306 Theory of the Pneumatic Paradox. 



Experiment II. A reservoir was made fire feet high and eight 

 inches square, and into one of its vertical sides, six or eight inches 

 from the bottom, was set, hke a pane of glass in a window, a 

 square piece of tinned sheet iron, to the middle of which had 

 been soldered, so as to have one end exactly even with its inner 

 surface, an English smooth-drawn or triblet tube, twelve inches 

 and a half long, and three eighths of an inch in diameter. This 

 kind of tube was selected on account of its being, from the man- 

 ner in which it is made, perfectly cylindrical. The one used in 

 this experiment, which was the longest in proportion to its diam- 

 eter I could obtain in this city, had a very highly polished inte- 

 rior surface, and was in every respect suited to afford exact re- 

 sults. A lateral hole, one line in diameter, which was now di- 

 rected upwards, had been previously made in the middle of it, 

 with great care to have no protrusion on the inside, and not to 

 impair in the slightest degree its cylindrical shape. The outer 

 end of the tube being stopped with a cork, and a strip of oiled 

 silk tied ovev the hole, the reservoir was filled with water. On 

 removing the oiled silk, a vertical jet rose to the height of forty 

 seven inches. The reservoir being kept full, and the cork with- 

 drawn, the water flowed through the tube, and the jet, not ceas- 

 ing as in the experiment of Bossut, assumed an oblique direction, 

 making an angle with the horizon of forty or forty five degrees. 

 The greatest height it attained above the level of the tube, which 

 had been previously adjusted in a horizontal position by means 

 of a spirit-level, was twelve and a half inches, and the distance 

 to which it was projected before descending to the same level, 

 was twenty four inches. 



Experiment III. To one end of a tube, similar in every respect 

 to the one used in the preceding experiment, was soldered, in 

 order that water flowing through might entirely fill it throughout 

 its whole extent, an ajutage nearly of the form of the vena con- 

 tracta. The accompanying fig- Fig. 4. 



ure represents, though imper- 

 fectly, the tube with the ajut- 

 age soldered to a piece of tinned 

 sheet iron, which was set in the 

 side of a reservoir, and every 

 thing arranged as in the preceding experiment. The end of the 

 tube being stopped, and the water allowed to issue through the 



