Spongy iron is prepared by the reduction of hæmatite ore with fusion, so that the iron is obtained in a porous and finely-divided condition. The Rivers Pollution Commissioners found spongy iron to be “a very active agent, not only in removing organic matter from water, but also in materially reducing its hardness, and otherwise altering its character.” It is a powerful oxidising agent, some of the water being decomposed, and hydrogen set free, and the oxygen acting upon any organic matter present. It also removes lead from water. As already seen, it does not, however, fulfil the primary object of water, by depriving it of any microbes contained in it. =Magnetic carbide of iron= is obtained by heating hæmatite ore with sawdust. Its action is similar to that of spongy iron. The =Pasteur-Chamberland filter= consists of a cylinder of unglazed fine porcelain made from a well-baked Kaolin of a certain degree of porosity and hardness. (Fig. 10.) The water passes through the porcelain from without inwards, and with the pressure of 1½ to 2½ atmospheres which is usually present in the pipes of a water-service, passes through at the rate of about three quarts per hour. The filter can easily be cleaned by brushing it in a stream of hot water, or by subjecting to the heat of a Bunsen burner. The filtration is entirely mechanical, the filtered water being quite freed of microbes. No chemical action takes place. [Illustration: FIG. 10. PASTEUR-CHAMBERLAND FILTER. A.—Outlet of filtered water. B.—Pasteur tube. C.—Metal tube containing unfiltered water. D.—Unfiltered water delivered through tap. ] The =Berkefeld filter= is cylindrical like the Pasteur-Chamberland filter, and is used in the same way. It is made of infusorial earth, which is soft and friable and liable to break. The cylinder becomes gradually worn thin by cleaning, and it then ceases to filter efficiently. Its sole advantage over the Pasteur-Chamberland filter is the more rapid rate of filtration; and against this is to be set the greater liability to fracture and the lack of continuance of efficient filtration. Woodhead and Wood in the report already quoted, state: “The Berkefeld filter appears to have the largest pores among the efficient filters, as is evidenced by the fact that the water organisms were not apparently weakened, that more species of organisms appeared in its filtrate, and that lowering the temperature to 11° C. did not prevent their appearance. The Pasteur-Chamberland filter, on the other hand, at 11° C. was able to give an apparently sterile filtrate for a prolonged period.” More recent experiments have shewn that pathogenic (disease-producing) microbes contained in water after awhile grow through the substance of a Berkefeld filter, and that this does not happen with a Pasteur-Chamberland filter. The latter is therefore preferable.