The Napier frequency (frequency of maximum absorption per wavelength) in oxygen and in air is shifted to higher values by water vapor according to a quadratic law that has different coefficients in the two cases. It is shown that proper treatment of the relevant equations leads to the correct experimental relationship between them. From this relationship, all the various velocity constants in the equilibrium equations may be determined numerically. It turns out that nitrogen has a somewhat stronger effect than oxygen on the relaxation of water vapor, and that the rate of resonant interchange of energy between oxygen and water vapor is, in fact, smaller than the rate of relaxation of water vapor by itself. It is also shown that the postulate of dimer formation in water vapor as an explanation of the quadratic relationship leads to a contradiction with experiment.