150 PROCEEDINGS OF THE AMERICAN ACADEMY. 



becomes negligible with a given current density, the deposition of 

 hydrogen interferes with the exact determination of the electrochemical 

 equivalent, even before hydrogen is actually evolved in bubbles. Hence 

 the use of a small cathode is dangerous beyond a limit which depends 

 upon the accuracy required. With small current density an approximate 

 correction may be made for the area of the cathode. 



5. Some possible causes of error in the silver voltameter are obviated 

 by a simple device ; and with the help of this apparatus the older 

 methods are compared with one another. The danger of a fallacy in 

 such a comparison is pointed out. 



6. The electrochemical equivalent of silver appears to be about 

 0.0011172 gram per ampere per second. 



7. The electrochemical equivalent of copper appears to be between 

 0.00032915 and 0.00032925 gram per ampere per second. 



8. The discrepancy between the chemical atomic weights of copper 

 and silver and their electrochemical equivalents is explained ; hence 

 Faraday's law is verified with two cathions more precisely than before. 

 Conversely, assuming Faraday's law to hold rigidly, and the value 63. GO 

 to represent the true atomic weight of copper, the agreement furnishes 

 evidence of the accuracy of the new method. 



Harvard University, Cambridge, Mass., 

 October 14, 1899. 



