STATEMENT OF PROBLEM IT IS NOT known what happens in the cortex when the reaction patterns of a growing child are laid down and then become established as enduring foundations of behavior. However, it is highly probable that these events are not the result of anatomic changes in the cortex; it is hard to imagine dendrites or axons changing position so as to set up new connections. The establishment of patterns is more likely to depend on a physiologic process, conceivably facilitation. Hence there is a chance that the neural process of pattern fixing may be reversible. Indeed, it is known that when psychotherapy is successful the patterns do change. Old ones disappear, to be replaced by new ones, or old ones may so change as to look like new ones. The present study began as an attempt to answer the following questions: If the neural arrangements underlying reaction patterns