CAPILLARY RIPPLES. 49 
in fact, to make an experiment with a magnet upon it. At first one 
would say, judging by the common information that is to be found in all 
books dealing with capillary phenomena, that a liquified wire could 
not exist for an instant. The fact is that all writers who have 
dealt with capillary forces, myself among the number, take the greatest 
pains to show that a cylindrical column of liquid launched into space 
possesses a form which cannot exist except for a moment. The two 
photographs that I showed here last year, and I will do so again if 
I may, illustrate what happens when a liquid cylinder is launched into 
space; and our good friend the diadema spider has come to our aid to 
provide an illustration. Here (exhibiting a slide) I have a photograph 
of one of those diadema spiders resting on her web; I took it by about 
two minutes exposure with limelight. What I want to illustrate is 
this: that while these radial webs are smooth, and necessarily for the 
purpose of scaffolding, these other finer webs that go round and round 
in an irregular spiral are beaded. The next slide (exhibiting the same) 
I took with the microscope of part of that very web, and it shows the 
exact structure of the web as it exists when you see it magnified. If 
you watch a spider spinning one of those webs where there is a dark 
background and a bright sun shining you will see that the circumfer- 
ential webs when spun are shiny, and in the reflected light you 
recognise a polished wire; but gradually the shininess vanishes and a 
frosted appearance takes its place. The extremely viscous liquid with 
which the line is in effect wetted furnishes, of course, that brilliant 
reflecting cylinder which we at first see. But the liquid cylinder cannot 
exist ; it is an unstable form, and it breaks up into microscopic beads of 
an extremely regular pattern alternately large and small, as Plateau, 
the blind Belgian Physicist, originally showed. Now if you take a 
wire of aluminium and send an electric current through it so strongly 
as to melt it, the liquid wire should not be able to exist. I shall ina 
moment connect the ends of this piece of aluminium wire with the 
terminal of the battery in the next room which is providing the electric 
light. Aluminium is a good conductor of heat and electricity, but in 
spite of that, owing to the enormous current, the wire will rapidly be 
raised to the melting point. But aluminium of all metals oxidises so 
violently that it is a most dangerous thing to mix it with materials that 
contain oxygen because they may produce the most violent explosions. 
Nevertheless, melted aluminium remains bright in a crucible ; it needs 
no flux; itis spoilt by a flux. The result is that the wire does not 
necessarily oxidise through and take fire as one would expect. But 
one would say if it remains a bright metal cylinder, and is absolutely 
fluid, the form is an unstable form, it will break up into beads in the 
same way that the spider’s web, or rather the liquid upon it, broke up. 
But there is some oxide, and there is an excessively thin film on the 
surface, not enough to see, but sufficient to prevent the wire at any 
part getting thinner than anywhere else; there is enough to prevent 
the unstable form breaking up, and therefore for a few seconds the wire 
remains suspended between the posts, white hot and absolutely fluid ; 
and having got this white hot wire absolutely fluid we have for the first 
time the means of examining feeble magnetic forces in the most 
