Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2024 Sep 15;105(6):1843–1849. doi: 10.1111/jfb.15930

Is the diel cycle of routine metabolic rate in mummichog (Fundulus heteroclitus) affected by near‐infrared lighting used for visualizing behavior of fishes at night?

Annie M Trembley 1, Lauren E Rowsey 1, Ben Speers‐Roesch 1,✉
PMCID: PMC11650951  PMID: 39279054

Abstract

The metabolic rate of a freely moving fish (routine metabolic rate) is tightly coupled with volitional movement (spontaneous activity), both of which commonly show strong daily cycles linked to the species‐specific diel activity pattern. Mummichog (Fundulus heteroclitus), an important estuarine fish in the north western Atlantic Ocean, are historically reported as diurnal (i.e., more active during daylight). Our recent laboratory studies on a Bay of Fundy population, however, showed a free‐running (i.e., similarly active daytime and night‐time) or even nocturnal (i.e., more active at night‐time) diel activity pattern. In the laboratory, near‐infrared (NIR) illumination is commonly used with a NIR‐sensitive camera to visualize fish activity across the light–dark periods of the day. Because NIR light is close to the visible light spectrum and certain fishes show sensitivity to NIR, the use of NIR with mummichog possibly could disturb the animals and obscure the identification of their true diel activity pattern. We aimed to determine if NIR illumination (940 nm wavelength) influences the diel activity pattern of mummichog. We used measurements of routine metabolic rate (oxygen consumption rate, MO2) as a proxy for activity, as evaluating the effect of NIR requires treatments where NIR lights are off, which precludes visualization and direct assessment of fish activity at night‐time. We measured routine MO2 of mummichogs over 6 days, exposed to either NIR off–on–off (2 days for each off or on period) or the opposite sequence of NIR on–off–on (to control for time‐dependent effects). NIR lights did not influence the diel cycle of routine MO2, and activity by proxy, in mummichog. Thus, NIR illumination is a suitable method to visualize mummichog during light–dark diel cycles. Routine MO2, and presumably activity, was similar or higher during night‐time periods compared to daytime periods, confirming a free‐running or nocturnal activity pattern for at least certain populations of mummichog.

Keywords: behavior, diel activity pattern, fish, metabolic rate, near‐infrared light

1. INTRODUCTION

The diel cycle of activity in fishes is a ubiquitous behavioral response to light intensity changes over the 24‐h day (Reebs, 2002). Fish species can be nocturnal, diurnal, crepuscular, or have a free‐running activity, but diel activity patterns are frequently plastic (Reebs, 2002). For example, Atlantic salmon (Salmo salar) show a change in foraging activity from diurnal to nocturnal in response to decreasing temperature, independent of photoperiod and season (Fraser et al., 1993). Activity patterns are intimately linked to important life processes, including foraging, reproduction, and predator avoidance. Activity itself is also energetically costly, so diel cycles of activity will correlate with routine metabolic rate, and activity can be modified to conserve energy (Middleton et al., 2024; Nilsson et al., 1993; Reeve et al., 2022). Thus, knowing the diel activity pattern of a species is fundamental to understanding its biology and predicting ecological interactions.

Diel activity patterns can be identified in the field using telemetry or inferred from capture frequency in passive devices, such as traps; in the laboratory, spontaneous locomotor activity (spontaneous activity) can be monitored using video analysis or by fish‐induced disruptions of sound or light waves (Reebs, 2002). Monitoring fish in the dark (e.g., at night) is commonly facilitated by near‐infrared (NIR) light illumination, which is beyond the visible light spectrum but allows fish to be visualized using NIR‐sensitive video cameras (Hinch & Collins, 1991; Mussen & Peeke, 2001; Pautsina et al., 2015; Reeve et al., 2022; Zhou et al., 2017). The use of NIR is expected to result in minimal or no disturbance of the fish, as, like most vertebrates, fish are generally thought to be unable to see NIR wavelengths of light (Kusmic & Gualtieri, 2000; Pautsina et al., 2015; Zhou et al., 2017). However, at least some fish species apparently can see NIR light to some extent (Hartmann et al., 2018; Matsumoto & Kawamura, 2005; Meuthen et al., 2012; Shcherbakov et al., 2013). Shcherbakov et al. (2013) suggested that the ability to see NIR light may relate to the turbidity of the water of the fish's natural habitat, as turbidity increases light attenuation, and the longer wavelengths of NIR can penetrate further in the water. Even if NIR‐sensitive species may only perceive it dimly, the laboratory use of NIR light has the potential to confound the accurate determination of their diel activity patterns. One previous study indicates that NIR light (840 nm wavelength) does not entrain circadian clock gene expression in zebrafish (Dekens et al., 2017), but the effect of NIR on diel activity or metabolic rate patterns in fishes has not yet been investigated.

The mummichog or common killifish Fundulus heteroclitus (L. 1766) is a common estuarine fish ranging from Newfoundland to Florida and a model species in organismal and environmental biology (Burnett et al., 2007). Kavaliers (1980) reported that mummichog (from Massachusetts) are diurnal based on 20–40 days of laboratory activity measurements of individually or group‐housed fish using an ultrasonic system. Field studies in New England and Delaware have also suggested predominantly diurnal foraging and activity, although some activity can occur at night (Allen et al., 1994; Halpin, 1997; Weisberg et al., 1981). Overall, the accepted consensus to date is that mummichog are diurnal. Recently, however, our laboratory study (Reeve et al., 2022) of individually housed mummichogs from a Bay of Fundy population showed similar activity during daytime and night‐time (i.e., free‐running) or more activity at night (i.e., nocturnal) across a range of temperatures from 14 to 2.5°C in an open arena with daily feeding. Unlike the earlier studies, however, Reeve et al. (2022) used NIR lights (940 nm wavelength) to illuminate the fish. To our knowledge, it is unknown if mummichog can see NIR light, but they are sensitive to certain peculiarities of light (Bressman et al., 2016); if mummichog can see NIR light, then use of NIR lights by Reeve et al. (2022) may have created brighter conditions at night‐time (i.e., more similar to daytime, when visible lighting was on), which could have stimulated abnormally higher nocturnal activity.

Given the uncertainty about the mummichog's true diel activity pattern or their sensitivity to NIR light used for laboratory studies of their behavior, the objective of the present study was to determine if the diel activity pattern in mummichogs is influenced by NIR illumination. We investigated this by measuring routine metabolic rate (specifically, routine oxygen consumption rate, MO2) as a proxy for activity in mummichog over 6 days during which NIR lights were sequentially turned off or on for 48‐h periods. Activity and MO2 are strongly positively correlated in mummichog (Reeve et al., 2022), so measuring MO2 allowed us to infer activity levels during all experimental periods, including when NIR lights were off during night‐time and the fish could not be visualized. If NIR lights influence activity levels in mummichog, we predicted that MO2 would be similar between daytime and night‐time or higher at night‐time when NIR lights were on; when NIR lights were off, MO2 would be higher during the daytime and lower at night‐time. Alternatively, if there was no effect of NIR light, the fish would show the same diel cycle throughout the experiment regardless of whether the NIR lights were on or off.

2. METHODS

2.1. Animals

Mummichog (F. heteroclitus) of mixed sexes, weighing 4.2 ± 1.1 g (mean ± SD), were collected on August 15, 2022, from Sam Orr Pond, Bocabec, New Brunswick, using minnow traps. Fish were transported to University of New Brunswick, Saint John, and kept in holding tanks for at least 8 weeks prior to the experiment; experiments were carried out from October 17 to November 30, 2022. The 165‐L holding tanks (86 cm diameter, 29 cm depth) were supplied with re‐circulating, filtered, and aerated seawater maintained at 14 ± 0.6°C using a commercial chiller (1/2 horsepower EcoPlus, Morr Inc., Commerce, CA, USA). Fish were fed dry pellets on Monday, Wednesday, and Friday mornings between 9:00 a.m. and 10:00 am (1.8 mm CLEAN Assist, Skretting, St. Andrews, New Brunswick, Canada). Mummichog were under a 10 h:14 h light–dark photoperiod, with the light period including a simulated sunrise and sunset (30 min each). This photoperiod was selected because it matched the one used in our earlier work, showing a nocturnal or free‐running diel activity in mummichogs (Reeve et al., 2022); by keeping the photoperiod the same, we could focus on whether the use of NIR light (as in Reeve et al., 2022) could have influenced the diel activity cycle. Fish collections were approved by Fisheries and Oceans Canada, and experiments were approved by the Animal Care Committee at UNB Saint John following the guidelines of the Canadian Council on Animal Care.

2.2. Experimental system for measuring the diel cycle of MO2 with or without NIR illumination

The experimental system was similar to the one used by Reeve et al. (2022) and consisted of five circular clear acrylic respirometers (10.1 cm internal diameter, 2.5 cm internal height, 299 ± 22 mL total volume, including re‐circulation loop) submerged in a clear acrylic tank containing re‐circulating aerated seawater at 14 ± 0.6°C. Respirometry details are provided below. Water in the clear acrylic tank was supplied from a sump (containing biological and mechanical filters) through a commercial chiller (1/3 horsepower Arctica chiller, JBJ Chillers, St Charles, MO, USA) for temperature control and then drained back to the sump. The acrylic tank rested on a thin sheet of translucent white acrylic (1.5 mm thickness for light diffusion), and this combined structure was raised on clear acrylic cylinders. Three commercially available NIR light‐emitting diode (LED) illuminators (each: 940 nm, 140 × 5‐mm LED, 45° beam angle, 15 W) were distributed 7.5 cm adjacent and 3.5 cm (to top of illuminator) below the raised acrylic tank with the NIR illuminators shining horizontally toward the tank. This is a standard method by which IR illumination is used to visualize fish in laboratory experiments; specifically, the fish is silhouetted against the diffused, but bright, NIR light from below, and this silhouette can be recorded with a NIR‐sensitive camera and later manually or automatically tracked to assess movement or behavior (Reeve et al., 2022).

2.3. Experimental protocol

For each experimental trial, four fish were removed from their holding tanks and placed independently into a respirometer in the experimental system (one respirometer was left empty to estimate background respiration; see below). Fish were allowed a 24‐h period of adjustment within the respirometers before the experimental trial began (see below). All fish were fasted for a total of 48 h prior to each trial and were not fed for the duration of the experiment to ensure that specific dynamic action would not confound our results. The weight and length of the fish were measured at the end of the experimental period. The experimental room was illuminated throughout the experiment with overhead fluorescent lights on a 10 h:14 h light–dark photoperiod, including a 30‐min simulated sunrise and sunset.

Four experimental trials were run while MO2 was continuously measured, with two trials for each of two NIR lighting sequences (four fish per trial for a total of eight fish for each NIR lighting sequence). The trials were run one after the other across several weeks. For trials 1 and 3, NIR lights were initially off for 48 h (two complete daytime–night‐time cycles), then were turned on for 48 h, and then were turned off again for 48 h. Trials 2 and 4 were carried out with the opposite sequence: NIR lights initially were on for 48 h (two complete daytime–night‐time cycles), then off for 48 h, then on for 48 h. By carrying out trials with two opposing sequences of NIR illumination, we controlled for potential effects of time spent in the respirometer (including lack of feeding during the trial). One fish from trial 4 was excluded from further analysis because of an oxygen probe malfunction, resulting in n = 7 for the NIR on–off–on sequence, and n = 15 for the entire study.

2.4. Measurement of routine metabolic rate

Routine metabolic rate was estimated by measuring whole‐animal oxygen consumption rate (MO2, mg O2 kg−1 h−1) using automated intermittent‐closed respirometry, as described in Reeve et al. (2022). Briefly, the respirometers each had two barbed fittings to create a re‐circulation loop driven by a Eheim pump and two barbed fittings for inflow and outflow of flush water from a separate Eheim pump (Eheim 300, 5 1 min−1, under a low flow secured with an adjustable clamp to provide consistent flow). The respirometers were sealed with a clear acrylic lid fitted with an O‐ring and stainless steel washers and wingnuts. The outer walls of the respirometers were wrapped in black plastic to reduce visual disturbance. Each respirometer was equipped with an oxygen probe fitted inline on the re‐circulation loop, and two temperature probes were placed in the water‐bath, allowing measurement of temperature‐compensated oxygen levels within the respirometers. The probes (PyroScience, Aachen, Germany) were interfaced with a computer for data recording using two four‐channel Firesting modules (PyroScience, Aachen, Germany). Flush periods were 4 min, and closed periods were 11 min. A blank respirometer without a fish was run simultaneously during each trial to correct for background oxygen consumption.

The slope of O2 decline was extracted over 9 min in each closed period using a custom R script (see acknowledgments for source) after the first and last 60 s of the closed period were excluded (RStudio version 2022.02.3 + 492). For one data file, slopes had to be extracted manually over the 9‐min segment within each closed period due to an inconsistent sampling interval (Excel, version 16.63.1, Microsoft Corporation, Redmond, WA, USA). MO2 was calculated from the slope of O2 decline after accounting for the respirometer volume, fish body mass, and background respiration, as described in Rowsey et al. (2024). The MO2 for each fish was calculated for every closed period throughout its trial. For each fish, any individual MO2 value (i.e., from any single closed period) that fell outside 2 SDs of the average of all that fish's MO2 values over the trial duration was considered outliers and excluded from further analysis. MO2 values were then averaged for each fish within three 3.33‐h periods during the 10‐h daytime (morning 08:30 a.m.–11:50 a.m., midday 11:50 a.m.–3:10 p.m., and late day 153:10 p.m.–6:30 p.m.) and four 3.5‐h periods during the 14‐h night‐time (early night 6:30 p.m.–10:00 p.m., midnight 10:00 p.m.–1:30 a.m., late night 1:30 a.m.–5:00 a.m., pre‐sunrise 05:00–08:30) across a given trial when NIR lights were either off or on, resulting in a single routine MO2 value per fish for each combination of time period of day and NIR light off or on (e.g., late night with NIR lights on). Our routine MO2 measurements represent routine metabolic rate (i.e., MO2 of animals undergoing spontaneous activity).

2.5. Statistical analysis

To determine if NIR lights or time period of the day significantly influenced routine metabolic rate, and activity by proxy, we used a generalized linear mixed‐effects model (GLM) in R using the function glmer (lme4 package). We tested whether routine MO2 values across the day were affected by the fixed factors NIR on/off, time period of the day, and lighting sequence (off–on–off or on–off–on) with fish ID nested as a random factor within trial. Significance was determined with the ANOVA function (car package), which calculated p‐values for the model using type‐II Wald χ2 tests. We found no significant effect of NIR lighting sequence (GLM: W1 = 0.03, p = 0.87) or interactions with other factors, and thus we removed sequence from subsequent models for simplification of comparisons. Our models showed temporal autocorrelation of residuals (Durbin–Watson test: p < 0.001), and we corrected for this by conducting model selection with AIC incorporating an autoregressive‐moving‐average (ARMA) model structure. The glmer function is not capable of using ARMA model structure, and so to account for temporal autocorrelation, we used the function lme (nlme package) to build our models. Residuals of the GLM and the linear models (i.e., the glmer and lme functions) were normally distributed and passed model assumptions. Post hoc analysis of the effect of time period of the day was conducted with the top model (i.e., the model with the lowest AIC value) by calculating estimated marginal means using the emmeans function (emmeans package) and then running Tukey's pair‐wise comparisons. p‐Values for pair‐wise comparisons were adjusted for multiple comparisons using the Holm–Bonferroni method. Significance was accepted at p < 0.05.

3. RESULTS

All routine metabolic rates (assessed as routine oxygen consumption rate, MO2) of individual fish measured throughout the experimental trials for each NIR light sequence are shown in Figure 1. As mentioned above, NIR light sequence had no effects on our data, so it was excluded as a factor in our main analysis of the effects of NIR illumination and time period of the day on routine metabolic rate. Routine metabolic rate was significantly higher at mid‐ and late day than during night‐time (Tukey's comparisons: p < 0.01) (Figure 2). Routine metabolic rate was unaffected by NIR illumination (GLM: W1 = 1.65, p = 0.20), and there was no interaction between NIR illumination and time period of the day (GLM: W6 = 1.64, p = 0.95); similar MO2 was observed at any given time of day, regardless of whether the NIR lights were on or off (Figure 2). Thus, NIR illumination did not affect the diel cycle of routine metabolic rate.

FIGURE 1.

FIGURE 1

Open in a new tab

All measurements of routine metabolic rate (assessed as routine oxygen consumption rate, MO2) in individual mummichog (Fundulus heteroclitus) over the experiment duration, which involved six diel cycles (days) and two distinct NIR illumination sequences that helped control for potential time‐dependent influences (two trials for each sequence): panel (a) NIR lights were initially off, then on, then off (n = 8); or panel (b) NIR lights were initially on, then off, and then on (n = 7). Translucent colored circles are the MO2 values of each individual fish (these appear darker where multiple individual values were similar and thus superimposed). Gray segments represent night‐time (visible lights off), and white segments represent daytime (visible lights on). Black vertical line represents near‐infrared (NIR) lights being turned on or off. These raw data were used to calculate routine MO2 during different time periods of the day with NIR illumination off or on (see main text and Figure 2).

FIGURE 2.

FIGURE 2

Open in a new tab

The effect of near‐infrared (NIR) light illumination on the diel cycle of routine metabolic rate (measured as routine oxygen consumption rate, MO2) in mummichog (Fundulus heteroclitus) (n = 15). Data are represented as boxplots, with X denoting the mean and horizontal line denoting the median. Upper and lower hinges are 75th and 25th percentiles, respectively. Upper and lower whiskers are upper/lower hinge ±1.5* interquartile range (IQR). The colored circles are the routine MO2 of the individual fish at each time period of the day for NIR off or on. The gray segment represents night‐time (visible lights off), and white segments represent daytime (visible lights on). Different lowercase letters denote statistical significance between time periods of the day (p < 0.05). There was no effect of NIR illumination, nor any significant interaction between the time of the day and NIR illumination (p > 0.05, generalized linear mixed‐effects model and Tukey's comparisons; see results text). See Figure S1 where values within an individual are connected by a line to highlight the individual responses.

4. DISCUSSION

To our knowledge, our study is the first to directly assess the effect of NIR illumination on diel metabolic or behavioral cycles in a fish. We found no evidence that NIR lighting (940 nm wavelength) affects the diel cycle of routine metabolic rate (routine MO2) and, by proxy, activity levels in mummichog. Our result is welcome news as NIR illumination is a powerful tool for direct and accurate behavioral analysis of fish under poor visible light (Pautsina et al., 2015; Reeve et al., 2022; Zhou et al., 2017). Our results provide circumstantial evidence that mummichog cannot see NIR light at the wavelength tested or at least not well enough for it to affect their diel cycles of activity and MO2. Although not directly tested in previous studies, NIR illumination also does not appear to perturb the well‐established diurnal or nocturnal activity patterns in a variety of other fish species (e.g., Atlantic salmon, Salmo salar; cunner, Tautogolabrus adspersus; pumpkinseed sunfish, Lepomis gibbosus; American eel, Anguilla rostrata) (Pautsina et al., 2015; Reeve et al., 2022), which is consistent with the consensus that most fishes cannot perceive light at NIR wavelengths (Kusmic & Gualtieri, 2000; Pautsina et al., 2015; Reebs, 2002; Zhou et al., 2017). On the contrary, certain fish species have been reported to be sensitive to NIR light to varying degrees, including zebrafish larvae (Danio rerio), common carp (Cyprinus carpio), guppy (Poecilia reticulata), and several cichlids (Pelvicachromis taeniatus, Oreochromis mossambicus, Oreochromis niloticus) (Hartmann et al., 2018; Matsumoto & Kawamura, 2005; Meuthen et al., 2012; Shcherbakov et al., 2013). Thus, there is a broad, albeit sporadic, phylogenetic occurrence of NIR sensitivity among teleosts, including a weak sensitivity in Poecilia reticulata (Shcherbakov et al., 2013), which, like mummichog, belong to order Cyprinodontiformes. Unfortunately, the ability of mummichogs to see IR light has never been established directly. The genomics and optics of visual cones in mummichog suggest a broad sensitivity to the visible light spectrum, which is consistent with their shallow water habitats (Flamarique & Hárosi, 2000; Lupše et al., 2022; Suliman & Flamarique, 2014). Among fishes, red‐sensitive cones, in particular, have a sensitivity closest to the NIR wavelengths, and having more red‐light gene copies creates the potential for more mutation and alternative function (Lupše et al., 2022) that could increase the potential to see wavelengths beyond the typical spectral capabilities. However, in mummichog, expression of red‐sensitive cones is similar in proportion to the other cone types that allow vision at other visible wavelengths (Lupše et al., 2022). Thus, overall, the existing data do not support NIR sensitivity in mummichogs, which is consistent with our findings, especially given the intense level of NIR light put out by our NIR lamps. Although direct studies on the NIR sensitivity of the mummichog retina is warranted, especially as their commonly turbid environment would seem to make NIR sensitivity beneficial (Shcherbakov et al., 2013), we conclude that NIR illumination at 940 nm is a suitable tool for studies on mummichog behavior and metabolic rate. Importantly, however, many commercially available NIR lamps emit at 850 nm, and caution is warranted when extrapolating our results to this shorter NIR wavelength, which is closer to the visible spectrum. Although one previous study showed no effect of shorter NIR wavelengths (840 nm) on circadian clock genes in zebrafish (Dekens et al., 2017), we recommend the use of NIR lamps that emit at 940 nm when the spectral sensitivity of the study species is unknown and, even then, this wavelength may be visible for certain species (e.g., Oreochromis tilapia [Shcherbakov et al., 2013]).

We found that MO2 of the mummichogs varied between night‐time and daytime, and routine MO2 was similar to or higher during night‐time compared to the daytime period, inferring a free‐running or nocturnal diel activity pattern for the mummichog from the studied Bay of Fundy population. Routine MO2 was highest during early night, and within daytime, it decreased from the morning onward, suggesting that greater activity during the daytime or night‐time phases of the day follows the daily changes in visible light and/or, in the case of morning, by entrainment to the morning feeding schedule used during holding. Collectively, these findings confirm our previous findings on the diel cycle of activity and MO2 of this population of mummichog (Reeve et al., 2022), including similar absolute levels of MO2 between the present study and Reeve et al. (2022). This nocturnal or free‐running activity pattern of our studied Bay of Fundy mummichog (Reeve et al., 2022; present study) contrasts with the only previous laboratory studies by Kavaliers (1980) (using direct activity measurement) and Borowiec et al. (2018) (inferred from the diel pattern of MO2) that indicated diurnality in mummichogs originating from Massachusetts and New Hampshire, respectively. Previous field studies also suggested a predominately diurnal activity pattern in mummichog, although activity at night has been noted (Allen et al., 1994; Halpin, 1997; Weisberg et al., 1981). Kavaliers (1980) monitored activity using an ultrasonic system, whereas we used NIR illumination; however, our findings indicate that the discrepancy in diel activity pattern compared to Kavaliers (1980) cannot be related to this methodological difference. Other factors must explain the discrepancy between our findings and those of Kavaliers (1980) or Borowiec et al. (2018). Photoperiod and salinity varied between the studies, but these are unlikely explanations for the discrepancy, because our other studies on our mummichog population showed that their diel activity pattern was unaffected by photoperiod or salinity (Senathirajah, 2020). We studied individually housed mummichogs, but this is unlikely to influence their diel activity pattern because Kavaliers (1980) observed a diurnal activity pattern in both individually housed and group‐housed fish, and in Reeve et al. (2022), the fish could see each other through their clear plastic housings. Kavaliers (1980) and Borowiec et al. (2018) used warmer temperatures (20–22°C) than those used by us (14°C) or Reeve et al. (2022) (≤14°C). Therefore, the influence of temperature on mummichog diel cycles should be further explored. Another possibility is that there are differences in diel activity patterns between the different mummichog populations that have been studied, as a result of distinct local environmental conditions (e.g., warmer or cooler habitats, higher or lower tidal flux, etc.) or other population‐specific factors (e.g., southern or northern subspecies genotypes (Burnett et al., 2007)). In this context, it is relevant that the field studies of Halpin (1997) and Weisberg et al. (1981) revealed some nocturnal foraging activity by mummichog (albeit less than in daytime), indicating that mummichog have inherent flexibility to be nocturnal under certain circumstances. Future research should explore intraspecific variation in diel activity patterns among mummichog populations, as well as the influence of water temperature.

AUTHOR CONTRIBUTIONS

Conceptualization: Annie M. Trembley, Lauren E. Rowsey, and Ben Speers‐Roesch. Methodology: Annie M. Trembley, Lauren E. Rowsey, and Ben Speers‐Roesch. Validation: Annie M. Trembley. Formal analysis: Annie M. Trembley. Investigation: Annie M. Trembley. Data curation: Annie M. Trembley. Writing—original draft: Annie M. Trembley. Writing—review and editing: Annie M. Trembley, Lauren E. Rowsey, and Ben Speers‐Roesch. Supervision: Ben Speers‐Roesch. Funding acquisition: Ben Speers‐Roesch.

FUNDING INFORMATION

Grants to Ben Speers‐Roesch: Discovery grant (National Science and Engineering Research Council of Canada); Young Investigator Award (Harrison McCain Foundation); Start‐Up and Equipment Funds (New Brunswick Innovation Foundation); University Research Fellowship (University of New Brunswick); John R. Evans Leadership Fund (Canada Foundation for Innovation).

Supporting information

Figure S1. The diel cycle of routine metabolic rate (measured as routine oxygen consumption rate, MO2) in mummichog (n = 15) when near‐infrared (NIR) lights were off (panel a) or on (panel b), with the responses of the individual mummichogs highlighted with a connecting line between each individual's MO2 values. The gray segment represents night‐time (visible lights off), and white segments represent daytime (visible lights on). X denotes the mean, and the triangle denotes the median. The data shown are the same as the mean, median, and individual data in Figure 2.

JFB-105-1843-s001.docx (65.3KB, docx)

ACKNOWLEDGMENTS

We thank Dr. Matt Gilbert, Ella Middleton, and Dr. Connor Reeve for technical advice. Dr. Tommy Norin provided the invaluable MO2 R script, which was originally written in Mathematica by Dr. Hans Malte and translated to R by Dr. Norin and Alexandre Mercière.

Trembley, A. M. , Rowsey, L. E. , & Speers‐Roesch, B. (2024). Is the diel cycle of routine metabolic rate in mummichog (Fundulus heteroclitus) affected by near‐infrared lighting used for visualizing behavior of fishes at night? Journal of Fish Biology, 105(6), 1843–1849. 10.1111/jfb.15930

REFERENCES

  1. Allen, E. A. , Fell, P. E. , Peck, M. A. , Gieg, J. A. , Guthke, C. R. , & Newkirk, M. D. (1994). Gut contents of common mummichogs, Fundulus heteroclitus L., in a restored impounded marsh and in natural reference marshes. Estuaries, 17, 462–471. [Google Scholar]
  2. Borowiec, B. G. , McClelland, G. B. , Rees, B. B. , & Scott, G. R. (2018). Distinct metabolic adjustments arise from acclimation to constant hypoxia and intermittent hypoxia in estuarine killifish (Fundulus heteroclitus). Journal of Experimental Biology, 221, jeb190900. [DOI] [PubMed] [Google Scholar]
  3. Bressman, N. R. , Farina, S. C. , & Gibb, A. C. (2016). Look before you leap: Visual navigation and terrestrial locomotion of the intertidal killifish Fundulus heteroclitus . Journal of Experimental Zoology A, 325, 57–64. [DOI] [PubMed] [Google Scholar]
  4. Burnett, K. G. , Bain, L. J. , Baldwin, W. S. , Callard, G. V. , Cohen, S. , Di Giulio, R. T. , Evans, D. H. , Gómez‐Chiarri, M. , Hahn, M. E. , Hoover, C. A. , & Karchner, S. I. (2007). Fundulus as the premier teleost model in environmental biology: Opportunities for new insights using genomics. Comparative Biochemistry and Physiology D, 2, 257–286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dekens, M. P. S. , Foulkes, N. S. , & Tessmar‐Raible, K. (2017). Instrument design and protocol for the study of light controlled processes in aquatic organisms, and its application to examine the effect of infrared light on zebrafish. PLoS One, 12, e0172038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Flamarique, I. N. , & Hárosi, F. I. (2000). Photoreceptors, visual pigments, and ellipsosomes in the killifish, Fundulus heteroclitus: A microspectrophotometric and histological study. Visual Neuroscience, 17, 403–420. [DOI] [PubMed] [Google Scholar]
  7. Fraser, N. C. , Metcalfe, N. B. , & Thorpe, J. E. (1993). Temperature‐dependent switch between diurnal and nocturnal foraging in salmon. Proceedings Royal Society London B, 252, 135–139. [Google Scholar]
  8. Halpin, P. M. (1997). Habitat use pattern of the Mummichog, Fundulus heteroclitus, in New England. I. Intermarsh variation. Estuaries, 20, 618–625. [Google Scholar]
  9. Hartmann, S. , Vogt, R. , Kunze, J. , Rauschert, A. , Kuhnert, K. D. , Wanzenböck, J. , Lamatsch, D. K. , & Witte, K. (2018). Zebrafish larvae show negative phototaxis to near‐infrared light. PLoS One, 13, e0207264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Hinch, S. G. , & Collins, N. C. (1991). Importance of diurnal and nocturnal nest defense in the energy budget of male smallmouth bass: Insights from direct video observations. Transactions of the American Fisheries Society, 120, 657–663. [Google Scholar]
  11. Kavaliers, M. (1980). Social groupings and circadian activity of the killifish, Fundulus heteroclitus . Biological Bulletin, 158, 69–76. [Google Scholar]
  12. Kusmic, C. , & Gualtieri, P. (2000). Morphology and spectral sensitivities of retinal and extraretinal photoreceptors in freshwater teleosts. Micron, 31, 183–200. [DOI] [PubMed] [Google Scholar]
  13. Lupše, N. , Kłodawska, M. , Truhlářová, V. , Košátko, P. , Kašpar, V. , Bitja Nyom, A. R. , & Musilova, Z. (2022). Developmental changes of opsin gene expression in ray‐finned fishes (Actinopterygii). Proceedings of the Royal Society B, 289, 20221855. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Matsumoto, T. , & Kawamura, G. (2005). The eyes of the common carp and Nile tilapia are sensitive to near‐infrared. Fisheries Science, 71, 350–355. [Google Scholar]
  15. Meuthen, D. , Rick, I. P. , Thünken, T. , & Baldauf, S. A. (2012). Visual prey detection by near‐infrared cues in a fish. Naturwissenschaften, 99, 1063–1066. [DOI] [PubMed] [Google Scholar]
  16. Middleton, E. K. , Gilbert, M. J. , Landry, T. , Lamarre, S. G. , & Speers‐Roesch, B. (2024). Environmental variation associated with overwintering elicits marked metabolic plasticity in a temperate salmonid, Salvelinus fontinalis . Journal of Experimental Biology, 227, jeb246743. 10.1242/jeb.246743 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Mussen, T. D. , & Peeke, H. V. S. (2001). Nocturnal feeding in the marine threespine stickleback (Gasterosteus aculeatus L.): Modulation by chemical stimulation. Behaviour, 138, 857–871. [Google Scholar]
  18. Nilsson, G. E. , Rosén, P. , & Johansson, D. (1993). Anoxic depression of spontaneous locomotor activity in crucian carp quantified by a computerized imaging technique. Journal of Experimental Biology, 180, 153–162. [Google Scholar]
  19. Pautsina, A. , Císař, P. , Štys, D. , Terjesen, B. F. , & Espmark, Å. M. (2015). Infrared reflection system for indoor 3D tracking of fish. Aquacultural Engineering, 69, 7–17. [Google Scholar]
  20. Reebs, S. G. (2002). Plasticity of diel and circadian activity rhythms in fishes. Reviews in Fish Biology and Fisheries, 12, 349–371. [Google Scholar]
  21. Reeve, C. , Rowsey, L. E. , & Speers‐Roesch, B. (2022). Inactivity and the passive slowing effect of cold on resting metabolism as the primary drivers of energy savings in overwintering fishes. Journal of Experimental Biology, 255, jeb243407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Rowsey, L. E. , Reeve, C. , Savoy, T. , & Speers‐Roesch, B. (2024). Thermal constraints on exercise and aerobic performance do not explain the use of dormancy as an overwintering strategy in the cunner (Tautogolabrus adspersus). Journal of Experimental Biology, 227, jeb246741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Senathirajah, E. J. (2020). The effect of salinity and photoperiod on the diel activity pattern of mummichogs (Fundulus heteroclitus). Honours thesis. University of New Brunswick Saint John. Retrieved from https://unbscholar.lib.unb.ca/items/714b6d3f-753f-4f22-a02c-d8fa89beddd2 [Google Scholar]
  24. Shcherbakov, D. , Knörzer, A. , Espenhahn, S. , Hilbig, R. , Haas, U. , & Blum, M. (2013). Sensitivity differences in fish offer near‐infrared vision as an adaptable evolutionary trait. PLoS One, 8, e64429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Suliman, T. , & Flamarique, I. N. (2014). Visual pigments and opsin expression in the juveniles of three species of fish (rainbow trout, zebrafish, and killifish) following prolonged exposure to thyroid hormone or retinoic acid. Journal of Comparative Neurology, 522, 98–117. [DOI] [PubMed] [Google Scholar]
  26. Weisberg, S. B. , Whalen, R. , & Lotrich, V. A. (1981). Tidal and diurnal influence on food consumption of a salt marsh killifish Fundulus heteroclitus . Marine Biology, 61, 243–246. [Google Scholar]
  27. Zhou, C. , Zhang, B. , Lin, K. , Xu, D. , Chen, C. , Yang, X. , & Sun, C. (2017). Near‐infrared imaging to quantify the feeding behavior of fish in aquaculture. Computers and Electronics in Agriculture, 135, 233–241. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1. The diel cycle of routine metabolic rate (measured as routine oxygen consumption rate, MO2) in mummichog (n = 15) when near‐infrared (NIR) lights were off (panel a) or on (panel b), with the responses of the individual mummichogs highlighted with a connecting line between each individual's MO2 values. The gray segment represents night‐time (visible lights off), and white segments represent daytime (visible lights on). X denotes the mean, and the triangle denotes the median. The data shown are the same as the mean, median, and individual data in Figure 2.

JFB-105-1843-s001.docx (65.3KB, docx)

Articles from Journal of Fish Biology are provided here courtesy of Wiley

RESOURCES