XV. ELECTRONS, NEUTRONS, AND ALPHA PARTICLES 523 



more convenient source of very high voltage electrons. Both sources 

 deliver pulses of electronic radiation. 



It is always of interest to extend the range over which a particular 

 variable in any investigation may be studied and, no doubt motivated 

 by this common scientific curiosity, a number of experiments have 

 been carried out with radiation delivered in impulses of extremely 

 short duration. Thus Kingdon, Zahl, Haskins, and Tanis (53) ex- 

 posed Drosophila eggs, Aspergillus spores, and wheat seedlings to 

 impulses of 0.3 to 3.5 r. of X radiation delivered in five microseconds, 

 and repeated once a second, but noted no significantly different re- 

 sponse from that to the same total dose of X radiation delivered at 

 ordinary dose rates. To produce these X rays an "impulse generator" 

 of the kind used by electrical engineers for testing the strength of 

 dielectrics exposed to electrical surges, and consisting of a bank of 

 condensers, applies the desired voltage to a special design of tube in 

 which instantaneous electron currents of the order of 400 amperes 

 are obtained by autoelectronic emission. If, instead of allowing the 

 electrons to fall on a target, they pass out of the tube through a thin 

 window, much higher radiation doses per impulse can be obtained. 

 Thus, Brasch and Huber (54) report having obtained a variety of 

 unusual biological effects when using electron beams derived from a 

 nominal 3.6 m.e.v. genei'ator giving impulses of 1 microsecond dura- 

 tion and mean current during the impulse of 40,000 amp. The dose 

 of radiation delivered per impulse was not stated but in some cases 

 the specimens may well have been exposed to as much as 100,000,000 

 r. in one microsecond. Unless steps are taken to expose specimens 

 only to electrons generated during a fraction of the total duration of 

 the discharge, very inhomogeneous electron beams are obtained in 

 this way since the condensers are completely discharged at each im- 

 pulse. 



Since doses of 100,000 r. and upward are generally convenient for 

 radiochemical studies, impulses of high voltage electronic radiation 

 may prove of considerable value in the field of chemical kinetics. 

 Clearly in those cases in which a biological effect stems from a par- 

 ticular ionization or excitation and is independent of changes in the 

 surrounding molecules, impulse radiation will show no characteristic 

 effects. It may well be, however, that some biological changes that 

 are correctly interpreted as derived from the effects of a single ion 

 cluster are nevertheless brought about with greater probability by 

 large clusters than by small. Effects such as chromatid break produc- 



