A study was conducted by Los Angeles City College to determine whether the early registration of spring 1972 was primarily responsible for the inflated enrollment figures for fall 1972 classes. A tabulation by course and department was provided by the college Data Center of the number of enrollments at the close of registration, at the close of the second week, at census week, and at the end of the semester. Persistence rates for "continuing" and "new and returning" students were compared. Persistence patterns for continuing and non-continuing students differed markedly between the first day of the semester and census week. From census week to the end of the semester, persistence rates for both groups were comparable to past years. It appears that many of the students who registered early did not actually appear for classes and their names were not removed from the rolls until census week. It is recommended that either early registration be abandoned, some means of affirming the early enrollment be employed, or departments study the drop rates and oversubscribe their courses appropriately. (KM)