Abstract
The formula developed by Kramers (and recently on different lines by Hebb) to represent the fine structure of the ground state of oxygen is compared with the experimental data. It is pointed out that the currently accepted values of the parameters are in error. The measurements are compared with a more accurate formula, which takes into account the fact that the rotation does not completely decouple the spin from the axis of the molecule, and values of the parameters are found which give excellent agreement. The rotational intensity distribution in the atmospheric O2 bands is recalculated on the assumption of incomplete decoupling; the intensities in the PP and RR branches are found to be unchanged, while those in QP and QR become more nearly equal than with complete decoupling, a result in harmony with the measurements.