366 PROCEEDINGS OF THE AMERICAN ACADEMY. 



The temperatures found thus are plotted, a e, Fig. 3, and give a very 

 smooth curve. 



T7ie Change of Resistance of the Sections on Reversal. 



This change was measured for all the sections from lead no. 4 to lead 

 no. 14, according to the general method as described on page 358. In 

 order to show the exact way of taking the observations, the readings for 

 one section are given in full below. The other sections are given only 

 in abstract below. 



Take section (9 to 10). With 72 ohms in resistance box A and 200 



ohms in box B, potential lead no. 9 balanced near the zero end of the 



potentiometer and no. 10 near the 100 end. The readings are marked, 



to show the order in which they were taken, by the small numbers in 



parentheses. 



T„ jxT n Direction of t j xt ia Direction of 

 Lead No. 9. ^ , Lead No. 10. ^ ^ 



Current. Current. 



5.412(1) S 91.200(2) S 



4.447 (3) N 90.044 (4) N 



5.418 (5) S 91.208 (6) S 



5.453 (7) N 90.050 (8) N 



5.422(9) S 91.204 etc. S 



4.451 N 90.050 N 



5.412 S 91.204 S 



4.451 N 90.046 N 



Means 



5.422 S 91.202 S 



5.417 S 91.204 S 



4.450 N 90.050 N 



Means, plus calibration correction of potentiometer : 



5.455 S 91.241 S 



4.488 N 90.089 N 



Hence a length on the potentiometer of 91.241 — 5.455 = 85.786 

 represents the resistance of section 9 to 10 when the current is in the 

 S. direction, and a length 90.089 — 4.488 = 85.601 represents the resist- 

 ance when the current is in the N. direction. So that the difference 

 between these lengths, that is 85.786 — 85.601 = 0.185, represents the 

 change of resistance on reversal of the current on the same scale that 

 the mean of the lengths, i. e. 85.69, represents the actual resistance of the 

 section. From column VI., page 369, the mean value of the resistance 



