Theory of the Pneumatic Paradox. 307 



lateral hole, the jet rose vertically, as in the preceding case, to the 

 height of forty seven inches. The end of the tube was then 

 unstopped, and the jet becoming inclined at an angle of forty or 

 forty five degrees with the horizon, attained the height of sixteen 

 or seventeen inches above the level of the tube, and the distance 

 of thirty inches before descending to the same level. The force 

 of the jet was not apparently diminished by inclining the reser- 

 voir, so as to give the tube an ascending or descending direction. 

 By diminishing the head of water, a corresponding diminution 

 was produced in the height of the jet. 



Experiments similar to the two last described, with the excep- 

 tion of the substitution of drawn leaden tubes, both with and 

 without an ajutage, and of the same dimensions with those em- 

 ployed by Bossut, gave results not materially different from the 

 preceding. The tubes were one eighth of an inch in thickness, 

 which interfered with the natural direction of the jet, and somewhat 

 diminished its height, and it was almost impossible to pare away 

 the lead about the holes sufficiently to obviate this obstruction, 

 without impairing the cylindrical shape of the tubes. On this 

 account, therefore, and because of the imperfection of their inte- 

 rior surface, experiments made with them are less to be relied on 

 than those which I have already described. They were suffi- 

 cient to show that mere difference of size produces no material 

 discrepancy of results. 



The foregoing experiments demonstrate a pressure against the 

 inner surface of horizontal and inclined tubes, independent of 

 that which arises from the column of water contained in them. 

 Of this pressure the lateral jet, though it proves its existence, 

 cannot be regarded as a measure. Its oblique direction, and the 

 consequent obstruction presented by the thickness of the tube, 

 together with the diminution of pressure arising from the expen- 

 diture of the jet itself, prevent it from reaching the height due 

 to the actual lateral pressure exerted by the flowing water, when 

 the hole is closed. 



Experiment IV. In order to determine whether lateral pressure 

 takes place in vertical descending tubes also, I procured a cylindri- 

 cal vessel of tinned sheet iron, thirteen inches in height and nine 

 inches in diameter ; to the centre of the bottom, where a perfora- 

 tion had been made, was soldered the triblet tube, furnished with 

 an ajutage, used in Experiment III, having two lateral holes, each 



