The best of these precipitants give a removal of only 65·8 per cent. of organic nitrogen, whilst the A. B. C. process shows a diminution of 58·9 only. It appears from the first and second reports of the Rivers Pollution Commissioners, that on an average the precipitation processes remove 89·8 per cent. of the suspended matters, but only 36·6 per cent. of the organic nitrogen dissolved in the liquid.
The effects of a soil upon sewage water passing through it are the following:—
1. The filtering property of the soil mechanically arrests and retains the suspended particles of the sewage.
2 and 3. The porosity and physical attraction of the soil lead to the oxidation of the organic matter contained in the sewage, as instanced in the discovery of nitrates and nitrites in the effluent water, which did not exist previous to filtration.
4. A chemical reaction takes place between the constituents of the sewage and those of the soil.
If the charges brought against the system of irrigation, viz. that it is detrimental to the health and comfort of those who reside near sewage farms cannot be denied, it seems pretty certain that, in most cases, any ill effects arising from the method may be traced to its defective management. The selection of the soil which is to receive the sewage is a highly important consideration. The best for this purpose seems to be a loose marl, containing oxide of iron and alumina; but sand, as well as chalk, are both said to answer excellently.
| Results of irrigation, in part per 100,000. | Percentage of dissolved Organic Pollution removed. | Percentage suspended Organic Pollution removed | |
| Organic Carbon. | Organic Nitrogen. | ||
| On fallow land at Chorley (adhesive loam) | 62·3 | 70·2 | 100· |
| At Edinburgh (both sand and clay) | 45·3 | 81·1 | 84·9 |
| Barking (gravelly soil) | 65·8 | 86·2 | 100· |
| Aldershot (light sand)— | |||
| Best result | 91·8 | 87·3 | 99·7 |
| Worst result | 69·9 | 82·9 | 87·7 |
| Average result | 80·9 | 85·1 | 93·7 |
| Carlisle (light loam) | 77·9 | 59·8 | 100· |
| Penrith (light loam) | 75·0 | 77·2 | 100· |
| Rugby (adhesive soil) | 72·3 | 92·9 | 96·0 |
| Banbury (principally clay)— | |||
| Best result | 87·8 | 91·3 | 96·0 |
| Worst result | 64·1 | 80·1 | 90·3 |
| Average result | 76· | 85·7 | 93·2 |
| Warwick (stiff clay) | 71·7 | 89·6 | 100· |
| Worthing (loam) | 42·7 | 85·3 | 100· |
| Bedford (light gravelly soil), average result | 71·6 | 81·3 | 100· |
| Norwood (clay), average result | 65·0 | 75·1 | 100· |
| Croydon (gravelly soil)— | |||
| Best result | 73·2 | 93·2 | 100· |
| Worst result | 61·6 | 90·4 | 100· |
| Average result | 67·4 | 91·8 | 100· |
If the soil be of a stiff clayey nature it must be broken up and mixed with sand, lime, or ashes. The upper parts must be comminuted
and rendered porous, and it must be efficiently and deeply drained. At Troedyrhiw, as we have seen, the effluent drain is six feet deep.
The sewer water should be poured over the land in as fresh a condition as possible, having been previously deprived of any solid or grosser parts by straining. At Carlisle, decomposition of the sewage during its flow is prevented by adding carbolic acid to it. Lastly, it is of the utmost consequence that the amount of land used as the filtering medium shall be large. Letheby has shown that where this precaution is neglected, not only is the purification of the sewage incomplete, but the plan becomes a public nuisance. The amount of filtering earth should not be less than one cubic yard for eight gallons of sewage in twenty-four hours, in properly prepared soils; in some soils more than a cubic yard is required.