138 
Proceedings of the Royal Society 
leaking througli the holes thus made, they are rendered quite 
air-tight by embedding the screw heads, which are sunk slightly 
below the surface of e, in red lead, and by laying, previous to 
screwing down, layers of red lead on both sides of the rubber sheet. 
The red lead should not, however, spread beyond the disk e on either 
side of the rubber sheet, so that the latter lies quite free except where 
it is held down by the disk. In h is fitted a single-bore rubber 
cork holding the brass tap /. This tap is simply an ordinary brass 
tap converted into a three-way tap by boring a hole g through one 
outer wall and through one wall of the plug, which enables the 
operator to establish communication between 
(1) The bell-jar and pump. 
(2) The outer air and both bell-jar and puuip. 
(3) The outer air and pump, the bell-jar being shut off. 
(4) The outer air and bell-jar, the pump being shut off. 
After fitting in the cork and tap, the knee-piece h (fig. 2) is screwed 
on. The object of this knee-piece will be explained later ou. If 
the mouth a be closed by a rubber stopper, and the tap / connected 
by a rubber tube with a water-pump or other exhausting apparatus 
as the pressure inside diminishes, the rubber sheet bulges up inside 
the bell-jar, and pressing against the lower edge closes up any 
interstices due to irregularity of its own surface, or to imper- 
fect grinding of the glass. A well-made water-pump will give 
within half an hour a high degree of exhaustion, and this without 
the use of any lubricant whatever. Where a very high degree of 
exhaustion is required, the application of a little grease outside 
round the lower edge of the bell-jar is advisable. The apparatus in 
this form can therefore be used for drying substances in vacuo, &c. 
Tor the purpose of rapid filtration such complete exhaustion is not, 
as a rule, required, and indeed is often positively detrimental, and 
defeats the object in view. 
All the essential parts of the apparatus have been now described, 
everything else which is required for filtration being either in 
ordinary laboratory use, or else can be made with but little time 
and trouble, and at almost no cost. 
When it is desired to collect the filtrate in a beaker or flask, the 
simplest arrangement is that represented in fig. 1. Fixed in a by 
means of a rubber stopper is an ordinary correct-angled funnel. 
