368 
hol largely diluted with water we may arrange to with- 
draw part of the alcohol at one particular place by 
promoting its rapid evaporation, and thereby increase 
the surface-tension of the liquid layer in that region by 
diminishing the percentage of alcohol which it contains. 
In this shallow tray, the bottom of which is of ground 
glass resting on white paper, so as to make the pheno- 
mena to be exhibited more easily visible, there is a 
thin layer of water coloured deep blue with aniline ; 
now, when I place on the water-surface a small quan- 
tity of alcohol from this fine pipette, observe the effect 
of bringing the alcohol-surface, with a surface-tension 
of only 25°5 dynes per lineal centimetre, into contact 
with the water-surface, which has a tension of 75 dynes 
per lineal centimetre. See how the water pulls back, 
as it were, all round the alcohol, forming a circular ridge 
surrounding a hollow, or small crater, which gradually 
Fic. 27.—Section of the air-bubble in a level tube filled with water, and bent 
so that its axis is part of a circle of large radius; scale is represented in 
Fig. <8. 
widens and deepens until the glass plate is actually laid 
bare in the centre, and the liquid is heaped up in a 
circular ridge around it. Similarly, when I paint with a 
brush a streak of alcohol across the tray, we find the 
water drawing back on each side from the portion 
of the tray touched with the brush. Now, when I incline 
the glass tray, it is most interesting to observe how the 
coloured water with its slight admixture of alcohol flows 
down the incline—first in isolated drops, afterwards 
joining together into narrow continuous streams. 
These and other well-known phenomena, including that 
interesting one, “tears of strong wine,” were described 
and explained in a paper “On Certain Curious Motions 
Observable on the Surfaces of Wine and other Alcoholic 
Liquors,” by my brother, Prof. James Thomson, read 
before Section A of the British Association at the Glagow 
meeting of 1855. 
I i} 
Fic. 28.—Represents a length of one centimetre for Figs. 24 to 27. 
I find that a solution containing about 25 per cent. of 
alcohol shows the “ tears” readily and well, but that they 
cannot at all be produced if the percentage of alcohol 
is considerably smaller or considerably greater than 25. 
In two of those bottles the coloured solution contains 
respectively I per cent. and 90 per cent. of alcohol, and 
in them you see it is impossible to produce the “tears” ; 
but when I take this third bottle, in which the coloured 
liquid contains 25 per cent. of alcohol, and operate 
upon it, you see—there—the “tears” begin to form 
at once. I first incline and rotate the bottle so as 
to wet its inner surface with the liquid, and then, 
leaving it quite still, I remove the stopper, and with- 
draw by means of this paper tube the mixture of air and 
alcoholic vapour from the bottle and allow fresh air to 
take itsplace. In this way I promote the evaporation of 
NATURE 
[August 19, 1886 
alcoho] from all liquid surfaces within the bottle, and 
where the liquid is in the form of a thin film it very 
speedily loses a great part of its alcohol. Hence the 
surface-tension of the thin film of liquid on the interior 
wall of the bottle comes to have a greater and greater 
value than the surface-tension of the mass of liquid in the 
bottom, and where these two liquid surfaces, having 
different surface-tensions, come together we have the 
phenomena of “tears.” There, as 1 hasten the evapora- 
tion, you see the horizontal ring rising up the side of the 
bottle, and afterwards collecting into drops which slip 
down the side and give a fringe-like appearance to the 
space through which the rising ring has passed. 
These phenomena may also be observed by using, in- 
stead of alcohol, ether, which has a surface-tension equal 
to about three-fourths of that of alcohol. In using ether, 
however, this very curious effect may be seen.! I dip the 
brush into the ether, and hold it near to but not touching 
the water-surface. Now I see a hollow formed, which 
becomes more or less deep according as the brush is 
nearer to or farther from the normal water surface, and it 
follows the brush about as I move it so. 
Here is an experiment showing the effect of heat on 
surface-tension. Over a portion of this tin plate there is 
a thin layer of resin. I lay the tin plate on this hot 
copper cylinder, and we at once see the fluid resin draw- 
ing back from the portion of the tin plate directly over 
the end of the heated copper cylinder, and leaving a 
1 See Clerk-Maxwell’s article (p. 65) on ‘* Capillary Attraction”’ (“‘ Ency 
clepxdia Britannica,’’ oth edition). 
