Columns 1 and 2 give the ratio of carbonic acid in the quantity of air which will produce no precipitate in half an ounce lime water: column 3 is the same as column 2, with the addition of 14:16 cubic centimetres, or half an ounce, to give the corresponding size of bottle and column 4 gives the size of bottle in ounces avoirdupois. It will thus be seen that different-sized bottles containing half an ounce of lime water will indicate approximately the ratio of carbonic acid in the air contained in them, by giving no precipitate when the bottle is well shaken. Thus, if a bottle of 10 oz. is used, and there is no precipitate, it will indicate that the ratio of carbonic acid does not exceed 06 per cent; or if one of 8 oz. is used, and there is also no precipitate, it will indicate that the ratio does not exceed 08, and so on. Dr. Smith says that "the lime water may be prepared of the same constant strength so closely that we may neglect the difference. Burnt lime is slaked with water and dissolved by shaking. It is then kept in a bottle to stand till it is clear. The bottle or bottles used should be wide-mouthed, so that they can be readily cleaned and dried, and the air to be examined may be made to enter them by inhaling the air contained in them through a glass or caoutchouc tube, care being taken not to breathe into the bottle." As a practical application of this method, which can be practised by any one, Dr. Smith proposes the following rule:"Let us keep our rooms so that the air gives no precipitate when a 10 oz. bottleful is shaken with half an ounce of clear lime water." 2. Organic Impurities.-To obtain an approximate estimate of the organic impurities, the air may be drawn. through, or washed, in a very dilute solution of potas The result is sium permanganate of known strength. stated in the number of cubic feet of air which it takes to decolorise 001 gramme of the potassium permanganate in solution. The method at present pursued by Dr. Angus Smith is somewhat different from this. He takes about 30 cubic centimetres of pure water, and adds to it a small amount of known solution of the potassium permanganate. This solution is shaken up with the air in the bottle; the air is then drawn out by a bellows-pump, and another bottleful washed, and so on until the whole colour is removed, or a sufficient amount to enable him to test the remainder, so as to be able to estimate the difference. The actual amount of oxygen taken is then calculated, and the results stated in grains per million cubic feet of air. 3. Ammonia. The estimation of the ammonia, organic and albuminoid, is a still more delicate and difficult process. The water used for washing the air must be perfectly pure, and should therefore be boiled with soda or potash before distillation. From 30 to 50 cubic centimetres are then introduced into a bottle of about 2000 cubic centimetres, and the washing of successive bottlefuls of the air to be examined is continued until the water is sufficiently charged with impurities. The testing afterwards is the same as that proposed by Messrs. Wanklyn, Chapman, and Smith, for organic impurities in water, and the results are stated in grains per million cubic feet of air. For further information concerning these methods of analysis, and for an account of numerous valuable experiments, see Dr. Angus Smith's work already quoted. SECTION IV.MICROSCOPICAL EXAMINATION. Suspended matters contained in the air may be collected by drawing the air through distilled water by means of an aspirator, by washing the air in distilled water, or by employing an instrument called an aeroscope. When either of the first two processes is employed, the suspended matters are merely allowed to subside, and are then removed to a glass slip for examination. The aeroscope invented by Pouchet consists of a funnel-shaped tube, ending in a fine point, beneath which is placed a slip of glass moistened with glycerine. . Both glass and tube are enclosed in an air-tight chamber, which is connected by tubing with an aspirator, so that when the stopcock of the aspirator is turned, and the water allowed to escape, the air is drawn through the tube, and impinges against the slip of glass moistened with glycerine, by which the suspended matters are arrested. SECTION V.-EXAMINATION AS REGARDS TEMPERATURE AND MOISTURE. 1. Temperature.-The various points connected with the temperature of the contained air, such as its equability, sufficiency, etc., are readily ascertained by a judicious distribution of thermometers throughout the space to be examined, and by comparing the outside with the inside temperature. The efficiency of the heating apparatus is of course best tested during very cold weather and in the night time. When open fire-places are used the temperature should be noted at the remote parts of the room, and if the air is heated before entering, it is advisable to take the temperature at the point of delivery, and to ascertain whether it is well diffused or not. 2. Moisture. The amount of watery vapour, or the hygrometricity of the contained air, may be determined by hair hygrometers, or by wet and dry bulb thermometers. The latter are the most convenient and reliable, but they should be distributed some two or three hours before the observations are made. The wet bulb is covered with muslin, over which there is twisted a small skein of cotton, which dips into a little vessel containing either distilled or rain water. The cotton should be boiled in ether, or soaked in a solution of sodium carbonate, to free it from fat, so that the water may readily ascend by the force of capillary attraction. Unless the air is saturated with moisture, the temperature of the wet bulb is always below the temperature of the dry, and the number of degrees of difference between them varies according to the amount of watery vapour present. This is generally represented in relative terms. For example, the point of complete saturation being assumed to be 100, any degree of dryness may be stated as a percentage of this, and can be conveniently ascertained by reference to the following table, which has been calculated from Mr. Glaisher's large tables (see Parkes' Manual of Practical Hygiene). table is read by taking the temperature of the dry bulb, and the difference between it and that of the wet bulb, and noting the number given at the intersection of the two columns. This number gives the relative humidity. The |