Theory of the Pneumatic Paradox. 313 



in mercury contained in a basin supported in the upper part of the 

 box. Having forced into a condensing chamber, by means of a 

 syringe, three or four times as much air as it naturally contained, 

 he directed a very rapid current through the tubes A and B, and 

 the mercury was depressed from C to D, more than two inches, 

 in consequence of the rarefaction which the current produced in 

 the air of the box. 



Experiment 11. In order to determine to what degree air can be 

 condensed by means of the mouth and lungs, I poured a quantity 

 of mercury into a glass tube, bent into the form of the letter U, 

 and furnished at one end with a brass cap, stop-cock, and mouth- 

 piece. The greatest difference that could be produced in the 

 height of the mercury in the two branches of the tube by blow- 

 ing through the stop-cock, was found, in a series of trials made 

 by half a dozen able-bodied men, to vary from two and a half to 

 nearly four inches. This result shows, that the degree of con- 

 densation producible by this means, ranges from a twelfth to less 

 than a seventh of an additional atmosphere. The condensation 

 producible by blowing through a tube presenting no obstruction 

 to the escape of air, is obviously much less. One of the individ- 

 uals mentioned above was able to exhaust fourteen fifteenths of 

 the air from a tube having the lower end immersed in mercury, 

 and the other furnished with a stop-cock, as was shown by the 

 rise of the mercury in it to the height of twenty eight inches. 



Experiment Ifl. Make a cylindrical tube of tissue paper, about 

 three fourths of an inch in diameter and five or six inches 

 long, and fit into one end of it a circular piece of wood, having 

 a hole in its centre one quarter of an inch in diameter. On blow- 

 ing through the hole the tube will collapse. On blowing through 

 a metallic tube of the same size, having a similar circular piece 

 of wood in one end, and lateral holes, the flame of a lamp will 

 be drawn into the holes. In both these cases the current from 

 the mouth communicates a share of its outward motion to the 

 lateral portion of air with which it comes in contact, and being, 

 as has been already shown, but slightly condensed, becomes more 

 rare, when it expands to fill the tube, than the external air ; which 

 is prevented from rushing in at the end of the tubes to restore 

 the equilibrium, by the impulse of the blast. 



Experiment IV. Lateral holes were made in a cylindrical brass 

 tube, about five inches long and three sixteenths of an inch in 



