RECTIFYING PROPERTY IN SILICON AND SELENIUM. 131 



Whenever the rectification was great the resistance was 

 extremely high, and the currents could not be measured to 

 a high degree of accuracy as the observations become very 

 irregular when the currents were increased, this being due 

 in all probability to heating effects.' Instead of thin sheets, 

 large masses of selenium were sometimes used, and with 

 these the rectifying property was observed when the cur- 

 rent entered and left through surfaces of unequal area, but 

 the differences were not so great as those sometimes 

 obtained with thin sheets. 



In all cases the current for any e.m.f. was less when the 

 point was at the higher potential than when at the lower 

 potential, and in a number of trials the conductance fell at 

 first when the e.m.f. was increased, the flow being in the 

 positive direction. In some experiments the currents were 

 too small to permit the existence of the minimum to be 

 detected ; when it occurred it was for an e.m.f. of not 

 more than one volt. 



Several trials were made with alternating electro-motive 

 forces of frequencies varying from 20 to 60, the currents 

 being measured with an Einthoven galvanometer which 

 enabled the maximum value of the current to be observed 

 for each direction. The effects were similar to those 

 obtained with continuous currents; for example, on one 

 occasion, a deflection in one direction of 3*5 divisions was 

 obtained before a movement of one tenth of a division was 

 detected on the other side, so that the substance must take 

 up the condition necessary that rectification may occur 

 within a very short time of the application of the e.m.f. 



It seemed to be of interest to ascertain whether other 

 substances of the many which possess the rectifying pro- 

 perty showed a minimum conductance such as that found 

 with selenium, and some experiments were carried out with 

 silicon. In all the trials with this material the conductance 



