APPENDIX.

Summary of results obtained by Colonel G. E. Waring, Junr., M.lnst. C.E., at Newport, Rhode Island, U.S.A.

All the figures are corrected for loss due to draining the straining tanks before aeration.

The percentage of purification is estimated from ammonia figures, and represents

1. In column marked A the percentage of nitrogen removed from the sewage after passing through the straining tank.

2. In column marked B the further percentage of nitrogen removed from the effluent after passing through the aerating tank.

The first, therefore, is a measure of the purification of the original sewage on leaving the straining tank. The second set of percentages is a measure of the further purification of the effluent by the aerating tank.

TANK No. 1 STRAINER.

Filtering surface To of an acre.

Filled with coarse broken

stone. Voids of stones after drainage=43 per cent. Capacity for sewage 450 gallons, being allowance for drainage.

TANK No. 2 STRAINER.

Filled with fine crushed stone. Filtering surface 1980 of an acre. Capacity for sewage 450 gallons, being allowance for drainage. Voids of stones after draining=44.1 per cent.

TANK No. 3 STRAINER.

Filled with coarse beach gravel. Filtering material surface 1980 of an acre. Capacity for sewage 350 gallons, being allowance for draining. Voids of gravel after draining=32.1 per cent.

TANK NO. 4 STRAINER.

Filled with fine beach gravel. Filtering surface 1980 of an acre. Capacity for sewage 350 gallons, being allowance for draining. Voids of gravel after draining=34.4 per cent.

TANK No. 5 AERATOR.

Filled with fine beach gravel. Filtering surface 12 of an Voids of gravel after draining=34.4 per cent.

acre.

TANK NO. 1 AERATOR.

Filled with coke (screening half). Filtering surface of Voids of coke=38.9 per cent.

anacre.

Prof. W. H. CORFIELD (London) regarded the experiments made by the Massachusetts Board of Health as very largely confirmatory of our English experiments in the filtration of sewage and its application to land. The experiments of Frankland and the practical results obtained by Bailey Denton and Baldwin Latham taught that the use of sand and gravel was quite effective as a filtering medium for sewage if properly applied. The experiments made in sewage filtration and published by the Sewage Committee of the British Association showed that filtration was as effective in Winter as in Summer. The Massachusetts experiments were valuable chiefly as affording confirmatory proof of our previous experience. The experiments referred to in Sir Douglas Galton's paper further evidenced the fact that filtration was even in Winter far more effective than chemical precipitation, and also that acid sewage could be treated if a layer of limestone were placed in the upper part of the filtering media. In water filtration it was shown that the results were more satisfactory if the scum which gradually formed on the top of the VOL. XVII. PART I.

B

filter was allowed to remain, but the Massachusetts experiments showed that the scum formed by sewage on land should be systematically broken up. Here no new fact was brought to light, for it was well known to the sewage farmer that the results of ground enriched by sewage manure were better if the surface of the land treated was broken up.

Mr. S. R. LowcOCK (Birmingham) agreed that the principles on which the Massachusetts experiments were based were perfectly well known in this country, and what they had done was to demonstrate practically that these principles were correct, and they were therefore most valuable. The theory of sewage treatment whether on land or by means of special filters is exactly the same, and the failures in sewage purification which not infrequently occur are due to neglect of the principles on which this theory is based. The error of those people who say that the engineer of to-day knows nothing (or very little) about sewage treatment is due to the fact that such persons know nothing of the extensive literature of sewage treatment which now exists, and do not read up or in other ways do not make themselves acquainted with the real facts. He was very pleased to know that in America Col. Waring had obtained similiar results to those he himself had obtained in this country, the only practical difference between Col. Waring's system and his own being in the method by which the filtering medium was aerated. If Col. Waring's straining tank were superimposed upon the aerating tank the arrangement would be almost identical with his own, but the aeration would be from the bottom upward to the middle layer, whereas in his own it was downward from the middle layer to the bottom. He quite endorsed the conclusions as to the clearing of the straining material, as he had found that the surface sand removed from his filters and merely turned over in the presence of air in the course of a very few days became perfectly clean and could be re-used. In comparing the results obtained by various systems, not only as between America and this country, but in different towns and districts in the same country, due consideration must be given to the different characteristics and strengths of the sewage treated, as these differences control the qualities which can be dealt with on any given area, and the measure of purification which can be obtained. His own experiments showed this very clearly; at Malvern Link, where the sewage is of a purely domestic character, he had obtained a purification of over 99 per cent. from the tank effluent to the filtrate, while at Wolverhampton with exactly similar arrangements the purification, calculated on the same basis, so far has only reached 85 per cent. This is entirely due to the character of the sewage, which is quite abnormal and extremely difficult to treat. There are a large number of galvanizing and other works in the district, and in spite of vigilance on the part of the Authorities large quantities of acid waste are discharged into the sewers, or, when the waste is treated at the works quantities of chloride of calcium are discharged. The result is shown in the following table by the unusually high amount of chlorine.

The Table shows a set of twelve analyses, six of tank effluent, and the corresponding six of filtrate, taken at various periods extending over twelve months, during which time the filter has been continuously at work. These samples are an average of twelve hours' flow in each case. The filter was started in October, 1894, and the first pair of samples were taken a month afterwards, and the others at intervals until the last were taken on October 8th, 1895. The rate of flow was 1,000,000 gallons per acre when samples four and five were taken, and 500,000 gallons when the other samples were taken. The analyses are interesting as showing the effect of the chlorides on the purification, as, although throughout the whole series the results, as judged by the oxygen consumed and the large increase in the nitrates, show continuous improvement; yet where the chlorine is high there appears to have been a retarding action which prevented the breaking down of the albuminoid ammonia, or combined it in some stable form, the nitrification reducing the free ammonia in larger proportions, while when the chlorides are lower the albuminoid ammonia is reduced in larger proportions. The nitrifying organisms are evidently very active, and this is further proved by the fact that none of the samples of filtrate have undergone putrefaction, but have almost uniformly developed a green growth. His own experience agreed with that of the Massachusetts experimenters and Col. Waring, that a very small quantity of air is sufficient, and that nothing is gained by an excess, and that where the aeration is sufficient very complete purification of the sewage is effected, and the filtering media are kept perfectly sweet and clean for apparently an indefinite period.

Mr. SCOTT-MONCRIEFF (London) said that when he commenced his experiments the reports of the Massachusetts Board of Health were not make known. As a matter of fact the first biological filter bed ever used in England had been constructed by him at his own house and practically applied in July, 1892, and since then many others had been adopted. His investigations had been carried out upon lines which differed greatly from those referred to in the paper and