BETWEEN THE FREEZING AND BOILING-POINTS. 
165 
the metallic mercury. In preparing the crystals for these cells it was deemed 
unnecessary to follow the old prescription, inasmuch as they were made from the 
purest anhydrous salt purchased from Merck. However, a few of the cells were 
made from crystals that had been treated when in solution with a small quantity 
of Hg 2 SOj,, but filtered out before re-crystallizing. 
In the table of comparison now given the inverted cells are expressed as difference 
from the mean of the five crystal cells X l3 X 2 , X 3 , X 10 , and X n . The temperature 
changes between 15° and 0°, and 15° and 30° are also given, as determined by the 
different students who made the cells. 
Table II.—Comparison of Inverted Cells to Mean “ Crystal ” (1897-98). 
Cell. 
Difference from mean 
crystal in mvolts. 
E.M.F. changes between 
15°—0°. 
15°—30’. 
XR n . 
+ 0-16 
+ 16-67 
- 19-45 - 
XR x ,. 
+ 0-17 
16-67 
19-46 
XR,3. 
+ 0-20 
16-69 
19-58 
xr ]6 . 
+ 0-19 
16-66 
19-60 
XR ir . 
+ 0-10 
16-67 
19-61 
XR 1S . 
+ 0-20 
16-69 
19-55 
XR in . 
+ 0-24 
16-60 
19-51 
Means 
0-166 
16-67 
19-54 
These tests are sufficient to show that the inverted cell gives a value somewhat 
in excess of the older crystal cells. The temperature change between 15° and 0° 
is also somewhat larger than the value given by the crystal cells, which was 
+ 16‘62 mvolts. The reason for this may possibly be, as I have already pointed out 
in another place, that the sensitive electrode (negative) is at the bottom of the cell 
and deeply immersed in the bath, whereas in the crystal cells the negative electrode 
is a zinc rod at the top of the cell, and although immersed below the level of the 
liquid in the temperature bath, may yet conduct an appreciable amount of heat and 
be at a slightly different temperature to that of the cell. 
I have used the term sensitive for the negative electrode because a small difference 
of temperature between it and the other parts of the cell influences the E.M.F. very 
considerably. This can be very forcibly shown in the case of an inverted cell by 
removing it from the 15° bath and standing it on a cold surface. This produces 
a larger change (increase) in the E.M.F. than if the complete cell were immersed at 
the cooler temperature ; and in a similar way for a higher temperature, the negative 
electrode being warmer than the other parts of the cell, the E.M.F. decreases more 
than it would were all the cell at the same temperature. Hence, in the crystal cell, 
