The performances of light-reared and dark-reared rats were compared on a variety of transfer tests following acquisition of a visual discrimination in which the critical difference between the patterns was the type of intersection or junction between lines. Significant differences due to rearing were observed in the initial rate at which the discrimination was acquired and, following 15 days of overtraining, on a subsequent series of transfer tests. The results suggest that the spatial integration necessary to recognize even as basic a relationship between linear elements as angle or junction involves mechanisms which require experience for their development and maintenance. RESUME Etude comparee de rats eleves a I'obscurite et de rats eleves a la lumiere lors de I'acquisition d'une discrimination visuelle et lors du rendement subsequent a un ensemble varie de tests de transfert lorsque la difference critique entre les patterns presentes est le type d'intersection ou de jonction entre les lignes. On observe des differences significatives dues au mode d'elevage dans le rythme initial d'acquisition de la discrimination de meme que, apres 15 jours de surapprentissage, dans une serie de tests de transfert. Les resultats suggerent que Integration spatiale necessaire pour reconnaTtre une relation aussi fondamentale que celle qui existe entre des elements lineaires percus comme angle ou points de jonction implique I'existence de mecanismes qui requierent une certaine forme d'experience pour leur developpement et leur maintien.In the past few years we have concentrated on obtaining an accurate description of the characteristics of the sensory coding system of the visually inexperienced rat. Evidence summarized previously (Tees, 1976) is consistent with the idea that some basic information processing capacities depend in only a limited way on visual experience for their development. These capacities might be roughly subserved under the rubric of feature detection (e.g., Sutherland, 1973).In the case of one of these, the ability to discriminate contour orientation, measurement of stimulus control for light-reared (LR) and dark-reared (DR) rats has involved extensive testing, and experience appeared to play little role in the ontogeny of this capacity (Bruinsma & Tees, 1977;