Fig. 2.

Schematic illustration of dNCL. Cross section through the caudal telencephalon (after Bottjer et al. 2000) showing RA and Ad within the arcopallium and dNCL in the dorsolateral nidopallium The projection from dNCL to Ad is topographic, as shown here by injection sites in midmedial (black) and midlateral (gray) Ad that produced corresponding clusters of retrogradely labeled cells in dNCL (note that the extreme lateral and medial regions of dNCL do not include labeled cells in this figure). The borders of dNCL are not apparent in Nissl-stained sections nor in our immunostained tissue. In experiment 1, we therefore extrapolated the location of dNCL based on the location of Ad and our knowledge of the path of axons from dNCL to Ad. We counted ZENK+ cells within a sampling window (gray box), the size and placement of which were chosen to be conservative in the sense that it would exclude extreme dorsal, ventral, medial, and lateral regions of dNCL. In this way we avoided counting immunoreactive cells outside the borders of dNCL. In experiment 2, we counted cells within dNCL by locating our sampling window over regions that included the highest density of retrogradely labeled neurons, depending on the quality and location of the injection site. Neurons outside dNCL are never labeled by injections into Ad (Bottjer et al. 2000).