BETWEEN THE FREEZING AND BOILING-POINTS. 
161 
For a cell about the mean temperature of 39°, with the hexahydrate crystals as 
solid phase, the formula 
E, = E 39 - 1-000 (t - 39°) - -0070 (t - 39°) 2 
was obtained. For temperatures above 30°, as the second formula shows, the values 
given by the first formula diverge from the observed values, due probably to 
a secondary change produced by the decomposition of the mercurous sulphate. If 
the first formula is corrected by the additional term 
- -00006 (t - 15 0 ) 3 , 
the calculated values from 30° to 40° C. are brought into very close agreement with 
the observed values. 
The Clark cells I have used in the present work are some of the original crystal 
cells described in the paper by Professor Callendar and myself, “ On the Variation 
of the Electromotive Force of the Clark Standard Cell,’’ already referred to. These 
cells have been in the laboratory since 1895, and frequent comparisons made of their 
E.M.F. with newer cells constructed at different times, both by myself and the 
advanced electrical engineering students. As these cells are the originals from which 
the temperature formulae already given were obtained, the constancy of the E.M.F. 
maintained to the present time is a matter of some surprise, considering the severe 
treatment they were subjected to during our earlier experiments. They were made 
in the generally accepted way in a test-tube, and sealed by means of a cork, on the 
top of which marine glue was melted. The life of such a cell is necessarily dependent 
on the speed with which the crystals commence to dry, and this fact has been raised 
against the use of moist crystals in place of a saturated solution. I have found 
however, that in point of usefulness our crystal cells have outlived several cells with 
saturated solution which were made at the same time. It appears that the crystals 
retain the moisture more tenaciously than the saturated solution does, so that whereas 
a solution may be reduced to one-third of its original bulk, with deposition of 
crystals, a mass of crystals retains its moisture without diminishing in bulk or 
uncovering the zinc rod. Owing to the dryness of the Montreal climate, the question 
of the slow evaporation of liquid from the cells is a serious one. Our cells have been 
re-sealed on one occasion by simply re-melting the marine glue, but apparently 
without harm except to one (X 3 ), which when left undisturbed for several months 
returned to its original normal condition, and is at present as good as the others. 
Several sealed cells, inverted cells, and a number of new crystal cells have since 
been made in the laboratory, and have served to check the constancy of the original 
crystal cells. Independently several cadmium cells were made in 1897, in the 
inverted form, which proved to be quite satisfactory, and a comparison of the mean 
of these cells was made with the mean of the crystal cells. These cadmium cells 
VOL. cxcix.—A. 
Y 
