86 Mr. J. Brown on the Theory of 



was observed, which decreased, at first rapidly, and in about a 

 quarter of an hour became constant at 1 0°. When the battery 

 was cut out of the circuit there was a strong secondary current, 

 and on dismounting the apparatus, it was found that copper 

 had been deposited on the parchment-paper in that vessel con- 

 taining the positive pole of the battery. 



From these observations it was concluded that the posi- 

 tively charged copper ions, not being able to pass the partition, 

 gave up their electricity in order to deposit in the metallic 

 form. Certain electrochemical actions are described in detail 

 as having occurred at the other portions of the circuit, 

 such as the passage of potassium atoms through the other 

 partition, but apparently no attempt was made to ascertain 

 if they actually took place. One has a right to expect that 

 some evidence of their occurrence should be given, but none 

 appears. Simply the conclusion is drawn that the foregoing 

 appears to show the correctness of the above-mentioned pro- 

 position, that the semi-permeable walls are for certain ions 

 permeable, and for others not. 



But it will, I think, be admitted that in the absence of any 

 investigation by analysis or otherwise of the actual products 

 at the porous walls, the phenomena actually described might 

 arise from a certain degree of conductivity in the matter pre- 

 cipitated in the parchment-paper. In fact, if the parchment- 

 paper were replaced by platinum sheet with either a small 

 amount of any kind of porosity or none at all, phenomena the 

 same as, or similar to, those actually observed would occur, 

 L e. the initial decrease of current, the secondary current, and 

 the deposit of metallic copper. 



It remained, then, to test the precipitate for conductivity. 

 I prepared it by adding potassium-ferrocyanide solution to 

 excess of copper- sulphate solution, filtered off and washed the 

 red-brown precipitate (which is said to strongly retain potas- 

 sium ferrocyanide, together with the copper salt), and dried 

 it in the air without heating. 



Into one limb of a small U-tube, about 9 millim. in diameter, 

 was dropped a short piece of carbon rod of nearly the same 

 diameter as the tube and connected to a platinum wire passing 

 up the other limb for connexion to battery and galvanometer. 

 On the end of the carbon was spread a layer of the precipitate 

 about 1 millim. thick. The whole was then placed in a small 

 glass jar containing strong sulphuric acid, and closed by a 

 paraffin plug through which passed a carbon rod that could be 

 brought down on the precipitate. After standing eight days 

 to dry the precipitate and glass surfaces, the upper carbon was 

 lowered into the tube, but not in contact with the precipitate, 



