taminating air is not so much due to occasional and temporary failure in the efficacy of the trap as to an almost constant absorption of sewer-air by the water on the sewer-side of the trap, and its subsequent discharge from the house-side. Dr. Fergus has made a series of experiments in a glass tube so bent as to resemble the ordinary "siphon" trap, and charged with water. Certain gases were evolved on what we may call the sewer-side of the trap (b), and tests were applied to ascertain whether the gases succeeded in passing through the water. The results as tabulated by Dr. Fergus are as follows:
|Sp. gr.||Source.||Test.||Time for reaction to show.|
|"||·50||"||Nessler||30 minutes. Ate through a small wire in less than 24 hours|
|Sulphurous acid.||2·25||"||Litmus||1 hour.|
|Sulphuretted hydrogen.||1·25||"||Lead paper.||3 to 4 hours.|
|Chlorine||2·50||"||Iodide of starch paper.||4 hours.|
|"||2·50||"||Litmus-water in trap.||Began to show in a few minutes. In half an hour the whole was bleached.|
|Carbonic acid.||1·50||Generated.||Lime-water in trap.||1½ hour.|
|"||1·50||"||Litmus suspended over water in trap.||3 hours.|
It was, however, urged that the results would probably be different if the trap were ventilated. A ventilating-shaft (c) was, therefore, inserted in the upper part of the bend on the sewer-side, and the experiments were repeated. "The results," says Dr. Fergus, "were much the same, except that the reaction was a little longer in showing itself."
Ordinary sewer-air may be taken to contain in every hundred parts about seventy-nine parts of nitrogen, nearly twenty of oxygen, not quite half a part of carbonic acid, and traces of sulphuretted hydrogen, marsh-gas, and ammonia. These gases, however, when inhaled in the proportions indicated, can hardly be regarded as materially affecting health. Sewer-air also contains organic matter in the form of vapor, and of definite particles; but doubts have been expressed as to whether these organic particles succeed in making their way through water-traps, and some carefully executed experiments of Dr. Neil Carmichael, of Glasgow, have gone far to show that they do not do so.
There are other ways, however, in which danger comes about. The water in traps is apt to be sucked out by siphon-action, as the