308 PROCEEDINGS OF THE AMERICAN ACADEMY 



2G lbs. broke the wire. 



In all the.se experiments, it is easy to see that the change iu resist- 

 ance at first is not at all proportional to the increase iu length. At the 

 close of some of the experiments, the resistance is almost as much as 

 it ought to be ; and perhaps, if there were no error in ob-servation or 

 calculation, the resistance at the breaking-point would be as much a-? 

 the law would make it. 



Experiment 5. — Tliis experiment was mmle with copper wire not 

 so thick as that used in the previous experiment. I measured the 

 total leno:tli,and tlie changes in Icnijth ; so tliat, although not the whole 

 length is under th(i stretching process, the resistances given correspond 

 more exactly to the lengths given. 



I r- R 



1.71 2.924 .109 



1.725 2.965 .1092 



1.755 3.08 .1104 



1.815 3.294 .1128 



1.84 3.385 .1128 



1.875 3.516 .114 



1.9 3.61 .114 



1.91 3.648 .114 



1.95 3.8 .1356 



1.99 3.98 ass 



P R R r^ 



J- = .738 jr = .791 'F> ri 



V ^'1 ^'1 'i 



K) lbs. broke the wire. I performed a second experiment with the 

 .same kind of wire, and with es.sentially the same rc.^^ults. 



Experiment G. — I next took German-silver wire. Tiie curve of 

 observations here will be found to almost coincide with the straight line 

 representing the law. This experiment is (piite curious when con- 

 trasted with the others. 



