22 Trans. Acad. Sci. of St. Louis. 



The result of the two determinations was 



b+T" 100 



b ~ 100.7855 



b-\-P' _ 50 



~1) ~ 50.3945 



= 0.99221 

 = 0.99217. 



It thus appears that doubling the angular velocity of the 

 tube does not affect the ])ressure at the axis of rotation, if 

 the linear velocity of the free end is held unchanged by 

 reducing the length to one-half its former length. This of 

 course also means that the pressure P at the outer end is 

 also unchanged under the same conditions. Eq. (1). 



In integrating the function contained in the denominator 

 of (2), the rectangular area whose length is in the one 

 case r = 50, and in the other 100, and whose breadth is 



(y2^.2 



■^^ -^ j— ^> could be left off from the diagram, since its 



/ r = 



integral value was in the one case 50, and in the other 100. 

 The curve could thus be drawn upon a large scale. The area 

 actually measured was therefore that corresponding to the 

 two fractional values 0.7855 and 0.3945. 



