The problem of deriving accurate land surface temeprature (LSTs) using infrared satellite measurements is considered. The radiative transfer equation over the land is solved by linearization of the Planck function in temperature and wavenumber in the same manner as is done for obtaining the so-called split-window algorithm for sea surface temperature estimation. The effects of the atmosphere and of surface emissivity are both considered in the derivation. By making further assumptions, an approximate split-window algorithm which shows explicitly how atmospheric absorption by water vapor and surface emissivity both affect the accuracy of LSTs is derived. A new algorithm has been formulated for use with National Oceanic and Atmospheric Administration advanced very high resolution radiometer data and is compared with previously published algorithms. It has a global form with coefficients that depend upon local conditions. Dual angle algorithms are also derived. These may be used to estimate LST using data from the along-track scanning radiometer on board the first European remote-sensing satellite. ¿ American Geophysical Union 1993