The intermissions usually recommended are 'sewage for six hours and rest for twenty-four hours,' but my belief is that the purifying action of the filter-bed ceases directly the filter is filled and water-logged. The intermissions must be perpetual. The supply of slop-water in a private house is essentially intermittent, and this perpetual intermission is the secret of the success of the methods I have indicated. Between nine at night and seven in the morning—ten hours out of every twenty-four—the flow of slop-water is usually nil. Between 8 A.M. and 11 A.M. is the time of the bulkiest flow, but even this intermits. A housemaid's pail with its three or four gallons will come once in ten or fifteen minutes, so that the filter is always being emptied, and as the water drains off the fresh air follows it. The water of a fixed bath is practically clean, and gives the filtration gutter a vigorous stir, which does nothing but good. The domestic intermissions are invaluable. When sewage is collected in a tank and is then allowed to flow _without intermission_ for six hours at a stretch, it is doubtful if the greatest purifying power is obtained from the filter. RAIN-WATER. Another point of great importance is the bestowal of rain-water. The usual method is to conduct the rain-water from the eaves by means of pipes which open directly into an underground sewer or empty over a gully which runs into an underground sewer. This underground sewer conducts the rain-water either to a main sewer or cesspool, and the important fact to be borne in mind is that the length of underground pipes, whose main function is to conduct rain-water, are nothing but prolongations of sewers or cesspools which conduct the gases of putrefaction to many points round the dwelling, either at the ground level or the roof level. There can be no reason why rain-water pipes should not end in a 'shoe,' and discharge over open gutters which might flow to a gully, if absolutely necessary, at a _distance_ from the house. The practice of taking rain-water direct into underground drains is a great cause of damp walls. A year or so ago the rain-water pipes of a country house well known to the author, which ran direct apparently into the underground drain, were examined. In every case the underground drain was broken and leaky, and in some places completely choked by the roots of plants and trees, while the rain-water got away as it could, and kept the foundations of the walls perpetually soaked.