Do the fluorescent red eyes of the marine fish Tripterygion delaisi stand out? In situ and in vivo measurements at two depths

Abstract

Since the discovery of red fluorescence in fish, much effort has been invested to elucidate its potential functions, one of them being signaling. This implies that the combination of red fluorescence and reflection should generate a visible contrast against the background. Here, we present in vivo iris radiance measurements of Tripterygion delaisi under natural light conditions at 5 and 20 m depth. We also measured substrate radiance of shaded and exposed foraging sites at those depths. To assess the visual contrast of the red iris against these substrates, we used the receptor noise model for chromatic contrasts and Michelson contrast for achromatic calculations. At 20 m depth, T. delaisi iris radiance generated strong achromatic contrasts against substrate radiance, regardless of exposure, and despite substrate fluorescence. Given that downwelling light above 600 nm is negligible at this depth, we can attribute this effect to iris fluorescence. Contrasts were weaker in 5 m. Yet, the pooled radiance caused by red reflection and fluorescence still exceeded substrate radiance for all substrates under shaded conditions and all but Jania rubens and Padina pavonia under exposed conditions. Due to the negative effects of anesthesia on iris fluorescence, these estimates are conservative. We conclude that the requirements to create visual brightness contrasts are fulfilled for a wide range of conditions in the natural environment of T. delaisi.

Keywords: Michelson contrast; coloration; visual communication; visual contrast; visual signal.

Figure 1

Tripterygion delaisi displaying its conspicuous red iris fluorescence at 30 m depth. Picture taken with Nikon D4, LEE 287 Double C. T. Orange filter, and manual white balance, without postprocessing (Nico K. Michiels). Note that LEE 287 is not a long‐pass filter (as is, e.g., LEE 105 Orange or LEE 106 Primary Red). It is used to correct a bluish cold‐white scene to a warmer spectrum in photography (C. T. = “Correct to Tungsten”). Combined with Manual White Balance, this results in pictures that show colors at depth, including fluorescence, as perceived by a human diver

Figure 2

(a) Iris radiance measurements taken with a radiospectrometer aiming at a secured, slightly anesthetised fish at 20 m depth. (b) Substrate transect device with reflectance standards in the center (left to right): black standard, shaded diffuse white standard (PTFE) and exposed diffuse white standard (PTFE). The latter was used for the calculations presented here. Spectral measurements were taken approx. 1 cm above each of the 10 cable binder tips (yellow spot) while pointing horizontally at the substrate. The length of the central black carrier is 22.5 cm. (c) Substrate radiance measurements were taken at 5 and 20 m depth using a calibrated spectroradiometer in a custom‐made underwater housing (BS Kinetics)

Figure 3

(a) Line plots showing mean relative radiance (prop.) of typical Tripterygion delaisi substrate types as a function of wavelength at 5 and 20 m depth (rows) under sun‐exposed and shaded conditions (columns). Values exceeding 1 (black dashed line, referring to the radiance of the exposed diffuse white standard) indicate substrates that emitted more light at a specific wavelength than was available in the light spectrum, a typical signature of strong fluorescence. (b) Pie charts showing relative abundance of substrate types measured at each combination of depth and exposure. For a detailed species list, see Appendix S1

Figure 4

Line plot showing iris relative radiance (prop.) of Tripterygion delaisi as a function of wavelength under exposed (left column) and shaded (right column) conditions at either 5 m (upper row) or 20 m depth (lower row). Blue lines represent means ± SD (shading) of all fish. Red lines indicate the maximum relative radiance averaged across individuals (n = 34). Values exceeding 1 (horizontal black dashed line) indicate that more photons were emitted by the fish iris at that wavelength than were available in the ambient spectrum, indicative of red fluorescence (assuming absence of specular reflection). The blue dashed curve shows the estimated brightness of the iris without clove oil anesthesia (see Methods)

Figure 5

Boxplots of chromatic contrasts between mean iris and mean substrate radiance per substrate type. Data points (in red) represent processed, individual measurements. The horizontal line in the graph indicates the threshold of color discrimination, set at 1 just‐noticeable difference (JND). Values above 1 indicate that a contrast is likely to be perceived by Tripterygion delaisi. The algal species covering the substrate are coded as follows: Cl, Chaetomorpha linum; Dl, Dictyota linearis; Hf, Halopteris filicina; Jr, Jania rubens; and Pf, Padina pavonia (see Appendix S1)

Figure 6

Boxplots showing achromatic contrast (Michelson contrast) between mean iris and mean substrate radiance per substrate type. Data points (in red) represent processed, individual measurements. The achromatic contrasts are unitless (see Methods). Values above or below the gray horizontal strip (C = 0 +/‐ 0.018) indicate that a contrast is likely to be perceived by Tripterygion delaisi with positive values indicating irides being brighter than the substrate. The algal species covering the substrate are coded as follows: Cl, Chaetomorpha linum; Dl, Dictyota linearis; Hf, Halopteris filicina; Jr, Jania rubens; and Pf, Padina pavonia (see Table 1 for statistical significances and Appendix S1)