Introduction
The evolution of the Antarctic notothenioid fishes in the cold waters of the Southern Ocean has resulted in a marked and shared intolerance to heat (Somero & DeVries Reference Somero and DeVries1967, Bilyk & DeVries Reference Bilyk and DeVries2011). Though most extant members of the suborder Notothenioidei inhabit the Southern Ocean, a few species are only found in the warmer waters around New Zealand and southern South America. Divided into two distinct groups, these include both the members of the three basal notothenioid families (Bovichtidae, Pseudaphritidae, Eleginopidae), which diverged prior to the origin of the cold adapted Antarctic notothenioid clade, and at least 16 species nested within the Antarctic notothenioids but now permanently found outside of Antarctic waters (Eastman Reference Eastman2005). Unlike the basal species, the ancestors of this latter group of secondarily temperate species presumably once shared the reduced heat tolerance noted among the endemic Antarctic species.
The shallow coastal waters of southern New Zealand are home to one such secondarily temperate species, the black cod (Notothenia angustata Hutton). Despite a residence time in cold temperate waters that may reach back 11 m.y.a., N. angustata continues to exhibit a number of cold adaptations characteristic of the endemic Antarctic species. Although they inhabit continually ice-free waters, this species retains some functional genes for antifreeze glycoprotein (AFGP) not found among the basal notothenioid families (Cheng et al. Reference Cheng, Chen, Near and Jin2003). At their environmental water temperatures, they also show intermediate levels of both cellular membrane lipid saturation (Logue et al. Reference Logue, DeVries, Fodor and Cossins2000) and ubiquitin conjugated proteins (Todgham et al. Reference Todgham, Hoaglund and Hofmann2007) between cold temperate and endemic Antarctic species. However, whether whole organism traits, like heat tolerance, show a similar pattern remains unknown.
Heat tolerance in fishes has typically been determined using two methodologies, either the measurement of their upper incipient lethal temperature (UILT), or their critical thermal maximum (CTMax) (Kilgour & McCauley Reference Kilgour and McCauley1986). The UILT, based on lethal dosage methodologies, measures tolerance as the temperature at which median mortality first becomes independent of the length of exposure, typically determined from exposure times ranging from a day to a week (Fry Reference Fry1947, Cossins & Bowler Reference Cossins and Bowler1987). The first study of heat tolerance in the Antarctic notothenioids used this methodology and found UILTs between 5 and 7°C in three species of high latitude nototheniids acclimatized to their natural freezing water temperatures (-1.9°C), which were the lowest UILTs then recorded (Somero & DeVries Reference Somero and DeVries1967).
The CTMax is an increasingly used alternative, which determines tolerance as the temperature at which a specimen becomes incapacitated from warming at a constant rate (Hutchison Reference Hutchison1961, Paladino et al. Reference Paladino, Spotila, Schubauer and Kowalski1980). As a measure of acute tolerance, CTMaxs always exceed the UILT for equivalently acclimated specimens (Kilgour & McCauley Reference Kilgour and McCauley1986). Determinations of CTMax also differ according to experimental parameters, most notably warming rate and starting temperature (Becker & Genoway Reference Becker and Genoway1979, Terblanche et al. Reference Terblanche, Deere, Clusella-Trullas, Janion and Chown2007). However, when care is used to ensure identical experimental conditions, CTMaxs are highly repeatable and useful for making comparisons within and between species. Additionally, this latter approach benefits from greater economy of specimens, time, and space when compared with the UILT, while producing results that still appear correlated to more chronic measures of heat tolerance. When the CTMaxs of 11 species of Antarctic fishes were recently determined from specimens acclimatized to their natural freezing water temperatures, these ranged from 11.95–16.17°C that, like measures of UILT, were well below CTMaxs reported from cold temperate and temperate fishes (Bilyk & DeVries Reference Bilyk and DeVries2011).
Despite the polar ancestry of N. angustata, this species largely shares its present-day geographic and bathymetric distributions with a basal notothenioid, the New Zealand thornfish (Bovichtus variegatus (Richardson)). In their modern thermal range these species inhabit water temperatures continually above those experienced by the endemic Antarctic notothenioids. While the endemic Antarctic species experience relatively little thermal variation, in Otago Harbour where the specimens used were collected, surface water temperatures have varied between monthly average highs of 14.2–18.2°C and lows of 5.6–8.6°C (Bev Dickson, personal communication 2010). The goal of this project was to compare CTMaxs between similarly acclimated specimens of the secondarily temperate N. angustata, the endemic cold temperate B. variegatus, and the endemic Antarctic N. coriiceps (Richardson). This was done to determine whether the heat tolerance of the secondarily temperate Antarctic notothenioid had increased to the extent that it matched that of the temperate B. variegatus which lack their Antarctic ancestry.
Materials and methods
Collection of specimens
Specimens of N. angustata were purchased in June of 2010 from the Portobello Marine Laboratory located on the Otago Peninsula. These had been collected with baited crayfish traps placed on the rocky bottom in shallow waters (10 m) around the Otago Heads. After collection, they were held in a large outdoor aquarium (10 x 10 m, 2 m deep) with a constant flow of local seawater. Specimens of B. variegatus were collected by hand net and by hook and line off the rocky shore and from tide pools along the Otago Peninsula and in the vicinity of the mouth of Bull Creek during June and July 2010. Prior to experimentation both species were moved into 390 L shallow indoor aquaria for one to two weeks where they received a constant flow of local seawater ranging in temperature from 6.7–9.0°C and averaging 7.9°C, mirroring the conditions of their winter environment. While in captivity after specimens of B. variegatus were isolated in covered plastic mesh cages to prevent escape and predation. Specimens of both species were held and treated in accordance with protocol 39/09 administered through the University of Otago Animal Ethics Committee.
Specimens of the Antarctic nototheniid N. coriiceps were collected during July and August 2008 by traps deployed from the RV L.M. Gould at sites along the western Antarctic Peninsula (WAP) and by hook and line from the shore of Anvers Island (64°S) near Palmer Station. Those collected aboard the ship were transported to Palmer Station in aquaria with a constant flow of ice laden surface waters but kept above -1.5°C through the use of a 300 W aquarium heater. At Palmer Station all specimens were allowed several days to recover from collection stress before the start of warm acclimation.
At both Portobello and Palmer Station specimens were exposed to their regular light cycles through windows in the aquaria facilities. Prior to experimentation all specimens of N. coriiceps and B. variegatus were weighed to the nearest tenth of a gram, and N. angustata to the nearest 10 g. To track individual specimens of N. coriiceps and N. angustata Floy tags (Floy Tag & Mfg. Inc) were inserted into the muscle just behind the first ray of the second dorsal fin. Due to their small size B. variegatus could not be tagged and instead individuals were kept separated by tank partitions.
Determination of environmental CTMax
Following a seven to ten day period of adjustment to captivity, environmental CTMaxs were determined for specimens of both of the New Zealand notothenioids. The environmental CTMax of a specimen is a measure of the heat tolerance present while acclimatized to their natural environmental temperatures, which were only determined for winter acclimatized specimens of N. angustata and B. variegatus.
CTMaxs were determined according to the protocol of Bilyk & DeVries (Reference Bilyk and DeVries2011) adapted from Paladino et al. (Reference Paladino, Spotila, Schubauer and Kowalski1980). To summarize, specimens were transferred to test aquaria at their holding tank temperature then after a ten minute adjustment period they were warmed at 0.3°C min-1 until the onset of a persistent loss of equilibrium. The temperature of this endpoint was taken as the specimen's CTMax and species’ CTMaxs were calculated as the arithmetic mean across all specimens at each acclimation temperature.
Following a specimen's persistent loss of equilibrium they were immediately returned to their original water temperature and their recovery monitored over 24 hours. On account of the large size of N. angustata, all CTMaxs were determined in a 60 L aquarium which afforded even the largest specimens sufficient space for free movement. Temperature homogeneity within this aquarium was maintained through vigorous aeration and the use of a recirculation pump.
Warm acclimation of N. coriiceps to 10°C
Eight specimens of N. coriiceps were warm acclimated at 10°C for three weeks. As attempts to directly transfer specimens from their environmental water temperatures (-1.5 to -1.0°C) to 10°C resulted in notable mortality, these were instead warmed in steps up to their final acclimation temperature. First, specimens were held at 6°C for five days, then 8°C for two days, and finally at 10°C for three weeks after which their CTMaxs were determined as previously described. Water temperature was maintained using a 300 W submersible heater and temperature controller activated solenoid valve which allowed the addition of cold local seawater only when their tank water temperature rose above the set point. During acclimation, their tank water was vigorously aerated to maintain oxygen saturation, specimens were fed white nototheniid muscle every three days to satiation with any excess removed, and the tank was cleaned of detritus daily with the sides brushed every three days to prevent fouling.
Warm acclimation of N. angustata and B. variegatus to 15°C
Following two to five days to recover from the determination of their environmental CTMaxs, specimens of N. angustata and B. variegatus were placed at 15°C. This water temperature was maintained using two 300 W aquarium heaters activated when water temperatures fell below their set point and a submersible pump added cold seawater from a reservoir tank when the water temperature rose above its set point. Over the next three weeks water temperatures ranged from 14.6–15.4°C cycling every 6 to 38 minutes. During acclimation, white fish muscle was provided twice per week with any uneaten food removed immediately, and to reduce fouling, tanks were cleaned of detritus every three days. After three weeks CTMaxs were again determined as previously described.
Statistical analysis
CTMaxs of N. angustata and B. variegatus were compared using a two-way ANOVA testing for significant differences between species, acclimation temperature, and any interaction between the two. The CTMaxs of 10°C acclimated N. coriiceps were compared with the environmental CTMaxs of N. angustata and B. variegatus using a one-way ANOVA with species as the independent variable, followed by a post-hoc Tukey test. For both, the ANOVA assumption of homoscedasticity was confirmed using Levene's test and the residuals were tested for normality.
Results
During this study all warm acclimating specimens appeared healthy showing no visible change in colouration, behaviour, skin abrasions, or gross weight, though none of the B. variegatus were observed to feed. While direct transfer of N. coriiceps from their environmental water temperatures (-1.5 to -1.0°C) to 10°C resulted in some mortality after three or four days, neither stepped acclimation of N. coriiceps to 10°C, nor direct acclimation of the New Zealand species to 15°C resulted in any mortality or visible distress.
In addition to the warm acclimation of the two New Zealand species, an attempt was made to cold acclimate these fishes to 4°C which would have allowed for a direct comparison to the CTMaxs of eight Antarctic species previously reported by Bilyk & DeVries (Reference Bilyk and DeVries2011). While three B. variegatus specimens readily tolerated 4°C, six N. angustata showed increasing pallor, refusal of food, and declining health ultimately leading to the abandonment of this part of the experiment. However, specimens of N. angustata have previously been cold acclimated to 2°C without issue over several weeks by DeVries (unpublished), and it is unclear whether the observed decline during this study was due to sub-optimal holding conditions or from stepping them down to 4°C too rapidly.
Environmental and warm acclimated CTMaxs for all three species are reported in both Table I and Fig. 1. Though some caution is warranted in comparing CTMaxs between acclimatized and acclimated specimens, the one-way ANOVA found significant variation between the environmental CTMaxs of N. angustata, B. variegatus, and 10°C acclimated N. coriiceps (F 2,17 = 90.38, P < 0.0001, r 2 = 0.91), with the post-hoc Tukey test noting the lowest CTMaxs in N. coriiceps. The two-way ANOVA on N. angustata and B. variegatus found a significant effect of both species (F 1,20 = 53.53, P < 0.0001) and acclimation temperature (F 1,20 = 40.15, P < 0.0001) on CTMax, though no significant interaction between the two (F 1,20 = 0.0004, P = 0.9718). The absence of an interaction between species and acclimation temperature indicates a similar increase in CTMax in both species following warm acclimation despite the lower CTMaxs of N. angustata.
Table I Comparison of CTMaxs between endemic Antarctic (N. coriiceps), secondarily temperate (N. angustata), and continually cold temperate (B. variegatus) notothenioids. Values are reported as mean ± their standard deviation. CTMaxs of notothenioid species investigated during this study. N. angustata and B. variegatus tested at 7–8°C were winter acclimatized specimens while all other specimens were warm acclimated to their described temperatures. WAP = western Antarctic Peninsula.
†CTMaxs for environmental acclimatized and 4°C acclimated N. coriiceps are reprinted from Bilyk & DeVries (Reference Bilyk and DeVries2011).
Fig. 1 Comparison of CTMaxs against acclimation temperature in four species of notothenioid fishes. All points represent the mean value with vertical bars denoting their standard deviation. Values for Trematomus bernacchii and Notothenia coriiceps at -1.9 and 4°C were taken from Bilyk & DeVries (Reference Bilyk and DeVries2011).
With the exception of a single specimen out of eight B. variegatus, there were no fatalities following the determination of CTMaxs and all of the New Zealand specimens readily recovered within several hours of their CTMax. Unlike the New Zealand species, 10°C acclimated N. coriiceps were directly returned to their environmental water temperatures after their CTMax and 50% failed to recover. This reduction in survivorship suggests that their gain in heat tolerance from warm acclimation came at a notable expense in cold tolerance raising it above the freezing point of seawater.
Discussion
A species’ heat tolerance often reflects its ancestral thermal environment with tolerance apparently driven by selection from the highest temperatures experienced over their evolutionary history (Mongold et al. Reference Mongold, Bennett and Lenski1996, Stillman & Somero Reference Stillman and Somero2000). The removal of positive selective pressure for heat tolerance during the evolution of the Antarctic notothenioids in the cold stable waters of the Southern Ocean has been suggested as a cause for these species’ modern low heat tolerance (Somero et al. Reference Somero, Fields, Hofmann, Weinstein and Kawall1998). While the ancestors of the secondarily temperate Antarctic notothenioids once presumably shared the low heat tolerance noted among modern endemic Antarctic notothenioids, their presence in cold temperate waters shows that they now have greater tolerance than species that have continually inhabited Antarctic waters. However, it has been unclear whether the upper thermal tolerances of these species are similar to fishes of non-Antarctic origin that share their current thermal habitat.
Comparing heat tolerances between the secondarily temperate N. angustata and the basal notothenioid B. variegatus
Although they inhabit the same cold temperate waters, the CTMax of N. angustata was significantly less than that of B. variegatus, both when compared between winter acclimatized specimens and between specimens warm acclimated to 15°C (Table I, Fig. 1). However, it is probable that the difference in heat tolerance between these species is understated when measured by CTMax as this fails to account for the effects of thermal lag on the much larger black cod.
Thermal lag is the time required for heat to be conducted from the periphery of an animal to the heat sensitive tissues in its body core. The greater thermal lag of larger animals means that it takes longer for heat to reach these sensitive sites and as a result they can resist warming longer simply by virtue of their size (Fry Reference Fry1971). For assays such as the CTMax that rely on continuous warming, the greater thermal lag of larger species therefore acts to inflate their measured tolerance relative to smaller species. As the tolerance of heat sensitive tissues is often of greater interest than the time needed for heat to reach them it is important to account for differences in thermal lag when comparing species that vary greatly in body size.
When thermal lag was directly compared between three species of Antarctic nototheniids warmed at 0.3°C min-1, core temperatures in N. coriiceps (791 ± 414 g) were 0.79°C cooler than Pagothenia borchgrevinki (Boulanger) (112 ± 39 g), and 1.11°C cooler than Trematomus bernacchii Boulanger (114 ± 42 g) (Bilyk & DeVries Reference Bilyk and DeVries2011). The difference in mass between the N. angustata and B. variegatus in this experiment was greater yet, with N. angustata specimens averaging 24 times the mass of B. variegatus (Table I) indicating that our determinations of CTMax are overstating the already low heat tolerance of N. angustata relative to B. variegatus.
While sharing a modern thermal environment with the endemic temperate B. variegatus, the lower CTMax of N. angustata occur along with the persistence of several biochemical and cellular cold adaptations found in endemic Antarctic species. As previously noted, at its environmental water temperatures this species shows incomplete homeoviscous adaptation of cell membranes (Logue et al. Reference Logue, DeVries, Fodor and Cossins2000) and intermediate lability of their cellular protein pool values which fall between cold temperate New Zealand and endemic Antarctic fishes (Todgham et al. Reference Todgham, Hoaglund and Hofmann2007). It also retains circulating AFGPs found in most endemic Antarctic notothenioids, though at vestigial levels insufficient to provide protection from freezing conditions (Cheng et al. Reference Cheng, Chen, Near and Jin2003). Unlike the endemic Antarctic species, N. angustata shows a modest inducible cellular heat shock response which is believed to have been lost in the endemic Antarctic species (Hofmann et al. Reference Hofmann, Buckley, Airaksinen, Keen and Somero2000, Clark et al. Reference Clark, Fraser, Burns and Peck2008) but less than that of B. variegatus (Hofmann et al. Reference Hofmann, Lund, Place and Whitmer2005).
Despite the lower CTMax of N. angustata compared to B. variegatus, both species showed an equivalent capacity to increase their CTMax in response to warm acclimation (Fig. 1). Recent investigations of endemic Antarctic species have shown notable organismal and systems level capacity for warm acclimation even with their comparably low heat tolerance (Podrabsky & Somero Reference Podrabsky and Somero2006, Franklin et al. Reference Franklin, Davison and Seebacher2007, Robinson & Davison Reference Robinson and Davison2008, Bilyk & DeVries Reference Bilyk and DeVries2011). As such, it is not surprising N. angustata appears to show a similar capacity to increase heat tolerance through warm acclimation. However, the plasticity of heat tolerance in N. angustata and B. variegatus was only tested over a small temperature range and it remains unknown whether these species share an equivalent response across their full acclimatory ranges.
Comparing the heat tolerance of the secondarily temperate Antarctic notothenioid N. angustata and the endemic Antarctic species N. coriiceps
Though N. angustata is found in the same cold temperate waters as B. variegatus, it is more closely related to endemic Antarctic species including its congener N. coriiceps. The latter is restricted to Antarctic waters below 54°S (Gon & Heemstra Reference Gon and Heemstra1990) which remain continually below 4°C throughout the year (Barnes et al. Reference Barnes, Fuentes, Clarke, Schloss and Wallace2006). When the environmental CTMaxs of winter acclimatized (7.9°C) N. angustata were compared to the CTMaxs of 10°C acclimated N. coriiceps, those of the N. angustata were significantly higher despite the slightly warmer acclimation temperature of N. coriiceps.
Averaging 3.3°C, this difference is larger than that observed between N. angustata and B. variegatus though it is probably partly overstated due to greater thermal lag in the larger N. angustata (Table I). However, unlike the comparison between N. angustata and B. variegatus, that fraction of the CTMax difference due to thermal lag is less because of the smaller difference in body size between N. angustata and N. coriiceps. A CTMax intermediate to cold temperate and polar fishes is consistent with the reduced presence of some polar characteristics in N. angustata, and with recent work showing an increase in standard and total metabolic rates when compared to N. coriiceps (Campbell et al. Reference Campbell, Fraser, Peck, Bishop and Egginton2007). The intermediate heat tolerance of N. angustata between B. variegatus and N. coriiceps remains despite an apparent lengthy period under selection for higher heat tolerance from the warmer water temperatures shared by the endemic cold temperate B. variegatus.
Although the method of dating the origin of the secondarily temperate Antarctic notothenioids remains in question, the divergence of N. angustata from endemic Antarctic species probably extends to at least the early Pliocene. The prevalent hypothesis for the appearance of Antarctic notothenioids in New Zealand waters is that their ranges extended northward during glacial maxima (Petricorena & Somero Reference Petricorena and Somero2007). During prior maxima, the Antarctic circumpolar front which forms the boundary of the Southern Ocean has shifted northward by up to five to ten degrees of latitude in open ocean areas (Nelson & Cooke Reference Nelson and Cooke2001). This could have brought the Antarctic Front near New Zealand allowing for the dispersal of the eggs and larva of N. angustata's progenitor. Given the persistence of Southern Ocean fronts between New Zealand and Antarctica and the associated temperature gradient, once N. angustata's progenitor arrived in New Zealand waters it would have experienced continually warmer water temperatures than Antarctic species (Kennett Reference Kennett1982, Nelson & Cooke Reference Nelson and Cooke2001, Sikes et al. Reference Sikes, Howard, Neil and Volkman2002).
The length of their isolation from Antarctic waters has been estimated at 11 m.y. by Cheng et al. (Reference Cheng, Chen, Near and Jin2003) using the divergence in the mitochondrial ND2 gene between the two New Zealand nototheniids. Though this places their origin near a glacial maximum, the reliance of this analysis on universal substitution rates may not allow it to withstand future scrutiny and other investigators have instead suggested a more recent origin during the late Miocene or early Pliocene (Eastman & McCune Reference Eastman and McCune2000, Petricorena & Somero Reference Petricorena and Somero2007).
Comparing the heat tolerance of 10°C acclimated N. coriiceps to other Antarctic species
Most prior studies of increased heat tolerance from warm acclimation in the endemic Antarctic notothenioids have been limited to acclimation temperatures of 4–5°C. This has been in deference to the apparent stenothermy of high-Antarctic species, where chronic exposure to temperatures as low as 5–7°C can prove fatal in specimens acclimatized to their natural freezing water temperatures (-1.9°C) (Somero & DeVries Reference Somero and DeVries1967). However, comparably greater heat tolerance is present among Antarctic fishes endemic to the WAP and more northerly Scotia Arc Islands such as N. coriiceps which may more readily acclimate to warmer temperatures (Bilyk & DeVries Reference Bilyk and DeVries2011).
While 10°C acclimated N. coriiceps have lower CTMax than comparably acclimated cold temperate fishes (Otto & Ohararice Reference Otto and Ohararice1977, Becker & Genoway Reference Becker and Genoway1979, Fangue & Bennett Reference Fangue and Bennett2003), including the winter acclimatized New Zealand notothenioids, these showed a notable increase in CTMax over environmental and 4°C acclimated specimens (Table I). At 10°C they similarly exceed the CTMaxs of seven other 4°C acclimated notothenioids, predominantly from high latitude waters (Bilyk & DeVries Reference Bilyk and DeVries2011). Both their survival and continued normal activity at this temperature show a broader thermal range than anticipated even among the fishes found in the relatively more thermally variable waters along the western Antarctic Peninsula.
Summer acclimatized specimens of the nototheniid Leptonotothen nudifrons (L.), which is restricted to more northerly Antarctic waters, warmed at 1°C per day resulted in the mortality of most upon reaching 8–9°C (Hardewig et al. Reference Hardewig, Peck and Pörtner1999). Similarly, Van Dijk et al. (Reference Van Dijk, Tesch, Hardewig and Pörtner1999) noted a cellular shift to anaerobic metabolism at 9–10°C in the phylogenetically distant Antarctic zoarcid, Pachycara brachycephalum (Pappenheim) when also warmed at 1°C per day, which suggests the limits of their heat tolerance lies near this temperature. However, while these studies warmed specimens slowly they may not have allowed sufficient time for the physiological and cellular processes that occur during acclimation to fully track rising temperatures as some high latitude Antarctic species required up to two weeks to fully manifest their increased organismal heat tolerance during warm acclimation at 4°C (Bilyk & DeVries Reference Bilyk and DeVries2011). The failure of N. coriiceps to directly acclimate to 10°C in this study would also suggest that while acclimating notothenioids to cold temperate water temperatures is possible it may require very slow rates of warming or staged increases.
Despite the survival of N. coriiceps at 10°C, it is unclear whether this species could survive the even warmer summer water temperatures experienced by the secondarily temperate species. If it could survive such extremes of heat, the capacity for activity, growth, and reproduction are under more stringent thermal constraints than survival (Fry Reference Fry1971) and the temperature limitations of these remain unknown in N. coriiceps.
Gains in heat tolerance among fishes from warm acclimation typically come at the expense of an even greater loss in cold tolerance (Bennett et al. Reference Bennett, McCauley and Beitinger1998). The mortality of warm acclimated N. coriiceps when returned to their natural freezing water temperatures indicates that such a loss in cold tolerance occurred. That acclimation to just 10°C costs them the ability to tolerate freezing Antarctic water temperatures would argue that the endemic Antarctic species have a particularly narrow thermal range despite the presence of plasticity in their heat tolerance.
Conclusion
Evolution in chronic cold has led to reduction of heat tolerance in the Antarctic notothenioids. While the Antarctic N. coriiceps shows a surprising ability to warm acclimate to 10°C its heat tolerance was still less than that of equivalently acclimated cold temperate and temperate fishes. Several members of the Antarctic notothenioids are now permanently distributed outside Antarctic waters even though their Antarctic ancestors presumably once shared the reduced heat tolerance now common throughout endemic Antarctic species. Our examination of heat tolerance in N. angustata, one such secondarily temperate notothenioid, is consistent with an increase in heat tolerance following their arrival in New Zealand waters. Heat tolerance can serve as an important physiological constraint on geographic distribution (Hochachka & Somero Reference Hochachka and Somero2002) and the capacity of the Antarctic notothenioids to adapt to rising water temperatures will probably be important for their future survival in the Southern Ocean in light of predicted future impacts of global climate change on the Antarctic ecosystem (Clarke et al. Reference Clarke, Johnston, Murphy and Rogers2007).
Acknowledgements
The authors would like to thank the staff of the University of Otago's Portobello Marine Lab, especially Bev Dickson. We would also like to thank the University of Otago's Department of Marine Science, Dr Miles Lamare, and Dr Brian Paavo for their assistance in New Zealand. This research was funded by the US National Science Foundation East Asia and Pacific Summer Institutes fellowship to K.T. Bilyk and US National Science Foundation Office of Polar Programs Grant NSF OPP-Ant-0231006 to A.L. DeVries and C-H.C. Cheng, with additional support from the Royal Society of New Zealand and the Department of Animal Biology at the University of Illinois Urbana Champaign. The constructive comments of the reviewers are also gratefully acknowledged.
