110 1 WE O RFY> ‘or 
lf there be any wind at the time of the phenomenon, 
the aerial funnel through which the water afcends, inftead 
of being perpendicular to the horizon, as it would be in a 
calm, might incline more or lefs to it, in proportion to the 
ftrength or weaknefs of the prevailing current of air: Or, 
inftead of continuing in one fpot, it might have a progref- 
five motion over the furface of the fea, in the direétion of 
the general current; both of which circumftances frequently 
take place. In either cafe it is natural to fuppofe, that both 
air and water would afcend {fpirally, as through the worm 
of a ferew, every current which fets in towards the centre 
receiving an oblique bias from the prevailing current. 
It fometimes happens, that after the fubfiding of a fpout, 
it is fucceeded by a fecond, and that by a third, either in 
the fame place, or at no great diftance from it. But this 
alfo is analogous to what we obferve upon the plunging 
of heavy bodies out of air into water. For, after the firlt 
fubfiding of the {mall column of water which is occafion- 
ed by it, and is above refembled to an icicle, the water 
again rifes and fubfides as at firft, though not in the fame 
degree, as may be concluded from thofe fainter concentric 
circles which expand from the fame centre after the fub- 
fidence of the firft column. ‘The fame thing which here 
takes place in water, may alfo take place in air, under 
fimilar circumftances. 
Since writing the foregoing, while I was endeavouring 
to contrive fome experiment to illuftrate the fubje&, a very 
fimple one was fuggefted to my mind, the fuccefs of which 
I think demonftrates the truth of the hypothefis introduced 
above to account for the firft afcent of the water in the 
{pout ; the event being precifely the fame as was expeCted 
before hand, and as ought to have taken place upon the 
principles above advanced. 
EXPERUM EN T. 
In a ftiff paper card I made a hole juft big enough to 
infert a goofe quill fo as that it might be fixed perpendi- 
cularly 
