220
Ron H. Douglas
4 Visual Pigments of Deep-Sea Fish
In fact the visual pigments of deep-sea fish are often cited as representing the classic
example of visual pigment adaptation to the environment. Downwelling sunlight is
rapidly attenuated with depth until, even in ideal conditions, at around I 000 m
insufficient light penetrates to allow vision in even the most sensitive fish (Dentor
1990). Usually, however, sunlight becomes visually irrelevant at shallower depths.
Not only is the intensity of this downwelling light reduced as it penetrates the water
column, its spectral composition also becomes increasingly restricted. Due to spectral
filtering by the water, at depth, light primarily consists of a narrow band of radiation
between 470 and 480 nm (Kirk 1983). For this reason, it was predicted more than 60
years ago (Clark 1936; Bayliss et al. 1936) that in comparison to species living nearer
the water surface, deep-sea fish should have visual pigments with their absorption
maximum shifted towards shorter wavelengths. Superficially at least, such a
hypsochromatic shift has been amply confirmed (Douglas et al. 1998a for review).
Figure 3 shows the A max values for all 175 species of deep-sea fish examined to
date. Whereas it is certainly true that deep-sea fish on average have visual pigments
absorbing around 470-490 nm, as might be expected based on the predominant
wavelength in their environment, a simple matching to downwelling sunlight cannot
be the whole explanation.
Firstly, the range ofAmax values (444 nm to in excess of580 nm), is much greater
than would be expected for a simple match to the background. Thus, animals caught
in a single trawl and hence occupying approximately the same depths in the same
geographical location, often have visual pigments whose A max values differ by 20 nm
or more. Secondly, for simple background matching one would expect a gradual blue
shift of visual pigments with depth. This is not the case (Douglas et a!. 1998a).
Thirdly, a simple sensitivity hypothesis cannot explain the presence of more than one
visual pigment within the retinae of several deep-sea species (Douglas and Partridge
1997; Douglas et a!. 1998a). Finally, theoretical modelling of the visual pigments
required to optimally detect residual sunlight (Partridge, pers. comm.; Douglas eta!.
1998a) indicates that visual pigments with maximum absorption at shorter
wavelengths than those observed in deep-sea species are required.
Does one therefore conclude that the absorption of the visual pigment has little to
do with the downwelling illumination? It is, for instance, possible that the point of
maximum absorption of the pigment is determined largely by ambient pressure,
phylogenetic constraints, the need to maximize the signal-to-noise ratio or thermal
stability (Douglas et al. 1998a for review).
It has also been suggested that the precise location of the A max of deep-sea fish
visual pigments is of little consequence due to their relatively broad absorption
spectrum resulting from their unusually high concentration within the outer segments
(Munz 1965; Bowmaker 1995).
Ron H. Douglas
4 Visual Pigments of Deep-Sea Fish
In fact the visual pigments of deep-sea fish are often cited as representing the classic
example of visual pigment adaptation to the environment. Downwelling sunlight is
rapidly attenuated with depth until, even in ideal conditions, at around I 000 m
insufficient light penetrates to allow vision in even the most sensitive fish (Dentor
1990). Usually, however, sunlight becomes visually irrelevant at shallower depths.
Not only is the intensity of this downwelling light reduced as it penetrates the water
column, its spectral composition also becomes increasingly restricted. Due to spectral
filtering by the water, at depth, light primarily consists of a narrow band of radiation
between 470 and 480 nm (Kirk 1983). For this reason, it was predicted more than 60
years ago (Clark 1936; Bayliss et al. 1936) that in comparison to species living nearer
the water surface, deep-sea fish should have visual pigments with their absorption
maximum shifted towards shorter wavelengths. Superficially at least, such a
hypsochromatic shift has been amply confirmed (Douglas et al. 1998a for review).
Figure 3 shows the A max values for all 175 species of deep-sea fish examined to
date. Whereas it is certainly true that deep-sea fish on average have visual pigments
absorbing around 470-490 nm, as might be expected based on the predominant
wavelength in their environment, a simple matching to downwelling sunlight cannot
be the whole explanation.
Firstly, the range ofAmax values (444 nm to in excess of580 nm), is much greater
than would be expected for a simple match to the background. Thus, animals caught
in a single trawl and hence occupying approximately the same depths in the same
geographical location, often have visual pigments whose A max values differ by 20 nm
or more. Secondly, for simple background matching one would expect a gradual blue
shift of visual pigments with depth. This is not the case (Douglas et a!. 1998a).
Thirdly, a simple sensitivity hypothesis cannot explain the presence of more than one
visual pigment within the retinae of several deep-sea species (Douglas and Partridge
1997; Douglas et a!. 1998a). Finally, theoretical modelling of the visual pigments
required to optimally detect residual sunlight (Partridge, pers. comm.; Douglas eta!.
1998a) indicates that visual pigments with maximum absorption at shorter
wavelengths than those observed in deep-sea species are required.
Does one therefore conclude that the absorption of the visual pigment has little to
do with the downwelling illumination? It is, for instance, possible that the point of
maximum absorption of the pigment is determined largely by ambient pressure,
phylogenetic constraints, the need to maximize the signal-to-noise ratio or thermal
stability (Douglas et al. 1998a for review).
It has also been suggested that the precise location of the A max of deep-sea fish
visual pigments is of little consequence due to their relatively broad absorption
spectrum resulting from their unusually high concentration within the outer segments
(Munz 1965; Bowmaker 1995).