In the first part, a brief historical description of the development of auroral morphology leading to the concept of the auroral substorm is described, partly based on the author's humble experience. In the second part, the major features of auroral substorms, such as the onset, poleward expansion, westward traveling surges, the omega bands, torches, and patches, are reviewed, adding many features not described adequately by Akasofu (1964). These features occur in the equatorward half of the oval, while the poleward half responds only passively to the equatorward half. All these features must be telling us fundamental magnetosphere-ionosphere coupling processes associated with substorms. If we can learn physics involved in each of the specific activities, it will become a powerful tool in studying magnetosphere-ionosphere coupling processes, taking advantage of the fact that the aurora is visible (or can be imaged), covering the entire polar region and thus over a large region of the magnetosphere. It is shown that we can now make a good progress by a recent breakthrough observation of substorm onset, which indicates that current reduction of the current sheet at about 8 R-E is an important cause of substorms, not by magnetic reconnection at a distant magneto-tail. It is hoped that 10 questions in the summary will be useful in advancing both auroral physics and magnetospheric physics.