PURIFICATION OF WATER DURING 1908 AND 1909. During 1908 and 1909 studies on the purification of polluted water by slow sand filtration, by double filtration, and by slow sand and mechanical filters with the aid of coagulants, have been continued. Systematic studies have been carried on concerning the relative efficiency of filters of equal depths and containing sand of the same effective size, but operated at differing rates, and of the effect of the use of calcium hypochlorite and other disinfectants in connection with water filtration. Tests have been made of a system of upward filtration, and also some experiments with filters constructed of broken stone and operated in a manner similar to trickling filters used in sewage purification. As in previous years, special attention has been paid to the work of the filters which purify the water supply of the city of Lawrence. LAWRENCE CITY FILTERS. The water supply of the city of Lawrence is purified by two slow sand filters. The older filter was constructed in 1893. Walls dividing it into three sections were built in 1902. The average depth of sand in this filter is about 4 feet, and the net filtering area, after deducting division walls, gate-chambers and lateral carriers, is about 2.2 acres. As originally constructed, this filter contained two different grades of sand, the filter immediately over the underdrains being composed of finer sand than the remainder of the filter. By reason of scraping, washing and replacing sand during seventeen years, the two grades of sand have become quite thoroughly mixed in the upper layers of the filter, and at the present time this upper sand has an effective size of approximately 0.25 millimeter. This filter is not covered and has an earth and hardpan bottom through which some ground water finds its way into the underdrains, mingling finally with the filtered water. The average rate of operation of this filter during the past two years has been about 1,250,000 gallons per acre daily. During 1906 and 1907 an additional filter was constructed. This filter is of concrete construction, has a tight bottom, is covered by a groined arch roof, is three-fourths of an acre in area and contains about 4.5 feet of sand of an effective size of 0.25 millimeter. It was first put into operation on Nov. 4, 1907, but the filtered water was not turned into the pump-well until Jan. 4, 1908. The rate of operation was at first 1,000,000 gallons per acre daily, but this rate was gradually increased during January and February, 1908, to 3,000,000 gallons per acre daily, at which rate the filter has since been operated. The effluents from both filters flow into the same pump-well, from which they are pumped into the distributing reservoir. In the accompanying tables are shown analyses of the Merrimack River as it flows to the filters, of the effluents from both filters, and of the mixed effluents, not only as pumped into the distributing reservoir but also as distributed to various points after passage through that reservoir. Comparing the results obtained with these two filters during the past two years, it will be noted that the new filter operated at a rate somewhat more than twice as great as that of the old, produced an effluent during 1908 of about the same chemical quality as that of the old filter, with two exceptions, — nitrification was somewhat less marked and there was present only slightly more than one-half as much iron. During 1909 the effluent of the new filter contained considerably less coloring matter, less free ammonia and only about one-third as much iron as the effluent of the old filter, the difference being due to the ground water that enters the underdrains of the old filter and is there mixed with the filtered water. As noted in previous years, the chemical character of the filtered water improved materially as it passed through the reservoir and distributing system of the city. The bacterial efficiency of the old filter during both years was greater than that of the new filter, this superiority being due in part to the much lower rate of operation of the old filter and in part to the fact that this filter has become thoroughly matured by many years of operation. The efficiency of both filters, however, was quite satisfactory. The average bacterial content of the Merrimack River water applied to these filters during 1908 was 6,500 per cubic centimeter. The effluent of the old filter contained on an average 90 bacteria per cubic centimeter, and the effluent of the new filter 120 bacteria per cubic centimeter, giving bacterial efficiencies of 98.6 and 98.2 per cent., respectively. During 1909 the river water averaged 7,400 bacteria per cubic centimeter, the effluent of the old filter, 60 bacteria per cubic centimeter and the effluent of the new filter 100 bacteria per cubic centimeter, giving bacterial efficiencies of 99.2 and 98.7 per cent., respectively. During both years. the B. coli efficiency of the old filter, operated at the lower rate, was considerably greater than that of the new filter. |