Review Environmental and functional limits to muscular ... · ommastrephid squid, migrates vast distances through the open ocean as part of its life cycle. For this reason, the most

Comparative Biochemistry and Physiology Part A 133(2002) 303–321

1095-6433/02/$ - see front matter� 2002 Elsevier Science Inc. All rights reserved.PII: S1095-6433Ž02.00162-9

Review

Environmental and functional limits to muscular exercise and bodysize in marine invertebrate athletes�

Hans O. Portner*¨

Alfred-Wegener-Institut fur Polar- und Meeresforschung, Okophysiologie, Postfach 12 01 61, D-27515 Bremerhaven, FRG¨¨

Received 11 June 2001; accepted 7 November 2001

Abstract

Many similarities exist between the key characteristics of muscular metabolism in marine invertebrates and thosefound in vertebrate striated muscle, even though there are important phosphagens and glycolytic end products that differbetween groups. Lifestyles and modes of locomotion also vary extremely among invertebrates thereby shaping thepattern of exercise metabolism. In accordance with the limited availability of integrated ecological and physiologicalinformation the present paper reports recent progress in the exercise physiology of cephalopods, which are characterizedby high rates of aerobic and anaerobic energy turnover during high velocity hunts or escapes in their pelagic environment,and a sipunculid worm, which mostly uses anaerobic resources during extended marathon-like digging excursions in thehypoxic marine sediment. Particular attention is paid to how lifestyle and oxygen availability in various marineenvironments shapes the use and rates of aerobic and anaerobic metabolism and acidosis as they depend on activitylevels and energy saving strategies. Whereas aerobic scope and, accordingly, use of ambient oxygen by blood oxygentransport and skin respiration is maximized in some squids, aerobic scope is very small in the worm and anaerobicmetabolism readily used upon muscular activity. Until recently, it was widely accepted that the glycolytic end productoctopine, produced in the musculature of these invertebrates, acted as a weak acid and so did not compromise acid–basebalance. However, it has now been demonstrated that octopine does cause acidosis. Concomitant study of tissue energyand acid–base status allows to evaluate the contribution of glycolysis, pH and free ADP accumulation to the use of thephosphagen and to the delayed drop in the Gibb’s free energy change of ATP hydrolysis. The analysis reveals speciesspecific capacities of these mechanisms to support exercise beyond the anaerobic threshold. During high intensityanaerobic exercise observed in squid, the levels of ATP free energy change finally fall to critical minimum levelscontributing to fatigue. Maintenance of sufficiently high energy levels is found at low but extended rates of anaerobicmetabolism as observed in the long term digging sipunculid worm. The greatest aerobic and anaerobic performancelevels are seen in squid inhabiting the open ocean and appear to be made possible by the uniform and stablephysicochemical parameters(esp. high O and low CO levels) that characterize such an environment. It is suggested2 2

that at least some squid operate at their functional and environmental limits. In extremely different environments, boththe worm and the squids display a tradeoff between oxygen availability, temperature, performance level and also, bodysize.� 2002 Elsevier Science Inc. All rights reserved.

Keywords: Anaerobic vs. aerobic metabolism; Aerobic scope; Body size; Cephalopod; Cold adaptation; Critical velocity; Deep sea;Muscle energetics; Giant squid; Gibb’s free energy of ATP hydrolysis; Intracellular pH; Marine invertebrate; Muscle metabolism;Octopine; Oxygen availability; Performance; Phosphagen; Squid; Sipunculid worm; Temperature

� This paper was presented in the session, ‘Physiology and Biochemistry of Exercise’, at the Society for Experimental Biology,April 2–6, 2001, Canterbury, UK.

*Tel.: q49-471-4831-1307; fax:q49-471-4831-1149.E-mail address: [email protected](H.O. Portner).¨

304 H.O. Portner / Comparative Biochemistry and Physiology Part A 133 (2002) 303–321¨

1. Introduction: muscular activity and mode oflife

There is tremendous diversity in the lifestylesand modes of locomotion among marine inverte-brates, thereby shaping the pattern of muscularmetabolism. Neglecting those forms that float inthe pelagic owing to neutral buoyancy mechanismsmodes of locomotion or muscular activity rangefrom swimming, walking, digging in the sedimentor just holding tight (e.g. by means of catchmuscles used for valve closures) with decreasinglevels of energy turnover in that order. Athletes inthe sense that animals regularly and extensivelyuse maximal levels of muscular performance aspart of their lifestyle can be found in severalgroups, however, the characteristics of muscularactivity in the context of environmental and life-style constraints have only been studied in veryfew cases. The present review tries to evaluatethese relationships in four examples, where suchanalysis appears possible. Three of those examplesare different squid species and one is a sipunculidworm. These four species live in various environ-ments, the three squid in different parts of thepelagic, ranging from inshore and coastal watersto the open ocean, whereas the sipunculid lives inpreliminary burrows in the intertidal and subtidalzone and is characterized by extended diggingexcursions during feeding in the marine sediment.These contrasting environments are characterizedby different degrees of stability of environmentalfactors, like oxygen and CO concentrations, salin-2

ity and temperature. A physiological approachallows to identify to what extent these environ-mental parameters represent limits to the levels ofperformance reached, as well as to other species-specific features like body size.

Whereas the sipunculid worm,Sipunculusnudus, is well adapted to life in the hypoxic marinesediment by means of low aerobic metabolic rateand oxyconformity on the one side and a moderateto high resistance to long term anaerobiosis on theother side (Portner et al., 1984b, 1985) squid¨display the highest rates of muscular energy turn-over among invertebrates and are found in diversemarine environments ranging from the pelagic ofshallow estuaries or shelf and open oceans to thedark and permanently cold deep sea including thepolar oceans(for general reviews see Wells, 1994;Portner and Zielinski, 1998). They are the sole¨large invertebrates to compete successfully with

marine vertebrates(fish and marine mammals) inthat they have conquered the pelagic zone andoccupy similar positions in ecosystems and foodwebs. Usually, squid living in shelf environmentsor in the illuminated upper water layers are mus-cular and not neutrally buoyant. Based on theirextreme rates of aerobic metabolism they ensuretheir position in the pelagic ecosystem by display-ing a level of activity which is comparable to, oreven exceeds, that of similar sized fish. In contrast,deep sea and some polar squid have developedneutral buoyancy by accumulating ammoniumsalts in separate fluid filled chambers and, thereby,likely lead a less expensive life in deep or coldoceans.

The mode of locomotion differs extremelybetween the squid and the sipunculid worm. Per-formance in the worms, which reach between 10and 80 g body weight depending on the grain sizeof the sediment(see below), can be studied bysimply placing them on the sediment surface there-by eliciting digging excursions of various lengths.The worm operates as a hydraulic machine whilepenetrating the marine sandy sediment(Truemanand Foster-Smith, 1976). In a simplified picturemajor forces are exerted by the musculature(cir-cular and longitudinal muscles) on the non-cham-bered coelomic fluid which is then used torepeatedly expel the introvert into the sedimentwhere it forms an anker allowing the rest of thebody to follow by contraction of the longitudinalmuscles.

Performance levels and energy metabolism insquid have been studied in swim tunnel experi-ments (O’Dor et al., 1990; O’Dor and Webber,1991), in greatest detail in the ommastrephid squidIllex illecebrosus and the loliginid squidsLoligopealei and, most recentlyLolliguncula brevis(Webber and O’Dor, 1985; Portner et al., 1991,¨1993, 1996; Finke et al., 1996). The majority ofstudies withI. illecebrosus or L. pealei have beencarried out on specimens up to 50 cm in lengthand weighing between 300 and 600 g, whereasinvestigations withLolliguncula used animals of6–35 g and 5–9 cm mantle length.I. illecebrosus,L. pealei, and L. brevis also differ in their choiceof habitat and way of life.L. pealei, like manyother loliginids, is more or less non-migratory andrestricted to coastal waters. The brief squid,L.brevis lives in warmer water and even enters veryshallow and brackish waters in the inshore envi-ronment. In contrastI. illecebrosus, like many

305H.O. Portner / Comparative Biochemistry and Physiology Part A 133 (2002) 303–321¨

Table 1Comparison of basal metabolic rates(mmolO kg h ) with active (i.e. maximum) metabolic rates at critical swimming velocitiesy1 y1

2

(modified after Portner and Zielinski, 1998)¨

Oncorhynchus Scomber Illex Loligo Lolligunculanerka scombrus illecebrosus pealei brevis

Temperature(8C) 15 10 15 15 20Weight (kg) 0.5 0.4 0.6 0.6 0.01Standard or resting metabolic rate(SMR) 1.8 1.5 8 4.5 22Active metabolic rate 21.4 8.3 48.7 38 36

Standard metabolic rate for the fish species and forL. brevis was evaluated from an extrapolation of oxygen consumption to zeroswimming speed. Metabolism of the larger squid,I. illecebrosus andLoligo pealei resting on the bottom is found below the standardmetabolic rate since squid are negatively buoyant and standard metabolic rate comprises the cost of maintaining position in the watercolumn. Active metabolic rate is the rate measured at the critical swimming velocity, except forL. brevis, where it is the maximumrate of oxygen consumption measured at the highest swimming speed(after O’Dor and Webber, 1991; Lucas et al., 1993; Finke et al.,1996; Wells et al., 1988).

ommastrephid squid, migrates vast distancesthrough the open ocean as part of its life cycle.For this reason, the most active cephalopods areincluded amongst the ommastrephids, rather thanthe loliginids.

Squid locomotion occurs by use of jet-propul-sion elicited by expelling water from the mantlecavity. The mantle musculature of squid is mainly(up to 90%) composed of circular musculatureinterspersed with narrow bands of radial musclefibres (running perpendicular to the mantle sur-face). In addition, there is a thin layer of connec-tive tissue on either side of the mantle(1% of themantle tissue), which is important for the attach-ment of the radial muscles and the elastic collagenfibres (0.5%) which act as springs, facilitating therefilling of the mantle, after a jet of water hasbeen expelled(Gosline and Demont, 1985). Incontrast to octopods, squid have no longitudinalmuscle fibres in the mantle.

The various types of muscle cells in squid(Boneet al., 1981; Mommsen et al., 1981) supportventilation or jet-propelled swimming. InI. ille-cebrosus with a mantle cross section of approxi-mately 6 mm, aerobic fibres (with 47%mitochondria) comprise 14%(0.8 mm) and 5%(0.3 mm) of the circular muscle on the outer andinner mantle surfaces, respectively. The inner sec-tion of squid mantle(81%) is composed of anaer-obic fibres (7% mitochondria). Themitochondria-rich circular muscle fibres, whichmaintain continuous movement of the mantle forventilation and constant swimming, function pri-marily at low swimming speeds(the maximumforce generated by these cells is lower than thatof the anaerobic fibres, since the number of power

producing myofibrils is reduced due to the greatnumbers of mitochondria present). Circular andradial muscle fibres poor in mitochondria and richin myofilaments mainly come into play when thesquid is either under attack or escaping so thathigh speed is essential. The radial cells and colla-gen fibres work together to speed the reinflationof the mantle and intake of water after a jet hasbeen expelled. When escaping attack, the mantlemay even become hyperinflated, in turn producingmaximum effect from the contracting circular mus-cle. As the circular muscle contracts, diagonalcollagen fibres increase the tension and act assprings to aid the re-filling of the mantle cavity.When fully contracted the outer perimeter of themantle is reduced by approximately 30%, and atthe same time the mantle wall doubles in thickness(Gosline et al., 1983; Gosline and Demont, 1985;Shadwick, 1994).

Whereas aerobic scope in the worm is negligible(see below) squid activity relies to a large extenton aerobic metabolism. Original comparisons ofoxygen consumption rates in squid and fish of asimilar size and mode of life used salmon as arepresentative high performance fish(O’Dor andWebber, 1986). Recent data available for scombridfish emphasize that the difference in metabolicrate between mackerel and(Atlantic) salmon issmall and, therefore, confirm the general conclu-sion that the metabolic rate of squid exceeds thatof comparable ectothermal fish species(Table 1).Only the development of heterothermy or regionalendothermy in scombrid fishes like tuna or inlamnid shark may lead to metabolic expenditureabove the rates seen in ectothermal fish and squid(Brill, 1996; Korsmeyer et al., 1996). In compar-


ison with the mode of swimming in ectothermicfish, oxygen consumption measurements show thatjet propulsion is far more costly(Table 1). Fishset in motion a large volume of water with theirundulating movement. The mechanical efficiencyis high, meaning that by far the greater proportionof the caudal fin movement is converted intoforward thrust(Wieser, 1986). Squid, on the otherhand, use a great deal of energy to expel arelatively small amount of water. Then, after eachcontraction, water(flowing in the opposite direc-tion to swimming) is sucked back into the mantlecavity. This creates turbulence which increaseswith speed and acts as a frictional force reducingthe efficiency with which muscle power is con-verted to propulsion. The inefficiency of this meth-od of propulsion is reflected in the extremely highmetabolic rate of muscular squids(Table 1).

This review deals with the factors limiting thedegree of muscular activity in the various squidand the sipunculid species which all display dif-ferent levels of aerobic(e.g. Table 1) and anaero-bic energy turnover during exercise and useanaerobic glycolysis and the phosphagen to variousdegrees. The paper also addresses the environmen-tal characters limiting performance consideringenergy saving strategies used in different habitats.

2. Transition to anaerobic metabolism

Previous swim tunnel experiments in fish andsquid focused on aerobic energy costs of swim-ming at various speeds and a determination of thecritical swimming speed, i.e. the maximum veloc-ity that a fish or squid can sustain for extendedperiods without evident signs of fatigue, or adetermination of oxygen debt accumulated beyondthis speed(e.g. Brett, 1964; Webb, 1971; Webberand O’Dor, 1985). For a study of anaerobic metab-olism, its capacities and effects on tissue energyand acid–base status, typically, resting andfatigued animals or the time course of recoveryfrom fatigue have been analysed(Portner et al.,¨1991, 1993, for reviews in fish see Milligan, 1996;Kieffer, 2000). The transition to anaerobic metab-olism with rising activity levels or the progressive,correlated changes in tissue energy and acid–basestatus have rarely been addressed(e.g. Goolish,1991; Burgetz et al., 1998). The reason for thescarcity of such data is that, on the one hand,online monitoring of aerobic energy expenditurewas feasible early on by use of oxygen sensors,

whereas, on the other hand, analyses of anaerobiccontributions and associated acid–base distur-bances depended on the cannulations of bloodvessels or the use of invasive techniques for tissuesampling and analyses and, therefore, sacrifice ofa large number of specimens during the course ofthe experiments. A technique for adequate invasiveanalysis of exercise induced transitions in intracel-lular acid–base status only became available in1990 (Portner et al., 1990; Portner 1990b), which¨ ïs another reason for the limited information avail-able. The use of P-NMR techniques has been31

limited by accessibility of equipment for suchanalyses and the fact that fish had to be quiescentduring the actual measurement such that onlinemonitoring was not possible(e.g. Burgetz et al.,1998).

The present review will use data from invasiveanalyses of correlated changes in tissue energy andacid–base status carried out during graded levelsof activity or extrapolate from data obtained inexercise protocols where animals displayed anaer-obic metabolism to various degrees(Finke et al.,1996; Portner et al., 1991, 1993, 1996 and unpub-¨lished). Future work in this area will be mucheasier as only very recently in June 2001, onlinenon-invasive monitoring of tissue energy andacid–base status at various swimming speeds andthe direct determination of the critical swimmingvelocity became possible in fish(Atlantic cod,Gadus morhua) in a swim tunnel, by adaptationand use of P-NMR techniques(H.O. Portner,31 ¨D.M. Webber, C. Bock, unpublished).

An analysis of the different anaerobic pathwaysused in a squid,L. brevis, led to the observationthat, with increasing swimming speed, octopine,a-glycerophoshate and succinate accumulationstart more or less simultaneously beyond a certainswimming velocity. This indicates that an anaero-bic threshold had been reached, i.e. the criticalswimming speed above which anaerobic metabo-lism contributes to energy production(Finke et al.,1996, Fig. 1). The more or less simultaneousaccumulation of these metabolites as well as theprogressive decrease in phospho-L-arginine levelsindicates that limited oxygen supply to mitochon-dria and the onset of anaerobic metabolism in bothmitochondria and the cytosol go hand in hand.ATP levels fell and adenylates were degraded,while glucose-6-phosphate accumulated in themantle muscle. This finding is quite opposite tothe situation found in other invertebrate and ver-


Fig. 1. Changes in octopine,a-glycerophosphate and succinateconcentrations in the mantle musculature of the brief squid,L.brevis sampled at different swimming speeds(after Finke etal., 1996). Lines represent second- and third-order polynomialfits and 95% confidence intervals.

tebrate species, where mitochondria remain aerobicwhen energy requirements in excess of aerobicenergy production are covered by anaerobic metab-olism during functional anaerobiosis(Grieshaberet al., 1994, for a recent example see Kemper etal., 2001). The observations in squid are in linewith a much lower degree of tissue capillarisationin squid than in fish(Bone et al., 1981) and theconclusion that a large fraction of the oxygenconsumption is covered by oxygen uptake via theskin on top of a maximum use of blood oxygentransport(see below). Here already, the impressionarises that exercise metabolism in squid is limitedby oxygen availability. This conclusion is support-ed by an oxygen efficient design of cephalopodmuscle metabolism(Hochachka, 1994).

Whereas squid use a significant aerobic scopeduring muscular activity(Table 1) the marine

worm S. nudus appears unable to raise aerobicmetabolic rate to a large extent on top of the costof ventilatory activity. Comparisons of analyses bydirect and indirect calorimetry(Hardewig et al.,1991) suggested that whatever spontaneous activ-ity was associated with anaerobic heat productionat virtually constant rates of oxygen consumption.Anaerobic metabolic rate during muscular exercisewas unchanged and similar metabolite levels werereached regardless of whether digging occurredunder an aerated or an oxygen deficient nitrogenatmosphere(unpublished data). These observa-tions are easily explained in a species which isactive during digging in hypoxic sediments and isshort in oxygen supply during any period ofmuscular activity except ventilation and primitiveswimming movements. Accordingly, metabolismduring exercise in the worm is in extreme contrastto the metabolic situation in squid, in that it isalmost exclusively fuelled by anaerobic metabo-lism and thus limited by anaerobic capacity onlyand not by aerobic scope which is virtually nil inthe worm.

3. Oxygen transport

Oxygen supply to tissues inS. nudus is sup-ported by cells containing haemerythrin which arefound in the large coelomic and smaller vascularcavities (Florkin, 1933; Manwell, 1960), as wellas by intracellular myo-haemerythrin(Hendricksonet al., 1985). The coelomic pigment is largely usedfor oxygen storage, displays no Bohr effect andbound oxygen is depleted during hypoxic exposurewhich the animal experiences regularly duringdigging excursions or during low tide(Portner etäl., 1985).

Oxygen transport to tissues in squid occurs byuse of haemocyanin(for review see Mangum,1990). This pigment is found in extracellularsolution as a macromolecule, consisting of tensubunits, with eight oxygen binding sites per unit(Miller, 1994). Cephalopods, and in particularsquid, have the highest concentrations of haemo-cyanin in the animal kingdom, with over 150 mgprotein ml blood. In spite of this, the level ofy1

bound oxygen(oxygen binding capacity) at 1–3mmol l , remains below the 4–5 mmol l foundy1 y1

in fish. For optimised oxygen transport, especiallythe haemocyanins of muscular squid display astrong pH dependence for oxygen binding, withBohr coefficients Dlog P yDpH less thany150


Fig. 2. Changes in haemocyanin percent oxygen saturation between control conditions and exercise in the squidI. illecebrosus, depictedin a pHysaturation diagram(Portner, 1990a). The shift in arterial blood pH between control conditions(Ca) and exercise(Ea) mirrors¨the protective base release into the blood(Fig. 3). Note the maximum use of haemocyanin bound oxygen in venous mantle blood(Ev,based on model calculations by Portner, 1994 using data compiled by Portner et al., 1991, 1993).¨ ¨

(Bridges, 1994). Highly pH dependent cooperativ-ity gives rise to anS-shaped O -binding curve2

with a particularly steep slope shifted to highlevels of P , i.e. oxygen affinity is low. WithO2

high P values the buffer range ofP is shifted50 O2

to high values supporting a highP gradient and,O2

in consequence, a high flow of oxygen betweenblood and mitochondria(Portner, 1990a, 1994; see¨Fig. 2 for a depiction of arterial and venous pHand P values in a pH saturation diagram). WithO2

Bohr and Haldane coefficients(y1 protonexchange at the protein level would cause pH torise between arterial and venous blood, unlessproduction of surplus CO occurs higher than2

expected from normal RQ values. This is onlypossible if a major proportion of the oxygen entersthrough the skin instead accounting for additionalCO production(Portner, 1994). The necessity for2 ¨complementary skin oxygen uptake is also under-lined by the observation that the use of bloodborne oxygen is maximal(Fig. 2), that the circu-latory system would not be able to pump therequired quantity of blood, and that the capillaris-ation of the mantle muscle is insufficient(seeabove). The mitochondrion-rich muscle cells aresituated on the inner and outer mantle surfacesand are supplied with oxygen directly through anextremely thin skin. Continuous movement ofwater over the inner and outer mantle surfaces

reduces the formation of oxygen poor boundarylayers. Oxygen uptake via the skin is not justdependent on passive diffusion, since the propor-tion provided by this form of respiration is toogreat (at times over 50% of the total oxygenuptake in the mantle ofI. illecebrosus). It may beenhanced by the movement of the tissue fluids,produced by the constant muscle activity(and therepeated 100% alteration in the cross section ofthe mantle wall mentioned above; Gosline andDemont, 1985, for further details see Portner,¨1994).

The large pH sensitivity of the haemocyaninand the maximum use of blood oxygen transport(Fig. 2) makes especially ommastrephids quitesensitive to fluctuating ambientP andP levelsCO O2 2

(Portner, 1990a, 1994). Due to their high oxygen¨requirement and the high pH dependence of oxy-gen binding, they rely on a constantly high levelof oxygen and a low level of carbon dioxide inorder to safeguard their oxygen transport. Thedanger of a drop in arterial pH during activeperiods is reduced by release of a surplus ofbicarbonate into the blood(Fig. 3), raising pHbetween control and exercise(as depicted in Fig.2) and thereby also increasing oxygen affinity(Portner, 1994, see below). As suggested aboveäll of these considerations are in line with alimiting role of ambient oxygen levels and indicate


Fig. 3. Comparison of non-respiratory proton quantities in blood and mantle tissue of the squidI. illecebrosus with those bound duringhaemocyanin deoxygenation(left, for blood) or those turned over in metabolism(right, for mantle tissue). A base excess found in theblood is mirrored by the net release of base from the mantle tissue(after Portner, 1994).¨

that these animals could only achieve such a highlevel of activity after their ancestors had lost theirshell(cf. Wells, 1994). However, as squid increasein size, the proportion of the oxygen supplyobtained via the skin must become less owing tothe short distances covered by diffusion. Thisseems likely, since the metabolic rate per unitweight drops and the animals can manage almostentirely on the oxygen supplied by the blood.Interestingly, the Bohr coefficient for giant squid(Architeuthis monachus) with a value of y0.8(Brix et al., 1989), is in excess ofy1, whichmeans that the pigment’sP buffer function isO2

nonetheless guaranteed. The oxygen uptake overthe skin is probably also lower inOctopus vulgarisand Sepia. With Bohr factors belowy1, thesespecies use an alternative and unusual method ofvenous acidification: their haemocyanin bindsCO together with oxygen in the gills and trans-2

ports it to the tissues(Lykkeboe et al., 1980).However, nothing is known about the molecularbasis of this particular method of CO binding.2

4. Environmental constraints for muscularactivity: oxygen and temperature

The discussion above already suggests that, inthe case of squid, oxygen availability to tissues

limits performance levels, whereas the wormalmost exclusively relies on anaerobic resourcesfor muscular exercise. As outlined below, oxygenlimitations already become effective under restingconditions in the case of the worm, whereas in thecase of squid, such limitations seem to be largelyeffective on aerobic scope. Not only in the worm’shabitat but also in some natural habitats of squidoxygen levels may fluctuate as reported for thebrief squid,L. brevis. These animals not only entershallow coastal waters characterized by larger fluc-tuations of environmental parameters like temper-ature and salinity, butL. brevis were also found inhypoxic bottom waters with aP of 2.1 kPa byO2

use of a remotely operated vehicle(Vecchione,1991). They were observed to be quite activeunder these conditions and may have entered thesewaters for feeding andyor escape excursions. Fur-thermore, catches by trawling indicated a highabundance in waters with aP between 4.3 andO2

8.6 kPa, however, animals were no longer presentwhen oxygen tensions fell below 2 kPa at temper-atures between 25 and 298C (Vecchione, 1991and personal communication).

Recently, the criticalP was determined forL.O2

brevis (Zielinski et al., 2000), the only analysisavailable for squid. The criticalP is character-O2

ized by the onset of anaerobic energy metabolism


and indicates oxygen limitations under restingconditions when animals exhibit standard metabol-ic rates (SMR) (Portner and Grieshaber, 1993).¨The data forL. brevis reveal a rather high criticaloxygen level(P ) of 8–10 kPa at 208C, whichc

apparently contradicts the findings that these ani-mals are able to enter and stay in hypoxic waterlayers. However, an energy saving mode of jetpropulsion may allow this species to dive intohypoxic waters(see below). This strategy willprolong the period during which anaerobicresources can be used. Nothing is known in thisrespect for other squid species. As discussed earlierthe maximum use of oxygen resources and thehigh sensitivity of blood oxygen transport suggestthat I. illecebrosus is intolerant to even moderatehypoxia although theP has not been quantified.c

Tolerance to fluctuating temperatures may beintimitely linked with the question of hypoxiatolerance. Insufficient capacity of ventilatory andcirculatory mechanisms to cover tissue oxygendemand are seen to define the first line of thermalintolerance(Portner et al., 1998; Portner, 2001,¨ ¨2002a). As outlined previously, the critical tem-perature, defined by the onset of anaerobic metab-olism, is just above the range of ambienttemperatures found in the natural environments ofL. brevis, whereas the upper limit found for anAntarctic octopod is considerably above ambient(Portner and Zielinski, 1998). Critical tempera-¨tures will decrease during conditions of progressivehypoxia(Portner, 2001). Considering the environ-¨mental data set available forL. brevis, theseconsiderations emphasize that these squid operateat their functional and environmental limits.

5. Energy savings in squids

As concluded above, the maximization of per-formance characteristics and oxygen demand intoday’s muscular squid was only possible after theancestors of today’s shell-less species lost theirshell and thereby, mechanical protection from pred-ators as well as buoyancy by means of the shell’sgas-filled chambers. These developments occurredunder increasing competition and threat of attackfrom vertebrates. As much as life in hypoxicenvironments limits aerobic energy turnover, aconclusion valid for the worm but also, moregenerally, for fish and cephalopods living, forexample, in the oxygen minimum layer(Childressand Seibel, 1998) the high O and low CO2 2

concentrations in the upper pelagic zone at highfood availability supported a maximization of ener-gy turnover and, correspondingly, growth rates.Accordingly, loss of the shell and development ofa constantly high level of activity went hand inhand although, at times, squids are able to rest onthe sea bed, as observed even in the ommastrephidI. illecebrosus.

Maximizing energy turnover applies strictly onlyto the most active forms, that is the ommastrephidssuch asI. Illecebrosus, which cover huge stretchesof open ocean when migrating or in search offood. Loliginids like L. brevis reach limits ofperformance at lower levels in environments char-acterized by larger fluctuations of abiotic parame-ters. L. pealei also lives in a ‘complex’ coastalenvironment and is not dependent on maintainingsuch high velocities when catching prey or escap-ing attack. Owing to shorter distances they also donot need the stamina thatI. illecebrosus does. Awell-developed fin is most useful in coastal waters,for instance close to the sea bed, when the abilityto navigate well at low speeds becomes important.The fin with its undulating beat bridges the inher-ent pauses in between jets(refilling phase) andsmoothes out and stabilises movement. In coastalenvironments squid can even make use of upwell-ing currents to counteract sinking. As suggestedfor Loligo forbesi the fin may take on an additionalrole as a hydrofoil(O’Dor et al., 1994). Since thesame physical principles apply as in fish locomo-tion (see above) the use of a fin saves energy. Themore developed the fin, the more it is involved inpropulsion. However, the degree of attainablethrust is much lower than for fish, since only asmall mass of muscle is involved and there is nostrengthening of the muscle comparable to theeffect of the spinal column in fish. Thus the finonly saves energy at low speeds. BothSepiaofficinalis andL. pealei use their fin at low speedswhich results in a lower energy requirement thanin I. illecebrosus (O’Dor and Webber, 1991). Asthe fin increases in size, it increases drag andenergy expenditure by hampering the stream-linedbody shape(cf. O’Dor and Webber, 1991; Hoar etal., 1994). Accordingly, when it is of prime impor-tance to reach high speeds, as is the case forI.illecebrosus, there is only a small fin whichenhances a stream-lined body shape. As a jet ofwater is expelled the remaining fin is flattenedagainst the mantle, thus reducing body drag.


A different strategy to save energy appears toexist inL. brevis (Finke et al., 1996). This loliginidlives in the brackish waters of the Gulf of Mexico.As a small coastal and inshore species it usuallydoes not need to cover long distances searchingfor prey or in escape responses, a fact reflected inits low critical swimming speed(Finke et al.,1996). At increased velocities, the animals oscil-late between periods of high and low muscularactivity as indicated by the recording of low andhigh pressure jets. The relatively small differencebetween resting and active oxygen consumption(Table 1) suggests that this behaviour reducestransport cost and permits a longer term net useof anaerobic resources when speed exceeds thecritical value or when the squid dive into hypoxicwaters(see below).

In the deep sea, there is not such close compe-tition from similar-sized vertebrates. Since fooddensity and temperatures are lower than in conti-nental seas, a very economical way of life iswarranted for all life forms eliminating the com-petitive pressure to maximize performance. Inmidwater fish and squid energy savings are sup-ported by neutral buoyancy mechanisms whichcounteract the danger in sinking into deeper watersand, in the case of squid, alleviate the necessity tocontinuously use jet propulsion for the mainte-nance of position in the water column. Some squidstore fat obtained from their diet in the midgutgland to increase their buoyancy. However, themajority of neutrally buoyant squid have alteredthe ionic composition of their body fluids tocompensate for the high density of their tissues(Clarke et al., 1979). Seawater from the openocean has a density of between 1.023 and 1.028kg l . Densities for whole specimens of 1.06 andy1

1.076 kg l have been recorded forL. pealei andy1

I. illecebrosus, respectively(O’Dor and Webber,1991). To compensate for the high specific densityof their tissues, deep sea species reduce the sul-phate concentration of part of their body fluidsand exchange a relatively large proportion of theirsodium chloride for ammonium chloride(Voightet al., 1994). A solution of ammonium chloride ofseawater osmolarity has a density of 1.01 kg l .y1

In order to achieve neutral buoyancy they thereforeneed to replace a large percentage of their bodyfluids with solutions containing ammonia. To avoidpotential ammonium toxicity, the ammonium chlo-ride solution in cephalopods is stored in fluid filledchambers which are completely separated from the

blood (for review see Voight et al., 1994). Due tothe vacuolisation of the mantle tissue, the giantsquid have a lower proportion of active muscle incomparison with their more agile shallow waterrelatives. For this reason, and due to the size-dependence of oxygen transfer(see above), thesegiant squid can never attain the same level ofactivity as the smaller more muscular species, aconclusion recently confirmed by Webber et al.(2000).

6. pH regulation and muscular energetics

The question arises how the constraints onactivity levels and aerobic energy turnoverimposed by environmental conditions extend tothe use and rate of anaerobic energy turnover. Ingeneral, muscular metabolism in marine inverte-brates has all the characteristics well known formetabolism in vertebrate striated muscle(e.g. Fitts,1994), however, glycolytic end products differentfrom lactate are characteristic for many molluscs,annelids and lower invertebrates. Opines areformed instead which are condensation productsof pyruvate with amino acids(e.g. alanine foralanopine, glycine for strombine, arginine for octo-pine, see Grieshaber et al., 1994, for review). Boththe squids and the worm form octopine duringmuscular exercise. During muscular activitybeyond the anaerobic threshold the formation oflactate is known to lead to acidification of thetissues and blood owing to proton formation in theglycolytic pathway (Hochachka and Mommsen,1983; Portner et al., 1984a; Portner, 1987).¨ Älthough opines by themselves are much weakeracids than lactic acid, the net generation of thepyruvate moiety in glycolysis prior to condensationalways causes acidosis(Portner, 1987). L-arginine¨consumed in octopine formation is provided bythe hydrolysis ofL-arginine phosphate(Eq. (1)and Eq.(2), stoichiometry for Eq.(1) is simpli-fied, cf. Portner, 1987).¨

y yPhospho-L-arginine(PLA )qMgADPqqH

q 2ymL-Arginine (L-Arg )qMgATP (1)

qL-Arginine (L-Arg) q0.5 glucoseqmoctopineqH (2)

Accordingly, octopine formation was shown tocause acidosis in squid muscle, in accordance withthe view that the net process of glycolytic proton


Fig. 4. Correlated changes in pH and octopine levels in twoi

squid species(I. illecebrosus, L. brevis) and a worm (S.nudus). Lines represent second- and third-order polynomial fitsand 95% confidence intervals.(Data by Portner et al., 1991,¨1996 and unpublished, for a discussion of buffer valuesb seetext, NB, non-bicarbonate, eff, effective).

production is not different between lactate or opineproduction(Portner et al., 1991, 1993, 1996).¨

Phosphagen breakdown and anaerobic glycolys-is are predominantly responsible for metabolicchanges in the intracellular acid–base status start-ing beyond the anaerobic threshold. Net ATPbreakdown contributes further protons. Metabolicprocesses which reduce acidosis during activitynot only include phosphagen degradation, but alsothe deamination of AMP or adenosine in the ATPdegradation pathway or the metabolism of dicar-boxylic acids(malate, aspartate) in the early stagesof anaerobic metabolism in the mitochondria,which occurs when excessive oxygen consumptioncannot be compensated for by increased oxygensupply(Portner, 1994).¨

Unlike lactate, octopine does not usually leavethe intracellular space. An in depth study compar-ing all metabolic protons with changes in theacid–base balance was able to track the fate ofmetabolically produced protons(Portner, 1994).¨Virtually all the metabolic protons remained withinthe muscle cells. During the onset of exercise, thetissue even lost base to the extracellular space,illustrated by an early deviation of the correlatedchanges in non-respiratory and metabolic protonquantities from the line of equality in Fig. 3. Thisdeviation indicates an early intracellular non-res-piratory acidosis in the mantle(DH -0)q

i,non-resp

when metabolic proton production does not yetcontribute (DH s0). This base deficit is mir-q

i,met

rored by a base excess in the blood maintainedduring fatigue which is largely unexplained byproton quantities bound during haemocyanin deox-ygenation. During increasing metabolic acidifica-tion the base excess in the blood was reduced, thistime mirrored by a lower slope of the correlatedchanges in non-respiratory and metabolic protonquantities compared to the line of equality. Thisdifference in slope led to a crossover at elevatedlevels of tissue acidosis and finally to a somewhatlarger metabolic compared to non-respiratory pro-ton quantities owing to some proton release fromthe mantle tissue.

As a corollary, modern shell-less cephalopodsappear to be different from most other animalsinvestigated, since they primarily regulate theextracellular, not intracellular, acid–base balanceand maintain similar values of extracellular andintracellular pH under control conditions(cf. Fig.2 and Fig. 4), with a much lower extracellular pHthan found in fish. This immediately suggests that

a value of extracellular and intracellular pH closeto 7.4 (at 15 8C) may have evolved for optimisedpH dependent protein, in this case haemocyaninfunction. Also, squid avoid acidification of theextracellular space, even when high levels ofoctopine are produced(Fig. 3). This strategybecomes understandable in the light of maximizedoxygen transport by the haemocyanin whichdepends upon maximized fine control of extracel-lular acid–base status and protection of haemocy-anin function from metabolic acidification(Fig.2). Initial base release enhances and secures arte-rial haemocyanin oxygen binding. Respiratoryacidification of the blood in the mantle tissue leads


Fig. 5. Correlated changes in muscle intracellular pH and phos-pho-L-arginine levels in two squid species(I. illecebrosus, L.brevis) and a worm(S. nudus). Lines represent second- andthird-order polynomial fits and 95% confidence intervals.(Data by Portner et al., 1991, 1993, 1996 and unpublished).¨

Table 2Maximum acidification and energetic parameters limiting performance of three squid species and a sipunculid worm during functionalanaerobiosis(after Portner, 1993; Portner et al., 1996 and unpublished)¨ ¨

Illex Loligo Lolliguncula Sipunculusillecebrosus pealei brevis nudus

DpHi y0.6 )y0.1 y0.57 y0.53DPLA y31.2 y22.5 y18.8 y12.0 mmol gy1

DL-arginine 14.9 28.0 0.0 2.0 mmol gy1

DATP y2.3 y3.8 y2.5 y1.3 mmol gy1

DADP 1.3 2.4 1.0 0.0 mmol gy1

DAMP 0.7 2.3 0.7 0.0 mmol gy1

DdGydj 13.8 14.0 12 6.0 kJ moly1

Minimum dGydj y42 y42 y45 y49 kJ moly1

to maximum use of haemocyanin bound oxygenon each return. The non-release of opines observedin the squid as well as in the sipunculid wormsuggests that H and end products are removed inq

situ during recovery(Portner, 1993). In squid¨mantle muscle the removal of octopine and Hq

appears to be tightly coupled(Portner, 1994; Port-¨ ¨ner et al., 1991, 1993, 1996).

At peak activity, L. pealei produced far lessoctopine thanLoligo vulgaris (Grieshaber andGade, 1976), I. illecebrosus or L. brevis andëxperienced less intracellular acidosis(Portner etäl., 1991, 1993, 1996). I. illecebrosus accumulatedup to 25mmol octopine g fresh weight, coinci-y1

dent with a reduction in intracellular pH by up to0.6 pH units (Fig. 4, Table 2), whereas inL.brevis, octopine accumulations up to 18mmoloctopine g fresh weight led to a maximumy1

deflection of intracellular pH by 0.57 units withless phosphagen hydrolysis counteracting the acid-ification than in I. illecebrosus (Fig. 5, Table 2).In S. nudus, changes in intracellular pH of up to0.5 pH units occurred, in correlation with octopinelevels of up to 14mmol g fresh weight at eveny1

less use of the phosphagen(Table 2).Work on isolated contracting muscle showed

that ATP content is initially maintained at theexpense of phosphagen breakdown leading tointracellular alkalosis(Eq. (1)). Glycolytic ATPand H production became predominant duringq

maintained muscle activity with the result thatpH fell below control levels(Chih and Elllington,i

1985). These results appear to contradict findingsin vivo, where a highly significant linear correla-tion between changes in intracellular pH and octo-pine accumulation started early on(Fig. 4, Portnerët al., 1991, 1996). The slope of this relationshipreflects the response of the tissue to glycolytic

acidification in vivo as it results from summedreactions of intracellular physicochemical and met-abolic buffering processes and, with the exemption


Fig. 6. Correlated changes in swimming velocity(mantlelengths s ) or muscle intracellular pH and the Gibbs’s freey1

energy change of ATP hydrolysis in a squid species(L. brevis)and a worm(S. nudus). Lines represent 2nd and 3rd orderpolynomial fits and 95% confidence intervals.(Data by Portnerët al., 1996 and unpublished).

of squid, proton equivalent ion exchange betweentissue and extracellular fluid. Evidently, the com-pensatory or additive processes affect proton bal-ance in such a way that they become involvedearly on and in addition to anaerobic glycolysis,without fully compensating for the acidosis. Theyeither take place at a constant rate, with nodisruption of the linear relationship, or are them-selves directly dependent on pH in a way that ai

change in the linear slope results for the wholerange of pH . The direct dependence ofL-argininei

phosphorylation on pH is indicative that releasei

of inorganic phosphate during phosphagen hydrol-ysis has a direct effect on the degree of acidifica-tion caused by octopine(Fig. 5). Within a certainpH range, the contribution of inorganic phosphateto buffering changes linearly with pH(Portner etäl., 1996). An effective buffer valueb results(Fig.4), which is higher (bNB) than that measuredunder control conditions and indicates how protonconsuming processes in metabolism collectivelyreduce the degree of acidification(Portner et al.,¨1996).

Although at first sight mantle tissues of theommastrephid squid,I. illecebrosus and the loli-ginid squid,L. brevis as well as the muscle of thedigging worm appear similar in their metabolicresponse to exercise above the anaerobic threshold,differences become obvious when the relationshipsbetween octopine levels and intracellular pH arecompared(Fig. 4). The slopes of the in vivotitration curve(effective buffer line) are closer tothe slopes of the non-bicarbonate buffer line inL.brevis and S. nudus than in I. illecebrosus. Mostof this difference can be explained by slowerphosphagen turnover inL. brevis and S. nudus inaddition to a reduced amount of phosphagen beingavailable(Fig. 5). Looking at the effects of controlpH on phosphagen levels, PLA levels were foundhigher inIllex than inLolliguncula andSipunculus.Following Eq. (1) this is likely supported byhigher control pH values(Fig. 5) and sufficientlyi

high L-arginine levels. Less PLA depletion and, asa consequence, less inorganic phosphate accumu-lation in L. brevis andS. nudus explains the lowerbuffering of glycolytic protons and is related tolower levels of free ADP than inI. illecebrosusand L. pealei (Finke et al., 1996; Portner et al.,¨1996). Obviously, L. brevis and S. nudus utilizeacidification rather than the accumulation of freeADP to elicit phosphagen transphosphorylationand ATP buffering, whereasI. illecebrosus uses

both. In L. pealei accumulation of total and freeADP was far more expressed(Table 2) and triggersphospho-L-arginine depletion(Eq. (1)) with lessacidosis contributing than in the other species.

The extensive breakdown of ATP to ADP inL.pealei illustrates that this species is less wellequipped to withstand greater physical demandsthan I. illecebrosus (Portner et al., 1993). This is¨what would be expected from the ‘quieter’ way oflife of a loliginid which extensively uses its fins.Evidently, the acidosis(in I. Illecebrosus, L. brevisand S. nudus) protected ATP from being largelydegraded and thereby supported extended muscularactivity (see also Fig. 7 below, and Hartmund andGesser, 1995).


Fig. 7. Model of the delayed drop in ATP free energy and theincrease in maximum swimming speed depending on the buf-fering of ATP levels by use of the phosphagen, calculated tobe elicited by each, the drop in pH , the accumulation of freei

ADP, the removal ofL-arginine during octopine formation and,finally, the combined action of all three parameters(modifiedafter Portner et al., 1996).¨

7. Parameters extending and limiting workintensity and muscle function

The onset of anaerobiosis beyond the criticalswimming speed appears as the key to an under-standing of fatigue in squid mantle muscle. A fallin pH affects the energy status of the muscle cellswhich is not only related to ATP levels butquantified as the Gibb’s free energy change ofATP hydrolysis(dGydj). This parameter is seento limit the function of cellular ATPases(Kam-mermeier et al., 1982; Portner, 1993; Combs andËlllington, 1996). The term ‘ATP free energychange’ is derived from various parameters includ-ing intracellular pH and cellular concentrations ofATP, free ADP, inorganic phosphate, and freemagnesium(Portner et al., 1996). Assuming con-¨stant levels of all of these parameters, pH by itselfcauses a rather small ‘drop’ in standard and invivo ATP free energy change values by approxi-mately 2 kJ mol with a minimum below pH 7.0y1

(Portner, 1993). In squid mantle and sipunculid¨muscle, a much larger change occurs which,though partly caused by the acidosis, tends tofollow the depletion of ATP, the phosphagen andthe accumulation of ADP and inorganic phosphate(Fig. 6). These relationships await closer investi-gation in vertebrates.

The minimum value for ATP free energy changemay contribute to the onset of contractile failureor fatigue independent of pH. This conclusion issupported by the observation that the respectivechange in energy status was similar in fatiguedI.illecebrosus and L. pealei although pH fell byapproximately 0.6 pH units inI. illecebrosus man-tle but only by less than 0.1 pH units inL. pealei(Portner et al., 1993, 1996; cf. Table 2). As a¨corollary, a drop in ATP free energy occurs to thesame extent in fatiguing muscle regardless ofwhether pH falls at the same time or not(cf. Table2). This low level may actually represent thelimiting level of ATP free energy change. At thesame time, an intracellular acidosis as seen inI.illecebrosus (and L. brevis) may help to protectATP from being degraded(Portner et al., 1993,¨1996). With a similar change and the same mini-mal value of ATP free energy change as inI.illecebrosus, more ATP was depleted inL. pealeiand more ADP and AMP accumulated. An acido-sis, although decreasing the actual value of thefree energy change of ATP hydrolysis to someextent (see above), is in most cases associated

with complementary ATP production by glycolysisand by triggering the use of phospho-L-arginine,thereby compensating for this apparent disadvan-tage and delaying the progressive depletion of ATPlevels and the fatal decrease in ATP free energychange.

Fig. 7 models these processes as they dependupon each individual parameter, the decrease inintracellular pH, the accumulation of free ADPand the removal of arginine by octopine in thebrief squidL. brevis (Portner et al., 1996). Unbuf-¨fered ATP depletion calculated as being equivalentto the measured depletion of high energy phos-phates(the sum of ATP and PLA levels) wouldcause the ATP free energy change to fall early tothe observed minimum and would not allow theanimals to reach a higher than the critical swim-ming speed. It becomes obvious that the accumu-lation of free ADP causes significant ATPbuffering and an increase in the maximum swim-


ming velocity but the effect is less than thetransphosphorylation caused individually by each,the formation of octopine and, even more so, thefall in pH . The combined and integrated action ofi

all mechanisms finally allows the animals to raiseperformance and exceed the critical swimmingspeed (1.5 mantle lengths s ) by a factor ofy1

approximately three(Portner et al., 1996).¨The model emphasizes to what extent buffering

of ATP levels is linked to a buffering of ATP freeenergy change values which are decreased alreadyby the accumulation of inorganic phosphate asso-ciated with phosphagen depletion. Actually, thisdecrease can be considered advantageous, sinceboth the acidosis and the accumulation of inorganicphosphate have some protective effect on ATPconcentrations. As a consequence, the minimumlevel of ATP free energy is reached at higher ATPcontents with the use of the phosphagen thanwithout (Fig. 7).

The accumulation of inorganic phosphate is verylikely involved in limiting the degree of muscularactivity or reducing the maximum swimmingvelocity, processes which we tend to interpret assigns of fatigue(Fitts, 1994). This may actuallystart to slow muscular performance at levels ofATP free energy higher than the minimum, limitinglevel. For example, fatigue or a reduction in motoractivity is seen inL. brevis and especiallyS. nudusat higher values of the Gibb’s free energy changeof ATP hydrolysis than inI. illecebrosus (Table2). This may be related toLolligunculas’s abilityto dive into hypoxic waters(Vecchione, 1991)where a long term utilization of energy stores isrequired at low activity levels(Portner et al., 1996,¨see below). The comparison of the squid and thedigging worm emphasizes that differences inanaerobic metabolic rate cause the differences inslopes(at unchanged linearity) in the octopineypH relationships(Fig. 4). The differences in met-abolic rate reflect a delayed depletion of thephosphagen and also less of a decrease in theGibb’s free energy change of ATP hydrolysis inL. brevis and even more soS. nudus compared toboth I. illecebrosus and L. pealei (Portner et al.,¨1996, Table 2). This relates to a low energyturnover mode of locomotion not only forL.brevis, but also for the worm, in accordance withthe energy saving strategy discussed above. Anaer-obic marathon digging byS. nudus in the oxygendeficient marine sediment is associated with aneven larger extent of protection of ATP free energy

change than in the brief squid(Table 2) suggestingthat such a strategy may be typical for hypoxiatolerant animals.

In both the brief squid and the worm, a trendbecomes visible to reduce performance at higherlevels of ATP free energy change compared toL.pealei andI. Illecebrosus (Fig. 6, Table 2). Fatigueor, more adequately, the inability to maintain max-imum swimming speeds or digging rates may thenno longer be related to the drop in ATP free energychange(Fig. 6) but may mirror the reduction inmuscular performance elicited by the combinedeffects of low pH and elevated inorganic phosphatelevels on muscular performance(cf. Fitts, 1994).Actually, the lower degree of proton buffering inL. brevis and S. nudus (Fig. 4) means that pHfalls more rapidly in these species and may causea protective decrease in performance before theenergy status reaches a critically low level. Adecrease in performance induced by acidosis istypically observed in many types of muscle tissue(Fitts, 1994; Hartmund and Gesser, 1995) and maybe protective during ischemia, long term environ-mental stress or extended periods of anaerobicexercise.

These considerations lead to additional insightsfor the three squid species and the worm(Table2). Reduced acidosis(as in L. pealei) poses ahigher threat of ATP degradation causing moresevere and, possibly, longer lasting symptoms offatigue at low energy levels. In contrast, a severeacidosis at reduced rates of phosphagen depletion(as in L. brevis or S. nudus) reduces performanceand may support the more economical and longterm use of anaerobic resources. Maximum anaer-obic ATP and phosphagen turnover combined witha high rate of glycolytic proton production(as inI. illecebrosus) use the advantages of acidosis(protection of ATP levels owing to maximumexploitation of the phosphagen) at maximum levelsof performance until the critical level of acidosisor energy status may be reached. Again, a com-parable analysis for vertebrates of different life-styles is not available.

As discussed above squid do not release signif-icant amounts of H and glycolytic end productsq

into the blood. InS. nudus the extent to whichprotons are released into the extracellular fluidsdepends upon the rate of anaerobic glycolysis. Atreduced levels of octopine formation(Portner,¨1986) more than 90% of the metabolic protonsleft the body wall musculature and reached the


extracellular space. This process would reduce thedrop in pH , delay the use of the phosphagen andi

extend the period during which exercise is possi-ble. There appears to be a tradeoff between ‘vol-untary’ low rates of anaerobic exercise whichavoid the intracellular acidosis and, thereby, allowfor extended exercise periods on the one side and,on the other side, the maximized use of anaerobicresources, possibly in a stress situation, whichleads to more severe acidosis, thereby, passivelylimits muscular performance and ‘enforces’ areduction in the use of anaerobic resources.

In conclusion, reducing the energy requirementsfor locomotion in hypoxic environments protractsthe depletion of anaerobic resources and alsodelays fatigue through a slower use of the phos-phagen. Development of an acidosis is almostunchanged, albeit at reduced rates of phosphagendepletion. The acidosis protects the levels andenergy contents of the adenylates(Gibb’s freeenergy change of ATP hydrolysis) and may sup-port a reduction of performance prior to fatiguewhich might be understood as a strategy to extendthe period of activity in hypoxic environments. Ifmetabolic expenditure is low to begin with, effec-tive pH regulation may contribute to a long termi

use of anaerobic resources as seen inS. nudus. Asa corollary, the capacity of pH regulation stronglyi

modulates the use of anaerobic resources. Progres-sive acidification first supports maximum levels ofactivity but then sets a limit to the level ofperformance and, finally, the period of anaerobicexercise.

8. Perspectives: tradeoffs between aerobic scope,muscular performance and body size

It has been concluded above that a species likeL. brevis is operating at environmental limits oftemperature and oxygen availability. The relianceon oxygen uptake via the skin and maximized useof blood oxygen transport further indicates that inhighly active muscular squid likeI. illecebrosus,the capacity to extract enough ambient oxygenoperates at the limits of ambient oxygen levels.Small body size favours oxygen uptake owing toimproved volume to body surface ratios. This mayactually be one explanation of whyL. brevisremains rather small in a warm environment whereits metabolic rate is high, where hypoxia eventsoccur regularly and where at approximately 258Cmaximum water oxygen levels at air saturation

only reach between 60 and 65% of the levelsfound at polar (0 8C) or deep sea(2–4 8C)temperatures.

The outline of an oxygen limited performanceand the maximized use of ambient oxygen inI.illecebrosus suggests that similar conclusions canbe drawn for this ommastrephid squid. The highperformance levels of this species not just takeadvantage of the constant environmental conditionsof the pelagic, open ocean waters but are onlypossible under these conditions. Adult individualsof this species are likely to operate at their maxi-mum body size since this size may be the maxi-mum possible which allows to extract significantamounts of oxygen from the water via the skin(Fig. 8).

Following this line of thought the large bodysize of giant squid may only be possible under thecold temperature conditions of the deep sea, whenoxygen solubility is high and metabolic energyrequirements are reduced for all species owing tofood and temperature constraints. This leaves moreroom for increasing body size, supported by lowSMR, low net metabolic scopes, and, also, energyefficient growth (Portner, 2002a,b; cf. Portner et¨ äl., 2001, Fig. 8). Accordingly, exposure to warmwaters may cause asphyxiation and death of theseanimals, because water oxygen levels fall andoxygen demand rises to levels beyond critical.Restriction of giant squids to cold water environ-ments is also reflected in the observation that theblood pigment is adapted to function at low tem-peratures only(Brix, 1983). As a corollary, maxi-mum body size in squid appears to be the resultof a tradeoff between maximum energy expendi-ture attainable in a certain environment, availablestrategies to maximize the use of ambient oxygenand the potential development and use of energysaving strategies without compromising the suc-cessful competition with pelagic vertebrates.

In a completely different environment, oxygenavailability very likely also limits the body size ofthe marine worm,S. nudus, which also relies oncutaneous respiration. According to the above,aerobic energy used by ventilatory activity reflectsthe maximum aerobic scope available to this spe-cies, which is larger in small than in large animals.On the one hand a minimized aerobic scope inoxygen poor environments will allow for the devel-opment of a larger body size, on the other handthe larger body size will cause oxygen limitations


Fig. 8. Modelled, semi-quantitative depiction of the constraints and tradeoffs imposed by constant or fluctuating oxygen availabilityand temperature, as well as the degree of eurythermality or euryoxia(reflected by the lengths and widths of the respective box for eachspecies), and, in consequence, temperature dependent aerobic scope, on body size in cephalopods and sipunculids(SMR: standardmetabolic rate, MMR: maximum metabolic rate, distinguished between polar, i.e. Antarctic and deep sea stenotherms, where no met-abolic cold compensation occurs, and cold adapted eurytherms, which display significant metabolic cold adaptation, based on Portnerët al., 1984b; Portner, 2002a,b). Reduced standard metabolism and net metabolic scope in polar and deep sea stenotherms, including¨the giant squid, combined with increased oxygen solubilities in cold waters are interpreted as preconditions for the development ofgigantism in cold waters. In eurythermal squids with maximized aerobic energy turnover in warm water or temperate environments,like I. illecebrosus and L. brevis, or in the sipunculid worms living in hypoxic environments, maximum body size is also interpretedto be set by the combined effects of limited oxygen availability and the degree of temperature fluctuations(for further explanationssee text. Arrows and question marks indicate uncertainties in knowledge of the actual level of eurythermality or euryoxia. With respectto euryoxia note that the figure emphasizes ambient oxygen maxima rather than the full range of oxygen, incl. hyperoxia tolerance. Itdoes not reflect that the level of hypoxia tolerance, also indicated by arrows pointing left, will vary with ambient temperature, cf.Portner, 2002a. A wide range of ambient oxygen levels is emphasized forL. brevis and suggested to be relevant in reducing maximum¨body size.).

to become effective already under resting condi-tions. Depending on whether individuals are foundin coarse or fine sandy sediments those found incoarse sands reach 3–5 times larger body sizes.Since the non-permanent burrow has only oneopening towards the surface water all ventilatorywater is pressed into the sand, a process whichfinds less resistance in coarse compared to finesediments. Large animals need to pump a largervolume of water and, therefore, experience thislimitation earlier than small animals. Large speci-mens also maintain lower levels of body fluidP at the same level of ambient oxygen reflectingO2

an increased level of oxygen limitation at large

body size(Portner et al., 1985) or, in other words,ä limitation of body size by oxygen availability.

Body size limitations by oxygen availabilityhave most recently been discussed for amphipodsand deep sea gastropods based on correlationsbetween maximum body size within those groupsand temperature dependent oxygen content inmarine and freshwater environments(Chapelle andPeck, 1999; McClain and Rex, 2001). The presentdiscussion provides some insight into the physio-logical mechanisms and tradeoffs behind thoseempirical observations. The model depicted in Fig.8 suggests that not just the rise in ambient oxygenlevels but also the decrease in SMR and net aerobic


scope at low temperature are most likely precon-ditions for a cold induced rise in body size.

As a corollary, environmental and lifestyle con-straints shape the level of aerobic scope availableto the marine invertebrate species discussed.Whereas oxygen limitations in the sediment mini-mize aerobic scope for exercise and thereby allowfor larger body sizes in the worm, aerobic scopeis maximized in squid owing to the necessity tocompete with vertebrates at similar performancelevels. Within their group, the elite athletes havemaximized aerobic performance up to the limitsset by environmental constraints. In a tradeoffbetween aerobic scope and body size this maylimit body size depending on environmental char-acteristics. In some species, body size is likely setto a maximum level that does allow maintenanceof just sufficient aerobic scope. Similar to aerobicscope, which limits the capacity for long term andlong distance cruising at high velocities, the rateof anaerobic performance depends on the maxi-mum energy turnover required for escape or attackin the case of ommastrephid squid or on the lengthof time when anaerobic metabolism is needed, e.g.during excursions into hypoxic environmentslinked to food uptake, in the case ofL. brevis andthe digging worm. It should be very stimulatingto test the applicability of these principles in otherorganisms as well including the group of verte-brates where such detailed analyses of environ-mental and lifestyle limitations to the capacity ofmuscular exercise, including the implications forbody size, have not yet been carried out.

Acknowledgments

The author wishes to thank Iris Hardewig forexercising the worms. Supported by grants fromthe Deutsche Forschungsgemeinschaft(Po 278).

References

Bone, Q., Pulsford, A., Chubb, A.D., 1981. Squid mantlemuscle. J. Mar. Biol. Ass. UK 61, 327–342.

Brett, J.R., 1964. The respiratory metabolism and swimmingperformance of young sockeye salmon. J. Fish. Res. Bd.Canada 21, 1183–1226.

Bridges, C.R., 1994. Bohr and Root effects in cephalopodhaemocyanins—paradox or pressure inSepia officinalis. In:Portner, H.O., O’Dor, R.K., MacMillan, D.(Eds.), Physi-ölogy of Cephalopod Molluscs—Lifestyle and PerformanceAdaptations. Gordon and Breach Publishers, Basel, pp.121–130.

Brill, R.W., 1996. Selective advantages conferred by the highperformance physiology of tunas, billfishes, and dolphinfish. Comp. Biochem. Physiol. 113A, 3–15.

Brix, O., 1983. Giant squids may die when exposed to warmcurrents. Nature 303, 422–423.

Brix, O., Bardgard, A., Cau, A., Colosimo, A., Condo, S.G.,˚Giardina, B., 1989. Oxygen binding properties of cephalo-pod blood with special reference to ambient temperaturesand ecological distribution. J. Exp. Zool. 252, 34–42.

Burgetz, I.J., Rojas-Vargas, A., Hinch, S.G., Randall, D.J.,1998. Initial recruitment of anaerobic metabolism duringsub-maximal swimming in rainbow trout(Oncorhynchusmykiss). J. Exp. Biol. 201, 2711–2721.

Chapelle, G., Peck, L.S., 1999. Polar gigantism dictated byoxygen availability? Nature 399, 114–115.

Chih, C.P., Elllington, W.R., 1985. Metabolic correlates ofintracellular pH change during rapid contractile activity ina molluscan muscle. J. Exp. Zool. 236, 27–34.

Childress, J.J., Seibel, B.A., 1998. Life at stable low oxygenlevels: adaptations of animals to oceanic oxygen minimumlayers. J. Exp. Biol. 201, 1223–1232.

Clarke, M.R., Denton, E.J., Gilpin-Brown, J.B., 1979. On theuse of ammonium for buoyancy in squids. J. Mar. Biol.Ass. UK 59, 259–276.

Combs, C.A., Ellington, W.R., 1996. Graded intracellularacidosis produces extensive and reversible reductions in theeffective free energy change of ATP hydrolysis in a mollus-can muscle. J. Comp. Physiol. 165B, 203–212.

Finke, E., Portner, H.O., Lee, P.G., Webber, D.M., 1996. Squid¨life in shallow waters: oxygen limitation of metabolism andperformance in exercising gulf squid,Lolliguncula brevis.J. Exp. Biol. 199, 911–921.

Fitts, R.H., 1994. Cellular mechanisms of muscle fatigue.Physiol. Rev. 7, 49–94.

Florkin, M., 1933. Recherches sur les hemerythrines. Arch.´ Ínt. Physiol. 36, 247–328.

Goolish, E.M., 1991. Anaerobic swimming metabolism of fish:sit-and-wait versus active forager. Physiol. Zool. 64,485–501.

Gosline, J.M., Demont, M.E., 1985. Jet-propelled swimmingin squids. Sci. Am. 252, 96–103.

Gosline, J.M., Steeves, J.D., Harman, A.D., Demont, M.E.,1983. Patterns of circular and radial muscle activity inrespiration and jetting of the squid,Loligo opalescens. J.Exp. Biol. 104, 97–109.

Grieshaber, M., Gade, G., 1976. The biological role of octopineïn the squid,Loligo vulgaris (Lamarck). J. Comp. Physiol.108, 225–232.

Grieshaber, M.K., Hardewig, I., Kreutzer, U., Portner, H.O.,¨1994. Physiological and metabolic responses to hypoxia ininvertebrates. Rev. Physiol. Biochem. Pharmacol. 125,43–147.

Hardewig, I., Addink, A.D.F., Grieshaber, M.K., Portner, H.O.,¨van den Thillart, G., 1991. Metabolic rates at differentoxygen levels determined by direct and indirect calorimetryin the oxyconformerSipunculus nudus. J. Exp. Biol. 157,143–160.

Hartmund, T., Gesser, H., 1995. Acidosis, glycolysis andenergy state in anaerobic heart tissue from rainbow trout. J.Comp. Physiol. 165B, 219–229.


Hendrickson, W.A., Smith, J.L., Sheriff, S., 1985. Structureand function of hemerythrins. In: Lamy, J., Truchot, J.-P.,Gilles, R.(Eds.), Respiratory Pigments in Animals. Spring-er, Berlin, pp. 1–8.

Hoar, J., Sim, E., Webber, D.M., O’Dor, R.K., 1994. The roleof fins in the competition between squid and fish. In:Rayner, J., Bone, Q., Maddock, L.(Eds.), Mechanics andPhysiology of Animal Swimming. Cambridge UniversityPress, UK, pp. 27–43.

Hochachka, P.W., 1994. Oxygen efficient design of cephalopodmuscle. In: Portner, H.O., O’Dor, R.K., MacMillan, D.¨(Eds.), Physiology of Cephalopod Molluscs—Lifestyle andPerformance Adaptations. Gordon and Breach Publishers,Basel, pp. 61–67.

Hochachka, P.W., Mommsen, T.P., 1983. Protons and anaero-biosis. Science 219, 1391–1398.

Kammermeier, H., Schmidt, P., Jungling, E., 1982. Free energychange of ATP-hydrolysis: a causal factor of early hypoxicfailure of the myocardium? J. Mol. Cell. Cardiol. 14,267–277.

Kemper, W.F., Lindstedt, S.L., Hartzler, L.K., Hicks, J.W.,Conley, K.E., 2001. Shaking up glycolysis: sustained, highlactate flux during aerobic rattling. Proc. Natl. Acad. Sci.98, 723–728.

Kieffer, J.D., 2000. Limits to exhaustive exercise in fish.Comp. Biochem. Physiol. 126A, 161–179.

Korsmeyer, K.E., Dewar, H., Lai, N.C., Graham, J.B., 1996.The aerobic capacity of tunas: adaptation for multiplemetabolic demand. Comp. Biochem. Physiol. 113A, 17–24.

Lucas, M.C., Johnstone, A.D.F., Tang, J., 1993. An annularrespirometer for measuring aerobic metabolic rates of large,schooling fish. J. Exp. Biol. 175, 325–331.

Lykkeboe, G., Brix, O., Johansen, K., 1980. Oxygen-linkedCO binding independent of pH in cephalopod blood. Nature2

287, 330–331.Mangum, C.P., 1990. Gas transport in the blood. In: Gilbert,

D.L., Adelman, W.J., Arnold, J.M.(Eds.), Squid as Exper-imental Animals. Plenum Press, New York, pp. 443–468.

Manwell, C., 1960. Histological specificity of respiratorypigments. II. Oxygen transfer systems involving hemery-thrins in sipunculid worms of different ecologies. Comp.Biochem. Physiol. 1, 277–285.

McClain, C.R., Rex, M.A., 2001. The relationship betweendissolved oxygen concentration and maximum size in deep-sea turrid gastropods: an application of quantile regression.Mar. Biol. 139, 681–685.

Miller, K.I., 1994. Cephalopod haemocyanins. A review ofstructure and function. In: Portner, H.O., O’Dor, R.K.,¨MacMillan, D. (Eds.), Physiology of Cephalopod Mol-luscs—Lifestyle and Performance Adaptations. Gordon andBreach Publishers, Basel, pp. 101–120.

Milligan, C.L., 1996. Metabolic recovery from exhaustiveexercise in rainbow trout. Comp. Biochem. Physiol. 113A,51–60.

Mommsen, T.P., Ballantyne, J., MacDonald, D., Gosline, J.,Hochachka, P.W., 1981. Analogues of red and white musclein squid mantle. Proc. Natl. Acad. Sci. USA 78, 3274–3278.

O’Dor, R.K., Webber, D.M., 1986. The constraints on cepha-lopods: why squid aren’t fish. Can. J. Zool. 64, 1591–1605.

O’Dor, R.K., Webber, D.M., 1991. Invertebrate athletes: trade-offs between transport efficiency and power density incephalopod evolution. J. Exp. Biol. 160, 93–112.

O’Dor, R.K., Portner, H.O., Shadwick, R.E., 1990. Squid asëlite athletes: locomotory, respiratory and circulatory inte-gration. In: Gilbert, D.L., Adelman, W.J., Arnold, J.M.(Eds.), Squid as Experimental Animals. Plenum Press, NewYork, pp. 481–503.

O’Dor, R.K., Hoar, J.A., Webber, D.M., et al., 1994. Squid(Loligo forbesi) performance and metabolic rates in nature.In: Portner, H.O., O’Dor, R.K., MacMillan, D.(Eds.),¨Physiology of Cephalopod Molluscs—Lifestyle and Per-formance Adaptations. Gordon and Breach Publishers,Basel, pp. 163–177.

Portner, H.O., 1986. Proton balance of anaerobic and post-änaerobic metabolism- interrelations with pH regulation inmarine invertebrates. Zool. Beitr. N.F. 30, 231–247.

Portner, H.O., 1987. Contributions of anaerobic metabolism to¨pH regulation in animal tissues: theory. J. Exp. Biol. 131,69–87.

Portner, H.O., 1990a. An analysis of the effects of pH onöxygen binding by squid(Illex illecebrosus, Loligo pealei)haemocyanin. J. Exp. Biol. 150, 407–424.

Portner, H.O., 1990b. Determination of intracellular buffer¨values after metabolic inhibition by fluoride and nitrilotri-acetic acid. Respir. Physiol. 81, 275–288.

Portner, H.O., 1993. Multicompartmental analyses of acid–¨base and metabolic homeostasis during anaerobiosis: inver-tebrate and lower vertebrate examples. In: Hochachka, P.W.,Lutz, P.L., Sick, T., Rosenthal, M., van den Thillart, G.(Eds.), Surviving Hypoxia: Mechanisms of Control andAdaptation. CRC Press Inc, Boca Raton, pp. 139–156.

Portner, H.O., 1994. Coordination of metabolism, acid–base¨regulation and haemocyanin function in cephalopods. In:Portner, H.O., O’Dor, R.K., MacMillan, D.(Eds.), Physi-ölogy of Cephalopod Molluscs—Lifestyle and PerformanceAdaptations. Gordon and Breach Publishers, Basel, pp.131–148.

Portner, H.O., 2001. Climate change and temperature depend-ënt biogeography: oxygen limitation of thermal tolerance inanimals. Naturwissenschaften 88, 137–146.

Portner, H.O., 2002a. Climate variations and the physiological¨basis of temperature dependent biogeography: systemic tomolecular hierarchies of thermal tolerance in animals. Comp.Biochem. Physiol. 132A, 739–761.

Portner, H.O., 2002b. Physiological basis of temperature¨dependent biogeography: tradeoffs in muscle design andperformance in polar ectotherms. J. Exp. Biol. 205,2217–2230.

Portner, H.O., Heisler, N., Grieshaber, M.K., 1984a. Anaero-¨biosis and acid–base status in marine invertebrates: a theo-retical analysis of proton generation by anaerobicmetabolism. J. Comp. Physiol. 155B, 1–12.

Portner, H.O., Kreutzer, U., Siegmund, B., Heisler, N., Griesh-äber, M.K., 1984b. Metabolic adaptation of the intertidalworm Sipunculus nudus to functional and environmentalhypoxia. Mar. Biol. 79, 237–247.

Portner, H.O., Heisler, N., Grieshaber, M.K., 1985. Oxygen¨consumption and mode of energy production in the intertidalworm Sipunculus nudus L.: definition and characterizationof the criticalP for an oxyconformer. Respir. Physiol. 59,O2

361–377.Portner, H.O., Boutilier, R.G., Tang, Y., Toews, D.P., 1990.¨

Determination of intracellular pH andP after metabolicCO2

inhibition by fluoride and nitrilotriacetic acid. Respir. Phy-siol. 81, 255–274.


Portner, H.O., Webber, D.M., Boutilier, R.G., O’Dor, R.K.,¨1991. Acid–base regulation in exercising squid(Illex ille-cebrosus, Loligo pealei). Am. J. Physiol. 261, R239–R246.

Portner, H.O., Webber, D.M., O’Dor, R.K., Boutilier, R.G.,¨1993. Metabolism and energetics in squid(Illex illecebrosus,Loligo pealei) during muscular fatigue and recovery. Am.J. Physiol. 265, R157–R165.

Portner, H.O., Grieshaber, M.K., 1993. Characteristics of the¨critical P (s): gas exchange, metabolic rate and the modeO2

of energy production. In: Bicudo, J.E.P.W.(Ed.), The Ver-tebrate Gas Transport Cascade: Adaptations to Environmentand Mode of Life. CRC Press Inc, Boca Raton FL, USA,pp. 330–357.

Portner, H.O., Finke, E., Lee, P.G., 1996. Effective Gibb’s freeënergy change of ATP hydrolysis and metabolic correlatesof intracellular pH in progressive fatigue of squid(Lolligun-cula brevis) mantle muscle. Am. J. Physiol. 271,R1403–R1414.

Portner, H.O., Hardewig, I., Sartoris, F.J., van Dijk, P., 1998.Ënergetic aspects of cold adaptation: critical temperaturesin metabolic, ionic and acid–base regulation? In: Portner,¨H.O., Playle, R.(Eds.), Cold Ocean Physiology. CambridgeUniversity Press, pp. 88–120.

Portner, H.O., Zielinski, S., 1998. Environmental constraintsänd the physiology of performance in squids. In: Payne,A.I.L., Lipinski, M.R., Clarke, M.R., Roeleveld, M.A.C.(Eds.), Cephalopod Biodiversity Ecology and Evolution. S.Afr. J. Mar. Sci. 20, 207–221.

Portner, H.O., Berdal, B., Blust, R., et al., 2001. Climateëffects on growth performance, fecundity and recruitmentin marine fish: developing a hypothesis for cause and effectrelationships in Atlantic cod(Gadus morhua) and commoneelpout(Zoarces viviparus). Continental Shelf Research 21,1975–1997.

Shadwick, R.E., 1994. Mechanical organization of the mantleand circulatory system of cephalopods. In: Portner, H.O.,¨

O’Dor, R.K., MacMillan, D. (Eds.), Physiology of Cepha-lopod Molluscs—Lifestyle and Performance Adaptations.Gordon and Breach Publishers, Basel, pp. 69–85.

Trueman, E.R., Foster-Smith, R.L., 1976. The mechanism ofburrowing ofSipunculus nudus. J. Zool. 179, 373–386.

Vecchione, M., 1991. Dissolved oxygen and the distributionof the euryhaline squidLolliguncula brevis. Abstract Bull.Mar. Sci. 49, 668–669.

Voight, J.R., Portner, H.O., O’Dor, R.K., 1994. A review ofämmonia-mediated buoyancy in squids(Cephalopoda: Teu-thoidea). In: Portner, H.O., O’Dor, R.K., MacMillan, D.¨(Eds.), Physiology of Cephalopod Molluscs—Lifestyle andPerformance Adaptations. Gordon and Breach Publishers,Basel, pp. 193–203.

Webb, P.W., 1971. The swimming energetics of trout. II.Oxygen consumption and swimming efficiency. J. Exp.Biol. 55, 521–540.

Webber, D.M., O’Dor, R.K., 1985. Monitoring the metabolicrate and activity of free-swimming squid with telemeteredjet pressure. J. Exp. Biol. 126, 205–224.

Webber, D.M., Aitken, J.P., O’Dor, R.K., 2000. Costs oflocomotion and vertic dynamics of cephalopods and fish.Physiol. Biochem. Zool. 73, 651–662.

Wells, M.J., 1994. The evolution of a racing snail. In: Portner,¨H.O., O’Dor, R.K., MacMillan, D.(Eds.), Physiology ofCephalopod Molluscs—Lifestyle and Performance Adapta-tions. Gordon and Breach Publishers, Basel, pp. 1–12.

Wells, M.J., Hanlon, R.T., Lee, P.G., Dimarco, F.P., 1988.Respiratory and cardiac performance inLolliguncula brevis(Cephalopoda, Myopsida): the effects of activity, tempera-ture and hypoxia. J. Exp. Biol. 138, 17–36.

Wieser, W., 1986. Bioenergetik. Energietransformationen beiOrganismen. Georg Thieme, Stuttgart.

Zielinski, S., Lee, P.G., Portner, H.O., 2000. Metabolic per-¨formance of the squidLolliguncula brevis during hypoxia:an analysis of the criticalP . J. Exp. Mar. Biol. Ecol. 243,O2

241–259.

Documents

Review Environmental and functional limits to muscular ... · ommastrephid squid, migrates vast distances through the open ocean as part of its life cycle. For this reason, the most