[ 96 ] 



diminifhed in the ratio of 140 to 86,4 j that is, in a ratio between 

 the ratios of the fquare root of 2 and the fquare root of 3 

 to I. 



But this method alfo is fubjed to inaccuracy ; for the motion 

 of the fluid is found not to be very regular towards the end of 

 the flux ; it will therefore be better to calculate the time in which 

 the vcflTel fliould empty itfelf to a certain depth, which is done 

 in the following manner : The times in which the whole and part 



would be evacuated are rcfpedively 7^x^/5 ^^^ o ^ v^ ~7 ' ^ 



denoting the height of the part. Therefore the difference is 



;=:— ;--^ X v^ A - v^ P, in fcconds. 



Thus let the altitude of water in a vcff'el wholly and in part 

 filled with water be 16 and 12 inches, the diameter of the cylin- 

 dric veflTel 5,74, and the diameter of the circular aperture ,2. 

 Then, by the foregoing formula, the time in which the water 

 Ihould fubfide, according to theory, from the height of 16 to 

 12 inches, would be 33 feconds. But the time actually found 

 by experiment is 53 feconds j therefore the velocity determined 

 by theory is to be diminiOied in the ratio of 53 to 33, or 1,6 

 to I, t. e. very nearly in the fame ratio as determined by a for- 

 mer experiment, in which the fpouting fluid was mercury. 



