[ S a6 ]

and for augmenting its power with its surface,, two or threegratings, or perforated plates, with apertures of differentsizes, will be necessary : one with very fine holes, much.smaller than those of the cullender, that nothing may passthrough the former which can be in danger of sticking inthe latter: another with larger apertures, for detainingweeds, and such other matters as would soon obstruct thefiner strainer.

I have tried other methods of procuring this dispersion,of the water, by making the throat of the funnel os differ-ent figures ; but with little success. Whether the throatwas made converging or diverging, in greater or less de-grees, there did not appear to be any material' differencein the effect of the machine. I introduced into the funnela cylindrical core, which was fixed in the middle, bymeans of pins projecting from it,, so as to leave a cir-cular aperture all round it •, and this core was fognetimessolid, and sometimes a pipe which reached above the fun-nel and carried down air into the middle of the jet below :but no other difference was observed, in either case thanwhat arose from the necessary diminution of the quantityof water. It is probable indeed* that by duly propor-tioning the core to the funnel, and the width of the pipeto the sheet of water falling round the core,, the effect, bythis division of the stream, would be made greater than anequal quantity of water would produce when'falling in onecolumn,* though the increase,, obtainable by this method,did not promise to be considerable enough to deserve the.troublesome investigation of the proportions. One trial,however, depending partly on this principle, appeared, ofsome importance to be made..

As the water machine of St. Pierre is said to have two>apertures in the bottom of the funnel, whose streams, asthey issue out, cross one another and are dashed into drops,

I tried.