[10] It will be observed that there is a marked difference between the effects of dry heat and moist heat. Moist heat is able to kill organisms much more readily than dry, owing to its penetrating effect on the capsule of the bacillus. Dry heat at 140° C. (284° F.), maintained for three hours, is necessary to kill the resistant spores of _Bacillus anthracis_ and _B. subtilis_, but moist heat at fifty degrees less will have the same effect. It is from data such as these that in laboratories and in disinfecting apparatus moist heat is invariably preferred to dry heat. For with the latter such high temperatures would be required that they would damage the articles being disinfected. Koch states the following figures for general guidance: Dry heat at a temperature of 120° C. (248° F.) will destroy spores of mould fungi, micrococci, and bacilli in the absence of their spores; for the spores of bacilli 140° C. (284° F.), maintained for three hours, is necessary; moist heat at 100° C. (212° F.) for fifteen minutes will kill bacilli and their spores. [11] Water from a house cistern is rarely a fair sample. It should be taken from the main. If taken from a stream or still water, the collecting bottle should be held about a foot below the surface before the stopper is removed. [12] The _cubic centimetre_ (cc.) is a convenient standard of fluid measurement constantly recurring in bacteriology. It is equal to 16-20 drops, and 28 cc. equal one fluid ounce. [13] The gelatine is reduced to liquid form by heating in a water-bath. Before inserting the suspected water it is essential that the gelatine be under 40° C, or thereabouts, in order not to approach the thermal death-point of any bacteria. [14] _Micro-organisms in Water_ (1894). [15] Report on the Micro-organisms of Sewage, Reports to L. C. C., 1894, No. 216. [16] Harben Lectures, 1896. [17] Report on the Metropolitan Water Supply. [18] The methods adopted for making a quantitative and qualitative examination of sewage are precisely analogous to those used in milk research. Dilution with sterilised water previous to plating out on gelatine in Petri dishes is essential (1 cc. to 10,000 cc. of sterile water, or some equally considerable dilution), otherwise the large numbers of germs would rapidly liquefy and destroy the film. Special methods must be used for the isolation of special organisms; phenol-gelatine, Elsner medium, indol reaction, "shake" cultures, Parietti broth, etc., must often be resorted to for special bacteria. Spores of bacteria may always be numerically estimated by adding the suspected water or sewage to gelatine, and then heating to 80° C. for ten minutes before plating out. This temperature removes the bacilli, but leaves the spores untouched.