t 2 95 ]

blast issued from the air pipe ; when the pipe was stopt,the water in the receiver was forced lower and lower, andwas soon driven entirely out, abundance of air bubbles fol-lowing it into the water in the outer vessel.

When the funnel was kept entirely full: though thestream was directed as before against its side, there werelittle marks of any air being carried down. And whenthe funnel was near empty, the effects were also incon-siderable j the vacuity in the middle of the spiral circum-volutions of the water seeming to reach to the bottom,so as to suffer the air to escape upwards through the hollowcolumn of water.

Water falling from a consderable height into a funnel •with

a pipe.

A leaden pipe, six feet high and an inch and a halfin diameter, was inserted into an air vessel, with the watergage already described. Into the top of the pipe was fixeda tin funnel, whose throat fitted close to it; and into thefunnel a stream of water was let fall, from a reservoir fivefeet above, in quantity sufficient to keep the funnel run-ning over. This apparatus represents Mariotte’s blowingmachine described in the third article of the precedingsection.

The water, divided by the fall, pushed down abund-ance of air with it: a strong blast issued from the blow-ing pipe, and the gage rose high. On raising up thefunnel a little, the stream that issued from it appeared allfrothy : as often as the funnel was lifted up, the gagefunk, the air, which had been driven in by the dash ofwater, escaping between the funnel and pipe : on lettingdown the funnel close, the gage immediately rose again.

Instead