436 Distribution of Pressure around Spheres in Viscous Fluid. 

 hemispheres of the two spheres at corresponding points. The 



beino- asymmetrical with 



the 



curve Demo- asymmetrical witn respect to tne line joining 

 them, only the figures for the first quadrant are given . 



Table YI. 



Angle. 



Outer 

 hemisphere. 



Angle. 



Inner 

 hemisphere. 



Dif. 

 ferenee. 



Eesultant 

 component. 



360° 



•5000 



o 







•5000 



•0000 



•0000 



330 



•2202 



30 



•1790 



•0412 



•0206 



300 



-•3480 



60 



-•4211 



•0731 



•0617 



270 



—•6340 



90 



-•7130 



•0790 



•0790 



Table III. gives the results for a viscous fluid. The curve 

 (fig. 12) is asymmetrical with respect to both axial planes, 

 and it is clear from the form of the curve that the pressure 

 in the inner hemisphere is greater than the pressure on the 

 outer hemisphere. The pressure on the inner hemisphere is 

 at 30°, 3*3 per cent, of the normal pressure at 0° in excess of 

 the pressure at 360°^ and at 60° it is 5"7 per cent, of the 

 normal pressure greater than the corresponding pressure at 

 300°. 



For a perfect fluid, therefore, two spheres moving with 

 constant velocity perpendicular to the line joining their centres 

 attract, and for a viscous fluid they repel. 



I have shown in a former paper "^ that when two particles 

 in a sound-wave are a certain critical distance apart they are 

 attracted when their line of centres is parallel to the stream- 

 lines and repelled when their line of centres is perpendicular 

 to the stream-lines. The spheres used in these experiments 

 were relatively large compared with particles or sphere that 

 w^ould form flutings in a sound-wave. The results, however, 

 agree with the results obtained with the smaller sphere in a 

 sound-wave. I hope soon to be able to determine the pres- 

 sure around spheres small enough to form flutings in a 

 sound-wave. 



The experimental work included in this paper was con- 

 ducted under the direction of Dr. Brace in the Physical 

 Laboratory of the University of ^S^ebraska, and my sincere 

 thanks are due to him for valuable suggestions during the 

 progress of the experiments^ and also for his assistance in 

 determining the curves of distribution for a perfect fluid. 



Physical Laboratorv, Case School of Applied Science, 

 Clevelaid, Ohio, April 23, 1903. 



* Phil. Mao-. Mav 1902. 



