The 10.7-cm solar flux (S) is widely used for upper-atmosphere studies as an index of the solar extreme ultraviolet radiation (EUV), whose variations are responsible for major changes in properties above 200 km. A model was developed to calculate the density as a function of S for various local times and altitudes from 200 to 800 km. The accuracy was checked by comparison of calculated rates of satellite orbital decay with observed values for 7 satellites. This comparison revealed systematic differences between the calculated and observed decay values. From the analysis of these differences, it is concluded that the “semi-annual” effect is not a real effect in 1958 to 1961; rather this effect arises because S is not an accurate index for the total EUV. Evidently, the EUV has two variable components: one radiated from active (sunspot) areas, correlated with S, and the other radiated as a background emission rather uniformly distributed over the entire sun and not correlated with S. A more accurate... Abstract The 10.7-cm solar flux (S) is widely used for upper-atmosphere studies as an index of the solar extreme ultraviolet radiation (EUV), whose variations are responsible for major changes in properties above 200 km. A model was developed to calculate the density as a function of S for various local times and altitudes from 200 to 800 km. The accuracy was checked by comparison of calculated rates of satellite orbital decay with observed values for 7 satellites. This comparison revealed systematic differences between the calculated and observed decay values. From the analysis of these differences, it is concluded that the “semi-annual” effect is not a real effect in 1958 to 1961; rather this effect arises because S is not an accurate index for the total EUV. Evidently, the EUV has two variable components: one radiated from active (sunspot) areas, correlated with S, and the other radiated as a background emission rather uniformly distributed over the entire sun and not correlated with S. A more accurate...