162 Prof Ayrton and Mr Haycraft on an Apparatus for 



With the excellent electrical measuring-instruments now 

 obtainable it is possible to measure either current or pressure 

 with an accuracy much greater than one per cent.; indeed the 

 Board of Trade undertake to measure them within one tenth 

 part of one per cent.* We can also measure the mass of water 

 with considerable accuracy : any error, even a large one, made 

 in determining the water-equivalent of the other bodies raised 

 in temperature becomes of small consequence when the water- 

 equivalent is added to the much larger and accurately 

 measured mass. The measurements (a), (b), and (c), there- 

 fore, give us little trouble and do not affect the design of the 

 apparatus. 



The case of (d) and (e) is different. In order to measure 

 a change of temperature by means of a thermometer with the 

 accuracy required, either the chauge must be fairly large or 

 the thermometer must be very sensitive ; but as it is of little 

 use to employ an exceedingly sensitive thermometer to mea- 

 sure the temperature of a liquid which is being locally and 

 rapidly heated, even if the stirring is very efficient, it is 

 necessary to use a fairly large rise of temperature. Similarly, 

 if the circuit is to be closed and broken by hand, and the 

 interval of time measured hy an ordinary stop-watch such as 

 would be found in a junior laboratory, there is a certain 

 minimum time required to give the required accuracy of 

 measurement. Also we must keep the ratio of heat lost to 

 heat generated during the experiment equal to the probable 

 error in the time or temperature-measurement; for as these 

 three quantities are interdependent, the best condition is to 

 make the percentage accuracy of the temperature measure- 

 ment, the percentage accuracy of the time measurement, and 

 the percentage heat lost of heat generated, equal. 



Let W be the maximum number of watts at our disposal ; 

 M the mass of water, including water-equivalents ; 



5 the area of surface of the containing vessel ; 



e the average emissivity of the cooling surface, or the 

 ratio of heat lost in calories per second to the surface 

 area, for 1° C. excess temperature ; 



<S) the minimum change of temperature that can be 

 measured to 1 per cent, under the prescribed con- 

 ditions ; 



T the minimum time, in seconds, between closing and 

 opening the switch that can be measured by the 

 stop-watch to 1 per cent. ; 



6 the rise of temperature of the water ; 



t the time during which the switch is closed ; 



* See Schedule to Final Report of Electrical Standards Committee, 

 1894. 



