358 Messrs. Threlfall and Pollock on the Clark Cell 



Two days after being set up, the mean E.M.F. of these 13 

 cells exceeded the mean E.M.F. of the 40 cells by 0*00045 

 volt. 



Considering that the paste had not been neutralized, the 

 agreement is closer than might have been expected. 



(3) A cell, generally called the " master cell," made up in 

 a glass cylinder about 12 inches high and If inch in 

 diameter. The zinc plate, which was just small enough to 

 pass down the cylinder, was cut from a sheet of ordinary 

 commercial zinc, and had been amalgamated before being 

 placed in the cell. The mercury surface was about three square 

 inches in area. This cell was generally used to oppose the 

 others. 



(4) Two cells set up in glass jars similar to the above, but 

 shorter and of greater diameter. Each of these cells exposed 

 a mercury surface of about 12 square inches, and a zinc 

 surface of nearly equal amount. As it was expected that 

 considerable solution of zinc would occur in these cells, the 

 plates were wrapped up in parchment paper. One of the 

 cells was called the " large cell," and the other the " hospital 

 cell," as it was actually in tentative use in the hospital during 

 the progress of the testing, and was occasionally brought 

 over by hand for comparison. 



The investigation of the peculiarities of these cells was 

 directed to the elucidation of the following points: — 



1. The discovery of the most probable value of the E.M.F. 

 of- the cell at rest with a view to finding its terminal P.D. 

 when yielding a given current. 



2. The effect of the extent of mercury and of zinc surface 

 in determining the maximum current which can be taken 

 from any cell with a given allowable amount of uncertainty. 



3. Does the value of A change with lapse of time reckoned 

 from the epoch of short-circuiting ? and if so what is the 

 best time for making an observation? 



4. Does the value of A change appreciably from day to 

 day, the external resistance through which the cells are short- 

 circuited remaining constant, or generally has the immediate 

 previous history of the cell anything to do with the value of 

 A in short-circuiting? 



5. Has change of temperature any considerable effect on 

 the value of A, all the other conditions remaining constant ? 



6. The reliability of the large cell to give *001 ampere 

 through a platinoid resistance assumed correct at 15° C. 



Method of Experimenting . — As already stated, the method 

 described by Lord Bayleigh (loc. eit.) has been used through- 

 out. All the measurements were made in a room with a 



