The auroral distribution and activity over a large portion of the northern polar region are examined on the basis of 58 Isis 2 scanning auroral photometer pictures. These large-scale pictures clearly show the dynamic variation in the location of the auroral oval. The invariant latitude ϑp of the equatorward boundary of the diffuse auroral belt at local midnight is found to be a useful parameter in characterizing the size of the auroral oval and auroral activities. As ϑp increases, the probability of seeing a quiet oval increases, and the probability of seeing substorms decreases. Based on their longitudinal extent (which appears to be related to their intensities), substorms roughly fall into two types, 'confined' and 'widespread,' although such a distinction is not always clear-cut. Both types of substorms can occur along a contracted oval (ϑp>66¿); however, confined substorms occur more often along a contracted oval than along a normal oval (66¿?ϑp?65¿) or an expanded oval (ϑp<65¿). It is shown that the AE index is often inadequate to identify substorms or to determine their onset times. This is especially true for substorms along a contracted oval and for confined substorms along a normal oval.