28 Dr. H. J. S. Sand on the Measurement of the 



the glass plate it rests on are both ground perfectly flat. 

 The latter is rectangular in shape, and the edges o£ the hole 

 d, drilled in it, are perfectly even and free from chips. The 

 plate is put in position in such a manner that the hole d 

 exactly forms the continuation of the tube C ; the way in 

 which it is held on the flanged glass cylinder H and the 

 electrode pressed against it by the specially designed clamp 

 K will be understood from the figs. 1 and 2, as also the 

 manner in which electrical connexion is made by means of 

 thin copper foil bent round the rubber washer w (fig. 2). 

 Practically no leakage took place between the electrode and 

 the glass plate, even when no grease was employed to make 

 a joint. Nevertheless, a small amount of liquid always creeps 

 up, and this seems to give rise to convection currents owing 

 to evaporation, more especially when alcoholic solutions are 

 employed. A very small amount of vaseline thickened with 

 a little paraffin-wax was therefore spread round the edge of 

 the electrode before it was placed in position. In order to 

 guard against its being accidentally moved backwards and 

 forwards over the plate, which would have caused the grease 

 to spread over its surface, a cardboard ring c was fixed on 

 the glass plate by means of paraffin -wax, and the electrode 

 rotated inside this ring to place it in position. The electrode 

 could easily be levelled both when used as in fig. 1 and as in 

 fig. 2, by placing a small spirit-level on the glass plate (fig. 2) ; 

 the retort-stand, to which the apparatus was attached, being 

 fixed on a levelling stand. 



The tube C, the end of which was bevelled, was fixed 

 centrally in the flanged cylinder H, and was held firmly in 

 position with regard to the latter, not only by the rubber 

 bung L, but also by means of the two clamps M x and M 2 

 (fig. 2), both of which are fastened to a common short rod N. 

 The clamp M x at the same time serves to support the apparatus 

 on its retort-stand. The tube C ends about \ mm. below 

 the glass plate, thus allowing sufficient room for a liquid 

 connexion between the auxiliary electrodes and the liquid 

 through which the working current is passing. 



The two auxiliary electrodes have already been referred to. 

 The place where the junction between the electrolyte examined 

 and the decinormal solution of KC1 of the standard electrode 

 is made, is indicated in both figures by J. The method of 

 filling the tubes F and and keeping the liquid in them by 

 means of cotton-wool plugs" before placing the tubes in 

 position will need no further explanation. 



Considerable difficulty was at first experienced in arranging 

 the anode to satisfy condition 2 (p. 23). The method finally 



