DILUTE SOLUTIONS AT THE FREEZING POINT. 337 



of these strong solutions ultimately used, make the use of glass harmless for the 

 purpose of storing them. 



A second stock solution (b) was then prepared from one of these (a) solutions. 

 About 2 grammes of (a) were weighed out in the small filling machine, and run into 

 a platinum bottle. Water of the same quality as that to be used as solvent in the 

 platinum cell was then added till a dilution of about one in forty was reached. 

 This gave to solution (b) a concentration of about one-hundredth normal, and 

 enabled a solution of about one hundred-thousandth normal to be obtained in the 

 cell by the addition of about *25 of a gramme of stock solution. 



The atomic weights used in calculating the equivalent concentrations were the 

 same as those taken by KOHLRAUSCH for his latest results. The following list shows 

 their values : 



Hydrogen .... 1*008 Potassium .... 39'14 



Carbon 12*00 Chromium .... 52*14 



Nitrogen 14*04 Manganese .... 55*0 



Oxygen 16*00 Iron 56*02 



Sulphur 32'06 Copper 63*6 



Chlorine 35'45 Barium 137*4 



SECTION 9. On the Method of Experimenting. 



The method of conducting a series of measurements was as follows : 



The cell was thoroughly washed, water being put in and sucked out with an air- 

 pump several times. The copper tank was then filled with broken ice, and about 

 250 grammes of water of the best quality were run into the cell, which was then left 

 for several hours till the temperature of the contents had sunk nearly to zero. It 

 was found advisable to fill the cell the evening before an experiment was made, and 

 leave the water in it for the night. The level of the water was adjusted with the 

 emptying apparatus, and its temperature lowered, if necessary, by evaporating ether 

 through the spiral coils, till it stood within one-tenth of a degree of zero. The 

 resistance was then measured at intervals of five minutes, the screw being worked 

 between each observation. 



The conductivity of the solvent having reached a practically constant value, a 

 small quantity (usually about *25 of a gramme) of stock solution (b) was placed in 

 the filling machine, and run into the cell. As soon as it had entered, the screw was 

 turned till mixture was complete. This, it was found, was usually the case after 

 about six revolutions of the hand wheel, but more were generally given. A resist- 

 ance measurement was then made and repeated at intervals while the weighings 

 necessary for the next experiment were being performed. The value of the resistance 

 should be practically constant throughout. 



A second quantity of stock solution (b) was then added and the process repeated, 



VOL. cxciv. A. 2 x 



