CLARK CELL AS A SOURCE OF SMALL STANDARD CURRENTS. Ill 



same on consecutive clays and is independent, to a very high 

 degree of accuracy, of tlie previous treatment of the cell. The 

 same property is exhibited by Clark cells of all sizes ; but the 

 currents which can be constantly maintained are smaller, the 

 smaller the cells. The law regulating the fall of Electromoti-se 

 force was investigated with the result that if S E is the fall of 

 Electromotive force then very approximately for the same cell 

 SE = A + Bg where A and B are constants, and g is the 

 current. Generally, the larger the -cell the greater is the current 

 that can be taken out of it, in about the ratio of the areas of the 

 zinc and mercury surfaces. In small cells it was noticed that if 

 the current was too large the fall of E.M.F. was very difterent in 

 different cells. It was found that this did not depend on the amount 

 of zinc surface exposed (the mercury surface being constant), nor 

 on the thickness of the layer of zinc sulphate solution, nor on the 

 amount of solid zinc sulphate present. On the other hand the 

 thickness of the layer of mercurous sulphate seemed to produce a 

 marked effect. This led to a tentative theory of the gradual fall 

 increasing with the time when two great currents were employed. 

 In fine it is shewn, that taking tlie value of the mean electro- 

 motive force of Clark cells at 1.5C as 1-43500 volts, the E.M.F. 

 of a cell having an area of zinc surface of about ten square 

 inches is 1-43570 volts. The probable error of about two hundred 

 comparisons under all sorts of circumstances is about -00005 

 volts. The fall of E.M.F. on short circuiting the large cell through 

 1423 legal ohms, is in the mean -00526 volts. During two months 

 the greatest observed fall was -00664 volts, and the least was 

 •0045 volts. The behaviour of the cells was compared with that 

 of Daniel cells, and it was shewn that the advantage of Clark 

 cells lies in the two following peculiarities : 1st. the E.M.F. before 

 short circuiting is very constant and definite : 2nd. after short 

 circuiting there is an instantaneous fall which thereafter remains 

 constant. In the Daniel cell the fall of E.M.F. increases with the 

 time, during at all event several liours ; in the Clark cell the 

 diminution of E.JM.F. is practically over after two minutes. A 

 number of curves establish this result. It follows from the numbers 

 that the E.M.F. of a large Clark cell short circuited through 1423 

 legal ohms is 1-4304 volts. Since no appreciable change in the 

 E.M.F. is caused by increasing the resistance slightly, we may 

 say that a Clark cell in circuit with 1430-4 legal ohms (of which 

 about five are supplied by the cell) will give a current of -001 

 amperes with a probable error of about J^ per cent. 

 '■'- This property had been applied to the construction of a 

 galvanometer for the Pi'ince Alfred Hospital. The coil was 

 movable with respect to the suspended arrangement and two 

 positions were marked. A current of '001 amperes was always 

 available for testing the instruments ; any change of sensitiveness 

 could be set right by a special arrangement of the controlling 

 magnets. 



