The agreement between observations and theory at low latitudes is better than expected (F. Henyey, personal communication), because it is unlikely that the GM model should be accurate close to the ...