Reaction before Complete Equilibrium. 77 



Galvanometer (G.\ — For determinations of freezing- and 

 boiling-points an astatic Kelvin galvanometer is very con- 

 venient. As velocities of reaction were to be measured 

 where the indications of the galvanometer ought to be, if 

 possible, instantaneous, and as it was desired to avoid 

 magnetic effects, a sensitive Ayrton and Mather was used. 

 The resistance of the coil is 12*2 ohms. 1 ^ooo~uoo am P ere gives 

 at a distance of 1 metre a deflexion of 9*5 mm. 



The speed with which the galvanometer reaches the point 

 of rest has been photographed. This velocity is = ca. 103 cm. 

 per minute, when the mirror is removed 1 cm. from the 

 point of rest f , and when for the calculation of C the equation 



-- =C(t — t) is used; and to — t indicates centimetres. (It is 



ClT 



evident that the velocity with which my Pt- thermometer 

 assumes the temperature of the liquid must be greater than 

 103° C. per minute when its temperature is removed 1° from 

 the temperature of the liquid, because the same equation, 



__ =zC(to — t), is used, and the centimetres of the scale 



UT 



indicate the degrees of temperature. ) One or two dry 

 cells have been used so as to get for 1 mm. of the bridge 

 about 1 mm. deflexion on the scale. A great deal of 

 trouble was experienced in steadying the galvanometer, 

 owing to the great traffic in the adjacent streets, and to the 

 machinery always at work close by. An iron plate of about 

 7 or 8 mm. thick was fixed to two iron strips, one of them 

 holding the lamp and scale ; the centre of gravity is in the 

 middle of the plate. The galvanometer is suspended in the 

 air by indiarubber bands in a wooden box (</), and this is so 

 placed on the iron plate that the centre of gravity of the 

 whole system (iron plate, wooden box, and galvanometer) 

 coincides with the suspension of the coil and mirror. The 

 whole is floating on mercury in the wooden trough (T), which 

 again rests on sheets of indiarubber. These 1 arrangements, 

 however, did not suffice to reduce the vibration of the mirror 

 to the desired limit. This was only achieved when the thin, 

 flexible, copper piece on which the coil is suspended was 

 replaced by a solid inflexible one. In this way the vibrations 

 on the scale were reduced to the limits of almost +2 or 3 

 tenths of a millimetre, and the reading could be made with 

 accuracy to one or two tenths of a millimetre. There can be 

 no doubt that under more favourable conditions the results 

 obtained with these arrangements would have been much 

 better. 



