1135The nomogram may be duplicated with the aid of Table I. Lay off parallel, linear scales for barometric pressure and factor F of the desired size and spacing. Locate five points on the temperature scale by the intersection of lines joining corresponding points on the P and F scales. Draw the support for the T scale and subdivide it by linear interpolation. RESULTSAnalyses for only three compounds are reported in this paper, as no changes were made in the standard combustion procedure.The results shown in Table I1 are indicative of the usual precision of routine analyses.A number of benzoic acid and dextrose samples were burned by the standard Dumas method in an effort to determine the consistency of the nitrogen blank. A very limited amount of data is available in the literature on the size of this correction. Only the usua I sweeping precautions were observed.The amount of gas collected when 50 ml. of carbon dioxide were passed through the apparatus corresponded to 0.007 mg. of nitrogen. Ten samples of benzoic acid, varying in size from 3 to 7 mg., were burned, sweeping with the same volume of carbon dioxide. The average nitrogen equivalent was 0.020 mg., with an average deviation of 0.003 mg. Four samples of dextrose (4.8 to 5.2 mg.) were treated in the same way. The blank was slightly lower, averaging 0.018 mg. of nitrogen, with an average deviation of 0.001 mg. The value of the blank used in calculating the analyses of Table I1 was 0.020 mg. of nitrogen in each case.These two compounds were chosen for blank estimation because of their great difference in physical and chemical characteristics. The small difference in magnitude of the two blanks perhaps may be explained by the greater occlusion of air in the voluminous benzoic acid.