OF THE FREE DIATOMACEiE. 
27 
the bottom, the others being about eight inches apart. 
Each hole is provided with a short tube, of glass or gutta 
percha, and a stopcock, or spring tap, so that water may 
be drawn into vessels placed below, through either of the 
holes, and in any quantity required. At the apex of the 
cylinder a funnel, g , is placed, and to it is attached a narrow 
glass tube, 7i, of sufficient length to reach almost to the bottom 
of the cylinder. To the lower extremity of the tube, h , a 
kind of mouthpiece of glass is joined (a short piece of 
india-rubber tubing will effect the junction), drawn out 
into a very fine opening. Above the fhnnel, g , either 
placed on a stand or fixed against the wall, is a tin vessel, 
j, with a short pipe, Jc , leading into the funnel. This vessel 
is for the reception of water. 
The method of using this little apparatus is as follows : 
— The Diatomacea^, roughly freed from mud and sand, are 
poured into the cylinder, a, b, and immediately sink to the 
bottom. The funnel, g , being replaced, the tap, Jc , is 
opened, and water begins to flow down the inner tube, li. 
The effect of this is to cause a grand disturbance among 
the imprisoned Diatomacese, which are forced up the sides 
of the cylinder, a, b , the lightest of course being the 
highest. On opening the uppermost, c, of the four side 
sluices, the water, loaded with Diatomacese, issues into a 
wineglass, or some similar vessel, and this should be 
continued as long as the fluid remains thick. As soon as 
it appears clear and bright, let the next tap, d , be opened 
(the upper one still being allowed to discharge its contents), 
and the water caught in a second glass. This second glass 
will then contain all those Diatomacese which were too 
heavy to rise to the level of the uppermost tap, c, but at 
the same time too light to sink below the level of the 
third tap, e. And so the process is continued until each 
of the taps has been opened. 
It will be well to bear in mind, that the flow of water 
from the reservoir, can be regulated very nicely, and the 
quantity discharged lessened or increased, by means of the 
