Every experimenter who has been compelled to repeat the weighing of a somewhat complicate piece of apparatus, with an interval of several hours between, knows how many inaccuracies he is exposed to if he is compelled to take into calculation the changes of temperature and pressure, and the moisture on the surface of the apparatus. After fighting all these difficulties, and frequently in vain, the experimenter begins to mistrust every result that depends only on difference in weight, and to prefer those methods whereby the substance to be estimated can be isolated, so that it can be seen and handled, weighed or measured, in a free state, and in its own natural condition. The classical experiments of Thenard, of Th. de Saussure, of Messrs. Boussingault, on the quantity of carbonic acid in the air, are well known to every one: they need only to be organized, repeated, and multiplied. J. Reiset, who has conducted a long and tedious series of experiments on this subject, has adopted a process that seems to offer every guarantee of accuracy. The air that furnishes the carbonic acid is aspirated through the absorption apparatus by two aspirators of 600 liters capacity. The temperature and pressure of the air are carefully measured. The carbonic acid is absorbed by baryta water in three bulb apparatus. The last bulb, which serves as a check to control the operation, remains clear, and proves that no binoxide of barium is formed. The baryta water used is titrated before and after the operation, and from the difference is calculated the quantity of carbonate formed, and hence of the carbonic acid. These tedious experiments, which varied in duration from 6 to 25 hours, require at least two days of continuous labor. They were repeated 193 times by Reiset in 1872, 1873, and 1879. They were made in still weather, and in violent winds and storms. The air was taken at the sea-shore, in the middle of the fields, on the level earth, during harvests, in the forests, and in Paris. Under such varied conditions, the quantity of carbonic acid varied but little; the numbers obtained were between 2.94 and 3.1, which may be taken as a general average of the carbonic acid in the air. The quantity of carbonic acid in the free atmosphere is tolerably constant, which must necessarily be the case according to Schloesing's proposed relation between the bi-carbonate of lime in the sea and the carbonic acid in the air. The only cause that seems at all competent to change the geological quantity of carbonic acid in the atmosphere is the formation of fog. As the aqueous vapors condense, they collect the carbonic acid; and the foggy air, as a rule, is more heavily laden with this gas than ordinary air. It is not surprising that there is less carbonic acid in the air collected on clear summer days, in the midst of clover, etc., that is in an active reducing furnace; if anything is surprising, it is that the quantity of carbonic acid does not sink below 2.8.