Appraising Proposed Federal Standards for Water Resources

during the 3 years before ECT were impaired, but memories acquired before that time were not affected. The results therefore appear to confirm the hypothesis, originally formulatedby Ribot (12), that the susceptibility of a memory to disruption is inversely proportional to its age. The validity of this conclusion depends, of course, on how well the remote memory test satisfies the requirements it was designed to meet. The most important of these is that names from each time period must be forgotten at the same average rate. Although this point is difficult to establish definitively with any retrospective method, considerable effort was made to meet this requirement by minimizing various kinds of sampling bias (6). If resistance of memory to amnesic treatment can indeed increase over a period of years, then there must be two distinct consequences of the passage of time. First, material in memory becomes resistant to disruption over the years. Second, the same material becomes gradually more difficult to recall. Thus, names from 1971 to 1972 were remembered rather well, but were forgotten after ECT. Conversely, names from earlier time periods were remembered rather poorly, but were not affected by ECT. The neural substrate of memory apparently changes with the passage of time, such that resistance develops as forgetting occurs. The frequent observation that retrograde amnesia can cover a time period of minutes, hours, or days has usually been interpreted to indicate that the memory storage process is labile for only a short period of time after learning and is then consolidated into a more stable state (1). The findings reported here indicate, however, that retrograde amnesia can sometimes cover a period of years. Accordingly, the development of resistance to amnesic treatment need not depend on the rapid transition from a labile to a stable memory process. The development of resistance can presumably depend on gradual changes in neural mechanisms that subserve enduring, relatively stable modifications in function. LARRY R. SQUIRE PAMELA C. SLATER PAUL M. CHACE Department of Psychiatry, Veterans Administration Hospital, University of California School of Medicine, San Diego 92161