( 1084 ) • 
D, B,G A/td that tin fimil Pillars tf Threads of Water 
Dd, and Ee^ do not Jl' Je dorrn to F f, and G c, and fo go 
(^uite do!vn, feem to he owing to their Qohe^on with the 
Pillar A 4, which is fufiainA hy the Capillarji Tu^e A: 
, For if you break off the (aid Tube at D the whole Water 
will prefently fink donn. 
As this Solucttm was very different from what I had 
before given, and the Reputation of that Gentleman, 
whole great Knowledge in Experimental Pliilofophy is 
generally known, was fufficient to give weight to any 
of his Opinions, 1 thought my felf under an Obliga- 
tion to examine his account of the Experiment, in or- 
der either to demonffrate its infufficiency, or to retradJ 
my own Solution. Accordingly at the next meeting 
of tltc Society, I produced the foflowing Experiment. 
The Funnel, AFGBC, Fig whole lower part 
B C F G, was Cylindrical to a confiderablc height, and 
W’hofe top was drawn out into a fine Tube at A, 
being fill’d with Water to the height BF, fo that the 
furface of the Water F G, did not reach to the arch- 
ed part of the Funnel, I touch’d the end A with a wet- 
ted Finger, whereby a fmall quantity of Water being 
infinuated into the Capillary Tube at A, the Water 
contain’d in the Funnel W’as fufpended above the Le- 
vel of the Water in the Cillcrn D E, as in the fornier 
Experiment. 
In this Experiment it is manifeft, that the little 
Columns, into which we may luppofe the Cylinder of 
W^ater, FGBC, to be divided, are no way fuffaiii’d 
by the Attra6lion of the arched part of the Glafs above 
ibem, fince they have no contadi with it. Nor is 
there any fuch middle PiUar of Water, which, by its 
contad the Tube at top. is both fuftain’d it lelf, 
and helps to fupport the Pillars about it. Upon the 
