MARINE ECOLOGY PROGRESS SERIES Vol. 168: 163-186, 1998 Published July 9 Mar Ecol Prog Ser - Tentacle structure and filter-feeding in Crisia eburnea and other cyclostomatous bryozoans, with a review of upstream-collecting mechanisms Claus ~ielsen'v*,Hans Ulrik ~iisgird* 'Zoological Museum (University of Copenhagen), Universitetsparken 15, DK-2100 Copenhagen, Denmark 'Research Centre for Aquatic Biology (Institute of Biology. Odense University),Hindsholmvej 11, DK-5300 Kerteminde, Denmark ABSTRACT- The upstream-collecting filter-feeding mechanisms occurring in many aquatic organisms are not adequately descnbed. Tentacles of Crisia eburnea and other cyclostome bryozoans, with only lateral and laterofrontal ciliary bands, are among the least complicated upstream-collecting systems among metazoans. SEM and TEM revealed that the tentacles have 2 rows of very closely set lateral cilia and 1row of stiff laterofrontal cilia on each side. The shape of the basal membrane and the longitudinal muscles indicate that the tentacles are specialized for flicking movements directed towards the centre of the tentacle crown. Video observations of C. eburnea feeding on Rhodomonas cells showed charac- tenstic velocity gradients around the tentacle crown. Particles in the central current galn the highest velocity at the entrance to the tentacle crown from where the speed decreases to zero in front of the mouth. Usually the path of a particle deviates from the downward course to a more outwards directed course (between the tentacies),where they may be irclppeci by iile iiiiel iarnied by the stiff Ieteiofron:~! cilia; the tentacle then makes a flick that brings the particle into the central current and further down towards the mouth. A survey of the literature shows that a similar mechanical filter mechanism occurs also in gymnolaemate bryozoans and their cyphonautes larvae, but that the particle-collecting mecha- nism of larvae and adults of phoronids, brachiopods, pterobranchs, and enteropneusts is different. The differences in structure and functlon between the tentacles of ectoprocts and those of phoronids, bra- chiopods and pterobranchs support the idea that the 2 types of tentacle crowns are not homologous. KEY WORDS: Ectoproct feeding . Ciliary f~lter. Sieving . Bio-fluid-mechanics . Video observations Ultrastructure Phylogeny INTRODUCTION structure and function of the 2 systems are quite differ- ent and it seems impossible to bridge the gap between Ciliary filter-feeding mechanisms occur in many the 2 types of particle-collecting systems (Nielsen aquatic organisms. Strathmann et al. (1972) recognized 1995, Nielsen et al. 1996). 2 main types, opposed and single band systems, which In the downstream-collecting ciliary bands, com- were called downstream-collecting and upstream- pound cilia on multiciliate cells transfer particles from collecting bands, respectively, by Nielsen & Rostgaard the water current in a not fully understood way (e.g. (1976). A thorough description of the various types of Shimeta & Jumars 1991, Mayer 1994) to the down- ciliary systems has extensive importance for the basic stream side of the band, where they are taken over by understanding of ciliary filter-feeding in aquatic inver- a band of separate cilia which transports the particles tebrates and of phylogenetic interrelationships of the towards the mouth (Strathmann et al. 1972, Nielsen bilaterian phyla. Superficially, the arrangement of cilia 1987). Downstream-collecting bands are characteristic in ciliary bands may appear rather similar, but both of trochophora larvae of gastropods, bivalves, poly- chaetes, and entoprocts, and of adult sabellid poly- chaetes and some rotifers (e.g. Strathmann et al. 1972, Nielsen 1987, Gallager 1988). The mitraria larva of the O Inter-Research 1998 Resale of full article not permitted 164 Mar Ecol Prog SE polychaete Owenia has separate cilia on monocillate no conclusive experimental studles have been carried cells, but this is the only known exception among the out to verify the sievlng theory. The prevailing view downstream-collecting species. has been that particles are retamed upstream from the In the upstream-collecting ciliary bands, separate lateral cllia by a local reversal of beat of the lateral cilia cilia on mono- or multiciliate cells divert the particles (Strathmann 1973, 1982, LaBarbera 1984, Hart 1996). from the main current and concentrate them on the However, the video observations of Riisgdrd & Man- upstream side of the band where they move towards riquez (1997) on 15 species showed that when a par- the mouth. Upstream-collecting bands are found in ticle deviated from the downward course (towards the larvae of ectoprocts, phoronids, brachiopod.~,echino- mouth) to outwards (between the tentacles) it was derms, and enteropneusts, and on the tentacles of stopped at the frontal side of a tentacle. The trapped adult ectoprocts, phoronids, brachiopods, and ptero- particle was sometimes (in some species) seen to move branchs, and in the bivalve gill (Nielsen 1987). The steadily down the tentacle surface towards the mouth, mechanisms involved in separating food particles from possibly driven by the frontal cilia. But more fre- the currents set up by the ciliary bands in upstream- quently, a tentacle flicking pushed the particle towards collecting organisms are poorly understood. the central current of the tentacle crown to be carried The feeding apparatus of adult ectoproct bryozoans down towards the mouth. These observations support consists of a ring of ciliated tentacles which form a the hypothesis of a laterofrontal filter which strains the funnel with the mouth at the centre of its base. Three water while the centra! current, created by the special types of ciliary rows may be found on the tentacles: pump-design of the tentacle crown, and the action of lateral, frontal and laterofrontal rows. The lateral cilia, flicking tentacles in co-operation clean the filter and which are set on multiciliate cells, produce a water cur- transport the trapped particles toward the mouth. Fur- rent passing outwards between the tentacles, which ther, the measured particle-r~t~ntior?efficie_n_cy X.vas results in the formation of a central current directed found to support the assumption of a mechanical straight down the tentacle crown towards the mouth laterofrontal filter in the ectoprocts. This description of with the more lateral parts of the current deviating out the particle capture process decisively departs from between the tentacles (Ryland 1976). The ciliation of previous work as it does not involve a momentary the frontal cells of the tentacles varies widely in differ- reversal in the beat of lateral cilia as theorized by ent groups, but their possible role in the feeding pro- Strathmann (1973, 1982) and Strathmann et al. (1972). cess is not well known. The importance of the frontal cilia for the transport has The particle-capture mechanism in upstream-collect- remained uncertain, but preliminary observations on ing phyla is not adequately described. Bullivant pro- Crisia eburnea and Heteropol-a nlagna (Nielsen 1987) posed that food particles become separated from the indicate that cyclostomatous bryozoans lack frontal deflected water currents by an impingement mecha- cilia. The simple cyclostome ciliary design with only 2 nism (Bullivant 1968). This hypothesis was supported bands, viz. lateral and laterofrontal, makes this one of by Gilmour (1978),but rejected by most other authors the least complicated upstream-collecting systems because the food particles generally have the same among metazoans. Therefore, we decided to perform a specific grav~tyas the water (Strathman.n 1973). In- stu.dy on cyclostomes wlth the main emphasls on tenta- stead, Strathmann proposed that particles are cap- cle structure and video observations of the food partl- tured by local reversal of the beat of the lateral cilia, cle capture and transport mechan~smin the intact ten- whlch should transport the particles to the frontal side tacle crown. Especially the suggested mechanical of the tentacle and then towards the mouth (Strath- laterofrontal filter and the tentacle flicking mecha- mann et al. 1972, Strathmann 1973, 1982, Strathmann nism, perhaps triggered by the putative sensory latero- & McEdward 1986),and this hypothesis was supported frontal cilia, appeared to be of considerable funclional for example by the observations of Hart (1991, 1996). and phylogenetic interest. It has been suggested that the single row of stiff laterofrontal cil~aof ectoproct tentdcles, about 20 pm long at Intervals of 3 to 5 pm (Gordon 1974, Nielsen MATERIAL AND METHODS 1987),may affect particle capture as sensors by initiat- Ing tentacle flicking or reversal of the beat of lateral 'risia eburnea Linnaeus, 1758 was collected from cllia, and/or act as a sieve which strains the suspended red algae, depth about 17 m, Grollegrund, Northern food particles (Bullivant 1968, Winston 1978). Strath- Dresund, Denmark, 1976 to 1978, from red algae from mann & McEdward (1986) reported that the latero- a tide pool at Menai Strait, Wales, UK, 1997, and from frontal cilia of the cyphonautes larva act as a sieve. red algae, 12 to 15 m, NW of Flatholmen, Gullmars- Gordon et al. (1987) summarized the conflicting hypo- fjorden, Sweden, 1997. Material of the following spe- theses concerning particle retention, but until recently cies was collected from San Juan Islands. WtZ, USA: Nielsen & Riisgbrd. Filter-feeding mechanisms 165 Heteropora magna O'Donoghue & O'Donoghue, 1923 position