90 Comparison of the Blowpipe of Hare and Brook. 



gases to flow decidedly to the focus, and with such rapidity 

 as to keep up a constant supply. As there is nothing to 

 confine the stream of gas — no reacting force, particularly 

 on the sides of the stream, except simply the pressure of 

 the atmosphere, will not the effect of great pressure be, to 

 dissipate laterally, a portion of the gases, so that they will 

 not arri^^e at the focus, either burning out of it, or possibly, 

 even to some extent, escaping combustion. 



In Brook's instrument also, were a high pressure deemed 

 of advantage, it is evidently ver}'' transient, as, the moment 

 the gases begin to flow, it is diminished, probably, in a ge- 

 ometrical ratio, and it is very soon reduced to a pressure 

 not superior to that exerted in Mr. Hare's. From the small 

 capacity of the reservoir, the entire duration of one charge 

 of gas is also very short, and it is burnt out in a minute or 

 two. It is true, it is possible in some measure to compen- 

 sate for this defect, by employing an assistant to inject gas 

 constantly while the flame is burning, but this, besides the 

 iiiconvenience of the thing, would probably occasion vibra- 

 lions in the flame. 



The reservoir for Brook's blowpipe may, indeed, be made 

 ijlronger and larger to any extent — but who (considering the 

 evident hability to explosion which, after all the ingenious 

 contrivances to prevent it, does occasionally happen) would 

 wish to be in the vicinity of even a cubic foot of these gase?^ 

 which, when kindled, would, of course, exert, in their ex- 

 plosion, a force proportioned to the resistance of the walls 

 of the reservoir, and might therefore propel fragments, witi^ 

 oven fatal velocity.* 



" Tbnt lliis inforeiicc is not unfounded, will perhaps appear from the fo!- 

 lovving fact. We had been accustomed (borrowing the experiment from 

 1he late Dr. Woodhouse, of Philadelphia) to mix fulminating mercury anri 

 oxy muriate (chlorate) of potash by agitating them on an open paper by a 

 fculher; they were tiicn gently poured into the bowl of a common tobaccc- 

 pipe, and exploded by |)ouring sulphuric acid from a long pole ; the explo- 

 Mon was always very violent, but no iiarni lioppened from the fine frag- 

 ments, into which tlie pipe bowl was always broken. 



On one occasion, we substituted for the pipe bowl, a common tinplate can 

 die extinguisher, placing it with the vertex, of course, downwards, in an iron 

 f-andlestick, which was intended to actas a support. We were aware, that 

 liy the explosion, the tin recipient would be torn, but we imagined that the 

 fiagtneats would hardly be projected, or, if they were, it was concluded 

 that, on account of their thinness, they would not be propelled to any 

 distance, nor with sufficient momentum, to do any harm. But when the 

 cxplosiun occurred, the iron candlestick was split, although no part of i' 



