336 Frederick Guthrie on certain 
several cases ; and therefore these top, or richest, amalgams 
oan scarcely be compared. Again, the shape of the bottom of 
the tube with its capillary &c. puts the lowest or poorest out 
of court. But as the contents of the lower, irregular part of 
the tube is not more than a third of the volume of the unit 
measure, it is only the very lowest amalgam that need be 
rejected. 
In each case there were twelve full unit vessels drawn off, 
and in each case a fraction of a thirteenth, which last contained 
the drainage from the metal. 
Through the kindness of Dr. Hodgkinson a number of 
these amalgams were analyzed in the chemical laboratory by 
Messrs. Adie, Gahan, and Grange, to whom I am therefore 
indebted. These three gentlemen analyzed the zinc, lead, and 
tin amalgams respectively. The metals were determined in 
the following manners: — 
Lead. — The amalgam dissolved in nitric acid and evapo- 
rated with sulphuric acid, and the residue either ignited 
directly or after washing with dilute alcohol (as sulphate of 
lead). 
Tin. — The amalgam dissolved in nitric acid, evaporated to 
dryness, and ignited (as metastannic acid). 
Zinc. — (a) By dissolving in nitric acid, evaporating to dry- 
ness, and igniting ; or (J3) by separating the mercur}- as sul- 
phide and the zinc as sulphide, and igniting (both as oxide of 
zinc). 
§ 32. In Table I. the results of such determinations are 
given, so that the proportion of the errors of analysis may be 
compared with the true diffusion in each, and the difference 
of diffusion in the three cases. 
At the end of the experiment the cylinders of tin and lead 
presented nothing remarkable in appearance. On standing a 
couple of months the upper part of the lead cylinder has 
become as hard as zinc, though there is no sensible deforma- 
tion. The zinc cylinder swelled considerably in the tube ; 
and when left to itself afterwards, though drained from the 
mercury, it continued to swell and crack, and ultimately fell 
to pieces like a " lime-light " lime cylinder when slaked. Two 
cones with their apices towards the centre of the cylinder 
were formed at top and bottom ; the cracking otherwise was 
for the most part in radial planes. 
In Plate V. the percentages of metal in the several amal- 
gams of the three metals are given graphically, and without 
rounding off or other interpolation. The abscissa? reckoned 
from the left are distances from the bottom; the ordinatesare 
the corresponding percentages of the respective metals. 
