344 BELL SYSTEM TECIIXICAL JOURNAL 



maxima vary in position for the same element willi tlie condition 

 and mode of preparation of the surface. 



Theory of the Schroteffekt}^ T. C. Fry. The current from a 

 vacuum tube is composed of discrete particles of electricity which 

 emerge according to no regular law but in an accidental, statistical 

 fashion. The current therefore fluctuates with time. If the fluctua- 

 tions are amplified sufficiently they may be heard in a telephone 

 receiver as "noise" — a type of noise which is due to the mechanism 

 of electron emission itself and nfit to outside interference. This 

 noise is called the "SchrotefTekt." 



The effect is of certain importance from the telephone standpoint, 

 for it appears that signals, the intensity of which is lower than that 

 of the accidental current fluctuations, can never be rendered intel- 

 ligible liy vacuum tube amplification since the noise due to the sta- 

 tistical fluctuations of space current would be amplified to the same 

 extent and would mask the signals. Fortunately, however, the 

 effect is much less pronounced under operating conditions than it is 

 under the conditions which are most favorable for laboratory study. 

 This is due to the fact that the presence of space charge under operat- 

 ing conditions smooths out the electron stream to a very material 

 extent, and thus reduces the tube noise. The limitation imposed 

 upon amplification is therefore not serious. 



The present paper deals with what we have termed "laboratory 

 conditions" as distinct from "operating conditions." Its principal 

 result, arrived at by theoretical consideration, is: That if the elec- 

 trons are emitted independently of one another the intensity of the 

 noise in the measuring instrument is 



S=VU'\, 



where v is tlu' luiiiibiT of t'it'ctroiis I'liiillt'd per unit liiiii' and ?c'i is 

 the average over all electrons of the energ\' that each would have 

 caused to be dissipated in the measuring tle\ice if not other had eviT 

 been emitted. 



When this formula is ajjpiicd to llic ty|)i' nf sinipK tuned circuil 

 that was considered by earlier writers, it leads to substantially the 

 the same results as they had obtained. It is more general than these 

 earlier results, however, and rests on less questionable methods of 

 derivation. It is, in fact, more general than the problem of the 

 SchrotefTekt itself and applies ecjually well to the absorption of 

 energ>' from any type of accidental disturbance which satisfies the 

 condition that the individual electromoti\-e impulses occur inde- 



" Journal of Franklin Institute, Vol, 109, p. 203, 1925. 



