There is, of course, no such thing as a sewer gas having a definite and distinctive composition, and the nature of the mixture of gases in sewers is constantly varying according to season, temperature, etc. The tendency which sewer air has to cause disease depends in part upon certain gases, in part on minute particles of solid or semi-solid matter which are suspended in the air. In rare instances the sewers also contain illuminating gas, derived from leakage of gas-pipes in the vicinity. These gases produce debility, headache, loss of appetite, etc. As found in sewers and soil-pipes, they are so diluted that they are not absorbed by the water of a trap and given off on the other side to a sufficient extent to produce an evil effect. The air in a soil-pipe which is not ventilated is much more impure than that of the ordinary sewer, since the process of decomposition is constantly going on in the slimy coat which lines the interior of the pipes; and it is for this reason that it is so important to secure thorough ventilation of all the soil-pipes in a building. When this ventilation is secured, the proportion of dangerous gas in the pipes becomes very small, and the amount absorbed by the water in traps is almost inappreciable. The chief danger to life from sewer and soil-pipe air arises from the presence of minute particles of organic matter, dead and living, the so-called germs. Danger to life from these germs cannot be entirely removed by dilution, as can be done with gases. It has been found by the experiments of Dr. Carmichael and Dr. Wernick that an ordinary water-trap entirely prevents the passage of these germs, and that organic putrescible fluid will remain unchanged when exposed only to the air immediately {190} above such a trap. A pin-hole or minute sand-crack in the soil-pipe, or a very slight defect in a joint, is far more dangerous than a trap. The forms of disease produced by sewer air and its contents are more especially diphtheria, typhoid fever, and ill-defined disorders of the throat and digestive organs. It is possible that the germs of other specific diseases, such as scarlet fever, may be at times transmitted through sewer air, but such transmission must be very rare. While it is true that the germs of the specific diseases are very rarely present in sewer air, the house system of sewerage must be arranged as if they were always present, in order to obtain security. It must also be remembered that a system originally well planned and properly constructed will not always remain so; the pipes will corrode, the joints will become loosened, the valves will become clogged, and whenever alterations or repairs are made there is always danger of injury. Bearing these points in mind, the method of investigating a system can be readily understood.