On Water and A ir. 
519 
1880J 
planes with the lead weight attached. The receiver is 
placed air-tight upon a circular plate, and connected to an 
air-pump by the tube T. By means of the pump we have 
exhausted the receiver of air, and the lower plane, with its 
lead weight attached, is hanging there, adhering firmly to 
the upper plane. Had a weight of 100 lbs., instead of 
20 lbs., been attached to the lower plane, it would have been 
sustained by the powerful attraction of the two surfaces, so 
that this clinging force, whatever it is, is independent of the 
pressure of the atmosphere. It is, in point of fact, part and 
parcel of the force which causes the individual molecules of 
the air itself to cling together. If you bring the two sur- 
faces closely together the molecular cohesion comes into 
play, and holds the plates together. This experiment shows 
that in this particular, at all events, Boyle was not corredt 
in his conclusions ; but the number of things in which he 
was correct is simply wonderful. He observed, as I have 
said, those changes in the height of the mercurial barometer, 
and he observed many other things. Usually the credit of 
proving that sound cannot pass through a vacuum was 
ascribed to a most meritorious man, and far be it from me 
to say a word against him, — to Hawksbee, in 1709, — but 
nearly half a century previously Boyle actually put his 
watch under the receiver of his air-pump, and he found that 
when he exhausted the receiver the ticking of his watch 
ceased to be heard ; so that nearly half a century before the 
time of Hawksbee, Boyle had proved this fadt. This error 
on his part, in his conclusion as to the cause of the clinging 
of the plane surfaces, is a mere trivial thing compared with 
the amount of truth which he established. 
Now I have done referring to the transmission of the 
pressure through the water towards this little Cartesian 
diver. I do not know whether the experiment which I am 
now going to make is destined to succeed, because I had not 
the opportunity of making the experiment beforehand, having 
only one of these bottles at hand ; but I want to see whether 
the transmission of a shock through this water will be suffi- 
cient to produce a striking mechanical effedt which will be 
the means of showing the transmission. Here is one of 
what are called Rupert’s drops. It is a drop of glass cooled 
quickly, and drawn out into a long tail. I his glass is in 
such a state that the slightest crack or rupture, such as the 
breaking off of the long tail, causes the whole thing to fly 
into powder. As I said, this rupture of the Rupert’s drop 
is accompanied with a powerful mechanical shock. Here is 
a bottle of water (b, Fig. 41) ; I do not know whether it is 
