The real work of the contact filter is carried on during the period of “resting empty,” that is, after the effluent has been withdrawn from the bed. While the effluent fills the beds, much of the suspended solid matter, together with a large proportion of the bacteria contained in the sewage, adheres to a gelatinous film which has formed on the surfaces of the stones or other materials forming the beds. This interval of “resting full” should usually be about two hours. Then, when the liquid portion is withdrawn from the bed, air is drawn in between the stones, enabling the nitrifying or aërobic bacteria to do their work of breaking down both the suspended and the partially dissolved organic matters which have been contained in the sewage and which have adhered to the filter material. It is believed, that some oxidation of that portion of the organic matter which is in true solution is also accomplished when the effluent passes over the gelatinous covering of the stones by reason of the oxygen which has been absorbed by this covering. The interval when the bed is “resting empty” should be considerably longer than the combined intervals when the bed is filling, “resting full,” and emptying. For this reason there should be a series of from three to five beds in order that it will not be necessary to turn the effluent from the settling tank continuously onto one bed, which would result in the clogging of this bed with suspended matters. The additional third (or fifth) bed also gives opportunity for allowing each bed in turn to be thrown out of use for intervals of a week or so at a time, which is also necessary to keep the beds up to their proper efficiency and obviate the necessity of cleaning or renewing the filter material oftener than once in seven or eight years. [Illustration: PLAN OF SEWAGE DISPOSAL WORKS FOR MR. CHARLES L. A. WHITNEY ALBANY, N.Y. ] In Fig. 46 is shown in plan and section a sewage-disposal plant for the residence of Mr. Charles L. A. Whitney, of Albany, N. Y., consisting of a settling tank, dosing chamber, and contact beds. This plant is designed to serve twenty-five persons, although the settling tanks have a capacity for double the amount of sewage on the usual basis of design. The depth of filtering material in the beds should preferably be four or five feet, although, where operating head or fall is limited, this depth may be decreased to three feet. The floor of the bed should slope toward the outlet end at a rate of about one-eighth of an inch per foot. Various materials are used to form the body of the filter, such as broken stone, coke, broken brick, and furnace slag, but the material used should not be such as will disintegrate readily, and for this reason broken limestone, from one-half inch to one and one-half inches in size, with perhaps two-inch stones for the bottom six inches of the bed surrounding the underdrains, is most suitable for small plants.