﻿496 Mr. J. B. Henderson on the Effects of Magnetic 



in the magnet and taking the mean ; and on the whole these 

 four points can have a possible error of about 2 per cent. 



To determine the influence of temperature on the resist- 

 ance in the magnetic field two methods were made use of. 

 First, the temperature was kept constant, or as constant 

 as possible, and the resistance taken in different fields ; and 

 second, the field was kept constant and the temperature 

 varied, or the temperature-resistance curves in different fields 

 determined. The latter method proved much more successful 

 than the former, as it was much easier to keep the field con- 

 stant than the temperature, and the particulars and results of 

 the latter method only will be given. 



For the fields the current from the accumulators alone was 

 used, and by taking currents up to about 7 amperes constant 

 fields up to 23,000 C.G.S. were obtained. 



In the experiments a flow of water at 0° C. was sent ove 1 * 

 the poles and spiral, and the resistance taken at the lowest 

 temperature reached, which was when the galvanometer 

 became steady, and usually represented about 9° C. The 

 flow of water was then stopped and the temperature of the 

 bath slowly raised by a small flame to 80° C, the temperature 

 and resistance of the spiral being read off at intervals. The 

 constancy of the field throughout the experiment was roticed 

 by observing no change of current through the magnet, as 

 shown by an ampere-meter. 



"In this way temperature-resistance series were obtained 

 for the fields as given in Table III., varying from zero to 

 23,000, in all nine different series, which are represented 

 graphically in PL XVI. fig. 8. 



From the diagram it will be seen that the inclination of 

 these curves to the horizontal diminishes as the field increases, 

 being positive at first, then zero, and then negative ; but at 

 the same time it is also evident that with increasing field the 

 curves deviate more and more from the straight line form, to 

 which the curve in zero-field or the ordinary temperature- 

 coefficient curve approximates ; so that those curves which 

 have a horizontal part within the temperature limits of these 

 experiments are only horizontal for a small distance, which 

 represents a minimum value of the resistance. Such mini- 

 mum values occur at higher and higher temperatures as the 

 field increases. Thus, for example, the curve for a field of 

 9600 has its minimum value about 35°, that for 11,500 about 

 50°, and the one for 14,300 is horizontal at 85°. Thus, 

 reasoning from analogy, it would seem that all the curves 

 have a minimum value, the higher the field the higher being 

 the temperature at which the minimum occurs, and vice versa. 



PI. XVI. fig. 9 represents five curves for the temperatures 



