1. The topographic pattern of re-established retinotectal projections following various surgical manipulations of the optic tectum was studied in adult goldfish with neurophysiological mapping methods. 2. Immediately following excision of the caudal half of the tectum, a piece of the tectal tissue was dissected from the remaining rostral half-tectum, and then reimplanted to the same half-tectum after either 180 or 90 degrees anticlockwise rotation around the dorsoventral axis in the first experimental group. 3. A majority (twenty-one out of twenty-three) of these operated fish, in which the reimplanted tectal tissue degenerated, showed no sign of a field compression: only the nasal half of the visual field (with a localized partial scotoma corresponding to the area of the degenerated reimplant) projected on to the remaining intact area of the rostral half-tectum. 4. In seven fish, the re-established visual projections on to the 180 or 90 degrees rotated reimplants showed a corresponding localized 180 or 90 degrees rotation with reference to the other projections on to the surrounding intact area of the same half-tectum. Only one of these seven fish showed also a compression in the re-established projections from the entire visual field on to the operated half-tectum with the 90 degrees rotated reimplant. 5. When a field compression was induced first in the intact rostral half-tectum following excision of the caudal half, and then a piece of the 90 degrees rotated tectal tissue was reimplanted later within the rostral half-tectum, the previously induced field compression persisted, regardless of whether the reimplanted tissue degenerated or survived. In the latter case, the compression in the re-established visual projections on to the surviving reimplant occurred according to the original topographic polarity of the 90 degrees rotated tectal tissue. 6. A field compression could also be induced within a rotated tectal reimplant, which retained its original polarity, as follows. A piece of the tectal tissue was dissected from the central area of the whole tectum, and then reimplanted after either 180 or 90 degrees rotation. When the reimplanted tectal tissue became reinnervated later, the caudal half of the operated tectum (including the posterior half of the reimplant) was excised. The re-established visual projections on to the remaining part of the halved reimplant within the rostral half-tectum showed later a field compression in accordance with the original topographic polarity of the 180 or 90 degrees rotated tectal tissue. 7. These results provide direct evidence for the compatibility between the retention of original topographic polarity by a reimplanted tectal tissue and the capability of the same tectal tissue to readjust to a disparity in size. 8. Histological examination of the operated half-tectum with a reimplant, stained by a modified rapid Golai method, revealed that the reimplanted tectal tissues retained highly organized cytoarchitectonic structures...