CONTRIBUTIONS FROM THE JEFFERSON PHYSICAL LABORATORY, 



HARVARD COLLEGE. 



ON THE LENGTH OF THE TIME OF CONTACT IN THE 

 CASE OF A QUICK TAP ON A TELEGRAPH KEY. 



By B. Osgood Peirce. 



Presented February 14, 190G. Received March 13, 1906. 



In the use of a sensitive mirror, needle, ballistic galvanometer, in 

 cases where a moving-coil galvanometer is objectionable, and the 

 familiar sliding-coil-on-a-long-magnet arresting apparatus unsuitable, 

 it is sometimes desirable to control the needle by currents of short 

 duration, sent through either a coil of the galvanometer or an auxiliary 

 coil, in one direction or the other, by aid of a pair of reversing keys. 

 The current through the controlling circuit during a comparatively 

 long depression of the key must be strong enough to check quickly a 

 large swing of the needle, and yet the operator must be able to bring 

 the needle practically to rest when its whole range, as measured by the 

 indication on its scale, is only a small fraction of a millimeter. To do 

 this quickly and surely requires some skill in making short key con- 

 tacts, as every person who has tried it knows, and different observers 

 find it well to use different strengths of current in the controlling cir- 

 cuit. In order to be able to plan wisely some apparatus, I have lately 

 found it desirable to learn about how short a contact the average 

 observer can make, either with a telegraph key or with a thimble on 

 the finger, and, through the kindness of a number of my colleagues 

 and fi-iends, I have been able to get the results given below. 



The measurements were made according to a well-known principle, 

 by determining either the fi'actional part of the whole charge of a given 

 loaded condenser lost during the key contact by short circuit through 

 a given resistance, or the fractional part of a full charge gained by an 

 originally empty condenser, through a given resistance, when a fixed 

 electromotive force was applied during the time of closure of the key. 

 It is well known that with proper choice of the apparatus this proce- 



