Inquiries into the Principles of Liquid Attraction. 87 
Fig. 4.) Two of these globules, when brought within the distance 
of about an inch of each other, rushed together with a force which 
was constantly increased as they approached. (See Fig. 5.) These 
globules produce a very sensible depression of the liquid, and equally 
avoid the pin and the sides of the cup. 
These experiments were all made about twenty years since, and 
from that period, my inquiries have been occasionally directed to the 
discovery of those substances which are attended with an elevation 
of the liquid around them, and to those which have a depression. In 
my first experiments, vegetable leaves seemed to depress liquids, 
and glass and metals to elevate them ; but by giving them differ- 
ent positions, and consequently different angles of inclination to 
the surface, I discovered that all substances, when their angles of in- 
clination are sufficiently acute, are attended with an elevation of the 
liquid; and when sufliciently obtuse, with a depression. A green 
vegetable leaf, placed perpendicularly in water, has on both sides a 
depression, but give the leaf a suitable inclination to the surface, and- 
the liquid is elevated on the side of the acute angle, and depressed 
on the side of the obtuse angle. 
A fine needle, also, carefully placed on water, is known to swim. 
Now the convex surface of the needle, (the needle being considered 
a cylinder,) presents no difference of angles to the surface of the 
liquid, on its revolving. y By the angles which the convex surface 
makes with the surface of the liquid, is meant, the angles which are 
made with the surface, by the tangents, drawn to all the different 
points in a circular ring, taken about the needle, and produced to 
meet the surface of the liquid. For it is well known that a line from 
any point in this circular ring, makes the same angle with the surface 
of the liquid, which the tangent of that point makes with the same 
surface. The same explanation is also applicable to the convex 
surfaces of the globules of mercury. But when these globules, or 
the needle, descend into the liquid, the tangents of these convex 
surfaces, and consequently the convex surfaces themselves, make all 
the varieties of angles with the surface of the liquid, first the acute, 
and then the shine and, as it has already been remarked, when 
any substance ronkits, angles with the surface sufficiently acute, it is 
always attended with an elevation of the liquid ; and whenever suffi- 
ciently obtuse, with a depression. This is the case in the descend- 
ing of the needle ; and it is proved to be the case, by inserting the 
point of a needle in a small piece of wood, of equal length with the 
