372 
Proceedings of the Royal Society 
abundance than the colourless. The coloured, by their prepon- 
derance and slight downward tendency in shed blood, retain the 
lighter colourless corpuscles in any position in which they may 
have been placed, unless at the immediate surface, where there is a 
possibility of escape. All observations on this subject must be 
made on freshly-drawn blood, or on the blood as it circulates 
through the vessels. 
With these data a series of experiments was shown on the 
circulation of bodies through tubes. The circulating liquid was 
water, and the bodies employed for circulating in it were made of 
wax and different heavy colouring matters* Experimenting first 
with spheres, it was found that, if a series of these be taken made 
of wax and a metallic powder, so as to toe specifically considerably 
heavier than water , the smaller the sphere is the longer it takes to 
pass cdong the tube. This is explained entirely upon the relationship 
between the specific gravity of the liquid to the objects suspended in 
it. The sphere, when dropped into the tube in which the water is 
circulating, is acted upon mainly by two forces, the one is its 
specific gravity acting in this case in a down- 
ward direction, and which may be represented 
in magnitude and direction by the line AB 
(fig. 1 ) ; while the other force is comprised 
in the strata of water acting in direction AC. 
If the lines AB, AC then represent in 
magnitude and direction the two forces which influence the sphere 
when dropped into the current, it is evident that the direction 
which the sphere will take, immediately after being immersed, will 
be that of the diagonal AD ; and consequently the heavy sphere 
always comes, in the course of time, to the bottom 
of the tube. When in this situation, the smaller 
it is the more it is placed in the slow strata of 
liquid circulating at the periphery of the tube, 
while the larger it is, within bounds which will 
Fig. 2. allow it freely to rotate, the more of the swift 
axial filaments of liquid impinge upon it. Hence the large moves 
faster than the small sphere. Fig. 2 will illustrate this, where 
TTT represents a section of the tube, and abc the spheres of different 
sizes. With spheres of specific gravity lighter than water the 
