From the difference in weight of 1 liter of oxygen and 1 liter of nitrogen, a loss in weight of a gas containing a mixture of oxygen with a small per cent of nitrogen would actually represent a somewhat larger volume of gas than if pure oxygen were admitted. The differences in weight of the two gases, however, and the amount of nitrogen present are so small that one might almost wholly neglect the error thus arising from this admixture of nitrogen and compute the volume of oxygen directly from the loss in weight of the cylinder. As a matter of fact, it has been found that by increasing the loss in weight of the cylinder of oxygen containing 3 per cent nitrogen by 0.4 per cent and then converting this weight to volume by multiplying by 0.7, the volume of gas admitted is known with great accuracy. This method of calculation has been used with success in connection with the large chamber and particularly for experiments of short duration. It has also been introduced with great success in a portable type of apparatus described elsewhere.[27] Under these conditions, therefore, it is unnecessary to make any correction on the residual volume of nitrogen as calculated at the beginning of the experiment. When a direct comparison of the calculated residual amount of oxygen present is to be made upon determinations made with a gas-analysis apparatus the earlier and much more complicated method of calculation must be employed. CRITICISM OF THE METHOD OF CALCULATING THE VOLUME OF OXYGEN. Since the ventilating air-current has a confined volume, in which there are constantly changing percentages of carbon dioxide, oxygen, and water-vapor, it is important to note that the nitrogen present in the apparatus when the apparatus is sealed remains unchanged throughout the whole experiment, save for the small amounts added with the commercial oxygen--amounts well known and for which definite corrections can be made. Consequently, in order to find the amount of oxygen present in the residual air at any time it is only necessary to determine the amounts of carbon dioxide and water-vapor and, from these two factors and from the known volume of nitrogen present, it is possible to compute the total volume of oxygen after calculating the total absolute volume of air in the chamber at any given time.