Mr Kemp's Descrvptiim of' an Improved Blowpipe. 341 



a quarter of an inch in diameter. The one, C c, terminates at 

 the under surface of the cork c, and is open into the vessel ; ^ it 

 is bent, as shown in the figure, and is brought at its extremity 

 C to a fine point, for emitting a stream of air to urge the flame. 

 The other tube, D rf, also passes through the cork, and is termil 

 nated into a bottle-like tube E ^, a sufficient space being left be- 

 tween the under part of the tube D d, and the bottom of thie 

 bottle E e, for the passage of the air, which is forced in at the 

 other extremity of the tube at D. The small bottle-Hke tube 

 E e contains a portion of mercury, below the surface of which 

 the tube D d terminates, after passing through a cork at the 

 upper part of it E, and to which it is firmly attached by means 

 of a sealing of wax ; two small openings^ f^ are cut in the. cork, 

 to allow the air to pass freely through, into the large glass- 

 vessel. 



A small stand F, for supporting a spirit-lamp, is attached to 

 the neck of the vessel, which may be raised or lowered by a 

 screw, to allow the flame to be acted, upon by a stream of air, as 

 it issues from the tube C 



In using this blowpipe, all that is necessary is to adjust the 

 lamp, by means of the screw and stand, till it be opposite the 

 tube C. 



If we then blow into the tube D d, we will force in a quan- 

 tity of air, which passing through it, and the mercury contained 

 at the bottom of it, in the bottle E ^, rises and is forced througli 

 the openings f^f into the large vessel. This additional quan- 

 tity of air within the vessel, exerts a pressure on the surface of 

 the mercury, and raises a column of it in the tube D d, which, 

 acting as a valve, shuts the communication between the internal 

 and external air, and effectually prevents any of it from return- 

 ing through the tube D tZ, while, at the same time, the condensed 

 air within the glass-vessel, from its greater elasticity, flows 

 through the other tube C upon the flame of the lamp ; and as a 

 period of from one to two minutes elapses before it arrives at 

 the density of the atmosphere, a continued stream of air may be 

 made to act on tlie flame, by occasionally blowing into the tube 

 D, while the operator has the free use of both hands, which 

 is of some importance in experimenting on minute })ortions of 

 matter. 3 



