of Edinburgh, Session 1883 - 84 . 
421 
riment was made with the copper disc with the view of ascertaining 
whether an increase of temperature had an effect upon the amount 
of electricity produced by the whisk. The disc was heated to such 
a temperature that it could not he held in the hand with any degree 
of comfort ; the deflections were — 
100, 100, 120, 85, 170, 170, 105, 115, 155, 165, 180, 145, 120, 
180, 200, 
giving an average of 140, compared with 126 before heating. 
When a brass disc was heated in a similar manner, the average 
deflection was increased in much the same proportion. It is not 
improbable that the effect is due to the heat of the disc increasing 
the insulation of the brush (see p. 428). 
On the 29th hTovember the surfaces of the discs were polished 
with a fine sand paper. The several observations were — 
Copper, 215, 220, 210, 150, 167, 165, 140, 165, 165, 215, 145, 180, 
215, 185, 140. 
Zinc, 35, 70, 80, 70, 90, 40, 110, 75, 45, 110, 70, 85, 90, 65, 75, 
110 . 
The average for the copper is 178 ; for the zinc 45, if the first read- 
ing is not taken into account and the last is, and 42 if the first is 
and the last is not. The smallness of the first deflection is to be 
attributed, I believe, to a few of the minute particles of zinc or of 
sand produced in the cleaning process still remaining on the surface 
of the metal. Many of the subsequent experiments were made with 
the view of clearing up this effect, and the still greater effect gene- 
rally produced when zinc is polished with emery paper. 
The second series of observations were made with coins and a 
platinum capsule, while discs of other metals were being prepared. 
A half-sovereign (11/12 fine gold), a sixpence (37/40 fine silver), and 
a cent (88 Cu to 12 Ni), are of nearly equal diameter, and therefore of 
nearly equal capacity. If the cent be taken as 100, the sovereign 
is 123, and the sixpence 71. The second comparison was obtained 
by taking a florin and a penny (95 Cu to 4 Sn and 1 Zn) ; the 
result obtained by supposing the bronze equivalent to copper gives 
a result for silver (109), which agrees well with the best strict com- 
parison (107). The result of 4th December, gold to silver as 193 to 
