24 Experiments and Observations 
When the wire was very fine, about the +80th of an inch in 
diameter, its heat increased in very combustible mixtures, so as 
to explode them. The same wire in less combustible mixtures 
only,continued bright red, or dull red, according to the nature 
of the mixture. 
In mixtures not explosive by flame within certain limits, these 
curious phenomena took place whether the air or the inflamma- 
ble gas was in excess. 
The same circumstance occurred with certain inflammable 
vapours. I have tried those of ether, alcohol, oil of turpentine 
and naphtha. There cannot be a better mode of illustrating the 
fact, than by an experiment on the vapour of ether or of alcohol, 
which any person may make in a minute. Let a drop of ether 
be thrown into a cold glass, or a drop of alcohol into a warm 
one. Let a few coils of wire of platinum of the 1-60th or 1-70th 
of. an inch be heated at a hot poker or a candle, and let,it be 
brought into the glass ; it will in some part of the glass become 
glowitig, almost white hot, and will continue so as long as a 
sufficient quantity of vapour and of air remain in the glass. 
When the experiment on the slow combustion of ether is made 
in the dark, a pale phosphorescent light is perceived above the 
wire, which of course is most distinct when the wire ceases to 
be ignited. This appearance is connected with the formation of 
a peculiar acrid volatile substance possessed of acid properties. 
The chemical changes in general produced by slow combus- 
tion appear worthy of investigation. A wire of platinum intro- 
duced under the usual circumstances into a mixture of prussic 
gas (cyanogen) and oxygen in excess became ignited to white- 
ness, and the yellow vapours of nitrous acid were observed in the 
mixture. And in a mixture of olefiant gas non-explosive from 
the excess of inflammable gas, much carbonic oxide was formed. 
I have tried to produce these phenomena with various metals ; 
but I have succeeded only with platinum and palladium ; with 
copper, silver, iron, gold, and zine, the effect is not produced. 
Platinum and palladium have low conducting powers, and small 
capacities for heat, compared with other metals; and these seem 
to be the principal causes of their producing, continuing, and 
rendering sensible these slow combustions. 
I have tried some earthy substances which are bad conductors 
of heat ; but their capacities and power of radiating heat appear 
to interfere. A thin film of carbonaceous matter entirely de- 
stroys the igniting power of platinum, and a slight coating of 
sulphuret deprives palladium of this property, which must prin- 
cipally depend upon their increasing the power of the. metals to 
radiate heat. i 
Thin laminz of the metals, if their form admits of a free cir- 
culation 
