“An examination of the residue obtained by evaporating large quantities of open septic tank effluent (filtered through paper), shows that the mineral matters largely consist of iron oxide, from the decomposition of organic compounds of iron, and calcium sulphate. Among the volatile constituents have been detected ammonium carbonate, mercaptan-like compounds of very offensive smell, acetic and butyric acids. No evidence of the presence of amines could be found in the residue on evaporation, but by distilling large volumes of the liquid and carefully analysing the platinum salts obtained from the distillate, the presence of amines is indicated. “Research in this direction is being continued; careful comparison especially will be made of the products obtained by evaporation and distillation of crude sewage and septic tank effluent respectively. “The evidence, however, points to a breaking down of albuminoid and cellulose matter in the septic tank into simpler and to some extent volatile compounds. The reactions are probably hydrolytic in character, ammonia, amines, carbonic acid, water, and possibly alcohol, being produced. “A further quantity of organic matter also disappears as methane, nitrogen and hydrogen.” [Sidenote: Must aerobic fermentation in all cases be preceded by anaerobic fermentation?] It will be clear from the foregoing, that the changes going on in a septic tank are entirely different from those brought about in contact beds, and the question whether a septic tank is a necessity for the subsequent contact bed treatment, or whether it is a distinct disadvantage, can only be definitely settled when we know whether aerobic fermentation, i.e. decomposition, must in all cases be preceded by anaerobic fermentation, i.e. putrefaction, and to what extent, or whether such a succession of changes is not necessary. [Sidenote: At Manchester contact beds accustomed to septic tank effluent did not at once purify raw sewage.] It is interesting to note in connection with this point, that during the Manchester experiments it was established that contact beds, which have become accustomed to septic tank effluent, will not at once purify comparatively fresh sewage. (_d_) _Velocity of Flow through Tank._—The velocity of flow through the septic tank is of great importance, as on it depends the size of the installation. It seems to have become a habit to express this velocity by the length of the sojourn of the sewage in the septic tank—for instance, “the flow of sewage through the tank was such that it would fill it in twenty-four hours”; but as all tanks vary in size, and as in consequence the distance which has to be traversed by the sewage from the entrance to the exit in twenty-four hours is different in nearly every case, such a habit is, to say the very least, very misleading.