Three experiments were performed to investigate the effects of certain kinds of early experiences upon the later behavior, primarily social, of Peking ducklings. 1. Experiment I investigated whether the age at which exposure to a green cube occurred and whether any exposure that occurred had any effect upon the later preference shown by the Ss for the cube versus a duck model. Also varied was whether Ss were raised as pairs or as isolates. Subjects exposed to the cube during the age for imprinting showed stronger following of the cube than did Ss exposed during the age for avoidance. However, these groups did not differ in their later preference for the cube in the choice test. Whether Ss were raised as pairs or isolates did not affect their following scores or their preference scores but had a marked effect upon their later flocking behavior. Pair raised Ss flocked readily, while isolate reared Ss ran from each other. However, isolate reared Ss that were exposed to a moving cube during the critical period for imprinting showed stronger flocking behavior than other Ss. The fear shown by the isolate reared birds was readily overcome by 22 hours confinement in groups of 3 or 4 in a 3 foot cubicle. 2. Experiment II investigated the effects of primacy versus recency upon later preference behavior. Subjects were exposed daily to a moving cube or a companion duckling, some Ss experiencing the cube the first five days and the duck the next 5 days. Other Ss experienced them in the reverse order, while a third group had no exposure the first 5 days and were exposed to the cube the next 5 days. All Ss were then given a choice between the moving cube and a live duckling. The results of this experiment showed that exposure to an object early in life leads to better following of another object later than does no early exposure. Also there was evidence that both primacy and recency are important, but, other things being equal, a live duckling is a stronger stimulus for approach behavior than is a green cube. 3. Experiment III was designed to determine whether the age at which exposure to another duckling occurred would affect the rate at which a S would later run to get to a stimulus duckling. It was found that the age of exposure was not nearly so important as the length of the exposure-test interval. Apparently the shorter this interval, the faster the S will run. 4. In general it was found that imprinting is not necessary to later social behavior, since birds raised in isolation show flocking behavior after confinement in close quarters. There was also evidence that imprinting is not necessarily object specific for Ss exposed during the critical period for imprinting can transfer to other objects more successfully than Ss not so exposed. There was some evidence that, other things being equal, a S is likely to prefer the imprint object to other objects. The question of relative potency of stimulus objects was raised when Ss showed a marked preference for a sibling duck over a green cube, regardless of the pattern of previous exposure to the two objects. If Ss have not had the opportunity to imprint and establish social behavior before the fear response occurs, it is still possible to socialize these Ss by confining them for a period of time in close quarters. The social behavior thus established is not apparently different from that established during the critical period for imprinting. In fact, the fertile eggs, from which the subjects of these experiments were hatched, came from a flock of ducks which had been raised as isolates until 25 days of age.