The use of the word ''aurora'' for many different observations at high latitudes has limited the concepts involved; this is particularly true for F region irregularities. Observations setting the position of the auroral oval (Starkov and Fel'dshtein, 1970) were made using primarily the 555.7-nm green line, which is emitted predominantly at E layer heights. These observations have shown that the change in position of the auroral oval for low values of Kp as a function of sunspot cycle is of the order of 1¿ to 2¿ between sunspot maximum and sunspot minimum. However, irregularities in the F region show much larger solar cycle variations in the locations of the equatorward boundary, typically 10¿. A review of scintillation data indicates that at a given auroral latitude, the scintillation activity increases with sunspot number. In addition, for a constant scintillation intensity, the equatorward boundary moves to lower latitudes as sunspot maximum is approached. We review existing spread F studies and show that for quiet geomagnetic conductions, there is lower occurrence during years of low sunspot numbers than during years of high sunspot numbers. However, the spread F index, related to Δ f/f0F2, is higher during years of low sunspot number than during years of high sunspot number. We demonstrate that this apparent dichotomy can be reconciled by using a new method of normalizing the spread F index by the maximum electron concentration of the F layer. We briefly discuss the possible explanations for the observed solar cycle variations of irregularity occurrence in terms of the absolute values and gradients of electron concentration and the E region conductivity. ¿ American Geophysical Union 1995