Fig. 2.

Anatomical evidence from marmoset visual cortex for an upper quadrant and two lower quadrant representations bordering dorsal V2 anteriorly (based on data from Jeffs et al., 2013). Diagram of unfolded and flattened V1, V2 and third-tier visual cortex showing the location of injection sites (colored ovals with black outline) and of transported cell label (filled ovals); intra-areal label is omitted. Insets: visual field maps of the location of the injection sites (small circles) and transported label in V1 (shaded colored regions). (A) Evidence for an upper quadrant representation bordering V2d. Closely spaced injections of four different neuroanatomical tracers across the full width of upper field DM resulted in cell label in upper field V1 and V2 that progressed from these areas' HM representation (blue ovals resulting from the blue injection site) to their VM representation (red and green ovals arising from the red and green injection sites, respectively). This demonstrated that the injection sites resided in a region representing the upper quadrant. That this region abutted V2d, rather than V3d, was further demonstrated by the location of transported blue label at the HM representation of V1d, indicating that the injection site that produced it (blue) straddled the HM representation at the border between DM and lower field V2. Had the blue injection resided at the border between V3d and DM, which represents the VM, the resulting blue label would have, instead, resided at the lower VM representation of V1 (at the location of the blue arrow). While injections straddling the HM representation at the border between V2d and V3d, which represents the HM, would also produce label at the HM representation in both upper and lower field V1 and V2 (Jeffs et al., 2009), the progressively more anterior injections would be expected to produce label in lower, rather than upper field V1 and V2, had they resided in V3d. These data demonstrated that upper field DM directly abuts V2d without an interposed area V3. (B) Evidence for two patches of lower quadrant representation bordering V2d. Seven closely spaced injections of different tracers (only 3 are shown in the cartoon for clarity) were made across the full width of V2d, as demonstrated by the topography of transported label in V1d, which showed an orderly progression from the lower VM representation, at the border between V1d and V2d, to the HM representation in V1d. These tracer injections also produced two orderly progressions of transported label abutting V2d anteriorly (marked as 1 and 2), which were mirror reversals of the tracer injection site sequence. A third label reversal (marked as 3) was located well anterior (1.5 mm) to reversal 1, and its topography was consistent with the retinotopic organization of area DA demonstrated by Rosa and Schmid (1995). There was no evidence for a label reversal posterior to reversal 1.