Journal of Sea Research 44 (2000) 169±193 www.elsevier.nl/locate/seares Review Comparative ecophysiology of active zoobenthic ®lter feeding, essence of current knowledge

H.U. RiisgaÊrda,*, P.S. Larsenb

aResearch Centre for Aquatic Biology (Odense University), Hindsholmvej 11, DK-5300 Kerteminde, Denmark bFluid Mechanical Section of Department of Energy Engineering, Technical University of Denmark, Building 403 DK-2800 KGS. Lyngby, Denmark Received 17 March 2000; accepted 28 August 2000

Abstract The present contribution gives an overview of current knowledge of a comprehensive and steadily growing research ®eld. The ®rst section deals with water pumping and particle retention mechanisms in ciliary and muscular ®lter feeders. The second section examines the biological ®lter pumps in order to assess adaptation to the environment. Filter-feeding benthic inverte- brates have evolved ®lter pumps to solve common basic problems. This has led to a large degree of similarity between otherwise distant standing species, which makes comparative studies interesting and important. The present review of zoobenthic ®lter feeding aims at accentuating such recognition. q 2000 Elsevier Science B.V. All rights reserved.

Keywords: water pumping; particle retention mechanisms; ciliary and muscular ®lter feeders; bioenergetics; minimal scaling; adaptation to the environment

1. Introduction areas are typically between 1 and 10 m3 water per square meter area per day (RiisgaÊrd, 1991a, 1994; Active ®lter-feeding benthic invertebrates are Petersen and RiisgaÊrd, 1992; RiisgaÊrd et al., 1995, important elements in coastal ecosystems where 1996c,e, 1998; Lemmens et al., 1996; Vedel, 1998), they sustain life by removing suspended food particles but higher rates may be found for dense mussel beds from the water (Jùrgensen, 1966, 1975, 1990; Davies that can ®lter more than 100 m3 m22 d21 (Jùrgensen, et al., 1989; Mùhlenberg, 1995; Mann and Lazier, 1980, 1990). 1996; Wildish and Kristmanson, 1997). Quanti®cation of rates of energy transfer between Biological and physical processes are particularly the trophic components in a marine ecosystem is closely coupled in coastal waters. For example, water dif®cult and complex, and the role of zoobenthic depth, which affects the importance of wind and tidal ®lter feeding in the exchange processes is only just mixing, strongly in¯uences the grazing impact of coming to light. However, it is clear that energy benthic ®lter-feeder populations, which may be transfer through ®lter-feeding invertebrates is often considerable in shallow waters (Of®cer et al., 1982; an important route from the pelagic to the benthic Cloern, 1982). Population ®ltration rates in these components of coastal ecosystems (Baird and Ulano- wicz, 1989; RiisgaÊrd et al., 1996b; RiisgaÊrd, 1998a; Arntz et al., 1999). Recent work has produced * Corresponding author. evidence of the role of dense benthic ®lter feeding E-mail address: [email protected] (H.U. RiisgaÊrd). populations in regulating plankton production in

1385-1101/00/$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S1385-1101(00)00054-X 170 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193

Fig. 1. Types of capture mechanisms in active suspension feeding marine invertebrates. (a) Collar-sieving in sponges; two choanocytes with parallel ¯agella act as a peristaltic pump; Fl ¯agellum; Cf, collar of microvilli acting as a ®lter. (b) Cirri trapping in mussels; three gill ®laments are shown; F, frontal cilia; LFC, laterofrontal cirri; L, lateral cilia. (c) Ciliary sieving in bryozoans; upper: cross section of tentacle; LFC, laterofrontal cilia acting as a sieve above the water pumping lateral cilia; lower: ®lter consisting of stiff laterofrontal cilia has stopped a particle which in¯uences the streamlines for creeping ¯ow through the ®lter. (d) Cross section of tentacle from a ciliary downstream-collecting animal showing how the compound lateral cilia, CLC catch up a particle and transfer it to the frontal side of the tentacle. (e) Upper: rectangular mucus net of an ascidian; lower: lateral cilia, L in the pharynx slits pump water through the net; frontal cilia, F transport the mucus net with retained food particles. (f) Upstream retention in a brachiopod; upper: cross section of ®lament, indicating path of particle from through current into surface current; lower: short segment of lophophore with ®laments. Figs. modi®ed from: (a) Larsen and RiisgaÊrd (1994); (b) Silvester and Sleigh (1984); (c) Nielsen and RiisgaÊrd (1998); (d) RiisgaÊrd et al. (2000); (e) RiisgaÊrd and Larsen (1995); (f) Jùrgensen et al. (1984). littoral systems, including control of phytoplankton of pelagic production is still poorly understood blooms and other aspects of relevance for eutrophi- (Lemmens et al., 1996; Gili and Coma, 1998). cation problems (Nixon, 1995; RiisgaÊrd et al., 1995). The ®lter feeders which have adapted to a life in a But in habitats with a rich and diversi®ed ®lter- nutritionally dilute environment show great variations feeder fauna consisting of sponges, cnidarians, in the design of feeding mechanisms (Rubenstein and bryozoans etc. the role of these species as consumers Koehl, 1977; Chapman 1968; Foster-Smith, 1976a,b, H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 171 1978; Brown, 1977; LaBarbera, 1984, 1990; Jùrgen- Mùhlenberg, 1981; Mùhlenberg and Kiùrboe, 1981; sen, 1990; Shimeta and Jumars, 1991; Vogel, 1994; Jùrgensen, 1990, 1993, 1996; Ward et al., 1993, RiisgaÊrd and Larsen, 1995; Nielsen and RiisgaÊrd 1998a,b). Depending on context, different expressions 1998; GruÈnbaum et al., 1998). During the last decade for water processing by the ®lter-feeding animals are investigations of ®lter-pumps and their properties used: pumping rate ˆ ®ltration rate (volume ¯ow have been intensi®ed, including sponges, polychaetes, through the animal's ®lter per unit time); clearance bivalves, ascidians, bryozoans and other groups of rate ˆ volume of water cleared of suspended particles small ciliary ®lter feeders. The studies have aimed per unit of time. For particles that are retained with at giving a more precise description of water pumping 100% ef®ciency by the animal's ®lter, clearance rate systems and particle capture mechanisms, but also at is equivalent to ®ltration rate. characterising the pumps quantitatively, and assessing energy costs of water pumping as reviewed by Riis- 2.1. Types of particle capture mechanisms gaÊrd and Larsen (1995). The present contribution gives an overview of the Filter feeders (here used synonymous with suspen- essence of current knowledge of a comprehensive and sion feeders) extract their nourishment from suspended steadily growing research ®eld. The paper consists of micro-organisms and detritus in the surrounding water. two parts. Part 1 deals with water pumping and parti- The term particle capture mechanism is here used to cle retention mechanisms in ciliary and muscular ®lter describe the process of extraction of suspended particles feeders. Part 2 examines the biological ®lter-pumps in from water which is actively pumped through some kind order to assess the adaptation to the environment. The of ®lter device which separates the particles from the ®lter-pumps have evolved to solve common basic water so that they can be consumed by the animal. problems which has also led to a large number of Passive ®lter feeders, which utilise the natural ¯ow to resemblances between otherwise distant standing bring particles in contact with feeding structures species. This makes comparative studies interesting (Fauchald and Jumars, 1979; Shimeta and Jumars, and important, and the present review of ®lter feeding 1991; Sponaugle, 1991; Loo et al., 1996; Shimeta and in benthic marine invertebrates aims at accentuating Koehl, 1997), are not treated in the present review. such recognition. Switching between deposit and suspension feeding in coastal zoobenthos has recently been reviewed by Riis- gaÊrd and Kamermans (2000). 2. Part 1: Water pumping and particle retention Six types of food particle capture mechanisms have so far been recognised in marine benthic ®lter feeders: In active suspension feeders, water is pumped (1) collar-sieving in sponges, where a beating ¯agel- through a ®lter device which separates the food parti- lum creates the feeding current through a collar-®lter; cles from the water so that they can be consumed by (2) cirri trapping in bivalves with gill ®laments the animal. Extraction of suspended particles from the possessing laterofrontal cirri that beat against the water is based on different capture mechanisms. A through current; (3) ciliary sieving in bryozoans, review of methods for capturing particles in benthic where particles are retained by a ®lter formed by a animals feeding on suspended particles has previously band of stiff cilia upstream of the water pumping be given by Wotton (1994) and RiisgaÊrd and Larsen lateral cilia; (4) ciliary downstream collecting in a (1995). The present contribution deals with the six diverse group of ®lter-feeding invertebrates, where types of mechanisms found in the active ®lter feeding food particles are caught up by compound cilia, marine invertebrates depicted in Fig. 1. The paper which also create the current, and then transferred to does not attempt to cover all aspects of ®lter feeding, the downstream side for collection; (5) mucus net such as post-capture transport and sorting of particles ®lter feeding in some species of polychaete worms, although such aspects may at times be of crucial ascidians, gastropods and lancelets, where ciliary or importance, e.g. for certain polychaetes (Nicol, muscular pumps drive water through a mucus net 1931), ascidians (Jùrgensen, 1966), and bivalves ®lter which retains the suspended food particles; and (6) feeding at high seston concentrations (Kiùrboe and ciliary upstream collection in, e.g. brachiopods, 172 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 where particles are retained on the upstream side by a The sponge pump was analysed by Larsen and water transporting ciliary band. The different types are RiisgaÊrd (1994) and compared with the choano- presented in the following sections, and general prin- ¯agellate pump in order to identify prerequisite ciples are at times exempli®ed by way of speci®c properties of the basic pump units which have species. enabled the development of large sponges. The pumpanalysiswasbasedonexperimentally 2.2. Collar sieving measured back-pressure pumping-rate characteristics of the demosponge Haliclona urceolus and on math- Collar sieving is restricted to sponges, which are ematical-hydraulic modelling. A curved characteristic simple, multicellular animals whose entire body is was found for H. urceolus, and the maximal pressure specialised for suspension feeding. The basic element rise (at zero ¯ow) which could be delivered by the for pumping and ®ltering the surrounding water is the sponge was about 2.4 mm H2O. To unveil the pump- choanocyte, a specialised cell with a ¯agellum that ing principle and to propose a pump model for pumps water through a collar of microvilli acting as sponges Larsen and RiisgaÊrd (1994) considered the a ®lter (Fig. 1a). The structure of the choanocytes is free-living choano¯agellate Monosiga as representa- the same in all sponges, and among the metazoa, tive of a sponge-choanocyte because the two cell sponges are unique in feeding by means of choano- types are structurally and functionally identical: a cytes. Choanocytes are structurally and functionally ¯agellum pumps water through a collar of microvilli identical to the choano¯agellates (Fjerdingstad, acting as a ®lter. Knowing the ¯agellum length, beat 1961a,b; Laval, 1971; Hibberd, 1975; Bergquist, frequency, wavelength and amplitude the pump head 1978) that ®lter free-living bacteria from the water. of the choano¯agellate was estimated to be 0.076 mm The striking similarity of choano¯agellates to the H2O. Because this pump head is insuf®cient to handle choanocytes of sponges has given rise to speculations the pressure drop in a sponge it was suggested that the about the evolutionary relationship between the two closely spaced ¯agella in the choanocyte chambers of groups (Barnes, 1987). It has been theorised (Lead- sponges might all together act as a peristaltic pump beater, 1983) that sponges have evolved from colonies being able to create the necessary pump pressure to of choano¯agellates and that the small and simple overcome the resistance in the extensive canal system. ascon-type sponge has given rise to the more It was further argued by Larsen and RiisgaÊrd (1994) advanced sycon-type from which the large and more that the basic pump units in a `standard' demosponge successful complex leucon-type sponge evolved. are the choanocyte chambers, constituting 30±50% of

Table 1 Water ¯ow velocity immediately before impact onto mucus-net in various ®lter feeders. M is the muscular pump; and C the ciliary pump

Species Flow velocity (mm s21) Pump type Reference

Polychaetes Chaetopterus variopedatus 1.5 M RiisgaÊrd (1989) Nereis diversicolor 0.2±0.27 M RiisgaÊrd et al. (1992) Gastropods Crepidula fornicata 0.8 C Jùrgensen et al. 1984) Ascidians Ascidia mentula 0.37 C Jùrgensen et al. (1984) Styela clava 0.3 C RiisgaÊrd (1988a) Appendicularians Oikopleura vanhoeffeni 0.56 M AcunÄa et al. (1996) Thaliaceans Pegea confoederata 1.7 M Bone et al. (1991) Salpa maxima 6 M Madin (1990) Cephalochordates Branchiostoma lanceolatum 0.09±0.34 C RiisgaÊrd and Svane (1999) H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 173 the wall structure separating inhalant and exhalant 1996d; Smaal and Twisk, 1997; Charles et al., 1999; canals, and further, that all pump units operate in Cranford and Hill, 1999). Especially the particle parallel and at essentially the same ¯ow and pressure capture mechanism in mussels has been debated; but rise. considerable progress and also a high degree of agree- One crucial difference between sponges and ment in understanding this mechanism seem to have choano¯agellates is the ability of the former to feed been achieved in recent years (RiisgaÊrd et al., 1996d; on phytoplankton which is retained and digested Beninger, 2000; Silverman et al., 2000; RiisgaÊrd and (phagocytosis) in the extensive inhalant channel Larsen, 2000). system (Kilian, 1952; Simpson, 1984) before the In the blue mussel, Mytilus edulis, bands of lateral water is ®nally ®ltered through the 0.1 mm collar cilia produce the main water transport through the inter- slits of the choanocytes. Few workers have focused ®lamentary canals of the gill (Fig. 1b). Near the entrance on the differences in the physiological performances to the canals, particles are separated from the main that may exist between sponges and other ®lter-feed- currents and transferred onto the frontal surface by the ing invertebrates. Ecophysiological investigations of action of the laterofrontal cirri that have a ®xed pattern sponges are few and many questions remain largely of alternating beat. Interest in understanding ®lter feed- unanswered, although RiisgaÊrd et al. (1993) and ing by bivalves has led to video recording of water Thomassen and RiisgaÊrd (1995) carried out ecophy- motion inferred from particle motion when particle siological studies in order to uncover the nature of motion is not restrained by interaction with objects sponges to enable a comparison with other, more (cilia) or by very small (1 mm) particles. The issue in advanced ®lter-feeding invertebrates. such studies has been to gain optical access while main- The growth, ®ltration and respiration of Halichon- taining undisturbed, natural feeding. Used techniques dria panicea were measured by Thomassen and Riis- include isolated gill-®lament preparations observed gaÊrd (1995). The maximum speci®c growth rate was through traditional light-microscopes with lenses 4% d21, and it was stressed that sponges have a low immersed in water (Nielsen et al., 1993; RiisgaÊrd et water pumping capacity, expressed as litres of water al., 1996d), high powered confocal laser scanning ®ltered per cubiccentimeter oxygen consumed (see microscopy (Silverman et al., 1996a,b, 2000); observa- Table 3, and Section 3.3), compared with other tions of intact bivalves by means of endoscopy (Benin- ®lter-feeding invertebrates, but compensate for this ger et al., 1992; Ward et al., 1993; Ward, 1996), or by by a high retention ef®ciency for even very small using a long working distance dissecting microscope particles, down to 0.1 mm. It was also found that the (RiisgaÊrd and Larsen, 2000). energy cost of growth was equivalent to 139% of the The role of laterofrontal cirri in particle capture by actual growth which is comparable with microorgan- M. edulis was studied by RiisgaÊrd et al. (1996d) using isms, but considerably higher than found for `true' gill ®laments mechanically adjusted by `stretching' to invertebrates (see Section 3.4). restore the normal inter®lament gap of about 40 mm of the intact mussel. When stimulated with serotonin 2.3. Cirri trapping both the beat frequency of the lateral cilia and the laterofrontal cirri were comparable with values Bivalves are the best studied ®lter feeders. The measured in intact, small, transparent mussels (Dral, mechanisms of water pumping and particle retention 1967; Jùrgensen and Ockelmann, 1991). The video have been reviewed and re-evaluated repeatedly over observations showed that suspended algal cells the years (Dral, 1967; Bayne et al., 1976a,b; Owen carried with the through current were stopped for a and McCrae, 1976; Winter, 1976; Mùhlenberg and while at the entrance to the inter®lament gap. Then the RiisgaÊrd, 1978, 1979; RiisgaÊrd and Mùhlenberg, path was reversed 1808, and the particles were trans- 1979; Jùrgensen, 1983, 1989, 1990, 1996; MeyhoÈfer, ferred to the frontal side of the gill ®lament to be 1985; Jùrgensen et al., 1990; Silvester and Sleigh, transported towards the marginal food groove by fron- 1984; RiisgaÊrd 1988a; Ward et al., 1993, 1998a,b; tal cilia. Transfer of particles from the through current Ward, 1996; Nielsen et al., 1993; Silverman et al., to the frontal current is made by the laterofrontal cirri 1996a,b; RiisgaÊrd and Larsen, 1995; RiisgaÊrd et al., or by the water currents they generate as they beat 174 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 against the current through an angle of 908. RiisgaÊrd et ®lter pump (Bullivant, 1968a,b; Gordon et al., 1987; al. (1996d) concluded that particles larger than 4 mm Ryland, 1976; Winston, 1978; Gordon et al., 1987; are stopped and transferred to the frontal side, GruÈnbaum, 1995, 1997; RiisgaÊrd and ManrõÂquez, whereas smaller particles either follow the ¯ow 1997; RiisgaÊrd and Goldson, 1997; Eckman and around the cirri or they are stopped by the cirri's Okamura, 1998; GruÈnbaum et al., 1998; Larsen et branching cilia. Observations of interactions between al., 1998; Nielsen and RiisgaÊrd, 1998; Lisberg and 1 mm latex particles and laterofrontal cirri described Petersen, 2000). by Silverman et al. (1996a,b) extend at higher resolu- The feeding apparatus in the bryozoans consists of tion the observations of RiisgaÊrd et al. (1996d), and a ring of extended ciliated tentacles which form a the explanation is in agreement with experimentally crown (lophophore), with the mouth at the centre of measured particle retention ef®ciency in mussels its base. In all bryozoans (except the cyclostomes, (Mùhlenberg and RiisgaÊrd, 1978; RiisgaÊrd, 1988b; Nielsen and RiisgaÊrd, 1998) three types of ciliary Jùrgensen, 1996). rows may be found on the tentacles: lateral, frontal, Video observations have recently been made on and laterofrontal. The lateral cilia produce the feeding intact M. edulis by RiisgaÊrd and Larsen (2000). currents of the lophophore. When the through currents Using a horizontal dissecting microscope the gills of pass between the tentacles outwards this results in the an actively feeding mussel were seen behind the formation of a relatively strong core current directed gaping valves when the animal was ®xed on its side straight down the lophophore to the mouth (RiisgaÊrd in an observation chamber. Video recordings were and ManrõÂquez, 1997). Particles in the through made through the inhalant opening and the approach currents are retained by the stiff laterofrontal ciliary speed of particles was measured to be about ®lter (Fig. 1c) and either transported downwards on 1mms21, in agreement with the speed predicted by the tentacles by means of the frontal cilia, or the parti- RiisgaÊrd et al. (1996d), and as particles approached cles are transferred to the core current by means of the gill their paths curved in the direction normal to inward tentacle ¯icks triggered by the arrested parti- the gill frontal surface previous to capture which were cles. The mechanism which triggers a tentacle ¯ick is to be expected because of the parallel arranged lateral- unknown, but particles that are not actually stopped cilia pumps. This streamline pattern resembles that by the ®lter may not trigger a ¯ick. Thus, a certain observed in a channel with porous walls subject to extra drag force from the arrested particle, besides the suction, where streamlines curve to become normal `background' force exerted on the laterofrontal cilia to the surface as it is approached. Such paths of parti- due to the through current, may be a prerequisite for cles approaching the gill surface were computed for a triggering a tentacle ¯ick. model problem simulating the space between two The same tentacle ¯ick mechanism was observed demibranchs by RiisgaÊrd and Larsen (2000). by Nielsen and RiisgaÊrd (1998) in the cyclostome The absence of the laterofrontal cirri in certain Crisia eburnea, which does not have frontal cilia. bivalve gills, i.e. the Microciliobranchia excluding Here, feeding apparently relies crucially on tentacle the Ostreidae (Owen and McCrae, 1976) indicates ¯icks and the principle of particle capture is especially that the laterofrontal cirri may be a speciality of easy to observe. Eight tentacles are arranged as a some bivalves. The basic particle capture mechanism funnel in front of the mouth, the lateral band consists in bivalves without laterofrontal cirri remains largely of two rows of very closely spaced cilia, and no fron- unknown, see also Owen and McCrae (1976), Benin- tal cilia are present. The row of 15 mm long, stiff ger and Le Pennec (1988) and Beninger et al. (1992). laterofrontal cilia have a spacing of 3±4 mm. When food particles move toward the funnel they get caught 2.4. Ciliary sieving by the ®lter formed by the laterofrontal cilia which, judging from their ultrastructure, are sensory and the Ciliary sieving in bryozoans (or ectoprocts) seems tentacle makes a ¯icking movement towards the centre to be the ®rst documented example of a mechanical of the funnel. Again, the relatively strong central ciliary ®lter in metazoans. Bryozoans are active ®lter current carries the particles downward to the mouth. feeders working an energy-consuming lophophore The retention ef®ciency has been measured in two H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 175

Table 2

Operating pressure of pump (DH), power output (P), total metabolic rate of standard animal (Rt) and overall pump ef®ciency (P/Rt) in various ®lter-feeders

Taxonomic group and species DH0 (mm H2O) P (mW) Rt (mW) P/Rt (%) References

Sponges Haliclona urceolus 0.673 0.677 80 0.85 RiisgaÊrd et al. (1993) Bryozoans Crisia eburnea 0.065 ± ± ± Nielsen and RiisgaÊrd (1998) Polychaetes Chaetopterus variopedatus 1.43 4.3 107 4.0 RiisgaÊrd (1989) Nereis diversicolor 1.49 2.1 70 3.0 RiisgaÊrd et al. (1992) Sabella penicillus 0.0224 0.451 112 0.403 RiisgaÊrd and Ivarsson (1990) Bivalves Mytilus edulis 1.0 10 900 1.1 Jùrgensen et al. (1986a), Jùrgensen et al. (1988) Ascidians Styela clava 0.3 2.3 891 0.26 RiisgaÊrd (1988a) cheilostomate bryozoans. It was found that particles Polychaetes are among the most frequent and .5 mm in diameter are retained with near 100% ef®- important marine invertebrates in benthic environ- ciency in both Celleporella hyalina (RiisgaÊrd and ments. The available information on feeding modes ManrõÂquez, 1997) and Electra crustulenta (RiisgaÊrd was reviewed by Fauchald and Jumars (1979), who and Goldson, 1997). This shows that the spacing noted that ®lter-feeding has been documented for at between the 0.2 mm diameter laterofrontal cilia may least ®ve families (Chaetopteridae, Oweniidae, Sabel- be about 5 mm in these bryozoans. lariidae, Fabriciinae, Sepulidae-Spirorbidae, Spioni- The main resistance to water ¯ow in bryozoans is dae). Some of the species are partially or wholly the ciliary sieving system. The pressure drop across ®lter feeders (Buhr, 1976; Taghon et al., 1980; Miller the laterofrontal ciliary ®lter is found to be DH ˆ et al., 1992; RiisgaÊrd and Kamermans, 2000). Below

0:065 mm H2O (see Section 3.2., and Table 2). The are given two examples of ciliary downstream collect- operating pressure of the ciliary pump in an isolated ing in two polychaetes together with an entoproct and bryozoan is therefore predicted to be quite low a cycliophore using the same feeding mode. This also although it may be considerable in bryozoans in underlines the use of ciliary downstream collection as encrusting colonies (Larsen and RiisgaÊrd, in prepara- a widespread basic feeding principle, used across tion), likely of the order of magnitude generally found taxonomic boundaries and adequately working in for ®lter-feeders (cf. Table 2). spite of often great variation in functional morphology of the feeding organs. 2.5. Ciliary downstream collecting Downstream collecting has been studied in adult specimens of the polychaetes Sabella penicillus (Riis- In ciliary downstream-collecting animals a ciliary gaÊrd and Ivarsson, 1990; Mayer, 1994), Spirorbis band transfers particles from the water current to the tridentatus, the entoproct Loxosoma pectinaricola, downstream side of the band, where the separated and the cycliophore Symbion pandora (RiisgaÊrd et particles are taken over by another band of cilia al., 2000). The ciliary bands show some general struc- which transports the particles toward the mouth. Cili- ture in all groups as appears from the descriptions ary downstream particle collecting is found in differ- given below. ent taxonomic groups of invertebrates (Nielsen, 1987, The interpretation of observations of structure and 1995), including meroplanktonic larvae of gastropods activity of the ciliary bands in Loxosoma pectinari- (Emlet, 1990; Hansen, 1991), bivalves (Strathmann cola as summarised by RiisgaÊrd et al. (2000) may and Leise, 1979; Gallager, 1988), and polychaetes serve as a general description of ciliary downstream (Emlet, 1990; Hansen, 1993). collecting. In L. pectinaricola video-observations of 176 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 the paths and velocities of particles showed that feeding on a dilute suspension of microscopic food suspended particles are accelerated with the water particles, including free-living bacteria, are given that enters the region swept by the compound lateral below: cilia and then become caught up by one and more likely by more of the compound cilia in their power 2.6.1. Chaetopterus variopedatus and other stroke (Fig. 1d). This element in the process, denoted chaetopterids `catch-up', accelerates a particle and rapidly moves it The mucus-net ®lter-feeding polychaete Chaetop- in a curved path to the midline at the frontal side of the trus variopedatus, which lives in a parchment-like pinnule. In this phase of the power stroke the particle tube, has been described in several papers, and differ- is pushed out of the main water current which moves ent aspects of water pumping and particle retention past the tentacle, and as the compound lateral cilia have been treated over the years (MacGinitie, 1939; come to rest in their angular motion so does the parti- Wells and Dales, 1951; Brown, 1977; Flood and cle and surrounding ¯uid. The lateral cilia from the Fiala-MeÂdioni, 1982; Jùrgensen et al., 1984; RiisgaÊrd, two sides of the tentacle may prevent escape of the 1989). particle. Finally, the frontal cilia carry the particle When the worm is feeding, a ¯ow of water is driven along the food groove to the mouth. The basic princi- through the tube in the antero-posterior direction by ple of particle catch-up as adopted for Loxosoma (and three muscular piston-like parapods in the middle Spirorbis, Sabella, and others) applies for the cyclio- region of the body. Two notopodia secrete a mucus phore Symbion although the principle of opposed cili- net-bag, which ®lters the water current. The posterior ary bands has instead been obtained by a ciliated end of the suspended mucus net-bag is rolled up into a mouth ring while the frontal cilia are replaced by a food ball within a ciliated cup-like organ, and is ciliated mouth cavity. ingested at intervals of 15±20 min. The mucus net is Based on literature data, own video-observations built of longitudinal parallel ®bre bundles and trans- and ¯uid mechanical considerations it was concluded verse ®laments forming a rectangular mesh (Flood by RiisgaÊrd et al. (2000) that the capture mechanism and Fiala-MeÂdioni, 1982) which ef®ciently retains in all ciliary downstream-collecting suspension suspended food particles down to ca. 1 mm (Jùrgensen feeders is based on the same basic principle described et al., 1984). by the catchup hypothesis. According to this hypoth- The family Chaetopteridae is well known in terms esis compound cilia constitute the pump which gener- of feeding modes. Besides the above-mentioned ates a ¯ow with suspended particles that enters the `piston-pump mucus-net' feeding mode of C. vario- ciliary region. In this region the same cilia, during pedatus, ®ve additionally species of chaetopterids use their power stroke, catch up with suspended particles ®lter-nets, similar to the ones produced by C. vario- and transfer the particles to a food groove, or a mouth pedatus (Barnes, 1964, 1965). In some instances a cavity. In the particle-size retention spectrum, the single net is employed (Telepsavus costarum, Ranza- lower limit presumably depends on spacing between nides sagittaria), but in other cases several nets are cilia that beat in phase, while the upper limit depends formed on successive segments (Spiochaetoptrus on cilia length, which may or may not allow particles oculatus, Phyllochaetopterus socialis, Mesochaetop- to enter the ciliary region. The last phase of transfer is terus taylori). In Spiochaetopterus, Telepsavus, and likely to involve interactions with other cilia systems Phyllochaetopterus the water current through the which act to prevent intercepted particles from escap- tube is generated by ciliary rings of the foliaceous ing with the main through ¯ow. notopodia, whereas Ranzanides and Mesochaetop- terus pump water through the tube by means of peri- 2.6. Mucus-net ®lter feeding staltic contractions of a number of segments of the middle body region. These examples illustrate the Filter feeding by means of an ef®cient mucus net ¯exibility by which ciliary pumps, muscular-piston which sieves the feeding current has independently pumps and muscular-peristalltic pumps are being evolved in a number of different taxonomic groups employed more or less indiscriminately even within of marine animals. Examples of this adaptation to closely related species that may use either one big or H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 177 several smaller mucus nets for straining the tubal was calculated by RiisgaÊrd et al. (1992) to be between water current. 0.34 and 0.41 mm H2O. It is not known whether ®lter- feeding is of primary importance for any population of 2.6.2. Nereis diversicolor this species, but when the phytoplankton concentra- The common polychaete Nereis diversicolor lives tion is above a certain lower `trigger level', N. diver- in a U-shaped burrow in soft sediment. N. diversicolor sicolor prefers to ®lter-feed, while below the trigger has been described as a carnivore and/or scavenger, level it switches to one of the alternative feeding but also as a suspension-feeder and a detritivore, feed- modes (Vedel and RiisgaÊrd, 1993; Vedel et al., ing partly by swallowing surface mud around the 1994; RiisgaÊrd, 1994; RiisgaÊrd and Kamermans, openings of the burrow (Wells and Dales, 1951; 2000). Goerke, 1966; Evans, 1971; RiisgaÊrd, 1994). The occurrence of a ®lter-feeding mechanism in N. diver- 2.6.3. Ascidians sicolor was ®rst described by Harley (1950), and later Ascidians pump water through the inhalant siphon con®rmed by Goerke (1966). Observations of ®lter- into the pharyngeal chamber and through the stigmata feeding behaviour were made by RiisgaÊrd (1991a) on into the atrium from which the water leaves the asci- N. diversicolor in glass tubes immersed in seawater. dian as a jet through the exhalant siphon. The pharynx When algal cells are added, the worm moves to one is perforated with small slits (stigmata), and ciliary end of the glass tube to ®x mucus threads to the glass tracts on either side of the stigmata create the feeding wall, forming the circular opening of a net bag. This current (Fig. 1e). When the water is pumped across funnel-shaped net bag is completed as the worm the pharynx wall, suspended particles are trapped on retreats down the tube. For a period after the bag is the mucus net continuously produced by the endostyle completed, the worm pumps water through the net by (Holley, 1986). Cilia on the papillae or longitudinal means of vigorously undulating movements of the pharyngeal bars transport the endless mucus net, with body (peristaltic pump). Particles suspended in the the retained food particles, to the dorsal side where it inhalant water are retained by the net and after a is rolled into a cord which is passed downwards into period of pumping the worm moves forward, swal- the oesophagus as an unbroken string (MacGinitie, lowing the net bag and its entrapped food particles. 1939; Millar, 1971; Fiala-MeÂdioni, 1978). Particles The structure of the food-trapping net of N. diversi- down to 2±3 mm are completely retained (Randlùv color was studied by RiisgaÊrd et al. (1992). Electron- and RiisgaÊrd, 1979; Jùrgensen et al., 1984), and elec- micrographs of the ®lter-net structure showed that the tron microscopic studies of the mucus net have net is composed of an irregular mesh made up of long, revealed that in the ®xed state, it is composed of relatively thick ®laments (up to 300 nm) inter- 10±40 nm thick ®bres arranged in rectangular meshes connected with a variety of shorter and thinner ®la- that vary between 0.2 and 0.5 mm in width and ments with diameters ranging from 5 to about 25 nm. between 0.5 and 2.2 mm in length (Flood and Fiala- The average mesh size was between 0.5 and 1.0 mm MeÂdioni, 1981). The ciliary pump of Styela clava was (which due to shrinkage during preparation may analysed by RiisgaÊrd (1988a). Based on direct represent only approximately 75% of the actual measurement of the pumping rate, the ¯ow velocity dimension of the intact net). Simultaneous clearance through the mucus ®lter was found to be 0.3 mm s21 measurements of different-sized particles by RiisgaÊrd and the pressure drop across the net estimated to be et al. (1992) showed that particles down to 2±3 mm 0.069 mm H2O. The effects of temperature and algal may be ef®ciently (near 100%) cleared from the concentration on pumping of water in Ciona intesti- water. Thus, the retention ef®ciency of N. diversicolor nalis have recently been studied by Petersen et al. is as high as found in a most obligate ®lter feeders (1999). (Jùrgensen et al., 1984). The main resistance to water ¯ow through a tube holding a ®lter-feeding N. diver- 2.6.4. Lancelets (amphioxus) sicolor is the mucus net. Following Silvester (1983), The most common of the transitional species of using the modi®ed Tamada±Fujikawa equation (see lancelets, which share characteristics with both Section 3.2.), the pressure drop across the mucus net vertebrates and invertebrates, is Branchiostoma 178 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193

Table 3 3 Maximum volumes of water ®ltered (Fmax,dm ) per cubiccentimeter O2 consumed at starvation (Rm, maintenance metabolism) for various ®lter- feeding invertebrates, and for comparison the deposit-feeding bivalve M. balthica and the deposit-feeding polychaete lugworm Arenicola marina

3 23 Taxanomic groups and species Fmax/Rm (dm cm O2) Reference

Sponges Halichondria panicea 2.7 Thomassen and RiisgaÊrd (1995) Mycale sp. 19.6 Reiswig (1974) Verongia gigantea 4.1 Reiswig (1974) Verongia ®stularis 9.7 Reiswig (1974) Tethya crypta 22.8 Reiswig (1974) Bryozoans Celleporella hyalina 68 RiisgaÊrd and ManrõÂquez (1997) Copepods Acartia tonsa 37 Kiùrboe et al. (1985) Polychaetes Sabella penicillus 354 RiisgaÊrd and Ivarsson (1990) Chaetopterus variopedatus 50 RiisgaÊrd (1989) Nereis diversicolor 40 RiisgaÊrd (1991a) Arenicola marina 0.4 RiisgaÊrd et al. (1996a) Bivalves Mytilus edulis 15±50 RiisgaÊrd et al. (1980) Mytilus edulis 18 Clausen and RiisgaÊrd (1996) Macoma balthica 0.4±1.6a Kamermans (1994), De Wilde (1975), Hummel (1985) Ascidians Ciona intestinalis 82 Petersen et al. (1995) Ciona intestinalis 13 Jùrgensen (1955) Lancelets Brachiostoma lanceolatum 79 RiisgaÊrd and Svane (1999)

a Filtration rate (F)ofM. balthica ˆ 5cm3 h21 for a 8 mg body dry wt individual (Kamermans, 1994; Kamermans, pers. comm.), or 6± 3 21 3 21 23 cm h for a 38 mg body dry wt individual (Hummel, 1985). Respiration (R) ˆ 0.014 cm O2 h for an about 25 mg dry wt M. balthica (De 3 Wilde, 1975). F/R ˆ 0.4±1.6 dm water ®ltered per cubic centimeter O2 consumed. lanceolatum, generally known as amphioxus. Since The water current is maintained through the animal the early contributions of Orton (1913) and Van by the activity of the lateral `pump' cilia of the bran- Weel (1937) a number of papers have described lance- chial bars. As particles pass along the pharynx they lets as ®lter feeders (Barrington, 1958; Olsson 1963; are drawn up against the internal wall of the branchial Welsch 1975; Baskin and Detmers, 1976; RaÈhr, 1982; basket and the particles become caught in the mucus RiisgaÊrd and Svane, 1999; Ruppert et al., 2000). As an ®lter-net. The endostyle in the ventral part of the phar- obligate ®lter feeder B. lanceolatum obtains its food ynx produces the mucus ®lter which is transported by straining off phytoplankton in the surrounding along the branchial bars by means of frontal cilia to water. Amphioxus lies buried in the bottom gravel the dorsal groove where the ®lters of two sides with with the ventral side turned upward and with the the captured particles are rolled together and trans- mouth opening free of the bottom. A feeding current ported posteriorly to the oesophagus (Barrington, enters at the anterior end of the animal between the 1958; Olsson 1963; Welsch, 1975; Baskin and buccal tentacles and ¯ows through the buccal cavity. Detmers, 1976; Flood and Fiala-MeÂdioni, 1981; Then the water ¯ows successively through the mouth, RaÈhr, 1982). the branchial basket (pharynx) and mucus ®lter-net, to Littleworkhasbeendonetodescribelanceletsas the atrium chamber, and ®nally out of the exhalant ®lter feeders. Until recently no data existed on either opening. ®ltration rates or the ef®ciency with which food particles H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 179 are retained in the mucus ®lters, but RiisgaÊrd and Svane the particles from the main water current and concen- (1999) further described and characterised B. lanceola- trate them on the upstream side of the band in a not tum as a true ®lter feeder. Simultaneous measurements fully understood way (Nielsen, 1987) (Fig. 1f). Ciliary of clearance of particles of different sizes showed that upstream collecting has been described on the tenta- particles (at least) down to 4 mm are retained by the cles of adult phoronids, brachiopods, and ptero- mucus ®lter nets with 100% ef®ciency (see also Ruppert branchs, and in larvae of phoronids, brachiopods, et al., 2000). echinoderms, and enteropneusts (Strathmann et al., 1972; Strathmann, 1973; Strathmann and Bonar, 2.6.5. Flow characteristics of mucus-net ®lter feeding 1976; Strathmann and McEdward, 1986; Nielsen, The lancelet B. lanceolatum is a cephalochordate 1987). In general, the mechanisms involved in separ- and closely related to the urochordates (or tunicates), ating food particles from the currents set up by the including the sessile ascidians, and the planktonic ciliary bands in upstream-collecting organisms are salps and the appendicularians (larvaceans), which poorly understood. This mechanism is correlated all feed with a mucus net secreted by the endostyle with the presence of only a single band of cilia (Alldredge and Madin, 1982; Holley, 1986; Madin, (`single-band system'), which appears to be responsi- 1990; Nielsen, 1995). Table 1 shows a comparison ble for both the water transport and particle retention between water ¯ow velocity just upstream of the as suggested by Strathmann et al. (1972). mucus net in various species of urochordates and the Upstream-collecting ciliary bands in larvae of phor- mucus-net ®lter-feeding polychaetes N. diversicolor onids, echinoderms, pterobranchs, enteropneusts, and and C. variopedatus, and the mucociliary ®lter-feed- brachiopods, divert the particles from the main water ing gastropod Crepidula fornicata. As seen from current and concentrate them on the upstream side of Table 1, there is a tendency for ciliary pumps to the water pumping ciliary band before the particles occur at relatively low velocities and muscular type are transported toward the mouth (Nielsen, 1987). pumps at the higher velocities, but type also depends In brachiopods the lateral ciliary bands of the tenta- on the pressure drop. Apparently, operating condi- cles create a water current from which particles tions are adjusted to the characteristic of the pump become de¯ected to the frontal side where the frontal (see Section 3.2). A mucus-net consumer may collect cilia transport the particles towards the mouth. Jùrgen- more food particles per unit area of net in a given time sen et al. (1984) measured the ef®ciency with which by increasing the velocity, but pressure drop increases particles of various sizes are retained by the brachio- linearly and pump-work quadratically with increasing pod Terebratulina retuso. It was found that the ef®- pumping rate (RiisgaÊrd and Larsen, 1995; AcunÄaet ciency of retention increased with particle size in the al., 1996), so depending on predisposition for adapta- 1±7 mm range studied, and that the largest 7 mm parti- tion there is a natural limit to the increase. cles were removed with about 70% ef®ciency from the water during one passage of the lophophore. 2.6.6. Other taxonomic groups Upstream collection has been studied in plankto- Another, less studied mucus-net ®lter-feeder is the trophic echinoderm larvae by Strathmann (1971), echiuroid worm Urechis caupo which inhabits a U- Strathmann et al. (1972) and Hart (1991), and in echi- shaped burrow and feeds by ®ltering a current of water noid larvae by Hart and Strathmann (1994). The set up by peristaltic contractions of the body through a capture of particles on the upstream side of the ciliated conical mucus net (Lawry, 1966; Chapman, 1968; band of the echinoderm larvae was observed to be Pritchard and White, 1981; Jùrgensen et al., 1986b). accompanied by a change in the direction of particle Finally, it can be mentioned that the small polychaete movement toward the circumoral ®eld. This was Pygospio elegans can ®lter by building a mucus net found to support the hypothesis that echinoderm within its tube (Hempel, 1957). larvae and other upstream collectors remove particles from suspension by a brief, localised reversal in the 2.7. Ciliary upstream collection direction of beat of cilia on the ciliated band. However, this hypothesis for particle capture by In upstream-collecting animals ciliary bands divert ciliary reversal, suggested also to be in action in 180 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 al., 1994). During recent years ®lter-feeding sponges, polychaetes, bivalves and ascidians have been char- acterised by analysis of the biological ®lter-pumps (RiisgaÊrd and Larsen, 1995). The studies have shown that low-energy pumps continuously process the surrounding water through ®lters appropriately dimensioned for coping with the phytoplankton concentrations of the environment. Thus, there is increasing evidence that ®lter feeding in marine envir- onments is based on the principle of `minimal scaling'. However, many energetic studies on e.g. ®lter-feeding mussels are based on the assumption that the ®ltration rate is physiologically regulated in response to variations in the concentration of phyto- plankton (see Section 3.5). To test the general validity of the above hypothesis for minimal scaling, the in¯uence of algal concentra- tion on feeding and growth in mussels and bryozoans has been studied by Clausen and RiisgaÊrd (1996) and Fig. 2. Biological pumps and analogous man-made pumps. RiisgaÊrd and Goldson (1997), respectively. It was found that in the presence of algal cells in the bryozoans (Strathmann, 1982; LaBarbera, 1984; Hart, concentration interval between a lower `trigger- 1996), has recently been rejected by Nielsen and Riis- level' and an upper `satiation-level' ®lter-feeding gaÊrd (1998), who reviewed the literature on upstream- mussels and ectoprocts are continuously utilising collecting mechanisms and pointed out that further their clearance capacity. Observations of the feeding investigations are needed. activity revealed that alterations in clearance rates as a response to either extremely low or very high algal concentrations are caused by valve closure in 3. Part 2: Filter-pumps and bioenergetics mussels and shutting down of the number of actively feeding zooids in bryozoans. 3.1. Continuous ®lter pumps and minimal scaling Near-bottom vertical pro®les of phytoplankton caused by a dense population of N. diversicolor Filter-feeding marine invertebrates must ®lter large have been observed in the ®eld by RiisgaÊrd et al. volumes of water to meet their food requirements. The (1996e). Water samples were simultaneously low concentrations of small suspended food particles collected at different heights above the bottom in the sea is the key to understanding the character- where N. diversicolor were present, and it appeared istics of the ®lter feeding. Presumably these charac- that a phytoplankton-reduced near-bottom water teristics evolved according to `a principle of minimal layer of 5±10 cm in thickness developed on calm scaling', whereby dimensions of the ®lter-pumps are days. That such depletion of phytoplankton in near- suf®cient to enable continuous feeding at low rates, bottom waters plays a signi®cant role for this worm rather than discontinuous feeding at correspondingly was demonstrated in ®eld-growth experiments high rates (Jùrgensen, 1975). This hypothesis appears performed with worms transferred to glass tubes to exclude physiological regulation of the ®ltration placed at different height above the bottom (RiisgaÊrd rate in response to nutritional needs when the algal et al., 1996e). An approximate 10 times reduction in concentration is above a certain lower `stimulation growth rate of bottom-dwelling N. diversicolor level' as in M. edulis (RiisgaÊrd and Randlùv, 1981) compared with that of worms elevated just 10 cm and Ciona intestinalis (Petersen and RiisgaÊrd, 1992), above the sediment surface indicates that extremely or above the `trigger level' in N. diversicolor (Vedel et meagre food conditions are prevalent near the H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 181

Fig. 3. Pump characteristics and de®nition of operating point. A pump produces an increasing volume ¯ow when the pressure rise over the pump is being reduced (i.e. the pump characteristic is decreasing). The canal system connected to the pump offers an increasing resistance when the volume ¯ow increases (i.e. the system characteristic is increasing). When plotted on the same graph, the point of intersection of these two characteristics Ð which simply states that the pressure rise delivered by the pump equals that required to maintain ¯ow through the system Ð de®nes the operating point (Qo, DHo) of the pump±system arrangement in terms of the resulting operating ¯ow (Qo) and operating pressure head (DHo) which can be read off the graph. sea¯oor. Such situations deserve more attention in future studies on ®lter-feeding benthos.

3.2. Energetic cost of ®lter-feeding

A ¯uid pump delivers a volume ¯ow (Q) and main- tains a pressure rise (DH) from suction side to pressure side. If the pump is not tight this causes a backward leakage ¯ow. Man-made pumps are usually based on machine constructions with rotary elements that via a crank can be transformed to swinging or backward Fig. 4. Upper: direct measurement of pumping rates at different and forward movements (e.g. displacement and fric- hydrostatic pressures imposed between inlet and outlet end of tion pumps). Biological pumps use periodic waving, glass tube with Nereis diversicolor inserted in the wall separating the inhalant (C ) and exhalant chamber (C ). Water level in C is swinging or pulsating movements and use elastic 1 2 2 monitored with a laser beam striking a mirror (m) ®xed on a tethered tissues for sealing and valves. ¯oating ping-pong ball. Pumping rate of the worm is equal to Pumps may be divided up into closed pumps volume of water collected in the beaker (b) when the laser de¯ection (ideally without internal leakage ¯ow) and open point is maintained in a constant position on the scale. Lower: back pumps (always with leakage). Among the closed pressure-pumping rate characteristic in 6 worms. Frequency of 21 types of pumps is N. diversicolor`s undulating move- water-pumping undulating body movements (strokes min )is indicated. ments in its tube (Fig. 2a) which is analogous to a peristaltic tubing pump (Fig. 2b). C. variopedatus uses ¯exible and moving muscular piston-like para- bryozoans. These pumps consist of ciliary bands pods (Fig. 2c) which are analogous to a piston pump (Fig. 2e) and are analogous to belt pumps (Fig. 2f). (Fig. 2d). Open types of pumps are found in ciliary In sponges the ¯agella are acting as a peristaltic pump suspension-feeders, e.g. bivalves, ascidians and (Fig. 2g). 182 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 For any ®lter-feeding organism it is possible to kinetic energy loss, for water ¯owing through a identify a pump and a system, consisting of canals, pump system with known dimensions (for more thor- ®lters, siphons etc. Studied separately, the pump and ough explanation and review, see RiisgaÊrd and Larsen the system can be described by the pump and system 1995). characteristics, respectively, which gives the pressure A few examples may illustrate the basic principles. change (DH,mmH2O) as a function of the volume For a ¯at screen consisting of parallel cylinders with a ¯ow (Q), see Fig. 3. A pump will deliver a decreasing diameter of d and spacing b forming a ®lter the pres- ¯ow when the pump is forced to supply an increasing sure drop (DH) may be estimated from the equation of pressure rise (or pump head, DHp). Further, the fric- Tamada and Fujikawa (1957), see also RiisgaÊrd and tional pressure drop and loss of kinetic energy (or total Larsen (1995, Eq. (16)): head loss of the system, DH ), increases with increas- s DH ˆ Knu=gd; 5† ing volume ¯ow (see Fig. 3). The total pressure drop in an animal's pump± where K ˆ 8t= 1 2 2lnt 1 t2=6†; t ˆ p d=b†; u the system is the sum of frictional resistance, inclusive velocity of upstream ¯ow, g the acceleration of grav- of pressure drop across the ®lter (DHf), kinetic pres- ity, and n is the kinematic viscosity of seawater. sure loss (DHk) due to creation of an exhalant jet to The main resistance to water ¯ow in e.g. bryozoans avoid recirculation of once ®ltered water at the is the ®lter consisting of stiff laterofrontal cilia narrowing outlet of the animal, and hydrostatic back (Fig. 1c). For an approximate estimate of pressure pressure (DH12) which under natural circumstances is drop (resistance) across this ®lter, Eq. (5) is a usable zero, but may be imposed in an experimental set-up model. Assuming d ˆ 0:2 mm; b ˆ 3:5 mm; and u ˆ (see Fig. 4), that is: 0:4mms21 the pressure drop across the ®lter is found to be DH ˆ 0:065 mm H O (Table 2). Similar equa- DH ˆ DH 1 DH 1 DH : 1† 2 s f k 12 tions to describe the pressure drop across for example At the operating point (see Fig. 3) the ¯ow adjusts a rectangular mucus-net has been given by RiisgaÊrd itself so that the pump pressure (DHp) exactly and Larsen, 1995, Eqs. (17) and (18) adapted from balances the total resistance of the pump system Silvester (1983) and Munson (1988), respectively).

(DHs), that is: Another important contribution to the total pump± system resistance may be the exhalant water jet leav- DH ˆ DH : 2† p s ing many suspension feeders through a constriction. Insertion of Eq. (1) in Eq. (2) gives: The pressure drop due to this kinetic energy loss may be calculated as: DHp ˆ DH12 1 DHf 1 DHk: 3† DH ˆ u2=2g; 6† The pump characteristic cannot be measured directly in the suspension feeders; but the back pressure- where u is the velocity of water leaving through the pumping rate characteristic can be experimentally constriction (e.g. siphon opening) estimated as pump- determined (Fig. 4) and, according to Eq. (3) used to ing rate/area of opening. By means of such standard lay down the pump characteristic when the system ¯uid mechanical equations and engineering principles characteristic is known. (see also, Vogel, 1994; Garby and Larsen, 1995) a At the operating point, the useful power received number of suspension-feeding invertebrates have by the water can be calculated as [pump been analysed and characterised in order to determine pressure] £ [volume ¯ow]: the operating point and the power output of the pumps. The normal operating pressure head in these P ˆ rgDH Q ; 4† o o pumps has been found to be low, varying between where r is the water density and g the acceleration 0.02 and 1.5 mm H2O in different species, see Table 2. of gravity. To illustrate the principles of calculation of useful The system characteristic can be calculated by pumping power output, also shown in Table 2, the means of an assortment of standard ¯uid mechanical blue mussel M. edulis may serve as an example equations for pressure drop, caused by friction and (Jùrgensen et al., 1986a, 1988, 1990; Jùrgensen H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 183 1990). Based on detailed information about structures, costs. This attracts the attention to possible prerequi- dimensions and water velocities, and using standard sites and limitation to active suspension feeding. From ¯uid mechanical equations the operating pressure has the above considerations and the energetic costs of been estimated to be DHo < 1mm H2O, and the bivalve suspension feeding by means of large specia- pumping rate for the `standard' 35 mm shell length lised gills it may be realised that the adaptation of N. mussel has been measured in the laboratory to be diversicolor to be a facultative suspension feeder is 3 21 Qo < 1cm s . According to Eq. (4) the power exceptionally cheap in terms of no conspicuous output can now be calculated to be: P ˆ rgDHoQo ˆ anatomic alterations and accompanying metabolic 103 £ 9:81 £ 1023 £ 1026 ˆ 10 mW: investments. This has enabled N. diversicolor to be The energy cost of ®lter-feeding can be evaluated as ¯exible so that it only switches from deposit surface the ratio of useful pumping power (P) and total meta- feeding to suspension feeding when the ambient bolic power expenditure (Rt). This ratio, the overall phytoplankton concentration is suf®ciently high to pump ef®ciency, P/Rt, has been estimated for repre- be pro®table. sentatives of different taxonomic groups and show that the useful pump work constitutes from 0.3 to 3.3. Adaptation to environment 4% of the total metabolic expenditure (see Table 2). Referring to powers estimated for M. edulis by Riis- To assess the adaptation of a ®lter-feeding animal gaÊrd and Larsen (1995) it may be argued that other to a certain environment it is of interest to know the measures of ef®ciency than the overall pump ef®- minimum food energy uptake (ingestion) of the ciency may be appropriate when characterising the animal needed to cover its maintenance metabolic energy cost of ®lter-feeding. Denoting by Rp the meta- energy requirement. The maintenance metabolic bolic rate of the part of organism responsible for need may conveniently be expressed as the respiration pumping action (i.e. bands of lateral cilia and the (Rm) measured as the amount of oxygen consumed by cells carrying the cilia), a minimum ef®ciency the starving animal. The ingestion may be expressed would be: Rp=Rt ˆ 59:5=900 ˆ 6:6%; rather than as the maximum volume of water that the animal can P=Rt ˆ 10=900 ˆ 1:1%; since mechanical and meta- pump through the ®lter device (Fmax) times the food bolic energy conversion as well as the metabolism particle concentration, and will thus depend on phyto- of cells carrying cilia must be included. Moreover, plankton concentration in the environment as well as if the ratio gill metabolic rate Rg=Rt ˆ 175=900 ˆ the particle retention ef®ciency of the animal's ®lter. 3 19% is used as a measure for energy cost it is seen The ratio Fmax/Rm expresses the dm of water pumped that the energetic costs of ®lter feeding by means of per cubiccentimeter O2 consumed and may be used as large specialised gills are considerable. Obviously, the a tool to characterise ®lter feeding. A typical Fmax/Rm- gill structures are expensive to maintain irrespective value of 10 dm3 of water pumped per cubiccentimeter of whether the mussel is pumping water or not. This of oxygen consumed was stated by Jùrgensen (1975) supports the hypothesis of `minimal scaling' which as a minimum to ensure the performance of ®lter implies that energy for functions other than pump feeders inhabiting inshore waters. This may also be work can only be justi®ed when the part of the organ- illustrated by the following example. To balance a ism responsible for the pumping action is dimen- metabolic energy requirement equivalent to the 3 sioned for continuous feeding. consumption of 1 cm O2 (equivalent to 20 J), assum- From an examination of the literature dealing with ing 100% particle retention ef®ciency and 80% assim- possible switching between deposit feeding and ilation ef®ciency of ingested food, a benthic ®lter suspension feeding in marine invertebrates, RiisgaÊrd feeder exposed to a realistic near-bottom phytoplank- and Kamermans (2000) found that it was striking that, ton concentration of 1.5 chl-a dm23 (equivalent to 2.5 apart from the occasionally (facultative) suspension Jdm23) must pump 20= 2:5 £ 0:8†ˆ10 dm3 of water. feeder N. diversicolor, all examples of switching are From the above Fmax/Rm reference value of 3 23 found among passive suspension feeders that have no 10 dm cm O2, and with a thorough knowledge of pump but passively strain food particles from the particle retention ef®ciency, it is now possible to near-bottom current without metabolic energetic evaluate the degree of adaptation of ®lter feeders to 184 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 different environments with different typical phyto- come from suspended material drawn down into the plankton levels. As a rule of thumb, without taking sediment during irrigation. Especially resuspended retention ef®ciency into consideration, if: Fmax/Rm , 10 material from the sediment surface may represent a 3 23 dm cm O2 this suggest that the animal is not a ®lter potential food source (RiisgaÊrd and Banta, 1998). feeder grazing on phytoplankton, Fmax/Rm $ 10 3 23 dm cm O2 indicates a true ®lter feeder, Fmax/Rm @ 10 3.4. Respiration and energetic cost of growth 3 23 dm cm O2 indicates a ®lter feeder that has adapted to very low phytoplankon levels. Table 3 shows the ratio of Respiration rate (R) as a function of body weight independent laboratory measurements of ®ltration rate (W) is usually expressed as R ˆ aWb: West et al. (Fmax) and respiration rate (Rm) in a number of ®lter- (1997) have recently proposed a `3/4 power law' for feeding invertebrates. It is notable that sponges with the allometric scaling for respiration rates, i.e. the their extremely ef®cient collar ®lters that capture free- exponent b ˆ 3=4: Empirically, the b-value is often living bacteria and other small particles down to about close to 3/4 (Wieser, 1994; Garby and Larsen, 0.1 mm have the lowest Fmax/Rm-values, whereas the 1995), but a b-exponent of 3/4 is not always the much less ef®cient polychaete Sabella penicillus has a case, particularly not during stages of signi®cant remarkably high Fmax/Rm-value suggesting an adapta- growth as pointed out by RiisgaÊrd (1998b). Thus, tion to very low phytoplankton concentrations. respiration rate as a function of size throughout the Unfortunately, relevant ®eld measurements of ontogeny of M. edulis, from early larval to the adult phytoplankton concentrations in the inhalant water, stage, during which the mussel increases its weight by or very near benthic ®lter feeders, are still missing a factor of 108, has revealed that the b-exponent is though of great importance for understanding the about 0.9 throughout the early pelagic larval stage adaptation of animals to their environment. But both (RiisgaÊrd et al., 1981) as well as during the juvenile the pump-characteristics and the Fmax/Rm-value of an stage up to a size of 1±10 mg tissue dry weight (Riis- animal may be used to settle uncertainty about a possi- gaÊrd et al., 1980; Hamburger et al., 1983), above ble ®lter-feeding mode of life, as appears from the low which the b-value decreases to about 0.7 in the adult value shown by the deposit-feeding bivalve Macoma stage (Hamburger et al., 1983). The reason for such a balthica, and by the example given below. change in the b-exponent was not immediately It has been proposed by KruÈger (1959, 1962, 1964, obvious, but research on the energetic cost of growth 1971) that the lugworm Arenicola marina can make a in various marine invertebrates has demonstrated that, living as a ®lter feeder, using the sand immediately in as growth decreases with age, the respiration front of the head as a ®lter to retain suspended food decreases because respiration and growth are particles in the ventilatory water. This assumption has connected through the energetic costs of growth. been investigated by RiisgaÊrd et al. (1996a) and Riis- M. edulis may serve as an example of how growth gaÊrd and Banta (1998). The maximal pressure head in¯uences respiration (Clausen and RiisgaÊrd, 1996). that can be delivered by the A. marina pump is The speci®c growth rate (m,d21)ofM. edulis, fed 200 mm H2O (RiisgaÊrd et al., 1996a). This is algal cells at different maintained concentrations, 30±150 times higher than found in true ®lter feeders was calculated according to the equation: m ˆ 21 (RiisgaÊrd and Larsen, 1995) which are also charac- ln Wt=W0†t where W0 and Wt ˆ mean body mass terised by pumping more than 10 dm3 of water per of the mussels on Day 0 and Day t, respectively. By cubiccentimeter of oxygen consumed (cf. Table 3). simultaneous measurements of respiration and growth Thus, the three ®lter-feeding polychaetes Sabella in mussels with body mass W the relationship between penicillus, C. variopedatus and N. diversicolor total respiration rate (Rt) measured on Day t and 3 pump 354, 50 and 40 dm of water per cubiccenti- growth rate (mW) was described as: Rt ˆ b 12b meter of oxygen consumed, respectively. These Rm 1 nmW or Rt=W ˆ a 1 nmW ; where Rm ˆ 3 23 b values may be compared with 0.4 dm cm O2 in aW is the maintenance (starvation) respiration rate, A. marina. This very low value excludes the lugworm and n is the energy cost per unit of growth. In the from making a living as a true ®lter feeder, although it mentioned work the energy cost of growth was esti- is possible that some fraction of its nutrition may mated by using b ˆ 0:66 (Hamburger et al., 1983) and H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 185

experimentally determined values of Rt and m which enabled the identi®cation of n (i.e. slope of regression b 12b line for Rt/W as a function of mW ). Thus, it was experimentally found that the energy cost of growth (n) was equivalent to 12% of the actual growth. This value may be compared with those found in a similar way for other marine invertebrates: 20±26% of the growth in the polychaetes N. diversicolor and N. virens (Nielsen et al., 1995), 21±23% in the ascidian Ciona intestinalis (Petersen et al., 1995), 19% in the copepod Acartia tonsa (Kiùrboe et al., 1985), and 139% of the biomass production in the sponge H. panicea (Thomassen and RiisgaÊrd, 1995). However, dependent on feeding conditions and thus growth rate, the respiration rate of the individual organism may vary considerably (Nielsen et al., 1995). This explains why the b-exponents may assume many different values which may sometimes be larger than 3/4: e.g. b ˆ 1:2inN. diversicolor (Nielsen et al., 1995), b ˆ 0:93 in H. panicea (Thomassen and RiisgaÊrd, 1995), and b ˆ 0:86 in jelly®sh Aurelia aurita (Frandsen and RiisgaÊrd, 1997). From the above examples it is clear that the `3/4 power scaling law' suggested by West et al. (1997) is not a biological law because respiration cannot be separated from growth (RiisgaÊrd, 1998b). The relation between water pumping and respiration in ®lter feeders has been reviewed by Jùrgensen et al. (1986b) and Jùrgensen (1990).

Fig. 5. M. edulis. (A) Speci®c growth rate (m) in ®eld experiments 3.5. Mytilus edulis Ð a case model of general 23 as a function of food concentration expressed as mg chl-a dm validity? (lower axis) and algal equivalents (Rhodomonas sp.) concentration (upper axis). Open symbols: maximum growth rates found for net- bag-transplanted mussels in the Limfjord (Denmark) by RiisgaÊrd A comprehensive literature deals with bivalves, and Poulsen (1981); closed symbols and regression line n ˆ especially the blue mussel M. edulis. The following 35; r2 ˆ 0:55† : growth data obtained for net-bag-transplanted section may serve as a short introduction to relevant mussels in Kerteminde Fjord/Kertinge Nor (Denmark) by Clausen bioenergetic considerations and further re¯ects on and RiisgaÊrd (1996), and by RiisgaÊrd et al. (1994), respectively. (B) possible physiological regulation of ®ltration rates in Speci®c growth rate as a function of algal concentration in labora- tory growth experiments without silt (open symbol) and with zoobenthic ®lter feeders. 5mgdm23 silt (closed symbol). The actual speci®c growth rates The energy balance of a ®lter feeder can be may be compared with the estimated growth rates shown by the expressed as: I ˆ G 1 Rt 1 E; where I ˆ ingestion; inserted unbroken line based on the growth model expressed by G ˆ growth (production), Rt is the total respiration Eq. (7) assuming 80% assimilation ef®ciency and exploitation of (sum of maintenance respiration, R , and respiratory the ®ltration capacity. It is seen that only at algal concentrations m above roughly 5 mg chl-a dm23, where the growth potential of 9.5% cost of growth, Rg), and E is excretion. The growth d21 is utilised, the growth model may need to accept reduced ®ltra- may also be expressed as G ˆ I £ AE 2 Rm 1 Rg† or tion rates to ®t the experimental growth data. Figs. from Clausen G ˆ F £ C £ AE† 2 Rm 1 Rg†; where AE ˆ IBE†=I and RiisgaÊrd (1996). is the assimilation ef®ciency, F the ®ltration rate, and C is the algal concentration. The growth of M. edulis was estimated according to 186 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 the latter equation by Clausen and RiisgaÊrd (1996). Likewise, it may be argued that the prevailing algal The maximum speci®c growth rate of a 100 mg dry concentration in the actual, inhalant water of mussels 23 weight (Wt) `standard' mussel was approximately may often be ,5 mg chl-a dm . On this background 21 mmax ˆ 9:5% d corresponding to a maximum it is relevant to make a note of measured phytoplank- growth rate Gmax ˆ Wtmmax ˆ 2255 mW: The main- ton concentrations in the sea. 3 tenance metabolism (Rm) was estimated at 0.1 cm The median concentration between March and 21 23 O2 h ˆ 570 mW; and the maximum ®ltration rate December varies between 1.7 and 7.7 mg chl-a dm 3 21 measured to Fmax ˆ 1:8dm h : The energetic cost in the open coastal areas around Funen (Denmark), of growth was found to be 12% of the growth (see and a median value of 5.1 mg chl-a dm23 based on Section 3.4), i.e. Rg ˆ 0:12G: Thus, the growth rate about 1400 data series collected in the period may be expressed as: G ˆ A 2 Rm 2 0:12G or G ˆ 1985±1991 has been found for Danish fjords and A 2 Rm†=1:12: Assuming AE ˆ 80% (Kiùrboe et al., coastal waters (Sand-Jensen et al., 1994). In the 1981; RiisgaÊrd and Randlùv, 1981; Poulsen et al., South San Francisco bay the phytoplankton levels 1982), the estimated actual growth rate as a function remain at 2±5 mg chl-a dm23 year around except for of algal concentration (C) may now be expressed by a few days during early spring (Of®cer et al., 1982). In the equation (Clausen and RiisgaÊrd (1996): the Menai Strait (Wales, UK) Blight et al. (1995) found that the chl-a concentration in 1993 remained G ˆ‰ F £ AE £ C† 2 R Š=1:12 7† max m low during the summer, near 1 mgdm23. In the lower or G ˆ bC 2 a: The concentration of algal cells intertidal zone in Port Erin Bay (Isle of Man, northern (Rhodomonas) needed to cover the energy cost of Irish Sea) Sanderson et al. (1996) found that the chl-a maintenance G ˆ 0† was found to be: C ˆ a=b ˆ concentration varied over the course of the tidal cycle 810 cells cm23 ˆ 1:0 mg chl-a dm23; and the from about 16 mg chl-a dm23 at low tide to approxi- concentration of algal cells necessary for maximum mately 3 mg chl-a dm23 at high tide (mean 23 23 growth was: C ˆ Gmax 1 a†=b ˆ 4420 cells cm ˆ 8 ^ 5 mg chl-a dm ). In the Marsdiep tidal inlet 5:5 mgchl-adm23: On the assumption that the maxi- (Dutch Wadden Sea) the annual average is about 23 mum ®ltration rate (Fmax) is continuously being 8 mg chl-a dm , while the monthly averages vary exploited, lower and upper chlorophyll a concentra- from 1 mg chl-a in winter to 30 mg chl-a during the tions de®ne the interval within which there is a direct spring peak period (CadeÂe and Hegeman, 1993; see correlation between phytoplankton concentration also Beukema and CadeÂe, 1997). Meeuwig et al. and growth. (1998) recorded a mean chl-a concentration of Fig. 5 shows the actual speci®c growth rate (m)of 2.4 mg chl-a dm-3 in ®fteen Canadian estuaries. M. edulis in both ®eld and laboratory experiments as a Dame and Prins (1998) compared 11 coastal ecosys- function of food concentration. The actual speci®c tems where the chl-a concentration ranged from 3 to growth rates may be compared with the estimated 22 mg chl-a dm-3 with a mean concentration of growth rates represented by the inserted line in 7.7 mg chl-a dm-3. From this it is obvious that the Fig. 5b based on the above growth model expressed phytoplankton biomass varies considerably between by Eq. (7). It is seen that the estimated growth different localities and no unequivocal statement describes the actual growth fairly well up to about about adaptation of the mussel ®lter-pump to the 5 mg chl-a dm23 (both with and without silt). But at environment can be made without reservations. But algal concentrations above roughly 5 mg chl-a dm23, it seems likely that in many marine areas the M. edulis where the growth potential of 9.5% d21 is utilised, the ®lter pump may be operating more or less continu- estimated growth needs to be based on reduced ®ltra- ously at the prevailing natural algal concentrations, tion rates (i.e. lower than the maximum ®ltration rate) which may be especially low in the near-bottom and/or assimilation ef®ciencies ,80%. In nature, water over mussel beds (Wildish and Kristmanson, however, the growth potential seems generally not 1984; FreÂchette and Bourget, 1985a;b; Loo and to be exploited and for that reason it can be argued Rosenberg, 1989; Jùrgensen, 1990; Butman et al., that in most waters there may not be a conspicuous 1994; Petersen et al., 1997; Dolmer, 2000a,b). need for physiological regulation of the ®ltration rate. The amount of available food for mussels may also H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 187 be dependent on vertical wind mixing of the water also Dolmer (2000b), who suggests the threshold column. Thus, Mùhlenberg (1995) measured that the concentration is about 0.4±0.5 mg chl-a dm23). phytoplankton biomass in the water column decreased In long-term laboratory feeding and growth during windy periods because the benthic mussels experiments Clausen and RiisgaÊrd (1996) observed were brought into contact with the entire phytoplank- that high algal concentrations, equivalent to about ton biomass during such events (see also Small and 17±31 mg chl-a dm23, caused closure of the valves Haas, 1997). In other situations benthic ®ltration may and for that reason reduced ®ltration rate (see also deplete the near-bed water layers and cause a vertical RiisgaÊrd, 1991b). The valve closure phenomenon gradient in the phytoplankton biomass (Dolmer, may be a response to overloading and interpreted as 2000a). The available phytoplankton for the mussels a protective mechanism with the same strength of is then likely to drop to ,5 mg chl-a dm23 where argumentation as used for assuming physiological physiological regulation of the ®ltration rate seems regulation of the bivalve ®lter-pump to nutritional to be irrelevant. Therefore, a physiological regulation needs (Winter, 1973; Widdows, 1976; Navarro and mechanism for ®lter-feeding mussels may be an Winter, 1982; Bayne et al., 1976a,b, 1977, 1987, appropriate requisite in only few situations (see also 1988, 1989, 1993; Widdows and Johnson, 1988; Dolmer, 2000b). Hawkins and Bayne, 1992; Willows 1992; Stenton- It is well known that very low algal concentrations Dozey and Brown, 1992; Navarro et al., 1994; Kree- may cause reduction of water pumping due to closure ger et al., 1995; Sprung 1995; Cranford and Hill, of the valves, and likewise, high algal concentrations 1999). RiisgaÊrd and Randlùv (1981) and RiisgaÊrd may cause closure of the valves (RiisgaÊrd and (1991b) pointed out that low growth rates obtained Randlùv, 1981; RiisgaÊrd, 1991b; Clausen and Riis- on bivalves in laboratory studies, as compared to gaÊrd, 1996). In a recent study by Dolmer (2000b) maximal growth rates in nature (see Jùrgensen, the feeding activity of M. edulis in Limfjorden 1990, Table 8) may be due to the use of very high (Denmark) was related to near-bottom current veloci- algal concentrations (Winter, 1973, 1976; Winter and ties and phytoplankton biomass in the near-bottom Langton, 1976; Tenore et al., 1973) which lead to water. The content of chl-a in the mussels and shell overloading of the feeding system, reduced ®ltration gap size were used as indices of ®ltration activity. At rate, and eventually reduced growth. The view of low current velocities (about 1±2 cm s21) the near- physiological control or regulation of ®ltration rate bed algal biomass was low; 40±70% of the mussels has in particular been contradicted by Jùrgensen had closed shells and they accumulated only a small (1990, 1996), who advocates that, when phytoplank- amount of chl-a in the body. But on days with higher ton is available, as a rule ®lter-feeding bivalves near-bottom current velocities (4±7 cm s21) only always utilise their ®ltration capacity for food uptake. 17±25% of the mussels had closed valves, and the Meticulous laboratory measurements are a neces- mussels accumulated a larger amount of chl-a. This sity for dealing with the physiology and bioenergetics con®rms that mussels in nature react by closing the of suspension feeding bivalves. Here the ®ltration rate valves if phytoplankton concentrations are below a is an important bioenergetic parameter and different certain lower threshold. The valve closing phenom- techniques may be employed for obtaining ®ltration enon seems to represent a real physiological adapta- rates (see review by RiisgaÊrd, 2001). tion to suspension feeding in extremely meagre situations: by reducing the water transport through the mantle cavity the bivalve reduces the oxygen References uptake which leads to reduced metabolism (Jùrgensen et al., 1986b). So far, no studies have attempted to AcunÄa, J.L., Deibel, D., Morris, C.C., 1996. Particle capture quantify how much time bivalves in nature experience mechanism of the pelagic tunicate Oikopleura vanhoeffeni. algal concentrations below the threshold level Limnol. Oceanogr. 41, 1800±1814. (presumably ,1 mg chl-a dm23 to judge from the Alldredge, A.L., Madin, L.P., 1982. Pelagic tunicates: unique herbi- vores in the marine plankton. BioScience 32, 655±663. above estimation of minimal algal concentration Arntz, W.E., Gili, J.M., Reise, K., 1999. Unjusti®ably ignored: re¯ec- needed to cover the maintenance metabolism; see tions on the role of benthos in marine ecosystems. In: Gray, J.S. 188 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193

(Ed.). Biogeochemical Cycling and Sediment Ecology. Kluwer Blight, S.T., Bently, L.T., Lefevre, D., Robinson, C., Rodriques, R., Academic, Dordrecht, pp. 105±124. Rowlands, J., Williams, P.J.leB., 1995. Phasing of autrotrophic Baird, D., Ulanowicz, R.E., 1989. The seasonal dynamics of the and heterotrophic plankton metabolism in a temperate coastal Chesapeake Bay ecosystem. Ecol. Monogr. 59, 329±364. ecosystem. Mar. Ecol. Prog. Ser. 128, 61±75. Barnes, R.D., 1964. Tube-building and feeding in the chaetopterid Bone, Q., Braconnot, J.-C., Ryan, K.P., 1991. On the pharyngeal polychaete Spiochaetopterus oculatus. Biol. Bull. 127, 397±412. feeding ®lter of the salp Pegea confoederata (Tunicata: Thalia- Barnes, R.D., 1965. Tube-building and feeding in chaetopterid cea). Acta Zool. (Stockh.) 72, 55±60. polychaetes. Biol. Bull. 129, 217±233. Brown, S.C., 1977. Biomechanics of water-pumping by Chaetop- Barnes, R.D., 1987. Invertebrate Zoology. Saunders, New York. terus variopedatus Renier: kinetics and hydrodynamics. Biol. Barrington, E.J.W., 1958. The localization of organically bound Bull. Mar. Biol. Lab. Woods Hole 153, 121±132. iodine in the endostyle of Amphioxus. J. Mar. Biol. Ass. UK Buhr, K.-J., 1976. Suspension-feeding and assimilation ef®ciency in 37, 117±126. Lanice conchilega (Polychaeta). Mar. Biol. 38, 373±383. Baskin, D.G., Detmers, P., 1976. Electron microscopic study on the Bullivant, J.S., 1968a. The rate of feeding of the bryozoan Zoobotr- gill bars of Amphioxus (Branchiostoma californiense) with yon verticillatum. NZ J. Mar. Freshwater Res. 2, 111±134. special reference to neurociliary control. Cell Tissue Res. 166, Bullivant, J.S., 1968b. The method of feeding of lophophorates 167±178. (Bryozoa, Phoronida, Brachiopoda). NZ J. Mar. Freshwater Bayne, B.L., Bayne, C.J., Carefoot, T.C., Thompson, R.J., 1976a. Res. 2, 135±146. The physiological ecology of Mytilus californianus Conrad. 1. Butman, C.A., FreÂchette, M., Greyer, R.W., Starczak, V.R., 1994. Metabolism and energy balance. Oecologia 22, 211±228. Flume experiments on food supply to the blue mussel Mytilus Bayne, B.L., Thompson, R.J., Widdows, J., Physiology:, I., 1976b. edulis L. as a function of boundary-layer ¯ow. Limnol. Ocea- Physiology: I. In: Bayne, B.L. (Ed.). Marine Mussels: Their nogr. 39, 1755±1768. Ecology and Physiology. Cambridge University Press, CadeÂe, G.C., Hegeman, J., 1993. Persisting high levels of primary Cambridge, pp. 121±206. production at declining phosphate concentrations in the Dutch Bayne, B.L., Widdows, J., Newell, R.I.E., 1977. Physiological coastal area (Marsdiep). Neth. J. Sea Res. 31, 153±161. measurements on estuarine bivalve molluscs in the ®eld. In: Chapman, G., 1968. The hydraulic system of Urechis caupo Fisher Keegan, B.F., Ceidigh, P.O. (Eds.). Biology of Benthic Organ- and MacGinitie. J. Exp. Biol. 49, 657±667. isms. Pergamon Press, Oxford, pp. 57±68. Charles, F., Amouroux, J.M., GreÂmare, A., 1999. Comparative Bayne, B.L., Hawkins, A.J.S., Navarro, E., 1987. Feeding and study of the utilization of bacteria and microalgae by the suspen- digestion by the mussel Mytilus edulis L (Bivalvia: Mollusca) sion-feeding bivalve: Callista chione. J. Mar. Biol. Assoc. UK in mixtures of silt and algal cells at low concentrations. J. Exp. 79, 577±584. Mar. Biol. Ecol. 111, 1±22. Clausen, I., RiisgaÊrd, H.U., 1996. Growth, ®ltration and respiration Bayne, B.L., Hawkins, A.J.S., Navarro, E., 1988. Feeding and in the mussel Mytilus edulis: no regulation of the ®lter-pump to digestion in suspension-feeding bivalve molluscs: the relevance nutritional needs. Mar. Ecol. Prog. Ser. 141, 37±45. of physiological compensations. Am. Zool. 28, 147±154. Cloern, J.E., 1982. Does the benthos control phytoplankton biomass Bayne, B.L., Hawkins, A.J.S., Navarro, E., Iglesias, J.I.P., 1989. in South San Francisco Bay? Mar. Biol. Prog. Ser. 9, 191±202. Effects of seston concentration on feeding, digestion and growth Cranford, P.J., Hill, P.S., 1999. Seasonal variation in food utilization in the mussel Mytilus edulis. Mar. Ecol. Prog. Ser. 55, 47±54. by the suspension-feeding bivalve molluscs Mytilus edulis and Bayne, B.L., Iglesias, J.I.P., Hawkins, A.J.S., Navarro, E., Heral, M., Placopecten magellanicus. Mar. Ecol. Prog. Ser. 190, 223±239. Desclous-Paoli, J.M., 1993. Feeding behaviour of the mussel Myti- Dame, R.F., Prins, T.C., 1998. Bivalve carrying capacity in coastal lus edulis: Responses to variations in quantity and organic content ecosystems. Aquat. Ecol. 31, 409±421. of the seston. J. Mar. Biol. Assoc. UK 73, 813±829. Davies, B.R., Stuart, V., deVilliers, M., 1989. The ®ltration activity Beninger, P.G., 2000. A critique of premises and methods in a of a serpulid polychaete population (Ficopomatus enigmaticus recent study of particle capture mechanisms in bivalves. (Fauvel) and its effects on water quality in a coastal marina. Limnol. Oceanogr. 45, 1196±1199. Estuar. Coast. Shelf Sci. 29, 613±620. Beninger, P.G., Le Pennec, M., 1988. New observations of the gills De Wilde, P.A.W.J., 1975. In¯uence of temperature on behaviour, of Placopecten magellanicus (Mollusca: Bivalvia), and implica- energy metabolism and growth of Macoma balthica (L.). In: tions for nutrition. Mar. Biol. 98, 61±70. Barnes, H. (Ed.). Ninth European Marine Biology Symposium. Beninger, P.G., Ward, J.E., MacDonald, B.A., Thompson, R.J., Aberdeen University Press, Aberdeen, pp. 239±256. 1992. Gill function and particle transport in Placopecten magel- Dolmer, P., 2000a. Algal concentration pro®les above mussel beds. lanicus (Mollusca: Bivalvia) as revealed using video endoscopy. J. Sea Res. 43, 113±120. Mar. Biol. 114, 281±288. Dolmer, P., 2000b. Feeding activity of mussels Mytilus edulis as Bergquist, P.R., 1978. Sponges. University of California, Berkeley related to near-bed currents and phytoplankton biomass. J. Sea and Los Angeles. Res. 44, 221±231. Beukema, J.J., CadeÂe, G.C., 1997. Local differences in macrozoo- Dral, A.D.G., 1967. The movement of the latero-frontal cilia and benthic response to enhanced food supply caused by mild eutro- mechanism of particle retention in the mussel (Mytilus edulis phication in a Wadden Sea area: food is only locally a limiting L.). Neth. J. Sea Res. 3, 391±422. factor. Limnol. Oceanogr. 42, 1424±1435. Eckman, J.E., Okamura, B., 1998. A model of particle capture by H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 189

bryozoans in turbulent ¯ow: signi®cance of colony form. Am. GruÈnbaum, D., Eyre, D., Fogelson, A., 1998. Functional geometry Nat. 152, 861±880. of ciliated tentacular arrays in active suspension feeders. J. Exp. Emlet, R., 1990. Flow ®elds around ciliated larvae: effects of natural Biol. 201, 2575±2589. and arti®cial tethers. Mar. Ecol. Prog. Ser. 63, 211±225. Hamburger, K., Mùhlenberg, F., Randlùv, A., RiisgaÊrd, H.U., 1983. Evans, S.M., 1971. Behavior in polychaetes. Q. Rev. Biol. 46, 379± Size, oxygen consumption and growth in the mussel Mytilus 405. edulis. Mar. Biol. 75, 303±306. Fauchald, K., Jumars, P., 1979. The diet of worms: a study of Hansen, B., 1991. Feeding behaviour in larvae of the opisthobranch polychaete feeding guilds. Oceanogr. Mar. Biol. Annu. Rev. Philine aperta. II. Food size spectra and particle selectivity in 17, 193±284. relation to larval behaviour and morphology of the velar struc- Fiala-MeÂdioni, A., 1978. A scanning electron microscope study of tures. Mar. Biol. 111, 263±270. the branchial sac of benthic ®lter-feeding invertebrates (Asci- Hansen, B., 1993. Aspects of feeding, growth and stage develop- dians). Acta Zool., Stockh. 59, 1±9. ment by trochophora larvae of the boreal polychaete Mediomas- Fjerdingstad, E.J., 1961a. The ultrastructure of choanocyte collars tus fragile (Rasmussen)(Capitellidae). J. Exp. Mar. Biol. Ecol. in Spongilla lacustris (L.). Z. Zellforsch. 53, 645±657. 166, 273±288. Fjerdingstad, E.J., 1961b. Ultrastructure of the collar of the choano¯a- Harley, M.M., 1950. Occurrence of a ®lter-feeding mechanism in the gellate Codonosiga botrytis (Ehrenb.). Z. Zellforsch. 54, 499±510. polychaete Nereis diversicolor. Nature, London 165, 734±735. Flood, P.R., Fiala-MeÂdioni, A., 1981. Ultrastructure and histochem- Hart, M.W., 1991. Particle captures and the method of suspension istry of the branchial sac of benthic ®lter-feeding invertebrates feeding by echinoderm larvae. Biol. Bull. 180, 12±27. (Ascidians). Acta Zool., Stockh. 59, 1±9. Hart, M.W., 1996. Deconstructing suspension feeders by analysis of Flood, P.R., Fiala-MeÂdioni, A., 1982. Structure of the mucous feed- ®lm and video. Invert. Biol. 115, 185±190. ing ®lter of Chaetopterus variopedatus (Polychaeta). Mar. Biol. Hart, M.W., Strathmann, R.R., 1994. Functional consequences of 72, 27±33. phenotypic plasticity in echinoid larvae. Biol. Bull. 186, 291±299. Foster-Smith, R.L., 1976a. Pressures generated by the pumping Hempel, C., 1957. UÈ ber den RoÈhrenbau und die Nahrungsaufnahme mechanism of some ciliary ®lter-feeders. J. Exp. Mar. Biol. einiger Spioniden (Polychaeta sedentaria) der deutschen Ecol. 25, 199±206. KuÈsten. Helgol. Wiss. Meeresunters. 6, 100±135. Foster-Smith, R.L., 1976b. Some mechanisms for the control of Hawkins, A.J.S., Bayne, B.L., 1992. Physiological interrelations, pumping activity in bivalves. Mar. Behav. Physiol. 4, 41±60. and the regulation of production. In: Gosling, E. (Ed.). The Foster-Smith, R.L., 1978. An analysis of water ¯ow in tube-living Mussel Mytilus: Ecology, Physiology, Genetics and Culture. animals. J. Exp. Mar. Biol. Ecol. 34, 73±95. Elsevier, Amsterdam, pp. 171±222. Frandsen, K.T., RiisgaÊrd, H.U., 1997. Size dependent respiration Hibberd, D.J., 1975. Observations on the ultrastructure of the choa- and growth of jelly®sh Aurelia aurita. Sarsia 82, 307±312. no¯agellate Codosiga botrytis (Ehr.) Saville-Kent with special FreÂchette, M., Bourget, E., 1985a. Energy ¯ow between the pelagic reference to the ¯agellar apparatus. J. Cell. Sci. 17, 191±219. and benthic zones: factors controlling particulate organic matter Holley, M.C., 1986. Cell shape, spatial patterns of cilia, and mucus-net available to an intertidal mussel bed. Can. J. Fish. Aquat. Sci. construction in the ascidian endostyle. Tissue Cell 18, 667±684. 42, 1158±1165. Hummel, H., 1985. Food intake and growth in Macoma balthica FreÂchette, M., Bourget, E., 1985b. Food-limited growth of Mytilus (Mollusca) in the laboratory. Neth. J. Sea Res. 19, 77±83. edulis L. in relation to the benthic boundary layer. Can. J. Fish. Jùrgensen, B.B., 1980. Seasonal oxygen depletion in the bottom Aquat. Sci. 42, 1166±1170. waters of a Danish fjord and its effect on the benthic community. Gallager, S.M., 1988. Visual observations of particle manipulation Oikos 34, 68±76. during feeding in larvae of a bivalve mollusc. Bull. Mar. Sci. 43, Jùrgensen, C.B., 1955. Quantitative aspects of ®lter feeding in 344±365. invertebrates. Biol. Rev. 30, 391±454. Garby, L., Larsen, P.S., 1995. Bioenergetics. Cambridge University Jùrgensen, C.B., 1966. Biology of Suspension Feeding. Pergamon Press, Cambridge. Press, Oxford. Gili, J.-O., Coma, R., 1998. Benthic suspension feeders: their para- Jùrgensen, C.B., 1975. Comparative physiology of suspension feed- mount role in littoral marine food webs. TREE 13, 316±321. ing. Annu. Rev. Physiol. 37, 57±79. Goerke, H., 1966. Nahrungs®ltration von Nereis diversicolor Jùrgensen, C.B., 1983. Fluid mechanical aspects of suspension feed- MuÈller O.F. (Nereidae, Polychaeta). VeroÈff. Inst. Meeresforsch. ing. Mar. Ecol. Prog. Ser. 11, 89±103. Bremerh. 10, 49±58. Jùrgensen, C.B., 1989. Water processing in ciliary feeders, with Gordon, D.P., Clark, A.G., Harper, J.F., 1987. Bryozoa. In: Pandian, special reference to the bivalve ®lter pump, Comp. Biochem. T.J., Vernberg, F.J. (Eds.). Animal Energetics, vol. 2. Academic Physiol. 94A, 383±394. Press, London, pp. 173±199. Jùrgensen, C.B., 1990. Bivalve Filter Feeding: Hydrodynamics, GruÈnbaum, D., 1995. A model of feeding currents in encrusting Bioenergetics, Physiology and Ecology. Olsen and Olsen, bryozoans shows interference between zooids with a colony. Fredensborg, Denmark. J. Theor. Biol. 174, 409±425. Jùrgensen, C.B., et al., 1993. A comment on Mechanisms of GruÈnbaum, D., 1997. Hydromechanical mechanisms of colony suspension feeding in bivalves: Resolution of current controver- organization and cost of defense in an encrusting bryozoan sies by means of endoscopy (Ward et al.). Limnol. Oceanogr. Membranipora membranacea. Limnol. Oceanogr. 42, 741±752. 38, 466. 190 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193

Jùrgensen, C.B., 1996. Bivalve ®lter feeding revisited. Mar. Ecol. Larsen, P.S., RiisgaÊrd, H.U., 1994. The sponge pump. J. Theor. Prog. Ser. 142, 287±302. Biol. 168, 53±63. Jùrgensen, C.B., Ockelmann, K.O., 1991. Beat frequency of lateral Larsen, P.S., Matlok, S., RiisgaÊrd, H.U., 1998. Bryozoan ®lter feed- cilia in intact ®lter feeding bivalves: effect of temperature. ing in laminar wall layers: ¯ume experiments and computer Ophelia 33, 67±70. simulation. Life Environ. 48, 309±319. Jùrgensen, C.B., Kiùrboe, T., Mùhlenberg, F., RiisgaÊrd, H.U., 1984. Laval, M., 1971. Ultrastructure et mode de nutrition du choano¯a- Ciliary and mucus net ®lter feeding, with special reference to ¯uid gelle Salpingoeca pelagica, sp. nov. Comparaison avec les mechanical characteristics, Mar. Ecol. Prog. Ser. 15, 283±292. choanocytes des spongiaires. Protistologica 7, 325±336. Jùrgensen, C.B., Famme, P., Kristensen, H.S., Larsen, P.S., Lemmens, J.W.T.J., Clapin, G., Lavery, P., Cary, J., 1996. Filtering Mùhlenberg, F., RiisgaÊrd, H.U., 1986a. The bivalve pump. capacity of seagrass medows and other habitats of Cockburn Mar. Ecol. Prog. Ser. 34, 69±77. Sound, Western Australia. Mar. Ecol. Prog. Ser. 143, 187±200. Jùrgensen, C.B., Mùhlenberg, F., Sten-Knudsen, O., 1986b. Nature Lisberg, D., Petersen, J.K., 2000. Clearance capacity of Electra of relation between ventilation and oxygen consumption in ®lter bellula (Bryozoa) in seagrass meadows of Western Australia. feeders. Mar. Ecol. Prog. Ser. 29, 73±88. J. Exp. Mar. Biol. Ecol. 62, 225±236. Jùrgensen, C.B., Larsen, P.S., Mùhlenberg, M., RiisgaÊrd, H.U., Loo, L.O., Rosenberg, R., 1989. Bivalve suspension-feeding 1988. The bivalve pump: properties and modelling. Mar. Ecol. dynamics and benthic-pelagic coupling in an eutrophicated Prog. Ser. 45, 205±216. marine bay. J. Exp. Mar. Biol. Ecol. 130, 252±276. Jùrgensen, C.B., Larsen, P.S., RiisgaÊrd, H.U., 1990. Effects of Loo, L-O., Jonsson, P.R., SkoÈld, M., Karlsson, OÈ ., 1996. Passive temperature on the mussel pump. Mar. Ecol. Prog. Ser. 64, 89±97. suspension feeding in Amphiura ®liformis (Echinodermata: Kamermans, P., 1994. Nutritional value of solitary cells and colo- Ophiuroidea): feeding behaviour in ¯ume ¯ow and potential nies of Phaeocystis sp. for the bivalve Macoma balthica (L.). feeding rate of ®eld populations. Mar. Ecol. Prog. Ser. 139, 143± Ophelia 39, 35±44. 155. Kilian, E.F., 1952. WasserstroÈmung und Nahrungsaufnahme beim MacGinitie, G.E., 1939. The method of feeding of Chaetopterus. SuÈsswasserschwamm Ephydatia ¯uviatilis. Z. Vergl. Physiol. Biol. Bull. Mar. Biol. Lab. Woods Hole 77, 115±118. 34, 407±447. Madin, L.P., 1990. Aspects of jet propulsion in salps. Can. J. Zool. Kiùrboe, T., Mùhlenberg, F., 1981. Particle selection in suspension- 68, 765±777. feeding bivalves. Mar. Ecol. Prog. Ser. 5, 291±296. Mann, K.H., Lazier, J.R.N., 1996. Dynamics of Marine Ecosystems. Kiùrboe, T., Mùhlenberg, F., Nùhr, O., 1981. Effect of suspended Biological±physical Interactions in the Ocean. Blackwell, bottom material on growth and energetics in Mytilus edulis. Cambridge, MA 1±394. Mar. Biol. 61, 283±288. Mayer, S., 1994. Particle capture in the crown of the ciliary suspen- Kiùrboe, T., Mùhlenberg, F., Hamburger, K., 1985. Bioenergetics sion feeding polychaete Sabella penicillus: videotape recordings of the planktonic copepod Acartia tonsa: relation between feed- and interpretations. Mar. Biol. 119, 571±582. ing, egg production and respiration, and composition of speci®c Meeuwig, J.J., Rasmussen, J.B., Peters, R.H., 1998. Turbid waters dynamic action. Mar. Ecol. Prog. Ser. 26, 85±97. and clarifying mussels. Their moderation of empirical chl:nu- Kreeger, D.A., Hawkins, A.J.S., Bayne, B.L., Lowe, D.M., 1995. trient relations in estuaries in Prince Edward Island, Canada. Seasonal variation in the relative utilization of dietary protein Mar. Ecol. Prog. Ser. 171, 139±150. for energy and biosynthesis by the mussel Mytilus edulis. Mar. MeyhoÈfer, E., 1985. Comparative pumping rates in suspension- Ecol. Prog. Ser. 126, 177±184. feeding bivalves. Mar. Biol. 85, 137±142. KruÈger, F., 1959. Zur ErnaÈhrungsphysiologie von Arenicola Millar, R.H., 1971. The biologyofascidians. Adv. Mar.Biol. 9, 1±100. marina. Zool. Anzeiger. 22, 115±120. Miller, D.C., Bock, M.J., Turner, E.J., 1992. Deposit and suspension KruÈger, F., 1962. Eksperimentelle Untersuchungen zur oÈkolo- feeding in oscillatory ¯ows and sediment ¯uxes. J. Mar. Res. 50, gischen Physiologie von Arenicola marina. Kieler Meeresforch- 489±520. ungen 18, 157±168. Munson, B.R., 1988. Very low Reynolds number ¯ow through KruÈger, F., 1964. Messungen der PumptaÈtigkeit von Arenicola marina screens. J. Fluids Engng 110, 462±463. L. im Watt. HelgolaÈnder Wiss, Meeresunters. 18, 70±91. Mùhlenberg, F., 1995. Regulating mechanisms of phytoplankton KruÈger, F., 1971. Bau und Leben des Wattwurmes Arenicola growth and biomass in a shallow estuary. Ophelia 42, 239±256. marina. HelgolaÈnder Wiss. Meeresunters. 22, 149±200. Mùhlenberg, F., Kiùrboe, T., 1981. Growth and energetics in LaBarbera, M., 1984. Feeding currents and particle capture mechan- Spisula subtruncata (Da Costa) and the effect of suspended isms in suspension feeding animals. Am. Zool. 24, 71±84. bottom material. Ophelia 20, 79±90. LaBarbera, M., 1990. Principles of design of ¯uid transport systems Mùhlenberg, F., RiisgaÊrd, H.U., 1978. Ef®ciency of particle reten- in zoology. Science 249, 992±1000. tion in 13 species of suspension feeding bivalves. Ophelia 17, Lawry, J.V., 1966. Neuromuscular mechanisms of burrow irrigation 239±246. in the echiuroid worm Urechis caupo Fisher and MacGinitie. J. Mùhlenberg, F., RiisgaÊrd, H.U., 1979. Filtration rate, using a new Exp. Biol. 45, 343±356. indirect technique, in thirteen species of suspension-feeding Leadbeater, B.S.C., 1983. Life-history and ultrastructure of a new bivalves. Mar. Biol. 54, 143±148. marine species of Protospongia. J. Mar. Biol. Assoc. UK 63, Navarro, J.M., Winter, J.E., 1982. Ingestion rate, assimilation ef®- 135±160. ciency and energy balance in Mytilus chilensis in relation to H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 191

body size and different algal concentrations. Mar. Biol. 67, 255± RaÈhr, H., 1982. Ultrastructure of gill bars of Branchiostoma lanceo- 266. latum with special reference to gill skeleton and blood vessels Navarro, E., Iglesias, J.I.P., Ortega, M.M., Larretxa, X., 1994. The (Cephalochorda). Zoomorphology 99, 167±180. basis for a functional response to variable food quantity and Randlùv, A., RiisgaÊrd, H.U., 1979. Ef®ciency of particle retention quality in cockles Cerastoderma edule (Bivalvia, Cardiidae). and ®ltration rate in four species of ascidians. Mar. Ecol. Prog. Physiol. Zool. 67, 468±496. Ser. 1, 55±59. Nicol, E.A.T., 1931. The feeding mechanism, formation of the tube, Reiswig, H.M., 1974. Water transport, respiration and energetics of and physiology of digestion in Sabella pavonia. Trans. R. Soc. three tropical marine sponges. J. Exp. Mar. Biol. Ecol. 14, 231± Edinb. 56, 537±597. 249. Nielsen, C., 1987. Structure and function of metazoan ciliary bands RiisgaÊrd, H.U., 1988a. The ascidian pump: properties and energy and their phylogenetic signi®cance. Acta Zool. (Stockh.) 68, cost. Mar. Ecol. Prog. Ser. 47, 129±134. 202±262. RiisgaÊrd, H.U., 1988b. Ef®ciency of particle retention and ®ltration Nielsen, C., 1995. Animal Evolution: Interrelationships of the rate in 6 species of Northeast American bivalves. Mar. Ecol. Living Phyla. Oxford University Press, Oxford. Prog. Ser. 45, 217±223. Nielsen, C., RiisgaÊrd, H.U., 1998. Tentacle structure and ®lter-feed- RiisgaÊrd, H.U., 1989. Properties and energy cost of the muscular ing in Crisia eburnea and other cyclostomatous bryozoans, with piston pump in the suspension feeding polychaete Chaetopterus a review of upstream-collecting mechanisms. Mar. Ecol. Prog. variopedatus. Mar. Ecol. Prog. Ser. 56, 157±168. Ser. 168, 163±186. RiisgaÊrd, H.U., 1991a. Suspension feeding in the polychaete Nereis Nielsen, N.F., Larsen, P.S., RiisgaÊrd, H.U., Jùrgensen, C.B., 1993. diversicolor. Mar. Ecol. Prog. Ser. 70, 29±37. Fluid motion and particle retention in the gill of Mytilus edulis: RiisgaÊrd, H.U., 1991b. Filtration rate and growth in the blue mussel video recordings and numerical modelling. Mar. Biol. 116, 61± Mytilus edulis Linneaus, 1758: dependence on algal concentra- 71. tion. J. Shell®sh Res. 10, 29±35. Nielsen, A.M., Eriksen, N.T., Iversen, J.J.L., RiisgaÊrd, H.U., 1995. RiisgaÊrd, H.U., 1994. Filter-feeding in the polychaete Nereis diver- Feeding, growth and respiration in the polychaetes Nereis diver- sicolor: a review. Neth. J. Aquat. Ecol. 28, 453±458. sicolor (facultative ®lter-feeder) and N. virens (omnivorous) Ð RiisgaÊrd, H.U., 1998a. Filter feeding and plankton dynamics in a a comparative study. Mar. Ecol. Prog. Ser. 125, 149±158. Danish fjord: a review of the importance of ¯ow, mixing, and Nixon, S.W., 1995. Coastal marine eutrophication: a de®nition, density-driven circulation. J. Environ. Mgmt 53, 195±207. social causes and future concerns. Ophelia 41, 199±219. RiisgaÊrd, H.U., 1998b. No foundation of a `3/4 power scaling law' Of®cer, C.B., Smayda, T.J., Mann, R., 1982. Benthic ®lter feeding: for respiration in biology. Ecol. Lett. 1, 71±73. a natural eutrophication control. Mar. Ecol. Prog. Ser. 9, 203± RiisgaÊrd, H.U., 2001. On measurement of ®ltration rate in bivalves 210. Ð the stony road to reliable data, review and interpretation. Olsson, R., 1963. Endostyles and endostylar secretions: a compara- Mar. Ecol. Prog. Ser. (in press). tive histochemical study. Acta Zool., Stockh. 44, 299±328. RiisgaÊrd, H.U., Banta, G.T., 1998. Irrigation and deposit feeding by Orton, J.H., 1913. The ciliary mechanisms on the gill and the mode the lugworm Arenicola marina, characteristics and secondary of feeding in Amphioxus, ascidians and Solenomya togata.J. effects on the environment. A review of current knowledge. Life Mar. Biol. Assoc. UK 10, 19±49. Environ. 48, 243±257. Owen, G., McCrae, J.M., 1976. Further studies on the latero-frontal RiisgaÊrd, H.U., Goldson, A., 1997. Minimal scaling of the lopho- tracts of bivalves. Proc. R. Soc. London, Ser. B. 194, 527±544. phore ®lter-pump in ectoprocts (Bryozoa) excludes physiologi- Petersen, J.K., RiisgaÊrd, H.U., 1992. Filtration capacity of the asci- cal regulation of ®ltration rate to nutritional needs. Test of dian Ciona intestinalis (L.) and its grazing impact in a shallow hypothesis. Mar. Ecol. Prog. Ser. 156, 109±120. fjord. Mar. Ecol. Prog. Ser. 88, 9±17. RiisgaÊrd, H.U., Ivarsson, N.M., 1990. The crown-®lament-pump of Petersen, J.K., Schou, O., Thor, P., 1995. Growth and energetics in the suspension-feeding polychaete Sabella penicillus: ®ltration, the ascidian Ciona intestinalis (L.). Mar. Ecol. Prog. Ser. 120, effects of temperature, energy cost, and modelling. Mar. Ecol. 175±184. Prog. Ser. 62, 249±257. Petersen, J.K., Schou, O., Thor, P., 1997. In situ growth of the RiisgaÊrd, H.U., Kamermans, P., 2000. Switching between deposit- ascidian Ciona intestinalis (L.) and the blue mussel Mytilus and suspension-feeding in coastal zoobenthos. In: Reise, K. edulis in an eelgrass meadow. J. Exp. Mar. Biol. Ecol. 218, (Ed.). Sandy and Muddy Shores: Ecological Comparisons. 1±11. Ecological Studies, Springer, Berlin (in press). Petersen, J.K., Mayer, S., Knudsen, M.AÊ , 1999. Beat frequency of RiisgaÊrd, H.U., Larsen, P.S., 1995. Filter-feeding in marine macro- cilia in the branchial basket of the ascidian Ciona intestinalis in invertebrates: pump characteristics, modelling and energy cost. relation to temperature and algal concentration. Mar. Biol. 133, Biol. Rev. Camb. Philos. Soc. 70, 67±106. 185±192. RiisgaÊrd, H.U., Larsen, P.S., 2000. A comment on experimental Poulsen, E., RiisgaÊrd, H.U., Mùhlenberg, F., 1982. Accumulation of techniques for studying particle capture in ®lter-feeding cadmium and bioenergetics in the mussel Mytilus edulis. Mar. bivalves. Limnol. Oceanogr. 45, 1192±1195. Biol. 68, 25±29. RiisgaÊrd, H.U., ManrõÂquez, P., 1997. Filter-feeding in ®fteen marine Pritchard, A., White, F.N., 1981. Metabolism and oxygen transport ectoprocts (Bryozoa): particle capture and water pumping. Mar. in the innkeeper Urechis caupo. Physiol. Zool. 54, 44±54. Ecol. Prog. Ser. 154, 223±239. 192 H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193

RiisgaÊrd, H.U., Mùhlenberg, F., 1979. An improved automatic RiisgaÊrd, H.U., Jensen, A.S., JuÈrgensen, C., 1998. Hydrography, recording apparatus for determining the ®ltration rate of Mytilus near-bottom currents and grazing impact of benthic ®lter-feed- edulis as a function of size and algal concentration. Mar. Biol. ing ascidians (Ciona intestinalis) in a Danish fjord. Ophelia 49, 52, 61±67. 1±16. RiisgaÊrd, H.U., Poulsen, E., 1981. Growth of Mytilus edulis in net RiisgaÊrd, H.U., Nielsen, C., Larsen, P.S., 2000. Downstream bags transferred to different localities in a eutrophicated Danish collecting in ciliary suspension feeders: the principle catch-up. fjord. Mar. Pollut. Bull. 12, 272±276. Mar. Ecol. Prog. Ser. 207, 33±51. RiisgaÊrd, H.U., Randlùv, A., 1981. Energy budgets, growth and Rubenstein, D.I., Koehl, M.A.R., 1977. The mechanisms of®lter feed- ®ltration rates in Mytilus edulis at different algal concentration. ing: some theoretical considerations. Am. Nat. 111, 981±994. Mar. Biol. 61, 227±234. Ruppert, E.E., Nash, T.R., Smith, A.J., 2000. The size range of RiisgaÊrd, H.U., Svane, I., 1999. Filter feeding in lancelets suspended particles trapped and ingested by the ®lter-feeding (amphioxus) Brachiostoma lanceolatum. Invertebr. Biol. 118, lancet Branchiostoma ¯oridae (Cephalochorda: Acrania). J. 423±432. Mar. Biol. Assoc. UK 80, 329±332. RiisgaÊrd, H.U., Randlùv, A., Kristensen, P.S., 1980. Rates of water Ryland, J.S., 1976. Physiology and ecology of marine bryozoans. processing oxygen consumption and ef®ciency of particle reten- Adv. Mar. Biol. 14, 285±443. tion in veligers and young post-metamorphic Mytilus edulis. Sanderson, W.G., Harding, S.P., Thorpe, J.P., 1996. An investiga- Ophelia 19, 37±47. tion of the effects of tidal ¯uctuations on the ability of seston as RiisgaÊrd, H.U., Randlùv, A., Hamburger, K., 1981. Oxygen food for intertidal Bryozoa. In: Gordon, D.P., Smith, A.M., consumption and clearance as a function of size in Mytilus Grant-Mackie, J.A. (Eds.). Bryozoans in Space and Time. edulis L. veliger larvae. Ophelia 20, 179±183. Proceedings of the10th International Bryozoological Confer- RiisgaÊrd, H.U., Vedel, A., Boye, H., Larsen, P.S., 1992. Filter-net ence, Wellington, New Zealand, 1995, 259. National Institute structure and pumping activity in the polychaete Nereis diver- of Water and Atmospheric Research, Wellington, p. 269. sicolor: effects of temperature and pump-modelling. Mar. Ecol. Sand-Jensen, K., Nielsen, S.L., Borum, J., Geertz-Hansen, O., 1994. Prog. Ser. 83, 79±89. Fytoplankton- og makrofytudviklingen i danske kystomraÊder RiisgaÊrd, H.U., Thomassen, S., Jakobsen, H., Weeks, J.M., Larsen, (Phytoplankton and macrophyte development in Danish coastal P.S., 1993. Suspension feeding in marine sponges Halichondria waters). Havforskning fra Miljùstyrelsen no. 30 (in Danish with panicea and Haliclona urceolus: effects of temperature on ®ltra- English summary). tion rate and energy cost of pumping. Mar. Ecol. Prog. Ser. 96, Shimeta, J.S., Jumars, P.A., 1991. Physical mechanisms and rates of 177±188. particle capture by suspension feeders. Oceanogr. Mar. Biol. RiisgaÊrd, H.U., Clausen, I., Mùhlenberg, F., Petersen, J.K., Olesen, Annu. Rev. 29, 191±257. N.J., Christensen P.B., Mùller, M.M., Andersen, P., 1994. Filtra- Shimeta, J., Koehl, M.A.R., 1997. Mechanisms of particle selection torer, planktondynamik og biologisk struktur i Kertinge Nor. by tentaculate suspension feeders during encounter, retention, Havforskning fra Miljùstyrelsen no. 45, Miljùministeriet, and handling. J. Exp. Mar. Biol. Ecol. 209, 33±46. Copenhagen (in Danish with English summary). Silverman, H., Lynn, J.W., Achberger, E.C., Dietz, T.H., 1996a. RiisgaÊrd, H.U., Christensen, P.B., Olesen, N.J., Petersen, J.K., Gill structure in zebra mussels: bacteria-sized particle ®ltration. Mùller, M.M., Andersen, P., 1995. Biological structure in a Am. Zool. 36, 373±384. shallow cove (Kertinge Nor, Denmark) B control by benthic Silverman, H., Lynn, J.W., Dietz, T.H., 1996b. Particle capture by nutrient ¯uxes and suspension-feeding ascidians and jelly®sh. the gills of Dreissena polymorpha: structure and function of Ophelia 41, 329±344. latero-frontal cirri. Biol. Bull. 191, 42±54. RiisgaÊrd, H.U., Berntsen, I., Tarp, B., 1996a. The lugworm Areni- Silverman, H., Lynn, J.W., Dietz, T.H., 2000. In vitro studies of cola marina pump Ð characteristics, modelling and energy particle capture and transport in suspension-feeding bivalves. cost. Mar. Ecol. Prog. Ser. 138, 149±156. Limnol. Oceanogr. 45, 1199±1203. RiisgaÊrd, H.U., JuÈrgensen, C., Andersen, F.é., 1996. Case study: Silvester, N.R., 1983. Some hydrodynamic aspects of ®lter feeding Kertinge Nor. In: Barker Jùrgensen, B., Richardson, K. (Eds.), with rectangular-mesh nets. J. Theor. Biol. 103, 265±286. Eutrophication in coastal marine ecosystems. American Silvester, N.R., Sleigh, M.A., 1984. Hydrodynamic aspects of particle Geophysical Union, Coastal and Estuarine Studies, vol. 52, capture by Mytilus edulis. J. Mar. Biol. Assoc. UK 64, 859±879. pp. 205±221. Simpson, T.L., 1984. The Cell Biology of Sponges. Springer, New RiisgaÊrd, H.U., JuÈrgensen, C., Clausen, T., 1996c. Filter-feeding York. ascidians (Ciona intestinalis) in a shallow cove: implications of Small, A.C., Haas, H.A., 1997. Seston dynamics and food avail- hydrodynamics for grazing impact. J. Sea Res. 35, 293±300. ability on mussel and cockle beds. Estuar. Coast. Mar. Sci. 45, RiisgaÊrd, H.U., Larsen, P.S., Nielsen, N.F., 1996d. Particle capture 247±259. in the mussel Mytilus edulis: the role of latero-frontal cirri. Mar. Smaal, A.C., Twisk, F., 1997. Filtration and absorption of Phaeo- Biol. 127, 259±266. cystis cf. globosa by the mussel Mytilus edulis L. J. Exp. Mar. RiisgaÊrd, H.U., Poulsen, L., Larsen, P.S., 1996e. Phytoplankton Biol. Ecol. 209, 33±46. reduction in near-bottom water caused by ®lter-feeding Nereis Sponaugle, S., 1991. Flow patterns and velocities around a suspen- diversicolor Ð implications for worm growth and population sion-feeding gorgonian polyp: evidence for physical models. J. grazing impact. Mar. Ecol. Prog. Ser. 141, 47±54. Exp. Mar. Biol. Ecol. 148, 135±145. H.U. RiisgaÊrd, P.S. Larsen / Journal of Sea Research 44 (2000) 169±193 193

Sprung, M., 1995. Physiological energetics of the zebra mussel Ward, J.E., 1996. Biodynamics of suspension-feeding in adult Dreissena polymorpha in lakes. II. Food uptake and gross bivalve molluscs: particle capture, processing, and fate. Inver- growth ef®ciency. Hydrobiologia 304, 133±146. tebr. Biol. 115, 218±231. Stenton-Dozey, J.M.E., Brown, A.C., 1992. Clearance and retention Ward, J.E., MacDonald, B.A., Thompson, R.J., Beninger, P.G., ef®ciency of natural suspended particles by the rock-pool 1993. Mechanisms of suspension feeding in bivalves: Resolu- bivalve Venerupis corrugatus in relation to tidal availability. tion of current controversies by means of endoscopy. Limnol. Mar. Ecol. Prog. Ser. 82, 175±186. Oceanogr. 38, 265±272. Strathmann, R.R., 1971. The feeding behavior of planktotrophic Ward, J.E., Danford, L.P., Newell, R.I.E., MacDonald, B.A., 1998a. echinoderm larvae: mechanisms regulation and rates of suspen- A new explanation of particle capture in suspension-feeding sion feeding. J. Exp. Mar. Biol. Ecol. 6, 109±160. bivalve molluscs. Limnol. Oceanogr. 43, 741±752. Strathmann, R., 1973. Function of lateral cilia in suspension feeding Ward, J.E., Levington, J.S., Shumway, S.E., Cucci, T., 1998b. Parti- lophophorates (Brachiopoda, Phoronida, Ectoprocta). Mar. cle sorting in bivalves: in vivo determination of the pallial Biol. 23, 129±136. organs of selection. Mar. Biol. 131, 283±292. Strathmann, R.R., 1982. Cine®lms of particle capture by an induced Wells, G.P., Dales, R.P., 1951. Spontaneous activity patterns in local change of beat of lateral cilia of a bryozoan. J. Exp. Mar. animal behaviour: the irrigation of the burrow in the polychaetes Biol. Ecol. 62, 225±236. Chaetopterus variopedatus Renier and Nereis diversicolor O.F. Strathmann, R.R., Bonar, D., 1976. Ciliary feeding of tornaria MuÈller. J. Mar. Biol. Assoc. UK 29, 661±680. larvae of Ptychodera ¯ava (Hemichordata: Enteropneusta). Welsch, U., 1975. The ®ne structure of the pharynx, cyrtopodocytes Mar. Biol. 34, 317±324. and digestive caecum of amphioxus (Branchiostoma lanceola- Strathmann, R.R., Leise, E., 1979. On feeding mechanisms and tum). Symp. Zool. Soc. Lond. 36, 17±41. clearance rates of molluscan veligers. Biol. Bull. 157, 524±535. West, G.W., Brown, J.H., Enquist, B.J., 1997. A general model for Strathmann, R.R., McEdward, L., 1986. Cyphonautes' ciliary sieve the origin of allometric scaling laws in biology. Science 276, breaks a biological rule of inference. Biol. Bull. 171, 694±700. 122±126. Strathmann, R.R., Jahn, T.L., Fonseca, R.C., 1972. Suspension Widdows, J., 1976. Physiological adaptation of Mytilus edulis to feeding by marine invertebrate larvae: clearance of particles cyclic temperatures. J. Comp. Physiol. 105, 115±128. by ciliated bands of a rotifer, pluteus, and trochophore. Biol. Widdows, J., Johnson, D., 1988. Physiological energetics of Mytilus Bull. 142, 505±519. edulis: scope for growth. Mar. Ecol. Prog. Ser. 46, 113±121. Taghon, G.L., Nowell, A.R.M., Jumars, P.A., 1980. Induction of Wieser, W., 1994. Cost of growth in cells and organisms: general suspension feeding in spionid polychaetes by high particulate rules and comparative aspects. Biol. Rev. 68, 1±33. ¯uxes. Science 210, 562±564. Wildish, D.J., Kristmanson, D.D., 1984. Importance to mussels of Tamada, K., Fujikawa, H., 1957. The steady two-dimensional ¯ow the benthic boundary layer. Can. J. Fish. Aquat. Sci. 41, 1618± of viscous ¯uid at low Reynolds numbers passing through an 1625. in®nite row of equal parallel circular cylinders. Q. J. Mech. Wildish, D., Kristmanson, D., 1997. Benthic Suspension Feeders Appl. Math. 10, 426±432. and Flow. Cambridge University Press, Cambridge. Tenore, K.R., Goldman, J.C., Clarner, J.P., 1973. The food chain Willows, R.I., 1992. Optimal digestive investment: a model for ®lter dynamics of the oyster, clam, and mussel in an aquaculture food feeders experiencing variable diets. Limnol. Oceanogr. 37, chain. J. Exp. Mar. Biol. Ecol. 12, 157±165. 829±847. Thomassen, S., RiisgaÊrd, H.U., 1995. Growth and energetics of the sponge Halichondria panicea. Mar. Ecol. Prog. Ser. 128, Winston, J.E., 1978. Polypide morphology and feeding behavior in 239±246. marine ectoprocts. Bull. Mar. Sci. 28, 1±31. Van Weel, P.B., 1937. Die ErnaÈhrungsbiologie von Amphioxus Winter, J.E., 1973. The ®ltration rate of Mytilus edulis and its lanceolatus. Pubbl. Staz. Zool. Napoli 16, 221±272. dependence on algal concentration, measured by continuos Vedel, A., 1998. Phytoplankton depletion in the benthic boundary automatic recording apparatus. Mar. Biol. 22, 317±328. layer caused by suspension-feeding Nereis diversicolor (Poly- Winter, J.E., 1976. A review on the knowledge of suspension-feed- chaeta): grazing impact and effect of temperature. Mar. Ecol. ing lamellibranchiate bivalves, with special reference to arti®- Prog. Ser. 163, 125±132. cial aquaculture systems. Aquaculture 13, 1±33. Vedel, A., RiisgaÊrd, H.U., 1993. Filter-feeding in the polychaete Winter, J.E., Langton, R.W., 1976. Feeding experiments with Myti- Nereis diversicolor: growth and bioenergetics. Mar. Ecol. lus edulis L. At small laboratory scale. I. The in¯uence of the Prog. Ser. 100, 145±152. total amount of food ingested and food on growth. Tenth Vedel, A., Andersen, B.B., RiisgaÊrd, H.U., 1994. Field investiga- European Symposium on Marine Biology, Ostend, Belgium, tions of pumping activity of the facultatively ®lter-feeding poly- September 17±23 1975, vol. 1, pp. 565±581. chaete Nereis diversicolor using an improved infrared Wotton, R.S., 1994. Methods for capturing particles in benthic phototransducer system. Mar. Ecol. Prog. Ser. 103, 91±101. animals. In: Wotton, R.S. (Ed.). he Biology of Particles in Aqua- Vogel, S., 1994. Life in moving ¯uids. The Physical Biology of tic Systems. CRC Press, Boca Raton, pp. 183±204. Flow. Princeton University Press, Princeton, NJ.