184 
Proceedings of the Royal Society of Edinburgh. [Sess. 
The explanation of the formation of these hollow globes would appear 
to be as follows. When the water flows out without circular motion, the 
lines of flow are all radial, and as these converge on the centre of the 
opening they give rise to a reduction in the area of the jet a short distance 
from the opening. But when the escaping water has a circular motion, 
whenever it escapes from the constraint of the surrounding water its 
centrifugal force tends to cause it to move outwards, and it would fly off 
Fig.i 
in spray but for the surface tension of the water. This tension, however* 
yields to the centrifugal force and the jet expands; but in expanding the 
water loses its angular velocity and therefore some of its centrifugal force.. 
In this process the water carries the expansion further than it can main- 
tain it, and the surface tension draws it back, but in so doing increases the 
angular velocity of the water, thus enabling it to form a second bulb ; and 
this see-saw between the circling water and its surface tension goes on as 
long as the circular motion continues. The loss of circular motion accounts 
for the diminishing diameter of the globes, and the increasing rate of fall 
of the water for their greater length lower down. 
In Part I on this subject it was pointed out that, when water escapes 
