanta per neutron. Aeber- 

 sold {66) found that the ionization produced in a small amber ioniza- 

 tion chamber by the y rays from berylhum bombarded by 6 m.e.v. 

 deuterons was about 15% of that due to the neutrons. This figure 

 may be taken to indicate roughly the ratio of y ray to neutron ioni- 

 zation in tissue exposed to the unfiltered beam. Filtration by 3 cm. 



* I am indebted to Mr. P. H. Flanders for computing the curves shown in this 

 figure. 



