At the Migration Research Workshop of the 52nd SEG Meeting in Dallas (1982), an elegant method of wave-equation migration, known as reverse-time migration, was presented by Whitmore. Shortly thereafter McMechan (1982, 1983) and Baysal (1982) described virtually identical approaches to migration. This, along with further development by Baysal et al (1983a,b) and Whitmore (1983), has sparked much interest and discussion. In May 1983, Loewenthal and Mufti also published a migration algorithm, quite different from the above, that they call "reversed time migration" and developed as early as 1977. In this short note I present a general principle underlying reverse-time migration and discuss the various reverse-time migration procedures in light of it.