He also found that humidity of the air had marked influence in rendering the smell of organic matter perceptible, even more than a rise of temperature. The sense of smell is doubtless aided in detecting impurities in the air, by the _besoin de respirer_, a feeling of oppression caused by the deficiency of interchange between the blood and air. The state of cleanliness of the room as well as of the persons in it influences smell; hence there may not be in particular instances exact correspondence between excess of carbonic acid and of organic matter. =Chemical Examination.=—The estimation of nitrogen and oxygen in air is usually unnecessary, as these vary but little. The oxygen is, however, reduced in frequently re-breathed air. The ill effects of an often-breathed atmosphere are due not only to deficiency of oxygen, but also to the addition of carbonic acid and organic matters, rendering difficult the interchange between oxygen and the blood. =The Estimation of Carbonic Acid= is of great importance, as under ordinary circumstances, its amount is a fairly exact indication of the amount of contamination in the air. =Pettenkofer’s Method.=—A carefully dried glass vessel containing a gallon of water is filled with the air to be examined, by emptying the water in the room, the air of which is to be examined. Fifty cubic centimetres of clear freshly prepared baryta water are then added, and the stopper of the bottle then replaced. It is then well shaken, and afterwards allowed to stand for an hour. The carbonic acid combines with part of the baryta to form barium carbonate; and the baryta water remaining is consequently diminished in alkalinity. Given the alkalinity of the baryta water before and after the experiment, and the difference will give the amount of baryta which has combined with carbonic acid. The alkalinity of the baryta is estimated by a standard solution of oxalic acid, of such a strength that 1 c.c. is the equivalent of 0·5 c.c. of CO₂. The indicator used in making this test is phenolphthalein, which colours baryta water red, but its colour disappears when neutralization is reached. The following example is taken from “Pakes’ Laboratory Text Book of Hygiene,” p. 292:— The jar is found to contain 3,950 c.c. As 50 c.c. baryta water were run into the jar, the air experimented on = 3,950-50 = 3,900 c.c. On titrating 25 c.c. of the original baryta water, 22·50 c.c. standard acid solution were required to neutralise it. The baryta water in the jar required 19·35 c.c. 22·50-19·35 = 3·15 c.c. = difference of acid used. But 1 c.c. acid = 0·5 c.c. CO₂ at 0° C. and 760 mm. of mercury. Therefore CO₂ taken up by 25 c.c. of baryta = 3·15∕2 = 1·575 c.c. As 50 c.c. were used the CO₂ absorbed by the baryta = 3·15 c.c. This was present in 3,900 c.c. of air. Therefore the CO₂ = 0·80 per cent.