110 



CONTRIBUTION TO THE STUDY OF 



its conductivity very difficult, and, as a result, the conductivities 

 given below for hydroxylamine itself may be only approximate. 

 The following table shows the change in conductivity pro- 

 duced in one-tenth and one-hundredth normal solutions of 

 hydroxylamine chloride by coated electrodes. The molecular 

 conductivities (fi) are expressed in terms of mercury at 18° 

 multiplied by 10^ : — 



N/io NHg OH CI. 



N/ioo NH3 OH CI. 



With current 

 passing. 



No current With current 



passing. passing. 



i 



No current 

 passing. 



Time. 



min. 



5 " 



10 " 



15 " 



20 " 



25 '• 



30 " 



35 " 



40 " 



45 " 



/'• 



991 

 10.33 

 1074 

 1106 

 II31 

 1149 

 1168 

 1186 

 1200 

 1213 



Time. 



miu 



15 



30 



45 



f^- 



991 

 1083 

 1134 

 1181 



It thus appears that decomposition takes place somewhat 

 more rapidly when the current is kept passing continuously 

 through the solution, than when it is turned off after each 

 reading. , 



On account of this change in composition, the conductivities 

 of the solutions below had to be determined in a slightly different 

 method from that ordinarily followed. A solution once used 

 could not be used again to make up another solution of less 

 concentration. This used a larger amount of each salt than 

 I had expected, and, as a result, none of the solutions made up 

 were saturated excepting the phosphate. One-tenth normal 

 solutions and those of greater concentration were made up by 

 weighing out the salt directly. One-hundredth normal solu- 



