FLO WAGE IN CAPILLARY OPENINGS. 139 



According to Poiseuille, the general formula for the flow through a 

 tube of circular section is 



net]) 



in which /is the discharge in cubic centimeters per second, a is the radius 

 of the tube, I its length, p is the difference in pressure at its ends in dynes 

 per square centimeter, and ft is the coefficient of viscosity of the liquid." 

 According to Slichter, "if A is the area of cross section, this formula may 

 be written 



J Snpil 



and the mean velocity of the fluid in the tube is given by 



"=^=(0.08979)^" 



In a triangular tube the flow per second is represented by the formula 

 and the velocity by the formula 





v= (0.02887)^ 



"The mean velocity for a circular tube of equivalent area of cross section 

 was found to be about 38 per cent more." 6 Slichter finds the volume and 

 velocity of flow in an elliptical cylinder to vary but slightly from that of 

 a circular tube. "Even an eccentricity of 0.866 will change the flow by 

 but 10 per cent, and an eccentricity of one-half will reduce the flow by 

 about one-half of 1 per cent. Thus it is clear that a slight change in the 

 shape of the cross section of a tube will change but slightly the flow 

 through it. Analogy warrants us in extending this truth to tubes having 

 other than elliptical sections. For example, we may conclude that the flow 

 through a tube whose section is an oblique triangle is given approximately 

 by the formula for a tube whose section is an equilateral triangle of the 

 same area, even though the shape of the section of the given tube differs 

 slightly, or even materially, from that of an equilateral triangle.'"' Further- 



a Slichter, C. S., Theoretical investigation of the motion of ground water: Nineteenth Ann. Rept. 

 U. S. Geol. Survey, pt. 2, 1899, p. 317. 

 & Slichter, cit, p. 319. 



