188 Mr. J. Spiller on the Composition of 
duct being usually collected from the flasks at intervals of two 
or three days. 
In this manner, during the remarkably brilliant days in June 
and July 1857, no less than forty-six gallons of the standard 
solutions were submitted to treatment, and a comparatively con- 
siderable quantity of the darkened material procured for in- 
vestigation. 
The appearances presented during these trials were often such 
as to indicate a reduction more complete than that which would 
probably be required on the subchloride hypothesis :—first, a 
thin pellicle of high metallic lustre, white as silver, was usually 
seen floating on the surface of the liquid, and the internal walls 
of the flasks were frequently coated with a film much resembling 
the condition of silver reduced by any of the so-called “ silver- 
ing” processes. The product also, although containing admixed 
chloride, was susceptible of a high degree of lustre on being 
burnished in an agate mortar. 
In colour the products of several experiments varied a little,— 
sometimes presenting a dark purple-grey appearance, at other 
times the grey was slightly tinged with green; and to this 
depth of colour is probably attributable the circumstance that 
sometimes, during exposure to the burning rays of a midsummer 
sun, the contents of the flasks attained a degree of temperature 
which should be recorded as a condition of experiment, although 
such heat may not be supposed to have exercised any special 
influence in determining an abnormal decomposition. On the 
24th of June, 1857, a hot sunny day with cloudless sky, the tem- 
perature of the dark liquids in three of the flasks reached 110°, 
115°, and 116° Fahr. respectively; whilst at the same time, 
3 p.M., a thermometer in the shade registered 83°, and only 91° 
with the sun shining freely on the mercury in the bulb; such 
high degrees of temperature were, however, unusual and never 
again observed. 
With a view to counteract the possibility of the material 
undergoing alteration by drying, it was constantly preserved 
under water, and in the moist condition submitted to numerous 
experiments for the purpose of ascertaining its constitution. 
On determining by chemical analysis the composition of an 
average product, it was found to contain a larger proportion of 
silver than the original white chloride, as will appear in the 
following comparison :— 
Composition of Found in 
white chloride. grey product. 
Give net gine > PORGy 6) aie bee 
Chiormess| ja 24:74. 4) sence c) BAGO 
100:00 100°0 
