The patentees claim that this funnel may be used for a variety of purposes besides that of ordinary filtration, such as vacuum filtering, washing precipitates automatically, dialysis, etc, besides being a great improvement on the usual pattern. The following is the result of a comparative experiment with the above funnel. Time required to filter 4 pints of liquid, - No. 1. Ordinary funnel, plain filter paper, 50 minutes. No. 2. Patent funnel, plain filter paper, 23 minutes. No. 3. Ordinary funnel, plaited filter paper, 8 minutes. The high price of the above funnel is its chief objection, otherwise it gives very good results.

The most perfect way perhaps of utilising a paper filter is that suggested by Dr. Symes. He makes a linen cone and attaches it at the top to a wooden ring resting on an earthen jar. The linen forms a support for the filter paper, and a suitable cover prevents evaporation.

There is one point to which great importance should be attached in the consideration of this subject, and that is that there is a certain material or combination of material best suited for the filtration of any given liquid, and much time may frequently be saved by carefully noting the filtering medium best adapted to each particular fluid. For instance, a strong infusion of poppy capsules precipitated by rectified spirit filters best through swansdown, liquid extract of bael through paper, a strong infusion of senna precipitated by rectified spirit through flannel, etc, and every liquid which presents any difficulty requires intelligent treatment according to its nature. One combination of filtering materials seems specially adapted to the filtration of syrups, flannel coated with raw paper pulp. Syrups pass through such a filter with comparative rapidity, and the filtrate is as brilliant as it is possible to obtain it even through paper of the finest texture. The bag should be made of very coarse flannel, and filled with a mixture of paper pulp and water which has previously been boiled until the pulp is quite disintegrated.

As the water runs through the bag the pulp is left as a uniform layer on its interior.

There are several methods of assisting filtration by producing a vacuum in the receiving vessel, e. g. with the Bunsen pump; also an arrangement for producing a partial vacuum by connecting the receiving vessel with a bottle full of water placed at a height and communicating with a similar bottle at a lower level. The upper bottle acts as an aspirator, and when empty the positions of the bottles can be reversed and the action rendered continuous. The pressure exerted by a column of liquid may be utilised in two ways, either to force the filtrate through in the usual direction, or to make it pass upward through the filtering medium. In the first case a reservoir is placed at a convenient height, from which proceeds a tube terminating in a ring or collar, to which the filtering bag is securely attached, the height of the reservoir determining the pressure on the contents of the bag. This method is very well suited for thick viscid liquids, which contain but little sediment, and which filter too slowly under ordinary atmospheric pressure.

In the second case the filter is attached to the short limb of a siphon, so that the liquid passes in an upward direction through the filter, and herein lies its great advantage, viz. that the solid portion of the liquor, instead of settling on the filter and choking up its pores, tends to move away from it towards the bottom of the vessel.

The apparatus employed for centrifugal filtration consists of a shallow metal box supported horizontally on an axis, and capable of being revolved with great rapidity by means of suitable gearing. Inside is a similarly shaped but rather smaller box made of perforated metal or gauze, and fitting into the outer case so as to leave a space all round. The inner case is lined with flannel, forming a bag into which the liquid to be filtered is poured. On setting the machine in motion, the liquid is powerfully forced against the sides of the flannel bag; the clear portion then passes through into the annular space between the two cases, and leaves the sediment behind in the bag. A high speed is necessary to obtain good results.

Of the various methods just described none lends itself more readily to the filtration of thick viscous liquids (never ending sources of difficulty when considerable quantities are concerned) than that known as "upward" filtration. In this, advantage is taken of the pressure of the atmosphere by using a long column of liquid to suck the filtrate through. The apparatus usually employed for carrying out this process consists essentially of au inverted open box or cone, over the mouth of which the filtering medium is stretched, the cone being connected at its apex with the short limb of a siphon. The advantages of this arrangement are that by lengthening the siphon leg the rapidity of filtration may be increased to any reasonable extent, and by its position the filtering medium is prevented from becoming choked up with deposit, the sediment tending to move away from the filtering surface rather than to settle upon it. If a liquid containing a suspended precipitate be allowed to rest, a zone at the surface becomes clear in a comparatively short time, whilst that near the bottom still remains turbid.

As a comparatively clear liquid filters much more quickly than a thick and muddy one, the position of the filtering cone (which must of necessity be placed near the bottom of the containing vessel) in an upward filtration arrangement is hardly correct, as the filtrate is drawn from a layer of liquid which is much more turbid than that at the surface. The correct position of the filtering cone

Is evidently at the surface of the liquid, so that only the clearest portion may hare to pass through the filter, and full advantage be taken of the clarifying effect of subsidence.

In order to overcome the difficulties encountered, Bird" devised the apparatus thown in Fig. 39 with a view of applying the principle of npward filtration, avoiding aiposure to air at any stage of the process, and keeping the filtering medium in the moat advantageous position, viz. at the surface of the liquid.

A is a stoneware jar, of about 2 gal. capacity, placed on a shelf at a height of 5-6 ft. above the vessel II. It is secured to a board C of suitable dimensions, perforated by a circular bole q. B is of wood, 3 in. wide, and also perforated by two holes x and r, C, A, and B are securely fastened together by a string or other suitable means. J is the filter proper, and consists of a circular box, closed at the top and open at the bottom, and about 3/4 in. teas in diameter than A. J is divided at the centre by a partition, which thus forms an air-tight chamber in the upper portion. The tube t passes through this chamber and communicates with the lower half of A, its upper end being connected to the glass tube H by rubber tubing. Over the open mouth of J is stretched the filtering medium, consisting of three layers, calico, paper, and flannel, the latter being on the outside. E is a bar of wood to which the glass tubes H and I are firmly attached, J, H, I, E forming a rigid system, partially counterbalanced at its centre of gravity by the weight F through the cord and pulley G, the whole being capable of free motion up and down, so that J rises and falls with the liquid in the interior of A. The tubes H and I work through the holes x, r, and q, which serve as guides.

The weight F should be such that when A contains no liquid J just descends freely to the bottom of the jar. K is a piece of rubber tubing connecting I and L, so as to allow of the free motion of I. L is a glass tube passing into the bottle M are required to draw off the filtrate from M.

Filtering Water Part 4 50030Upward filtration.

Upward filtration.

If it is desired to start filtration, A is filled with liquid, when, by the buoyant action of the air chamber in J, aided by the weight F, the filter rises to the surface. A cork is inserted in the bottom of L, and the end of the rubber tube V is removed from I; through V and I, I K L and H J are filled with liquid (preferably bright). The connection at V is again made and secured, and a layer of colourless heavy petroleum oil about 1/3 in. deep is poured on the two surfaces 0 of the liquids in A and M. As soon as the end of it is enclosed, filtration commences and goes on continuously. The oil is of course unnecessary in the case of liquids which do not suffer by exposure to air. Glass jars, furnished with stop* cocks at the bottom, may be substituted for the vessels A and M with considerable gain in convenience. All joints must be bound with waxed thread or wire, and thick rubber tubing used, to avoid collapse of its walls and consequent stoppage of the flow; the filter J should also be well varnished with shellac dissolved in methylated spirit.

The construction of this filter demands but a small amount of mechanical skill; it works continuously, requires but little attention, and perfectly protects the liquid passing through it from the action of the atmosphere at any stage of the operation.