164 Prof. Ayrton and Mr. Haycraft on an Apparatus for 



The average emissivity for small excess temperatures of a 

 glass vessel standing on a felt base and containing water was 

 obtained by taking cooling curves, the mean value being 

 0*000232 calorie per square centimetre of area per 1° C. excess 

 temperature. The water was kept at a uniform temperature 

 by means of a light wooden stirrer during these experiments. 



©, the minimum rise of temperature that can be measured 

 to one per cent., was taken as 2*5 degrees, as it was not con- 

 sidered advisable to rely on the temperature measurements to 

 more than ^ of a degree. The thermometer used is read 

 without the aid of a telescope, and is graduated in 20ths of a 

 degree. 



T, the minimum time that can be read to one per cent. ; was 

 taken asa minute and a half, as it was thought that an error 

 of nearly a second might be made in the measurement of the 

 time between closing and opening the switch, and stop- 

 watches are often a little doubtful as to their zero. 



From equation (2) we have at once 



M _ 0-24x300x90 

 2-5 



= 2592 cubic centimetres. 

 Substituting in equation (1), we have 



2 _ _2_ ^2592 1 _1 



* ~ 100 X 5-53 X -00023 X 90' 

 ^=1-55; 



or the accuracy of measurement is 0*64 of one per cent. The 

 rise of temperature adopted should therefore be 3*87 degrees, 

 the time being 2 minutes 20 seconds. The numbers actually 

 adopted in the experiments were 2000 cubic centim. of water 

 and a time of 2 minutes ; the smaller quantity of water being 

 adopted because the resistance of the immersed strip when 

 made proved to be rather less than ^ ohm, the watts taken at 

 30 amperes being 260 instead of 300. 



To ensure success in an experiment of this kind it is neces- 

 sary that the water be as uniformly heated as possible, and 

 that the stirring be very efficient; and after some considera- 

 tion the authors decided that the best of all plans would be to 

 use a movable conductor of considerable surface, and so 

 shaped that it might itself be used to stir the liquid. By this 

 means an exceedingly uniform rise of temperature may be 

 produced, every particle of water receiving heat direct from 

 the strip at practically the same rate. The following is a 

 description of the apparatus as constructed by the authors. 



A strip of manganin (chosen on account of its low tempera- 



