[ 13 ° 1 
or foiA^ times through the air, find them not only 
cold, but hardened by their fudden cooling, To as 
to be fit for cutting other parts of the fame fteel, 
which has been cooled more gradually. Now the 
velicles being exceedingly fmaller than the points of 
fuch drills, muft cool much quicker, and the in- 
cluded air be condenfed equal to the circumambient 
air almoft inflantaneoufly j in which ftate the veficle 
is not fitter for afcending than that portion of water 
was before the veficle was formed : for though the 
furface of the water is vaftly increafed by the form 
of the veficle, which may retard the defcent of it 
through the air, yet that form muft equally obftrudt 
its afcent ; and the fpecific gravity of the conftituent 
particles remaining the fame, I cannot fee how the 
increafe of furface can aid its afcent. And here I 
muft obferve, that it has not yet been explained, how 
water can be dilated fo, as to occupy eight hun- 
dred or a thoufand times thefpace, which it naturally 
does. For I think the greateft expanfion it fuffers 
by boiling (the greateft beat we can give it) is not 
more than a nineteenth or twentieth part of what it 
was before. 
But fuppofing thefe veficles formed and afcending, 
there is ftill a very neceflary power wanting, I mean 
repulfion, to prevent their running into contact (by 
their natural attradlion), and forming larger mafies 
in which ftate they would immediately defcend. 
Now let us fuppofe, that they afcend by impulfe y it 
will be very difficult to explain their motions, which 
are evident to fenfe. For if they afcend by impulfe, 
their motion muft be continually retarded by their 
gravity, and the obftrmftions of the aii*, until they 
4 ariive 
