11 X r , & ble of learning." Our results (2) and those of Silva et al. (1) do not suggest, as Deutsch implies, that there is a dissociation between LTP and learning, and that therefore these processes are not linked. The data (1-3) indicate that interference with LTP can impair spatial learning. We believe it is naive to think that deficits in LTP in the CA1 region of the hippocampus will eliminate spatial learning in an all-or-none fashion, because even rodents with hippocampal lesions or with

more »

... a complete pharmacological blockade of LTP can demonstrate some spatial learning, and in both the fynand a-CaMKII mutant mice, LTP in the CA1 region is impaired but not completely abolished. We would simply emphasize that our initial behavioral studies of transgenic mice bearing specifically engineered mutations were not meant to be a definitive description of the behavioral repertoire of these animals. Instead, they represent the first steps toward exploiting the power of target gene discruption in a combined molecular, physiological, and behavioral study of learning and memory. We believe that our studies (2), as well as those of Silva et al. (1), demonstrate the usefulness of gene targeting techniques for investigating the molecular components of the signaling pathways responsible for long-term potentiation as well as providing a new approach to the study of behavior.