Rotational analyses have been carried out for the (0,0) bands of the [Formula: see text] absorption systems of S16O2 and S18O2, from high dispersion plates taken with the gases at dry ice temperature. The rotational analysis of the (0,0) band of S16O2 given by Brand, Jones, and di Lauro is confirmed in general, but their values for the anisotropic electron spin fine structure constants are found to be in error. Our new values remove the discrepancy in the sign of the spin–spin interaction parameter β = E between the gas phase work and the solid state value given by Tinti. This discrepancy had been rationalized by Brand, Jones, and di Lauro in terms of a different choice of phases for the angular momentum operators, but this argument is shown to be incorrect. The spectrum of S18O2 confirms our new values for the spin constants in detail.