The thermal sensitivity of heart rate (fH) in fishes has fascinated comparative physiologists for well over a century. We now know that elevating fH is the primary mechanism through which fishes increase convective oxygen delivery during warming to meet the concomitant rise in tissue oxygen consumption. Thus, limits on fH can constrain whole-animal aerobic metabolism. In this Review, we discuss an increasingly popular methodology to study these limits, the measurement of pharmacologically induced maximum fH (fH,max) during acute warming of an anaesthetized fish. During acute warming, fH,max increases exponentially over moderate temperatures (Q10∼2-3), but this response is blunted with further warming (Q10∼1-2), with fH,max ultimately reaching a peak (Q10≤1) and the heartbeat becoming arrhythmic. Because the temperatures at which these transitions occur commonly align with whole-animal optimum and critical temperatures (e.g. aerobic scope and the critical thermal maximum), they can be valuable indicators of thermal performance. The method can be performed simultaneously on multiple individuals over a few hours and across a broad size range (<1 to >6000 g) with compact equipment. This simplicity and high throughput make it tractable in lab and field settings and enable large experimental designs that would otherwise be impractical. As with all reductionist approaches, the method does have limitations. Namely, it requires anaesthesia and pharmacological removal of extrinsic cardiac regulation. Nonetheless, the method has proven particularly effective in the study of patterns and limits of thermal plasticity and holds promise for helping to predict and mitigate outcomes of environmental change.

Identifying and characterizing spawning locations are paramount for the protection of critical fish habitats but can be challenging, particularly in remote locations. Using the underexplored oviduct-tagging technique, we aimed to identify the timing and location of spawning for wild Arctic char (Salvelinus alpinus) and lake trout (Salvelinus namaycush) in two high-Arctic lakes in Nunavut. Specifically, Innovasea V7 acoustic telemetry transmitters were inserted into the oviducts of 13 Arctic char and 4 lake trout, and the timing and location of tag expulsion were determined using a fine-scale positioning system. Twenty Arctic char and 20 lake trout were also tagged with abdominal V16 transmitters, and 10 of them were paired with the oviduct tags, to further study the behavior of individual fish during the spawning season. Oviduct tags from four Arctic char and one lake trout could be used to assess the timing and location of spawning. Spawning anadromous Arctic char drastically reduced their activity and remained proximate to their presumed spawning location immediately before and for months after spawning. In contrast, a non-anadromous (i.e., freshwater resident) Arctic char and a lake trout showed little to no reduction in activity around presumed spawning events. Because of the highlighted sedentary behavior of inferred spawning anadromous Arctic char implanted with both abdominal and oviduct tags, we could also infer potential spawning based on the behavior of individuals equipped only with abdominal tags. Spawning areas identified via telemetry also aligned well with Inuit knowledge of those lakes. This is the first field study to use acoustic oviduct and abdominal tags coupled with a fine-scale positioning system. Despite a limited success rate of ejection, the study reveals the strong potential of the method to study spawning habitat and timing, particularly in remote areas.

Female Pacific salmon often experience higher mortality than males during their once-in-a-lifetime up-river spawning migration, particularly when exposed to secondary stressors (e.g. high temperatures). However, the underlying mechanisms remain unknown. One hypothesis is that female Pacific salmon hearts are more oxygenlimited than those of males and are less able to supply oxygen to the body's tissues during this demanding migration. Notably, female hearts have higher coronary blood flow, which could indicate a greater reliance on this oxygen source. Oxygen limitations can develop from naturally occurring coronary blockages (i.e. coronary arteriosclerosis) found in mature salmon hearts. If female hearts rely more heavily on coronary blood flow but experience similar arteriosclerosis levels as males, they will have disproportionately impaired aerobic performance. To test this hypothesis, we measured resting (RMR) and maximum metabolic rate (MMR), aerobic scope (AS) and acute upper thermal tolerance in coho salmon (Oncorhynchus kisutch) with an intact or artificially blocked coronary oxygen supply. We also assessed venous blood oxygen and chemistry (cortisol, ions and metabolite concentrations) at different time intervals during recovery from exhaustive exercise. We found that coronary blockage impaired MMR, AS and the partial pressure of oxygen in venous blood (PvO2) during exercise recovery but did not differ between sexes. Coronary ligation lowered acute upper thermal tolerance by 1.1 degrees C. Although we did not find evidence of enhanced female reliance on coronary supply, our findings highlight the importance of coronary blood supply for mature wild salmon, where migration success may be linked to cardiac performance, particularly during warm water conditions.

In the recent monitoring guidelines released by the Arctic Monitoring and Assessment Program's Litter and Microplastic Expert Group, Arctic salmonids were recommended as an important species for monitoring plastics in Arctic ecosystems, with an emphasis on aligning microplastic sampling and analysis methods in Arctic fishes. This recommendation was based on the minimal documentation of microplastics in Northern fishes, especially Arctic salmonids. In response, we worked collaboratively with local partners to quantify and characterize microplastics in Arctic char, Salvelinus alpinus , and their habitats in a commercial fishery near Iqaluktuuttiaq (Cambridge Bay), Nunavut. We sampled Arctic char, surface water, and benthic sediments within their summer foraging habitat at Palik (Byron Bay). We found microplastics in 95 % of char with an average of 26 (SD +/- 19) particles per individual. On average, surface water samples had 23 (SD +/- 12) particles/L and benthic sediment <1 particles/g (ww) . This is the first documentation of plastic pollution in Arctic char and their coastal habitats. Future work should evaluate seasonal, temporal and spatial trends for long-term monitoring of microplastics in Arctic fishes and their habitats.

ABSTRACT Poleward winters commonly expose animals, including fish, to frigid temperatures and low food availability. Fishes that remain active over winter must therefore balance trade-offs between conserving energy and maintaining physiological performance in the cold, yet the extent and underlying mechanisms of these trade-offs are not well understood. We investigated the metabolic plasticity of brook char (Salvelinus fontinalis), a temperate salmonid, from the biochemical to whole-animal level in response to cold and food deprivation. Acute cooling (1°C day−1) from 14°C to 2°C had no effect on food consumption but reduced activity by 77%. We then assessed metabolic performance and demand over 90 days with exposure to warm (8°C) or cold winter (2°C) temperatures while fish were fed or starved. Resting metabolic rate (RMR) decreased substantially during initial cooling from 8°C to 2°C (Q10=4.2–4.5) but brook char exhibited remarkable thermal compensation during acclimation (Q10=1.4–1.6). Conversely, RMR was substantially lower (40–48%) in starved fish, conserving energy. Thus, the absolute magnitude of thermal plasticity may be masked or modified under food restriction. This reduction in RMR was associated with atrophy and decreases in in vivo protein synthesis rates, primarily in non-essential tissues. Remarkably, food deprivation had no effect on maximum oxygen uptake rates and thus aerobic capacity, supporting the notion that metabolic capacity can be decoupled from RMR in certain contexts. Overall, our study highlights the multi-faceted energetic flexibility of Salvelinus spp. that likely contributes to their success in harsh and variable environments and may be emblematic of winter-active fishes more broadly.

Climatic warming elevates mortality for many salmonid populations during their physically challenging up-river spawning migrations, yet, the mechanisms underlying the increased mortality remain elusive. One hypothesis posits that a cardiac oxygen insufficiency impairs the heart’s capacity to pump sufficient oxygen to body tissues to sustain up-river swimming, especially in warm water when oxygen availability declines and cardiac and whole-animal oxygen demand increases. We tested this hypothesis by measuring cardiac and metabolic (cardiorespiratory) performance, and assessing the upper thermal tolerance of coho salmon (Oncorhynchus kisutch) during sustained swimming and acute warming. By surgically ligating the coronary artery, which naturally accumulates arteriosclerotic lesions in migrating salmon, we partially impaired oxygen supply to the heart. Coronary ligation caused drastic cardiac impairment during swimming, even at benign temperatures, and substantially constrained cardiorespiratory performance during swimming and progressive warming compared to sham-operated control fish. Furthermore, upper thermal tolerance during swimming was markedly reduced (by 4.4 °C) following ligation. While the cardiorespiratory capacity of female salmon was generally lower at higher temperatures compared to males, upper thermal tolerance during swimming was similar between sexes within treatment groups. Cardiac oxygen supply is a crucial determinant for the migratory capacity of salmon facing climatic environmental warming.

ABSTRACT Migratory fishes commonly encounter large and rapid thermal variation, which has the potential to disrupt essential physiological functions. Thus, we acclimated wild, migratory Arctic char to 13°C (∼7°C above a summer average) for an ecologically relevant period (3 days) and measured maximum heart rate (ƒH,max) during acute warming to determine their ability to rapidly improve cardiac function at high temperatures. Arctic char exhibited rapid compensatory cardiac plasticity similar to past observations following prolonged warm acclimation: they reduced ƒH,max over intermediate temperatures (−8%), improved their ability to increase ƒH,max during warming (+10%), and increased (+1.3°C) the temperature at the onset of an arrhythmic heartbeat, a sign of cardiac failure. This rapid cardiac plasticity may help migrating fishes such as Arctic char mitigate short-term thermal challenges. Furthermore, by using mobile Arctic research infrastructure in a remote field location, the present study illustrates the potential for field-based, experimental physiology in such locations.

Warm acclimation in fish is often characterized by an increase in heat tolerance and a reduction in physiological rates to improve the scope to respond to additional challenges including further warming. The speed of these responses can determine their effectiveness. However, acclimation rates vary across levels of biological organization and are poorly understood in part because most research is conducted after an acclimation period of >3 weeks, when acclimation is presumed to be complete. Here we show that when rainbow trout were transferred from 10 to 18 °C, over 50% of the total reduction of maximum heart rate (ƒHmax) (i.e. the thermal compensation at moderate temperatures) occurred within 72 h, with further compensation occurring more gradually over the following 25 days. Also, the ability to increase ƒHmax with acute warming improved within 24 h resulting in a 30% rise in peak ƒHmax, but this ultimately declined again with prolonged (28 days) exposure to 18 °C. In contrast with some previous studies, upper critical temperatures for ƒHmax did not increase. Nonetheless, we demonstrate that rapid cardiac plasticity is possible in rainbow trout and likely blunts the impacts of thermal variation over relatively short timescales, such as that associated with heat waves and migration between water bodies.

Increasing heart rate (ƒH) is a central, if not primary mechanism used by fishes to support their elevated tissue oxygen consumption during acute warming. Thermal acclimation can adjust this acute response to improve cardiac performance and heat tolerance under the prevailing temperatures. We predict that such acclimation will be particularly important in regions undergoing rapid environmental change such as the Arctic. Therefore, we acclimated Arctic char (Salvelinus alpinus), a high latitude, cold-adapted salmonid, to ecologically relevant temperatures (2, 6, 10, 14 and 18 °C) and examined how thermal acclimation influenced their cardiac heat tolerance by measuring the maximum heart rate (ƒHmax) response to acute warming. As expected, acute warming increased ƒHmax in all Arctic char before ƒHmax reached a peak and then became arrhythmic. The peak ƒHmax, and the temperature at which peak ƒHmax (Tpeak) and that at which arrhythmia first occurred (Tarr) all increased progressively (+33%, 49% and 35%, respectively) with acclimation temperature from 2 to 14 °C. When compared at the same test temperature ƒHmax also decreased by as much as 29% with increasing acclimation temperature, indicating significant thermal compensation. The upper temperature at which fish first lost their equilibrium (critical thermal maximum: CTmax) also increased with acclimation temperature, albeit to a lesser extent (+11%). Importantly, Arctic char experienced mortality after several weeks of acclimation at 18 °C and survivors did not have elevated cardiac thermal tolerance. Collectively, these findings suggest that if wild Arctic char have access to suitable temperatures (<18 °C) for a sufficient duration, warm acclimation can potentially mitigate some of the cardiorespiratory impairments previously documented during acute heat exposure.

Despite immense concern over amplified warming in the Arctic, physiological research to address related conservation issues for valuable cold-adapted fish, such as the Arctic char (Salvelinus alpinus), is lacking. This crucial knowledge gap is largely attributable to the practical and logistical challenges of conducting sensitive physiological investigations in remote field settings. Here, we used an innovative, mobile aquatic-research laboratory to assess the effects of temperature on aerobic metabolism and maximum heart rate (f Hmax) of upriver migrating Arctic char in the Kitikmeot region of Nunavut in the central Canadian Arctic. Absolute aerobic scope was unchanged at temperatures from 4 to 16°C, while f Hmax increased with temperature (Q 10 = 2.1), as expected. However, f Hmax fell precipitously below 4°C and it began to plateau above ~ 16°C, reaching a maximum at ~ 19°C before declining and becoming arrhythmic at ~ 21°C. Furthermore, recovery from exhaustive exercise appeared to be critically impaired above 16°C. The broad thermal range (~4-16°C) for increasing f Hmax and maintaining absolute aerobic scope matches river temperatures commonly encountered by migrating Arctic char in this region. Nevertheless, river temperatures can exceed 20°C during warm events and our results confirm that such temperatures would limit exercise performance and thus impair migration in this species. Thus, unless Arctic char can rapidly acclimatize or alter its migration timing or location, which are both open questions, these impairments would likely impact population persistence and reduce lifetime fitness. As such, future conservation efforts should work towards quantifying and accounting for the impacts of warming, variable river temperatures on migration and reproductive success.

The Arctic climate is changing rapidly, yet predicting how aquatic species will respond to these changes remains challenging given the lack of empirical data for most high-latitude taxa. Acoustic telemetry has recently emerged as an important methodology for understanding horizontal and vertical space-use patterns in fishes. Here, we used acoustic telemetry to document marine habitat use and depth/temperature preference of 26 anadromous Arctic char Salvelinus alpinus within the Kitikmeot Sea region of the Canadian central Arctic over 4 yr (20132016). Most detections (similar to 70%) were within the top 3 m of the water column, and most were in estuarine (72.6%) vs. marine (27.4%) habitats. Arctic char preferred deeper waters later in the summer, but the temperature they occupied remained relatively constant throughout the marine feeding season (similar to 5-8 degrees C). Most Arctic char exhibited some degree of repetitive diving behavior, with individuals diving to 35 m. Diving activity increased later in the summer marine feeding season and is likely a response to the seasonal transition of their preferred prey to deeper waters as the season progresses. Finally, Arctic char preferred deeper waters with less ice cover and during the day, the latter suggesting potential diel patterns to marine habitat use. Finally, year-to-year variation in Arctic char depth and temperature use was very modest despite differences in climatic and ice conditions. This result suggests that habitat use is relatively fixed and may reflect their thermal and osmoregulatory physiology, which has important implications for forecasting the impacts of a changing Arctic on this economically valuable species.

Anadromy was documented in 16 lake trout, Salvelinus namaycush, from Canada's central Arctic using capture data and otolith microchemistry. For the first time, estuarine/marine habitat use was described for five individuals using acoustic telemetry. Age-at-first-migration to sea was variable (10-39 years) among individuals and most S. namaycush undertook multiple anadromous migrations within their lifetime. Telemetry data suggested that S. namaycush do not travel far into marine habitats and prefer surface waters (<2 m). These results further our collective understanding of the marine ecology of Arctic S. namaycush.

This study hypothesized that oxygen uptake (& x1e40;O-2) measured with a novel protocol of chasing rainbow troutOncorhynchus mykissto exhaustion inside a static respirometer while simultaneously monitoring & x1e40;O-2(& x1e40;O-2chase) would generate the same and repeatable peak value as when peak active & x1e40;O-2(& x1e40;O-2active) is measured in a critical swimming speed protocol. To reliably determine peak & x1e40;O-2chase, and compare to the peak during recovery of & x1e40;O(2)after a conventional chase protocol outside the respirometer (& x1e40;O-2rec), this study applied an iterative algorithm and a minimum sampling window duration (i.e., 1 min based on an analysis of the variance in background and exercise & x1e40;O-2) to account for & x1e40;O(2)dynamics. In support of this hypothesis, peak & x1e40;O-2active(707 +/- 33 mg O(2)h(-1)kg(-1)) and peak & x1e40;O-2chase(663 +/- 43 mg O(2)h(-1)kg(-1)) were similar (P= 0.49) and repeatable (Pearson's and Spearman's correlation test;r >= 0.77;P < 0.05) when measured in the same fish. Therefore, estimates of & x1e40;O(2max)can be independent of whether a fish is exhaustively chased inside a respirometer or swum to fatigue in a swim tunnel, provided & x1e40;O(2)is analysed with an iterative algorithm and a minimum but reliable sampling window. The importance of using this analytical approach was illustrated by peak & x1e40;O(2chase)being 23% higher (P < 0.05) when compared with a conventional sequential interval regression analysis, whereas using the conventional chase protocol (1-min window) outside the respirometer increased this difference to 31% (P < 0.01). Moreover, because peak & x1e40;O(2chase)was 18% higher (P < 0.05) than peak & x1e40;O-2rec, chasing a fish inside a static respirometer may be a better protocol for obtaining maximum & x1e40;O-2.

We examined cardiac pacemaker rate resetting in rainbow trout following a reciprocal temperature transfer. In the original experiment, performed in winter, 4°C-acclimated fish transferred to 12°C reset intrinsic heart rate after just 1 h (from 56.8±1.2 to 50.8±1.5 bpm); 12°C-acclimated fish transferred to 4°C reset intrinsic heart rate after 8 h (from 33.4±0.7 to 37.7±1.2 bpm). However, in a replicate experiment, performed in the summer using a different brood year, intrinsic heart rate was not reset, even after 10 weeks at a new temperature. Using this serendipitous opportunity, we compared mRNA expression changes of a suite of proteins in sinoatrial node (SAN), atrial and ventricular tissues after both 1 h and longer than 3 weeks for both experimental acclimation groups to identify those changes only associated with pacemaker rate resetting. Of the changes in mRNA expression occurring after more than 3 weeks of warm acclimation and associated with pacemaker rate resetting, we observed downregulation of NKA α1c in the atrium and ventricle, and upregulation of HCN1 in the ventricle. However, in the SAN there were no mRNA expression changes unique to the fish with pacemaker rate resetting after either 1 h or 3 weeks of warm acclimation. Thus, despite identifying changes in mRNA expression of contractile cardiac tissues, there was absence of changes in mRNA expression directly involved with the initial, rapid pacemaker rate resetting with warm acclimation. Importantly, pacemaker rate resetting with thermal acclimation does not always occur in rainbow trout.

Acute warming in fish increases heart rate (fH) and cardiac output to peak values, after which performance plateaus or declines and arrhythmia may occur. This cardiac response can place a convective limitation on systemic oxygen delivery at high temperatures. To test the hypothesis that autonomic cardiac regulation protects cardiac performance in rainbow trout during acute warming, we investigated adrenergic and cholinergic regulation during the onset and progression of cardiac limitations. We explored the direct effects of adrenergic stimulation by acutely warming an in situ, working, perfused heart until arrhythmia occurred, cooling the heart to restore rhythmicity and rewarming with increasing adrenergic stimulation. Adrenergic stimulation produced a clear, dose-dependent increase in the temperature and peak fH achieved prior to the onset of arrhythmia. To examine how this adrenergic protection functions in conjunction with cholinergic vagal inhibition in vivo, rainbow trout fitted with ECG electrodes were acutely warmed in a respirometer until they lost equilibrium (CTmax) with and without muscarinic (atropine) and β-adrenergic (sotalol) antagonists. Trout exhibited roughly equal and opposing cholinergic and adrenergic tones on fH that persisted up to critical temperatures. β-adrenergic blockade significantly lowered peak fH by 14-17%, while muscarinic blockade significantly lowered the temperature for peak fH by 2.0°C. Moreover, muscarinic and β-adrenergic blockers injected individually or together significantly reduced CTmax by up to 3°C, indicating for the first time that cardiac adrenergic stimulation and cholinergic inhibition can enhance acute heat tolerance in rainbow trout at the level of the heart and the whole animal.

ABSTRACT As fish approach fatigue at high water velocities in a critical swimming speed (Ucrit) test, their swimming mode and oxygen cascade typically move to an unsteady state because they adopt an unsteady, burst-and-glide swimming mode despite a constant, imposed workload. However, conventional rate of oxygen uptake (ṀO2) sampling intervals (5–20 min) tend to smooth any dynamic fluctuations in active ṀO2 (ṀO2active) and thus likely underestimate the peak ṀO2active. Here, we used rainbow trout (Oncorhynchus mykiss) to explore the dynamic nature of ṀO2active near Ucrit using various sampling windows and an iterative algorithm. Compared with a conventional interval regression analysis of ṀO2active over a 10-min period, our new analytical approach generated a 23% higher peak ṀO2active. Therefore, we suggest that accounting for such dynamics in ṀO2active with this new analytical approach may lead to more accurate estimates of maximum ṀO2 in fishes. Summary: Accurately estimating peak oxygen uptake as a fish approaches fatigue in a swimming test must account for oxygen uptake being dynamic despite a constant, imposed workload.

1. The Arctic is warming at twice the global average rate; how native and non-native anadromous fishes will respond remains largely unknown. Some native Arctic salmonids are already experiencing warm (>21 degrees C), physically challenging migratory river conditions and large diurnal temperature fluctuations (>10 degrees C). 2. We conducted field and laboratory experiments to determine how these extreme conditions may affect the capacity for migration in Arctic and temperate salmonids. 3.In adult migratory Arctic char, reflex impairment following a handling challenge increased with temperature, indicating more extensive fatigue. In Arctic char smolts, temperature did not affect initial critical swimming performance (U-max1), however, there was a threshold for repeat swimming performance (U-max2) near 20 degrees C, above which recovery was impaired. 4. Following a simulated diurnal warming scenario (11-21 degrees C), U-max1 increased in rainbow trout and remained constant in Arctic char as in the field while U-max2 remained constant in rainbow trout it was drastically reduced in Arctic char. Furthermore, at warm temperatures, Arctic char were unable to recover to routine levels of oxygen uptake after exercise, while rainbow trout were. Warming also had more pronounced effects on blood composition and plasma glucose and lactate concentration in Arctic char than in rainbow trout. 5. In general, rainbow trout, a temperate salmonid, had superior swimming performance, aerobic capacity, and warm tolerance than Arctic char, an Arctic salmonid. 6. The present flow and temperature regimes in some Arctic rivers may restrict migration of native salmonids by limiting their ability to recover from fatiguing exercise. Non-native, temperate salmonids are likely better suited to overcome these particular physical and thermal challenges.

The critical swimming performance (U-crit) of fish has typically been measured on individuals given a step-based test that ends in fatigue. Many of the fish given this test naturally travel in schools; a concern is that the U-crit test underestimates the natural performance of schooling fish, as there are hydrodynamic benefits to schooling. We addressed whether Ucrit was improved by schooling by giving zebrafish (Danio rerio) Ucrit tests individually or in groups of three, five, or ten. We found that fish swam faster in schools and that schools grew more cohesive as flow speed increased. The increased U-crit owed to a greater number of fish swimming longer at lower speeds. In support of this, fish with the greatest Ucrit tended to have lower lactate and greater expression of citrate synthase. Our results indicate that Ucrit may increase when fish are tested in groups and that this may relate to energy savings, as well as differences in individual physiology.

In the recent special issue pertaining to metabolic rates, Norin & Clark (2016) were given the difficult task of defining and providing guidance on the estimation of maximum aerobic metabolic rate (MMR) in fishes while also interpreting the ecological relevance of past investigations. As relatively new practitioners in comparative and conservation physiology, we were enthused by the prospect of having a clear set of comprehensive definitions and guidelines to work from. The authors certainly provided some useful guidance, i.e. by acknowledging the need for precision when measuring rate of oxygen uptake (ṀO2) and the importance of selecting the appropriate methodology for the species of interest. A more precise definition and thorough discussion of how to estimate MMR would, however, be beneficial. Furthermore, we were compelled to emphasize that great care is needed when using Q10 temperature coefficients to scale MMR. The authors defined MMR as ‘... the maximum rate of oxygen consumption (ṀO2max)... ’, and stated that it was ‘... typically used to represent the maximum rate of aerobic metabolism... ’. While later in the article the authors do address some significant issues with the use of ṀO2max to represent the maximum aerobic metabolic rate, these limitations and others that were not addressed are fundamental to a comprehensive understanding of MMR and should, therefore, receive further attention. The primary issue with this definition is that the rate of oxygen uptake from the external environment is not the same as metabolic rate, a fact well established by Fry (1947) (Claireaux & Chabot, 2016). Rather, the rate of oxygen uptake can only be assumed to be proportional to aerobic metabolic rate when a fish is in a steady state (Wood & Perry, 1985), but even then, mitochondrial uncoupling likely accounts for a portion of the cellular oxygen consumption (Rolfe & Brown, 1997; Hulbert & Else, 2004). During fatiguing or exhaustive exercise fish are not in a steady state, internal energy stores (O2 and high-energy phosphates) are depleted as they transition to anaerobic metabolism and are rapidly replenished post exercise (Scarabello et al.,

The swimming performance of fishes has generally been assessed using a stepped velocity test where the speed at fatigue is considered the critical swimming performance (U crit). Although this test was designed for fishes that swim in the water column, it has been applied to fishes that adhere to the substrate. Here we examined the extent to which substrate holding, slipping and swimming contributed to reaching U crit in an example substrate holding fish, the invasive round goby. A linear model indicated that each behavior contributed significantly to U crit, but that substrate holding was by far the biggest contributor (65.8 ± 3.9 % vs. 5.8 ± 0.9 and 28.4 ± 3.4 % slipping and swimming). We also used our behavioural analysis to determine the critical substrate holding speed (U hold: 28.6 ± 1.1 cm s−1). We conclude that the U crit test can be applied to substrate holding fish but that it is not just an indication of critical swimming speed as is often considered and must be interpreted with caution.