90 On the Decomposition of the Earths. (Jan. 
some expedient, whereby, making allowance for the probability of 
an explosion, the operator may be protected from injury. Various 
methods have been proposed ; such, for example, as that of having 
different reservoirs for the two gases; so that their union may only 
take place immediately preceding, or in the moment of their com- 
bustion ; which does not answer; because it is difficult, if not 
impracticable, to measure the discharge of the two gases; allowing 
exactly two portions of hydrogen gas to enter into combination with 
one of oxygen gas ; and unless this proportion be observed, the gas 
will not burn well. Another plan proposed was to surround the 
apparatus, either with folds ef cloth, or witha cage of stout wire, 
or with a case of malleable iron; all of which have their inconve- 
niencies, which it would be tedious to mention. The best method 
that I have yet tried was recommended by our Professor of Chemistry, 
the Rev. J. Cumming. Mr. Newman sent to me a blow-pipe con- 
structed according to Professor Cumming’s plan. It contains a 
small cylindrical chamber, which is to be half filled with water, 
through which the gas is made to pass in its passage to the jet of the 
apparatus. If an explosion take place, it extends only to the sur- 
face of the water, and does not communicate combustion to the 
compressed gas in the reservoir. Towards the bottom of the 
cylinder, for containing the water, there is placed a wire-gauze ; 
and there are other trivial cireumstances which tend to render the 
apparatus secure ; but these I shall not now particularly detail, be- 
cause Mr. Newman himself proposes to publish an account of this 
blow-pipe. Suffice it only to say, that, with all the advantages of 
this ingenious contrivance, an explosion will sometimes happen ; and 
one has actually happened; my own apparatus, thus constructed, 
having exploded this day. If during the partial explosions which 
extend to the surface of the water, this fluid be driven into the re- 
servoir, or if through any inattention in the operator the handle ot 
the syringe be drawn out, while the stop-cock below it is open, 
previously to the introduction of the gas into the reservoir, then, 
the air in the reservoir being partially exhausted, the water will rush 
into it, and an explosion becomes extremely probable; nor will the 
wire-gauze prevent it; as it has been proved by the explosion I 
this day witnessed. However, this new contrivance is a very good 
one, provided the operator do not attempt to exhaust the reservoir of 
atmospheric air, and will be at the pains to listen, and to ascertain 
whether the water boil, owing to the passage of the gaseous 
bubbles, before he ventures to ignite the gas. My object is, to 
suggest an expedient, whereby, whatsoever explosion may happen, 
whether using Mr. Newman’s original blow-pipe, or one of those 
made according to Professor Cumming’s improvement, the operator 
may be perfectly secure from danger. Such an expedient | have 
adopted, since I sent my last letter to you; and, in the security it 
offers, I have been enabled to continue my experiments; although 
I have witnessed two explosions with the utmost impunity. Its 
simplicity may perhaps recommend it; and that my description of 
it may be perspicuous, it will be sent to you accompanied by a 
drawing. (Plate LX.) 
