It is well known that the water-vapor continuum plays an important role in the radiative balance in the Earth’s atmosphere. This was first discovered by Elsasser almost 70 years ago, and since that time there has been a large body of work, both experimental and theoretical, on this topic. It has been experimentally shown that for ambient atmospheric conditions, the continuum absorption scales quadratically with the H₂O number density and has a strong, negative temperature dependence (T dependence). Over the years, there have been three different theoretical mechanisms postulated: Far wings of allowed transitions, water dimers, and collision-induced absorption. Despite the improvements in experimental data, at present there is no consensus on which mechanism is primarily responsible for the absorption. The first mechanism proposed was the accumulation of the far-wing absorption of the strong allowed transitions. Later, absorption by water dimers was proposed and this mechanism provides a qualitative explanation for the strong, negative T dependence. Recently, some atmospheric modelers have proposed that collision-induced absorption is one of the major contributors. However, based on improvements in the theoretical calculation of accurate far-wing line shapes, ab initio dimer calculations, and theoretical collision-induced absorptions, it is now generally accepted that the dominant mechanism for the absorption in the infrared (IR) windows is that due to the far wings. Whether this is true for other spectral regions is not presently established. Although all these three mechanisms have a negative T dependence, their T dependences will be characterized by individual features. To analyze the characteristics of the latter will enable one to assess their roles with more certainty. In this paper, we present a detailed study of the T dependence of the far-wing absorption mechanism. We will then compare our theoretical calculations with the most recent and accurate experimental data in the IR windows. The results of our calculations are found to agree very well with measurements in the 800–1200 cm⁻¹ region.We conclude from this work that the T dependence in the IR window region predicted by the far-wing theory is negative and moderately strong. Its pattern is not simple and it could vary significantly as the frequency of interest varies.