282 Win rich — A Study of the Time Interval, etc. 



the tube with the hand was sufficient to unbalance the bridge 

 to the extent of deflecting the galvanometer as high as 2o""\ 

 It was also very reliable, for it was possible to reproduce any 

 points on the curves at any time. If the plan of having the 

 heater on the outside of the tube is pursued, the construction 

 is very much simplified. The space inside the tube may be 



Table IV. 





For Curve I 



For Curve II 



For Curve III 





Heater 

 current 

 in milli- 

 ainperes 



Time in 

 sec. for 

 2"» defl. 



Heater 

 current 

 in milli- 

 amperes 



Time in 

 sec. for 

 2*"' defl. 



Heater 

 current 

 in milli- 

 amperes 



Time in 

 sees, for 

 2"" defl. 



Time 

 lag. 



1 



60-0 



220'0 



27-0 



196-6 



20-0 



226-7 



6-0 



2 



66-9 



172-4 



29-1 



166-6 



23-0 



160-5 



? 



3 



73-1 



150 



35-8 



112-4 



265 



124-8 



5-8 



4 



83-4 



107-0 



45-0 



73'7 



31-0 



92-7 



5-6 



5 



95-4 



82-0 



51-0 



59-5 



35-0 



74-8 



5-0 



6 



105-1 



63-0 



61-0 



45-0 



41-0 



61-8 



5-5 



7 



114-8 



52-4 



67*0 



39-2 



49-2 



441 



3-9 



8 



120-5 



48-5 



705 



36 6 



55-0 



37-0 



3-8 



9 



124-5 



45-0 



79-0 



32-0 



60-5 



31-6 



4-0 



10 



127-9 



41-5 



85-5 



28*2 



09-8 



27-0 



3-6 



11 







95-3 



24-8 



79-0 



23-5 



3-2 



12 







101-4 



23-0 



94-0 



19-1 



3-2 



18 







106-0 



21-3 



102.6 



16-8 



3-4 



14 







1 13-6 



19-7 



no-o 



15-7 



3 



15 







124-5 



17*8 



122-0 



14-4 



3-2 



16 







128-5 



17 



128-5 



13-8 



3-0 



made much smaller. All that is needed in the resistance is 

 that EF and DC as a whole shall be equal. 



Since the object of this paper was to study the time interval 

 only from the experimental standpoint, there are no quantita- 

 tive results given, which would be necessary for the construc- 

 tion of a practical relay. 



I wish to thank Professor A. W. Goodspeed for the liberal 

 allowances he made me in building the special apparatus 

 required, and permitting me to draw freely from the resources 

 of the laboratory. 



I also acknowledge my indebtedness to Dr. R. H. Hough, 

 who gave me many valuable suggestions, and was always ready 

 for consultation even to the point of investigation. 



Randal Morgan Laboratory of Physics, 



University of Pennsylvania, May 2, 1910. 



