Berichte zur Polarforschung, Alfred Wegener Institut fuer Polarforschung, Sonderheft 4 (1983) On the biology of Krill Euphausia superba, Proceedings of the Seminar and Report of Krill Ecology Group, Editor S. B. Schnack, Frontpage and article on pages 130 - 155
The more knowledge we accumulate about physiological and biological data of Euphausia superba the more it appears, that this animal is quite an "extreme" one. We might look upon it as the Elephant is in comparative physiological discussions.
Some characteristics of Euphausia superba:
Unusually high weight (60 times that of Euphausia pacifica).
Unusually high metabolism for an euphausiid of that size and environment (reflected in a respiration of 1 mg oxygen per gramm dry weight per hour, swimming speeds up to 60 cm per second and a reaction time of 40 millisecond). Unusually the high size step between krill and its food (see Tab. 1) .
Tab. 1 Size- and weight-relationship between krill and its food.
Phytoplankton Euphausia superba
length 6 micrometer 60 mm
lenth-relation 1 / 10 000
weight 2.2 * 10-10 g 1.5 g
weigth-relation 1 / 7 000 000 000
Unusual too is the relationship between energetics and size. Normally bigger animals have a lower specific metabolism (energy per body unit and time unit) than smaller have, with an exponent of 0.6-0.7 to the weight. This is shown in Fig. 1 by the solid line. The reasons for this will not be discussed here, for details see Bertalanffy (1951), Champalbert and Gaudy (1972), Conover (1960), Ivlev (1963), Kils (1982b), Pauly (1979), Winberg (1961). The abundant investigations of euphausiid respiration, however, all find exponents close to 1, what means, that respiration increases much too strong during the growth of the animal (Chekunova and Rynkova 1974, Hirche 1983, Lasker 1966, Rakusa-Suszczewski and Opalinski 1978, Segawa et al. 1979, 1982, Small and Hebard 1967, Voss 1982). There must be some parts in the summequation of energy which strongly increase with size or weight. One indication into this direction is described in detail in Kils (1982h) and summarized here: Due to an unusually high underwater-weight (the gravity of adult krill is 1.070 compared to 1.055 of "normal" pelagic animals) krill has to bring up a considerable amount of energy not to loose height. This part of energy for hovering grows even exponentially with weight, as indicated in Fig. 1.
Further hints into this direction are the findings of Antezana et al. (1982), who reports an even overproportional food-intake with increasing animal-size (see Fig. 1) and the well developed gills of adult E. superba (Alberti and Kils 1983).
All these extreme findings make it feasible to expect in krill highly effective functional principles and "welldesigned" morphological solutions.
This paper describes recent experiments of the Antarctic expedition of FS "Meteor" during austral summer 1980/81 around the Antarctic Peninsula, and it presents preliminary results, some regarding swimming, and some feeding.
Morphological investigations were performed by an especially developed macrophotography system with high speed flashes and lenses with high resolution. Some details were evaluated using a scanning electron microcope (SEM).
For in situ observations we used underwater TV-, film- and photo equipment, combined with a low-light-level tube. This equipment was not lowered directly from the ship but from a non-wave-following buoy.
The forces acting on the krill were measured in a flume (with dead animals), and the forces produced by the krill were measured by gluing a tiny steel-rood onto the carapace (of a living animal), or by indirect methods measuring the water acceleration and flow field produced by a swimming krill.
The reaction-time was estimated by synchronizing a flash to one frame of an already running high-speed film registration. The animals reacted with a beat of their telson, and the frames passed between the flash and this event were counted.
As in my opinion the krill is a constantly swimming organism, traveling at a relatively high speed through most of its lifetime, all dynamics should be investigated preferably in such a state. For this purpose a flume was constructed, in which the swimming krill held position relatively to the outside, thus making it accessible to the measurements. This micro flume (Fig. 2) was built of two perspex tube-sections, a bigger outer one (a), and a smaller inner one (i). At the top and bottom are sealed plates (p) with sealed lids (1) and sealed probe openings (o) . Water is only between the two tube-sections, forming a "water-ring". Into this ring the water is injected tangentially at the inner side through a slit (in), ranging from the top to the bottom. The water-outlet (out) is a net-screened window (5 * 25 cm), again at the inner side.
Fig. 2 Schematic top-view of the micro flume
As a result there is a water current running in the ring, which could be adjusted to the desired range by controlling the inflow. A very important detail is, that the inner and outer tube sections are arranged excentrically. Therefore the cross-section area of the water-ring changes from narrow at the one side (hs) to wide at the opposite side (1s). As the water mass-transport is constant at all sections of the tank, the current has to adopt, and so we find a high water-velocity at the narrow side, and a lower one at the wide side, and ? what is even more important ? there is a gradual increase in velocity in the areas between these two points (ms).
Engaging this method we were able to offer the krill a variety of water-velocities, so it could freely select an area with a condition suiting its demands. For an example let us consider a krill with a swimming-speed of 10 cm per second; at position (1s) we adjusted the current to 5 cm per second and at (hs) to 15 cm per second If this krill is near the area (1s), it swims relatively too fast and will change position ahead (10 - 5 = + 5) into an area, where it meets a higher current. If it starts out near the area (hs), its swimming-speed is relatively too low (10 - 15 = -5), it will fall back into an area, where water-current is lower. After a few seconds there is a balance between swimming-speed and water-current, and the krill stands still relative to the outside. For the investigator this state has several advantages:
For the animal this system has advantages too:
As they are mainly orientated in parallel to the walls, they nearly never hit against parts of the equipment. Such crashes must be quite a stress considering the normal way of living of this animal and its delicate antenna-system.
In an older publication (Kils 1982b) I postulated, that krill is capable ? from the energetic and morphological point of view ? to travel at a considerably high speed for extended periods of time; therefore we continued experiments into this field. The evaluation of the material has not been completed jet, but what can be said already is, that most of our healthy krill in the above described experimental setup kept up swimming-speeds in the range of 1.5 to 3. 5 bodylengths per second for more than a week ? through day and night. There was a tendency, that smaller krill travelled with a slightly higher relative speed than the bigger did. There are several other observations of good swimming-capability: Hamner (in press) reported similar speeds from diving observations, Marr (1962) observed a krill swarm swimming against a current at 18 cm s-1 for several hours, and in the experiments of Torres and Childress (1983), E. pacifica travelled quite often at speeds between 3 and 6 bodylengths per second.
Many of the dynamics in krill ask for a sensor to detect velocity and direction of the water flowing over the animal (filter-feeding activities or utilizing the hydrodynamic lift as described in Kils 1982b). This sensing seems to be done by the krill engaging its second pair of antenna. Morphologically this antenna is well suited for such a task: Its cross-section looks like a flat hydrofoil (Fig. 3). If the water flows towards such an "antenna-wing" at no angle of attack the antenna will keep its position; if there is an angle of attack, this will result in a deflection to the one or the other side (indicated by the arrows in Fig. 3 and 4).
Fig. 3 Cross section of the second antenna and the hydrodynamic force at different angles of attack (image can be animated in simple-flip-movie-mode by cklicking the arrows or + and - to change the flowfield)
Fig. 4 and 5 Front and side view of krill showing the position and posture of the second antenna and its deflection by the flow
A change in the water-velocity will cause a change in the hydrodynamic drag and as a result the antennas bend more backwards at higher velocities (demonstrated by the big arrows in Fig. 5, which gives a lateral view of the animal), sensing by this method the speed of swimming.
These findings are the preliminary results from the flow- channel experiments, which have been proven by the underwater TV- and film-observations to some extent. Whether we can find confirmation for this theory at the neurophysiological level must be shown by further experiments.
The measurements of the propulsion force produced by the krill showed a result, which, at first sight, seemed not to fit into the general picture. The metacronally beat-succession of the pleopods would theoretically allow the animal to produce a constant propulsion-force (for details see Kils 1982b). However, the measured forces showed quite an oscillation: The strongest propulsion is produced during the beat of the second pleopod, whereas the fifths pleopod produced only half that force. For a good travelling performance alone this would not be very beneficial, as this will result in a repeated acceleration and deceleration - from the energetic point of view a dissipation (this will be discussed later on).
The reaction time of krill to optical stimulus is, considering the environmental temperature, very short: 35 to 55 ms. This again is an indication of the extreme metabolism.
A very interesting morphological detail we found at the pleopods. For details of the function and dynamic of the pleopods see Kils (1982b). During their backstroke, the exo- and endopodites are spread out to form a deeply ventral and lateral reaching surface, like a paddle. Left and right pleopod do this synchronized. However, if the spreading to the sides would be too far, a gap in the middle could be the result. To limit this excessive spreading krill developed a connection between the endopodites of both sides, as drawn schematically in Fig. 6 (during the backward stroke): Two bars reach from the inner side of the endopodites to the middleline and form a joint there. This connection does not hamper the folding of the pleopods during the forward stroke, but limits the angle during the backward stroke. Possibly this device aids in pulling the endo- and exopodites apart to form a bigger paddle-area. This detail, though small, is another indication of the very efficiency of the swimming apparatus in krill.
Morphologically there are some more interesting points regarding this connection: It is a secondary reconnection of two extremities; analog would be a joint between the thombs of our left and right hand. In preserved animals this link is lost, and we are now investigating, how the holding mechanism in alive animals is performed. At what lifestage is this joint developed? What happens during moulting, as this structure it forms a ring?
The findings of workers in the field of feeding are quite contradictory: Some find a preference for small food particles and propose a selection, others find the antithesis (Antezana pers. comm., Barkley 1940, Meyer and El-Sayed 1983, Kawamura unpublished manuscript, Morris unpublished manuscript, Nemoto pers. comm.).
Our results of the evaluation of the high-speed macro-photo- registrations (following the path of particles and analyzing the dynamics) led us not to join a special party. However, there is not such one feeding-method but a variety of highly effective skills developed by this animal, so probably all authors are right. Depending on several conditions such as food density, food spectrum, size of krill and probably the energetic state of the animal, different methods are engaged to get the needed energy. In this short presentation it is not possible to describe each method and its morphological and dynamic details, but I want to sum up the basic principles to demonstrate the variability in krill feeding.
The morphology of the filtering basket has been described in detail by Barkley (1940) with recent detail investigation by Alberti and Kils (1980), McClatchie and Boyd (1983), Boyd et al. (in press), Kils (1982a).
The dominant part of the net area is formed by the filtersetae of the thoracopods and has a basic construction as shown in Fig. 7: The 1st-degree-setae (pointing from one thoracopod to the anterior one) carry two rows of 2nd-degree-setae, forming V-shaped micro-nets. The gaps of these nets are half-crossed by 3rd-degree-setae (Fig. 8), resulting in a meshsize smaller than 1 micrometer. To get an idea of the total net area: One would have to glue 7500 times the Fig. 7 together to display the whole net!
Fig.7 and 8 (top) Stereo Electron Microscope images of flter setae (magnification 530 and 6170)
Fig.9 and 10 Stereo Electron Microscope images of comb setae (magnification 130 and 1255)
Another important structure is formed by the "comb-setae", shown in Fig. 9. Over ca. 85% of their length they show a rather similar construction as the filter-setae do, but at the very end they carry a comblike device (Fig. 10). These two basic types of setae form two different kinds of nets: One very fine net with a relatively large net area formed by the filter-setae (meshsize 1-4 micrometer), and one coarse net with a comparatively small net area formed by the basal parts of the comb-setae (meshsize 25-40 um).
One feeding-behavior we often observed is a "pump-filtering" or "compression-filtering": The filtering basket is periodically opened and closed by swinging the thoracopods to the side and front, then pulling them towards the middle-line, and finally drawing them back towards the body. Such a pumping-behaviour is also reported by Antezana et al. (1982), Boyd et al. (in press), Hamner (in press), McWhinnie (pers. comm.) . This pumping is generally synchronized with the beat of the pleopods (frequency between 1.5 and 3 beats per second). The details happening during the course of one beat are manifold and will only be scratched here: Fig. 11 shows a schematic 3-dimensional view of the left side of the filtering basket during the opening phase.
Fig 11 Schematic 3 dimensional view of the left side of the filtering basket, looking from the outside obliquely down/forward (only two net areas are drawn in detail) - (click into image to go to higher resolution and real images - the interior of the filtering basket is on the left side, as indicated by the arrow)
At this stage the thoracopods travel forward-outwards. The arrangement of the comb-setae is at right angle to the direction of travel. As a consequence the water passes the 30 micrometer comb-setae-nets from the outside to the inside, and particles bigger than 30 micrometer are rejected, At the same time the very ends of the comb-setae travel through the meshes of the filter-setae-net, collecting particles caught there at the stroke before and moving them a little bit towards the mouth.
In the second phase (during the movement of the basket towards the middle-line) the filter-setae swing their free end onto the inside of the anterior thoracopod, acting like a valve and closing the net of the comb-setae. Inside the filtering-basket is now water containing all the particles smaller than 30 micrometer. During the movement of the thoracopods to the inside, this water body is compressed. Part of this water body is pressed out through the posterior comb-setae-nets of the last thoracopods, and the other portion is forced through the fine net of the filter-setae. The relation of those two water portions has not been estimated jet. As the tiny structures of the filter-setae bring about quite low Reynolds numbers, such a net has an extreme high drag, and it depends strongly on the pressure inside the filtering-basket, how much water really passes the fine filter-setae-net. For details of fluid-mechanics see Joergensen (1983). As net structures and food particles range in the size of micrometers, it is not unrealistic to expect, that electrostatic forces might play an important part in collecting particles at the filter surfaces.
To give an impression of the two nets in comparison with the ambient food in Fig. 12 the meshes of the nets are drawn schematically at the same scale as the food is.
Fig. 12 Size relationship between ambient food organisms and the filtering nets in Euphausia superba - Comb setae net: top horzontal bars - filter setae net: left lower corner grid
Anyhow, by using this method krill can reject selectively particles bigger than 30 micrometer on one side, and the 1 micrometer meshes are small enough on the other side, to catch even bacteria.
The above described pump-filtering could be observed when there was abundant food in the water, and in my opinion this is the filtering-method krill performs most of the time. The evaluation of the underwater-films confirmed this belief and so does the following: As the moment of the opening of the filter-basket is correlated with the highest drag (Kils 1982b), now it makes sense to have the propulsion force of the pleopods oscillating, and indeed the moment of highest thrust of the pleopods coincides with the moment of the opening of the basket. So if we take both, the swimming-dynamics and the filtering-dynamics into account, we now find a balanced and smooth force-budget a prerequisite for economic and continuous performance.
Another type of pump-filtering with a slightly different succession of thoracopod movements is similar to a "back-flushing" of the filter-setae-net: During the movement to the outside the water enters the basket across the 1st-degree- setae, while the V-shaped nets formed by the 2nd-degree-setae swing back like an opening valve. This behaviour was often correlated with very thick plankton conditions and a considerable decrease in swimming speed of krill.
At lower food availability we observed krill traveling with a steadily opened basket for quite a distance, followed by a couple of pump movements. Fig. 13 shows an animal at such a state.
Fig .13 Three dimensional view of the filtering basket from front-below, showing the posture during the phase of open tow. Right thoracopods drawn solid, left dotted. Second antenna and exopodites of the thoracopods omitted.
This figure has been drawn directly from a photographic registration, so that the proportions are realistic. This feeding-method will not reject particles bigger than 30 micrometer. It might be a "get all there is" method. Here the portion of water passing the filter-setae-net is certainly very small, as during the phase of open tow there is only a small pressure increase inside. At this low pressure condition the function of the filter-setae-net shifts: It predominately acts as a deflector to channel the water across the comb-setae-nets, but a though small portion of the water might still pass the fine net.
If there is nearly no food in the water, the basket is folded close to the body; this of course makes sense, as the open filtering-basket produces quite a drag (see Kils 1982b), consuming a lot of energy.
Summarizing the filtration process it can be said, that krill developed extreme skills, covering diverse methods, thus utilizing a wide variety of available plankton conditions. All principles for good net construction are fulfilled:
The smaller the mesh size:
Besides suspension-feeding krill has a variety of additional opportunities: Cannibalism has often been reported, and we filmed (in aquarium) several animals incorporating a colleague totally within few hours, preferably "white" krill or krill stuck in the old exuvie during moulting. Maybe other big zooplankton can be caught too, as the krill exhibits quite a skill in using the two sides of the basket to grab bigger objects.
The dactylopodites of the thoracopods carry rake-like structures, which show quite a different morphology than the normal setae: They are much stronger and increase in diameter from the tip to the base. Their static points to the fact, that they are well suited for grazing diatoms from the ice or other surfaces, a behaviour that has often been reported. Fig. 14 gives a summary of the different feeding methods.
The fascinating morphological structures of the filtering- basket, the unique functional principles of the filtration dynamics, the extreme performances in metabolism, the well developed gills, the high feeding rates and the susceptibility of the adults to unfavourable conditions are for me all indications, that the adult krill is spending its life at a kind of physiological frontier: Energy-expenditure is enormous on one side and highly efficient and adaptive energy-supply-systems are needed on the other. This is a living at high risk, as even a relatively small trouble or change in the environment can lead to a catastrophe. But krill surely is the most abundant animal of its size range in the world, so this species really can effort to take such a risk, and the profit is high: Living in a very attractive oceanic community.
If we consider krill to be a more or less constantly traveling organism, this has quite some implications to our general understanding of its biology. As the crusing speed is size-dependent, a mixed swarm will dissociate after some time in one swarm ahead consisting of big animals and one swarm behind consisting of small animals. Also, the synchronization in moulting, as reported by Buchholz (this volume), could be caused by the unability of freshly moulted individuals to keep up with a moving swarm, thus falling behind and then gathering again with the other "moulters" to form a new swarm. All individuals of this new swarm will then moult together the next time.
The investigations of Denys (pers. comm.) showed structures in the eyes of krill suspected of sensing polarized light. This might enable a swarm to navigate, to migrate over great distances into one direction. A report of Guzman (this volume), citing Kanda et al. (1982), supports such an idea: They followed two swarms over 46 and 116 miles traveling at a speed of 15 cm per second in a south general trend.
If we make the two assumptions, that we have animals migrating at constant direction and at a constant speed and take into account the hydrography of the Southern Ocean with its turbulences, we might develop a theory for the unevenly large-scale distribution of krill: In Fig. 15 the parallel arrows represent the swimming-vectors of evenly distributed swarms. The circular arrow-system represents the superimposing current-vectors of an eddy.
Fig 15 Hypothetic concentration effect of a superimposed eddy: Much higher concentrations of schools are in the shaded area
As the cruising speed of a krill swarm can be about 15 cm per second and the rotation-velocity of an eddy can be about 20-40 cm per second, the parameters are not too far from reality. And the assumption of having krill-swarms traveling for some period into one direction might not be too odd, as bird-swarms and insect-swarms do so. The additions of the two vector- systems are drawn as the solid arrows. In such a system after a while there will be much more krill swarms in the shaded area, and many of the new entering swarms will be caught in the eddy.
I am quite aware of the fact, that this model is very hypothetical, but I think it might be worth while to investigate into such directions to find out more about this fascinating animal and ocean.
Alberti G, Kils U (1980) The filtering-basket of Euphausia superba. ICES CN 1980/L:54
Alberti G, Kils U (1983) Light- and electron microscopical studies on the anatomy and function of the gills of krill (Euphausiacea, Crustacea). Polar Biol 1:233-242
Antezana T, Ray K, Melo C (1982) Trophic behaviour of Euphausia superba Dana in laboratory conditions. Polar Biol 1:77-82
Barkley E (1940) Nahrung und Filterapparat des Walkrebschens Euphausia superba Dana. Z f Fischerei und deren Hilfswissenschaften, Beiheft 1, Walforschung 1:65-156
Bertalanffy L von (1951) Theoretische Biologie Zweiter Band: Stoffwechsel, Wachstum. A. Francke AG Verlag, Bern, 418pp
Boyd C, Heyraud M, Boyd C (in press) Feeding of the Antarctic krill, Euphausia superba. J Crust Biol
Buchholz F (1983) Moulting and moult physiology in krill. (This volume)
Champalbert G, Gaudy R (1972) Etude de la respiration chez des copepodes de niveaux bathymetriques varies dans la region sud marocaine at cannarienne. Mar Biol 12:159-169
Chekunova VI, Rynkova TI (1974) Energy requirements of the Antarctic crustacean Euphausia superba Dana. Oceanology 14:434-440
Conover R ( 1960) The feeding behavior and respiration of some marine planktonic Crustacea. Biol Bull Nar Biol Lab, Woods Hole 119: 399-415
Guzman 0 (1983) Distribution and abundance of the Antarctic krill (Euphausia superba) in the Bransfield Strait. (This volume)
Hamner W (in press) Aspects of schooling in Euphausia superba. J Crust Biol
Hirche H (1983) Excretion and respiration of the Antarctic krill Euphausia superba. Polar Biol 1:205-209
Ivlev V (1963) Energy consumption during the motion of shrimps. Zool, Zh 1465/1471 (Engl. summary)
Joergensen C (1983) Fluid mechanical aspects of suspension feeding. Mar Ecol Prog Ser l1:89-103
Kanda K, Takagi K, Seki Y (1982) Movement of the larger swarms of Antarctic krill (Euphausia superba) population off Enderby Land during 1976-77 season. J Tokyo Univ Fish 68:25-42
Kils U (1982a) The unique position of krill in the Antarctic system. Joint Oceanographic Assembly (Abstracts). SCOR Department of Fisheries and Oceans, Scientific Information and Publication Branch, Ottawa, Ontario, Canada
Kils U (1982b) Swimming behaviour, swimming performance and energy balance of Antartic krill, Euphausia superba. BIONASS Scientific Series 3:1-121
Lasker R (1966) Feeding, qrowth, respiration and carbon utilization of a euphausiid crustacean. J Fish Res Bd Can 23: 1291-1317
Marr JWS (1962) The natural history and geography of the Antarctic krill (Euphausia superba Dana). Discovery Rep 32:33-464
McClatchie S, Boyd C (1983) A morphological study of sieve efficiencies and mandibular surfaces in the Antarctic krill, Euphausia superba. Can J Fish Aquat Sci 40:955-967
Meyer MA, El-Sayed SZ (1983) Grazing of Euphausia superba Dana on natural phytoplankton populations. Polar Biol 1:193-197
Pauly D (1979) Gill size and temperature as governing factors in fish growth: A generalisation of von Bertalanffy's growth formula. Berichte aus dem Institut fuer Meereskunde 63, Kiel
Rakusa-Suszczewski S, Opalinski W (1978) Oxygen consumption in Euphausia superba. Pol Arch Hydrobiol 3:633-641
Segawa S, Kato M, Murano M ( 1979) Oxygen consumption of the Antarctic krill. Trans Tokyo Univ Fish 3:113-119
Segawa S, Kato M, Murano M (1982) Respiration and ammonia excretion rates of the Antarctic krill, Euphausia superba Dana. Trans Tokyo Univ Fish 5:177-187
Small F, Hebard J (1967) Respiration of a vertically migrating marine crustacean Euphausia pacifica Hansen. Limnol Oceanogr 2:272
Torres J, Childress J ( 1983) Relationship of oxygen consumption to swimming speed in Euphausia pacifica. 1. Effects of temperature and pressure. Mar Biol 74:79-86
Voss J (1982) Sauerstoff-Verbrauch des Antarktischen Krill, Euphausia superba. Diplomarbeit, Universitat Kiel 1982
Winberg G ( 196 1) New information on metabolic rate in fishes. J Fish Res Bd Can Trans 362
antarctic expedition german krill biology behavior behaviour artikel krill.rutgers.edu scienceplus natureplus kinder university article schwimmen swimming respiration sauerstoff verbrauch energetic energy balance sinking sinken absinken nahrung neu neue nahrungsquellen antarktis expedition deutschland kiel uwe kils peter marschall arthur baker david tranter verstaendliche wissenschaft k12 k 12 K12 distance teaching remote education distance education kindergarten kinder garten schule grundschule grund schule gymnasium realschule real schule lehrer teacher material bild bilder mikroskop nahrungsnetze nahrungs netze oekosysteme microscope ecoscope unterwasser optik unterwasser kameras ferngesteuerte fahrzeuge rov rovs remotely controlled vehicles mikroskopieren cybermic virtual microscope virtuelles mikroskop scienceplus natureplus telepresence videoconference video conference