For all visually competent organisms, the driving force behind the adaptation of photoreceptors involves obtaining the best balance of resolution to sensitivity in the prevailing light regime, as an increase in sensitivity often results in a decrease in resolution. A number of marine species have an additional problem to deal with, in that the juvenile stages live in relatively brightly lit shallow (100–200 m depth) waters, whereas the adult stages have daytime depths of more than 600 m, where little downwelling light remains. Here, I present the results of electrophysiological analyses of the temporal resolution and irradiance sensitivity of juvenile and adult stages of two species of ontogenetically migrating crustaceans (Gnathophausia ingens and Systellaspis debilis) that must deal with dramatically different light environments and temperatures during their life histories. The results demonstrate that there are significant effects of temperature on temporal resolution, which help to optimize the visual systems of the two life-history stages for their respective light environments.
This article is part of the themed issue ‘Vision in dim light’.
The ability to relate an organism's visual physiology to its environment has proven to be a particularly interesting topic of research in the aquatic world, owing to the variety of underwater light environments. For many visually competent organisms, regardless of habitat, the evolution of their eyes involved obtaining the best balance of resolution to sensitivity. This balance, of course, changes with the prevailing light regime, as an increase in sensitivity nearly always costs a decrease in resolution (reviewed in [1,2]). At low light levels, background ‘noise’ is quite high owing to the large random fluctuations resulting from small photon numbers, as well as biochemical/thermal noise in the photoreceptors themselves. Any strategy that will improve the signal-to-noise ratio will therefore improve the sensitivity of the eye. The strategies involved and consequential adaptations in the visual systems depend on the habitat depth of the organism. Finding other organisms as a contrast against the extended background ambient light field is the primary visual task at depths where ambient light is still available. At shallower depths with a brighter ambient light field, sensitivity can be sacrificed for resolution, so spatial and temporal resolution tend to be high. As the ambient light diminishes, a larger eye with larger pupil or facets would be optimal, and both temporal and spatial summation need to be greater . At even deeper depths, where identifying bright point sources (bioluminescence) against a dark environment is the primary visual task, a large eye/pupil is still required, but spatial and temporal summation should be at a minimum .
A number of marine species have an additional problem to deal with, in that the juvenile stages live in relatively brightly lit shallow (100–200 m depth) waters, whereas the adult stages live at daytime depths of more than 600 m, where little downwelling sunlight remains. Several studies have demonstrated that optical adaptations play a significant role in optimizing crustacean photoreceptors for the best sensitivity/resolution trade-off for their particular habitat. These include shifts from apposition optics, in which resolution is emphasized over sensitivity, to superposition optics, in which light entering a large number of facets is superposed on a single rhabdom, providing a much brighter image than that supplied by apposition optics. In general, species with apposition optics are usually active in well-lit environments, whereas superposition optics are associated with nocturnal species or species from dim-light environments (reviewed in ). The transition from larval apposition to juvenile superposition optics increases with each larval moult in decapod crustaceans [5–10], meaning that superposition optics should already be present in shallow-living juveniles before they make their ontogenetic migrations to deeper water. However, several studies have shown that, while juveniles of ontogenetic migrating Gnathophausia ingens (Lophogastridae) and Systellaspis debilis (Oplophoridae) do have superposition optics, their eyes have a much smaller clear zone than the adults, which minimizes light transfer between ommatidia, decreasing sensitivity but increasing spatial resolution [11–13]. Conversely, in the non-ontogenetic migrator Notostomus gibbosus, where both adults and juveniles reside below 700 m, the juveniles have a clear zone that is proportionally, given the smaller eye, the same size as that of the adult eye . Further structural adaptations include much higher screening pigment densities in shallow-living juveniles versus deep-living adults [12,14].
Another way that animals have adapted to dim-light environments is by extending the period during which photons are sampled, i.e. increasing the temporal summation in their visual systems. This can dramatically improve photon catch [2,15–17], although at the cost of temporal resolution. Temporal resolution and sensitivity are intrinsically linked—the higher the temporal resolution, the lower the temporal summation and the lower the sensitivity (reviewed in ). The classic studies of Autrum [19,20] demonstrated a clear correlation between the retinal response dynamics of insect photoreceptors and their lifestyle and habitat. Predictably, diurnal species have better temporal resolution than nocturnal species [18–22]. Several studies on deep-sea species have demonstrated a similar trend towards greater temporal resolution in species with shallower daytime depth ranges [23,24]. Therefore, one might anticipate that shallower-living juveniles within a species might have greater temporal resolution, as they can afford to sacrifice sensitivity in these more brightly lit waters in order to maintain good temporal resolution. In addition, the fact that juveniles live at warmer temperatures than the deep-living adults could have a significant effect on their temporal-resolving power, as increases in temperature have been shown to significantly increase the temporal resolution of fly photoreceptors .
Presented here are results of the first studies on the temporal resolution of several life-history stages of ontogenetically migrating species, as well as the effects of temperature on their temporal resolution. The significance of this research is based on the fact that in order to understand animal distribution patterns in marine ecosystems, and model the effects of environmental perturbations on their survival, one has to understand the factors that control their behaviour and interactions under normal conditions. If the photosensitivity of shallow-living juveniles is optimized for their brighter-light environment, then environmental perturbations that decrease environmental light, such as pollution (through increased turbidity and/or phytoplankton blooms) and global warming (which may drive shallow juveniles to deeper water with a dimmer-light field) could have a significant effect on their survivorship. Many of these species are major components of the oceanic food web that drive the major commercial fisheries, and understanding whether and how they adapt to different light environments is critical to modelling what effect a perturbation in their light environment may have on their ability to survive.
(a) Animal collections and maintenance
Animals were collected on research cruises on the RV New Horizon and RV Wecoma in San Clemente Basin, California, and on the RV Pelican and RV Cape Hatteras in the Gulf of Mexico. All animals were collected with a 9 m2 Tucker trawl fitted with a light-tight, thermally insulated cod-end which was closed at depth, so that animals were never exposed to thermal or light shock. Instars 10–12 of the lophogastrid crustacean G. ingens (carapace lengths 19.7–49 mm), hereafter referred to as adults, and instars 3 and 4 (carapace lengths 8.6–14.2 mm), hereafter referred to as juveniles , were collected off California. The oplophorid crustacean S. debilis was collected in the Gulf of Mexico. Individuals of carapace lengths 7.9–9.5 mm were considered juveniles, and individuals of carapace lengths 14.9–17.1 mm were considered adults [27,28]. Animals were maintained in light-tight containers in a cold room at 6.5°C (G. ingens) or 8.5°C (S. debilis) until used in experiments.
(b) Experimental set-up
All work was conducted using an electrophysiological set-up shipboard. Animal maintenance and experimental set-up was conducted under dim red light. Animals were placed in chilled seawater in an insulated recording chamber, and mounted on a post such that their pleopods were free to generate respiratory currents across the gills, with the upper quarter of the animal's eye in air above the water. Antifreeze chilled by a Lauda E100 circulating water bath was pumped through cooling coils in the recording chamber in order to maintain the seawater at the correct temperature. Temperature in the bath was continuously monitored with a HH11A Omega digital thermometer with a submersible thermocouple placed next to the animal's body. Single-ended electroretinograms were recorded with subcorneal metal microelectrodes (12–15 mΩ, FHC, Inc.), and amplified with an FHC model XCell-3 microelectrode amplifier, used with a high impedance probe to eliminate electrode polarization artefacts . The seawater bath was grounded with an AgCl-coated wire. Light from an Ocean Optics regulated white light supply, filtered with a 490 nm filter, was transmitted to the eye through one branch of a bifurcated fibre-optic light guide (EXFO) and placed 3 mm from the eye such that the entire eye was bathed in diffuse light. Stimulus irradiance was controlled via a neutral density wheel controlled by a stepper motor, and duration was controlled with a Uniblitz shutter, both under computer control. AC recordings were digitized, displayed on a computer monitor using a LabView (National Instruments Inc.) data acquisition program and stored for later analysis.
(c) Irradiance sensitivity experiments
Voltage versus log irradiance (V/logI) curves were generated in dark-adapted eyes at a given temperature (6.5°C for G. ingens; 8.5°C for S. debilis) in order to compare sensitivity between juveniles and adults. The eye was stimulated with 100 ms flashes of increasing irradiances of 490 nm light, with interflash intervals determined by recovery to the dark-adapted response level of a dim test flash. The data were normalized to the peak response (Vmax) for each individual, and mean V/log curves were generated for each species' life-history stage. These curves were fit with the Zettler modification of the Naka–Rushton equation [30–33], where V/Vmax = Im/(Im + Km). K is the stimulus irradiance eliciting half the maximum response (Vmax) and logK is a relative measure of the eye's sensitivity. The dynamic range, defined as the log irradiance range between response limits of 5–95% Vmax, corresponding to the voltage bandwidth of the photoreceptor , was also calculated from these curves as an additional measure of photoreceptor sensitivity.
(d) Temporal resolution
The temporal resolution of the eye was quantified by determining (i) the maximum critical flicker fusion frequency (CFFmax) and (ii) the response latency to a test flash of given irradiance and duration. The CFF is defined as the highest stimulus rate at which the eye can produce an electrical response to each flash over any 0.5 s interval within the 1.5 s stimulus train. As this characteristic is strongly dependent on stimulus irradiance , the CFFmax, which is the highest flicker rate that the eye is capable of following at any irradiance, was used to compare temporal resolution between the different life-history stages. Light adaptation also affects temporal dynamics in some crustaceans , so care was taken to ensure that the eye remained fully dark-adapted for dark-adapted measurements. Experiments involved presenting the eye with square pulses of light with a constant 50% duty cycle (50 : 50 light : dark ratio) for 1.5 s, generated by placing a computer-controlled electromagnetic shutter in the light path. For subsequent stimuli, the flicker rate and irradiance were increased until three 0.5-log increases in irradiance did not result in a faster flicker fusion frequency. The greatest frequency at which the eye could respond to individual flashes was defined as the CFFmax. To ensure that the eye was fully dark-adapted between successive flicker stimuli, a dim 100 ms test flash that elicited a 100 µV response was presented after every flicker stimulus. The eye was allowed to re-dark adapt until the test flash response had recovered to 100 µV; under the brightest light stimuli, recovery took up to 1 h.
The response latency, defined as the elapsed time between the onset of the light stimulus and the onset of the photoreceptor response, is also an indicator of the speed of transduction. Because this characteristic also varies with irradiance, the response latency was measured for test flashes that were 50% of the maximum amplitude that the eye was capable of generating to the brightest test flash, as determined from the VlogI curves.
(e) Temperature effects
Experiments were conducted at 5.5°C, 7.5°C and 9.5°C for G. ingens, and 6.5°C, 8.5°C and 12.5°C for S. debilis to correlate with ambient environmental temperatures at their depth ranges (figure 1). The initial temperature at the start of the experiment was randomized, and subsequent experimental temperatures were raised or lowered as required to cover as many temperatures as possible. Optimally, each individual would be tested at every temperature, but the juveniles of both species proved to be less resilient than the adults, and did not survive the full 24 h required to complete the full ranges of temperatures. The condition of the eye was continuously monitored with the dim test flash described above, and once the response to the test flash diminished, the experiment was terminated. For juvenile versus adult comparisons, data were grouped for each life-history stage and analysed with a one-way ANOVA. For effects of temperature, a paired two-sample for means T-test was used, to eliminate possible effects of individual variability on data analysis.
V/logI data for G. ingens, obtained at 6.5°C, were normalized to Vmax for each individual, and mean data for juveniles and adults were modelled using the Naka–Rushton equation (figure 2a). LogK (the log of the irradiance needed to give a 50% response, and an indicator of relative sensitivity) was higher for adults (10.98) than juveniles (10.66), but these differences were not statistically significant (single-factor ANOVA; p = 0.29). The dynamic range for both juveniles and adults was 4.5 log photons cm−2 s−1. There were no significant differences in CFFmax or response latencies between juveniles and adults of G. ingens when tested at the same temperatures (figure 2b). These data include data from table 1, as well as additional individuals for which data were obtained at only one temperature.
The logK for adult S. debilis was significantly lower than for the juveniles (logK = 10.45 versus logK = 11.09; p = 0.005), indicating a higher sensitivity relative to the juveniles (figure 3a). Conversely, the adult stages had significantly higher maximum CFF rates than juveniles at all temperatures tested (figure 3b), indicating a higher temporal resolution, which is often associated with lower sensitivity owing to a shorter integration time . These data again include data from table 1 as well as additional data from individuals that were not used in the paired analyses. There were not enough data for statistical analyses of response latencies between the two life-history stages except at 8.5°C, where there was no significant difference in response latency. The dynamic range for both life-history stages was 4.0 log photons cm−1 s−1.
Temperature significantly increased the flicker fusion frequencies of adults and juveniles of both species at each 2°C temperature interval (table 1). This analysis included only those individuals for which data from two temperature regimes were obtained. As temperature increased from 5.5°C to 9.5°C, the CFFmax for G. ingens juveniles increased by 12 Hz for juveniles and 9 Hz for adults, and the response latency decreased by 13 and 15 ms, respectively. Similar effects were present for S. debilis with temperature increases from 6.5°C to 12.5°C, where CFFmax increased by 10 Hz for both adults and juveniles, and response latencies decreased by 26 ms.
4. Discussion and conclusions
There have been several studies examining differences in optics and visual pigments between early life-history stages and adults of several species of ontogenetically migrating crustaceans [12,34,35], but this is the first study, to the best of my knowledge, comparing the physiological sensitivity and temporal resolution of adults and juveniles, as well as the effects of temperature on temporal resolution.
The lophogastrid crustacean G. ingens undergoes one of the most extensive ontogenetic migrations known (summarized in ), with daytime depths of instars 3 and 4 (small juveniles) between 175 and 300 m, and between 650 and 750 m for instars 10–12 (large adults). The adults do not undergo true vertical migrations, but disperse between 400 and 900 m at night. A study of their visual pigments and opsin complements demonstrated no differences between adults and juveniles, but did indicate that the juvenile age class has substantially more migrating screening pigment than the adults , which makes sense for the juveniles living in a brighter-light environment. In addition, although the juveniles have superposition optics like the adults, they have a substantially smaller clear zone, leading to a significantly lower calculated sensitivity than the deep-living adults . While electroretinography cannot be used for determination of absolute sensitivity to light , it can be used to make relative comparisons of sensitivity. The logK, measured at 6.5°C for both stages, was slightly higher for adults (higher logK means lower sensitivity) than for juveniles (logK = 10.98 versus logK = 10.66; figure 2), but these differences were not statistically significant (p = 0.29; single-factor ANOVA). These results suggest that, in spite of having a larger eye with a larger aperture, the optical differences are not translated into significantly greater irradiance sensitivity in the adults. The results of the temporal resolution experiments support this conjecture. When measured at the same temperature, there was no significant difference in either the response latency or the CFFmax (figure 2). However, there was a significant effect of temperature in both adults and juveniles, with an increase in CFFmax and concurrent decrease in response latency with each 2°C increase in temperature. The daytime depth of the juvenile stages is between 200 and 300 m, with temperatures ranging from 7.5 to 9°C versus more than 650 m for the adults, with temperatures between 4 and 5°C. Although the juveniles do have superposition optics, it appears that the warmer temperatures at their daytime depths (figure 1) serve as a significant adaptation for the brighter-light environment at their normal daytime depths. Once they migrate to deeper depths as adults, the decrease in temperature is sufficient to significantly decrease temporal resolution (equivalent to an increase in integration time) and hence enhance contrast sensitivity.
Histological studies on the eyes of S. debilis juveniles demonstrate that, like the adults, they have superposition optics throughout most of the eye . However, they have a smaller clear zone , possibly resulting in a decrease of light absorption at the retina  and retain apposition optics in the dorsal part of the eyes, with no clear zone . Eyes of the adults are also larger than those of the juveniles, giving them a greater calculated optical sensitivity owing to size alone . The results of the physiological sensitivity studies mimic the structural studies, with the sensitivity of adult eyes being significantly higher than that of juvenile eyes. The logK, measured at 8.5°C for both stages, was significantly lower for adults than for juveniles (logK = 10.45 versus logK = 11.03; p = 0.005, single-factor ANOVA; figure 3). However, the CFFmax was significantly higher in adults than juveniles at all three temperatures measured, indicative of a higher temporal resolution, which is usually associated with a lower irradiance sensitivity measured electrophysiologically , but only if the size and structure of the eyes are the same, which they clearly are not. While response latency is usually inversely related to CFFmax, decreasing as CFFmax increases, there were no significant differences between juveniles and adults with respect to this parameter. As with G. ingens, temperature increases also significantly increased temporal resolution and decreased response latency in both juveniles and adults. Interestingly, CFFmax continued to be higher for adults than juveniles at all temperatures, which could be an adaptation for the vertical migrations undertaken by this species. In addition to being ontogenetic migrators, S. debilis also undergoes extensive vertical migrations on a nightly basis. The larger adults migrate up from their daytime depths of 600–800 m into the epipelagic zone (0–200 m) every night [28,37] with the smaller juveniles making similar migrations, but of a smaller range owing to their shallower daytime depths (figure 1). With the lack of light at night, one would anticipate that a lower temporal resolution would be optimal. However, the greatest component of the adult diet (juvenile diets have not been studied) is euphausiid crustaceans , which are bioluminescent. Therefore, an increase in temporal resolution would enable S. debilis to more easily track their bioluminescent prey. At the night-time depths of both life-history stages, the ambient temperature is between 20 and 24°C, considerably warmer than the maximum temperatures tested in this study, and it is likely that the temporal resolution is substantially higher at these temperatures than what is reported here. This would also be true at the daytime depths of the juvenile life-history stages, where ambient temperatures are 4–5°C higher than what was tested here. These data indicate that, like G. ingens, the ambient temperatures at the daytime depth ranges of the juveniles serve to increase the temporal resolution (decreasing integration time) and this serves as an additional adaptation for vision in a brighter-light environment.
Light adaptation has been shown to shorten the response time in most arthropods, which enhances the frequency response to capture fast visual events [18,21,38,39]. However, in most deep-sea crustaceans studied to date, with the exception of euphausiid species with bilobed eyes, the temporal resolution of their photoreceptors is not affected by light adaptation, and this has been shown to be true for adult S. debilis  and G. ingens  as well as the juveniles (TM Frank 2005–2007, unpublished data). In contrast to light adaptation, temperature has a significant effect on the response dynamics of the eye, similar to what was reported in the Antarctic amphipod Abyssorchomene plebs , flies  and tunas . The Q10 (6.5–12.5°C) ranged from 2.1 to 2.4 for adults and juveniles of S. debilis, and from 3.5 for the juveniles and 3.1 for the adults of Gnathophausia (5.5–9.5°C). These Q10s are near the range expected for biochemical reactions , suggesting that the increasing temperature accelerated the phototransduction cascade, improving the ability of these eyes to track moving targets, i.e. capture prey, at the expense of photon capture. These physiological effects of temperature on temporal resolution serve to help optimize photoreceptors of ontogenetically migrating species for the different light environments they encounter during their life histories. For the shallow-living juvenile species, in their (comparatively) brightly lit environment, having a higher temporal resolution is a benefit to tracking prey in an environment where animals are always moving relative to their prey. The reduced photon capture resulting from a higher temporal resolution is not a problem in this brighter-light environment. For the deeper-living adults, the reduced temperatures at their deeper depths increase photon capture in their light-limited environment. The vertically migrating adult S. debilis also benefit from the increased temporal resolution at their shallow night-time depths, as they specialize on bioluminescent prey, and can afford to give up contrast resolution in order to track prey that bioluminesce when mechanically stimulated.
However, these data also suggest that these temperature effects may have negative consequences for juvenile life-history stages in the presence of anthropogenic disturbances that decrease environmental light. Increasing levels of pollution may decrease ambient light (through increased turbidity), and has been found to decrease reaction distances in several species of fish . Global warming may eventually drive juveniles to deeper (and dimmer) water that meets their temperature tolerance. There have been no studies on nektonic crustaceans, but studies on demersal fish in the North Sea showed that the whole assemblage has deepened significantly at an average rate of 3.6 m decade−1, with some individual species deepening up to 10 m decade−1, tracking warming trends in temperature . As the authors state, fisheries exploitation has had major effects, but their analysis indicated that the depth response was clearly correlated with climate change and increasing ocean water temperature. The eyes of the juvenile life-history stages of the crustaceans in this study are physiologically adapted for higher temporal resolution in their current brighter environment. The anthropogenic disturbances mentioned above could affect their ability to find prey, and the survival of the juvenile life-history stages has been shown to have a significant impact on the species' ecological success in eight commercially exploited crustacean species .
Raw data for the figures and tables in this manuscript, including conductivity/temperature/depth temperature versus depth data, VlogI data, response latencies and flicker fusion frequencies for Systellaspis debilis and Gnathophausia ingens can be found at the Dryad Digital Repository: http://dx.doi.org/10.5061/dryad.16df2.
I have no competing interests.
This work was supported by a grant from the National Science Foundation (IBN-0343871) to T.F.
I thank the captains and crews of the RV ‘New Horizon’, RV ‘Wecoma’, RV ‘Pelican’ amd RV ‘Cape Hatteras' for their invaluable help with animal collections.
One contribution of 17 to a theme issue ‘Vision in dim light’.
- Accepted July 19, 2016.
- © 2017 The Author(s)
Published by the Royal Society. All rights reserved.