Zebrafish cone mosaic. In adult zebrafish, the cone mosaic is highly structured, with red and green double cones (DC) between blue (Long single, LS) and UV (Short single, SS) cones, with the red cone always next to the blue cone and the green cone always next to the UV cone (from Robinson et al., 1993, National Academy of Sciences, U.S.A. Reprinted with permission).

Zebrafish horizontal cells (HCs). HCs in zebrafish retina are morphologically distinct. H1-like cells are smaller in vertical profile and somal size compared to H2 and H3 cells, and all types have distinct dendritic connections with presynaptic photoreceptors (PRs). Reproduced with permission from Song et al. (2008).

HC responses. Stimulation of zebrafish eyecups with light of different spectral wavelengths evokes different responses from zebrafish HCs. (A,B) H1-type cells hyperpolarize to all wavelengths, giving an L-type response; while (C) H2 cells are biphasic, hyperpolarizing to G, B, and ultraviolet (UV) and depolarizing to R light. H3-type HCs, in contrast, display triphasic (D,E) and tetraphasic (F) responses to spectral stimuli (from Connaughton and Nelson, 2010).

Bipolar cells make diverse connections with PRs. Summary figure (reproduced with permission from Li et al., 2012) showing the different PR connections associated with morphologically-distinct types of BCs. Each BC was identified by axon terminal ramification within the six sublaminae (s1–s6) of the inner plexiform layer (IPL) and by dendritic connections with specific cone types. Most BC types make connections with multiple (>2) PR types, except for two types that exclusively contact green cones (at left of figure) Rod, contacts rods; R, red cones; G, green cones; B, blue cones; UV, UV cones. Shading reflects frequency of identification of a given type, with the darker (black) color indicating the most commonly observed type.

Color inputs to inner retina are segregated in the IPL. BC terminals in the larval zebrafish IPL can be segregated based on the spectral inputs they respond to across the visual scene. (Left) BC-terminal excitation by R, G, B, and UV stimuli are represented as red, green, blue and magenta dots located throughout the IPL; inhibitory responses are represented as black dots. The distribution of ON-responses varies based on the spectral stimulus and which part of the visual scene (outer horizon vs. “strike zone (SZ)”) is being viewed. The “SZ” represents looking forward and upward at prey items. (Right) Subsequent classification of the different spectral inputs from BCs into the IPL identified achromatic and color-opponent regions in this synaptic layer that were most evident outside the “SZ”. For the right panels: black dots = achromatic OFF-responses, white dots = achromatic ON-responses, red dots = color-opponent responses, gray dots = “other” (taken from Figure 3, Zimmermann et al., 2018).

Spectral responses of ACs are associated with specific cell types. Transient ON-OFF amacrine cells (AC) responses (A) were characteristic of bi-stratified ACs with dendrites in both the OFF- and ON-sublaminae (B). In contrast, sustained ON responses (C) were observed in ACs with a dendritic arbor monostratified in sublamina b(D). Spectrally multiphasic C-type responses were primarily observed in ON-cells, as shown in the biphasic response (E) from a monostratified AC with dendrites in s5 of sublamina b(F) (Torvund et al., 2017).

Zebrafish ganglion cells (GCs). Di-I labeling of GCs in whole mount tissue identified 11 morphological types, based on dendritic arborization patterns in the IPL. These patterns included processes restricted to 1–2 sublaminae, as well as more diffuse arborization patterns.Reproduced with permission from Mangrum et al. (2002).

GC spectral responses. Loose patch recordings from larval zebrafish GCs revealed a diversity of response types, most of which are spectrally multiphasic. Triphasic responses were the most abundant type identified and could be subdivided into three distinct spectral patterns. Bi-, tetra-, and pentaphasic responses were also recorded (Connaughton and Nelson, 2015).