At the beginning of the bacteriological work the water should be examined by means of the gelatine plate method. This consists in drawing up into a fine sterilised pipette a small quantity of the water and introducing it thereby into a test-tube of melted gelatine at a temperature below 40° C.[13] It will depend upon the apparent quality of the water as to the exact quantity introduced into the gelatine; about .5 or .1 of a cubic centimetre is a common figure. The stopper is then quickly replaced in the test-tube, and the contents gently mixed more or less equally to distribute the one-tenth cubic centimetre throughout the melted gelatine. A sterilised sheet of glass (4 inches by 3) designated a _Koch's plate_ is now taken and placed upon the stage of a levelling apparatus, which holds iced water in a glass jar under the stage. The gelatine is now poured out over the glass plate, and by means of a sterilised rod stroked into a thin, even film all over the glass. It is then covered with a bell-jar and left at rest to set. The level stage prevents the gelatine running over the edge of the plate; the iced water under the stage expedites the setting of the gelatine into a fixed film. When it is thus set the plate is placed upon a small stand in a moist chamber, and the whole apparatus removed to the room temperature incubator. A _moist chamber_ is a glass dish, in which some filter paper, soaked with corrosive sublimate, is inserted, and the dish covered with a bell-jar. By this means the risks of pollution are minimized, and moisture maintained. In all cases at least two plates must be prepared of the same sample of water, and it is often advisable to make several. They may be made with different media for different purposes, and with different quantities of water, though the same method of procedure is adopted. In a highly polluted water extremely small quantities would be taken, and, _vice versâ_, in pure water a large quantity. [Illustration: MOIST CHAMBER IN WHICH KOCH'S PLATES ARE INCUBATED] When we come to discuss the relation of disease organisms to water, particularly those causing typhoid fever, we shall learn that they are both scarce and intermittent. This point has been dwelt upon frequently by Dr. Klein, and it is clear that such a state of things greatly enhances the difficulties in detecting such bacteria, and he has proposed a simple procedure by which the difficulty of finding the _Bacillus typhosus_ in a large body of water may be met. [Illustration: HOT AIR STERILISER For the Sterilization of Glass Apparatus, etc.]