SEROTONIN IN PORIFERA? EVIDENCE FROM DEVELOPING TEDANIA IGNIS, THE CARIBBEAN FIRE SPONGE (DEMOSPONGIAE)

SIMON WEYRER, KLAUS RÜTZLER AND REINHARD RIEGER

Weyrer, S., Rützler, K. & Rieger, R. 19990630: Serotonin in Porifera ? Evidence from developing Tedania ignis, the Caribbean tire sponge (Demospongiae). Memoirs of the Queensland Museum 44: 659-665. Brisbane. ISSN 0079-8835.

Histochemical study of larvae and freshly settled juveniles of the Caribbean tire sponge Tedania ignis (Tedaniidae, Poecilosclerida) reveals evidence of serotonin-like immuno-reactivity, a possible indication for the presence ofprecursors ofnerve cells in this species. Already in the earliest stages ofits life, T. ignis is made up oftwo discernable cell types: monociliated cells arranged in quasi-epithelial fashion and covering the larva and the developing settled organism, and mesohylal cells (archeocytes). In the adult sponge, several mesohylal cell types can be distinguished which form a complex connective tissue. Serotonin-like immuno-reactivity demonstrated by us occurs only in two cell types: in sorne archeocytes ofthe parenchymella larvae, and in similar archeocytes and in a second, bipolar cell type ofthe settled, juvenile sponge. The discovery ofa neuroactive substance in cells of developing sponges before and after metamorphosis provides new insights into the origin and evolution of nerve and muscle cells in the Eumetazoa. 0 Pori/era, histochemistry, serotonin, Tedania ignis, larval development, evolution.

Simon Weyrer & Reinhard Rieger, Institute of Zoology and Limnology, University of Innsbruck. Technikerstrasse 25, A-6020 Innsbruck, Austria; Klaus Rützler (email: [email protected]), Department ofInvertebrate Zoology, National Museum ofNatural History, Smithsonian Institution, Washington D. C. 20560, USA; 18 January 1999.

Although sponges, one ofthe oldest metazoan Phagocytella hypothesis (see Hyman, 1951; groups, possess the greatest diversity of Ivanov, 1988; Rieger et al., 1991; Ax, 1995) biologically active compounds of any marine encouraged the authors to search for precursors phylum, the neurotransmitter serotonin of nerve and muscle cells in sponge larvae in (5-hydroxytryptamine, 5-HT) has been reported early developmental stages rather than in adults, a only once, in myocyte-like cells of Sycon neglected area ofresearch so far (Harrison & De ciliatum (Sycettidae, Calcarea) (Lenz, 1966). Vos, 1991; Woollacott & Pinto, 1995). Such Serotonin appeared early in the evolution of precursors of nerve cells and myocytes in eucaryotes. For example, it is used in chemical sponges could represent the first stage in the signal chains in Protista where, in a species ofthe evolution ofintegrative systems (e.g. Payans de ciliate Blepharisma, a serotonin-like substance is Ceccatty, 1974a, 1989; Mackie, 1990). known to function as a mating pheromone (Haldane, 1954; Miyake, 1984). It has also been MATERIALS AND METHODS shown that a number of lower organisms use serotonin as an internai messenger in their neuro­ Larvae of Tedania ignis (Durchassaing & transmitter-receptor systems (Carr et al., 1989) Michelotti) (Tedaniidae, Poecilosclerida, Demo­ and that sorne of these characteristics of spongiae) were sampled in the laboratory molecular structure that arose in unicellular seawater system of the Smithsonian Coral Reef organisms may have been inherited and modified Field Station at Carrie BowCay, Belize, in March bymetazoans(Mackie, 1990; Van Houten, 1990). 1994 and November 1995. Larval release was induced in adult specimens collected in the It seems obvious that nerve cells developed nearby mangrove ofTwin Cays (Rützler & Feller, gradual1y over a long period of time but the 1996) and maintained in aerated seawater by sequences ofchanges that must have occurred are exposing them to natural sunlight fol1owing a difficult to establish. Being a primitive outgroup 12-24hr period of dark adaptation (Woollacott, ofthe Eumetazoa, Porifera do not have neurons 1993). The larvae were kept in seawater-rinsed or myocytes that are present in organisms of glass dishes (lOcm diameter) and fixed higher levels'of organisation. A common immediately after release and 80-100hrs after phylogenetic hypothesis such as the Planula or attachment to the substrate. Ta pravide a 660 MEMOIRS OF THE QUEENSLAND MUSEUM

substrate suitable for fixation and removal for lie embedded among mesohyl cells (Fig. 2A). subsequent processing, the bottom of these Using a whole-mount fluorescence technique, dishes was coated with polymerised epoxy resin we found serotonin-like immuno-reactivity in (Spurr). special mesohyl cells ofboth larval andjuvenile Specimens were fixed in 4% paraform­ T. ignis. Spherical serotonergic cells appear to be aldehyde (PFA) in phosphate-buffered saline randomly distributed and occur alone or in (PBS; O.lM, pH 7.4) for 6-8hrs, rinsed in PBS, clusters (Figs lB, 2B). In one larva, for example, and. treated with Triton X-IOO (0.2%, 1hr) to 6 clusters of such serotonin positive ceIls were permeabilise membranes. After labelling with found, each composed of 2-4 single spherical the primary antibody (rabbit anti-serotonin, cells with a diameter of4-6J..lm. In some clusters IMMUNOTECH 0601; 2.5%) ovemight at 4°C, as weIl as in several single cells the nuclei are fluorescence-labelling was done for Ihr with a clearly visible and appear as non-fluorescent tetrarhodamine-isothiocyanate-(TRITC)­ regions (Fig. lB). conjugated secondary antibody (swine anti-rabbit; In the juvenile, settled sponge, a few bipolar DAKO, 1%) for 1hr. Specimens were then rinsed cells were found in addition to the spherical cells in PBS, whole-mounted (Gelmount) on slides, that superficially resemble bipolar neurons or and examined under a REICHERT Polyvar 'myocyte-like' ceIls (actinocytes) reported by epifluorescence microscope. Incubation in Bagby (1966) (Fig. 2B, C). These bipolar cells bovine-serum albumin-Triton (BSA-T) without have a maximum length of20-50J..lm. Both types primary antibody was used as the control for ofserotonin-positive cells (spherical and bipolar) nonspecific binding of the secondary antibody. appear to be located in the mesohyl as spicules Three larvae and three settled juvenile sponges can be seen ontop ofthe labelled cells (Fig. 2C). were sectioned (epoxy-embedded, 1J..lm thick, No information is yet available on whether stained with toluidin blue) and investigated in interactions between these morphological types detail. The immuno-staining of both larvae and ofserotonin-positive cells occur, nor do we know freshly settled sponges was carried out by the whether the bipolar cells differentiate from the labelled streptavidin-biotin method (LSAB kit; spherical type. Ifthese serotonergic cells are part DAKO). Histochemical staining of peroxidase ofan integrative system, both cellular commun­ with amino-ethyl-carbazole (AEC, substrate ication at a distance (spherical ceIls) or cell-eell buffer) was used to enhance visibility of the contacts (bipolar ceIls) could be expected. labelled cells. DISCUSSION RESULTS Our study is the tirst to report serotonergic cells Tedania ignis has a parenchymella larva in Demospongiae, a spherical type in both larva composed oftwo types ofcells (Bergquist et al., and post-metamorphosed sponge, and a second 1979). PeripheraIly, flagellated epithelium-like bipolar cell type that is exclusive to the post­ cells coyer the organism. This "epithelium" is larval developing organism . Up to now, monociliated and 10-25 J..lm high. The free­ serotonin was only demonstrated histo­ swimming larva exhibits coordinated ciliary chemically in myocyte-like cells ofCalcarea (see action. A distinct basal lamina and typical below). Among the most primitive Eumetazoa, eumetazoan apical junctional complexes are serotonin is weIl known to act as a neuro­ apparently lacking (but see below). Interior, appar­ transmitter (e.g. in Anthozoa, Umbriaco et al., ently motile mesenchymal cells (mesohyl ceIls) 1990). ActuaIly, 5-HT has a wide range of are arranged beneath the epithelium-like sheath functions in invertebrates, such as control of (Woollacott, 1990, 1993; Amano & Hori, 1994) regeneration processes in Planaria (Kimmel & (Fig. lA). The live larvae are ovoid and have a Carlyon, 1990) and ofbeat ofcilia in echinoderm size of 700-900J..lm long, 500-600J..lm wide, but embryo (Mogami et al., 1992), and as inhibitors the necessary Triton X-100 treatment weakens and activators of muscle of molluscs (Welsh, the cell membranes and larvae usually shrink and 1953; Twarog, 1988). As in Porifera, members of collapse (Fig. lA, B). In the juvenile, settled the phylum Placozoado not differentiate nerve or sponge to(}-- as in the' adult-the exopinacoderm muscle cells and are therefore counted among the which covers the ectosome acts as the protective most primitive eumetazoans (GreIl, 1974; Ax, layer. Inside the sponge, choanocyte chambers 1989; GreIl &, Ruthmann, 1991). Schuchert connected by canals lined with endopinacocytes (1993) demonstrated in Trichoplax adherens SEROTONIN IN PORIFERA ? 661

FIG. 1. Tedania ignis, histology of larva. A, Longitudinal section of an entire larva stained by toluidin blue showing epithelial-like celllayer (e) and dark-staining cells (archaeocytes, arrow) in c1usters near the posterior pole CP) (scale bar=lOOJlm). B, Serotonin-positive cells (s) are randomly distributed in the larval body; at least 6 c1usters of2-4 labelled cells are evident (one marked, asterisk). The nucleus in sorne ofthe cells is visible as a non-fluorescent region (arrow). As control for non specifie binding ofthe secondary antibody specimen were incubated in BSA-T without primary antibodies. (Scale bar=IOOJlm.) C, Nomarsky contrast view ofthe same specimen as in Fig. lB. The collapsed and shrunk appearance is due to a necessary Triton X-I00 treatment that weakens the cell membranes. (Scale bar=IOOmJl.)

(Placozoa) bottle-shaped cells (2.7--4Jlm) that tissue fluid of the sponge is regarded as a basic stain specifically with the neuropeptide RF­ precondition for the evolution ofnervous systems. amide. The author speculated about a possible Bandshaped, (epithelial-type) apical cell sensory function of the bottle-shaped cells junctiona1 complexes seem to be present in adult because the RF-amide positive cells were Porifera (e.g. apposed membrane junctions in 10calised at the margin ofthe disc-like body ofT. Bagby, 1970; simplejunctions in Ledger, 1975; adhaerens and the neuropeptide RF-amide is parallel membrane junctions in Green & regarded as functionally conservative in lower Bergquist, 1982; fig. 6 in Garonne & Lethias, invertebrates. 1990) in ultrastructural investigations ofTedania Much effort has been made toward identif)ring ignis, comparable structures seem to be evident attachment complexes between adjacent cells of (authors, unpublished). However, unequivoca1 the pinacoderm in adult sponges, as this layer clarification of the organisation of apical controls the sponge's internaI milieu which junctiona1 complexes is still lacking in the differs from the surrounding environment Porifera. It has been stated repeatedly that perm­ (Ledger, 1975; for review see Harrison & De Vos, anent junctiona1 complexes in epithelially 1991). This altered chemical composition in the organised cells, if present, are different in SEROTONIN IN PORIFERA ? 663

& De Vos, 1991). The ability to contract or valuable discussions and for critical reading of condense in response to endogenous events or the manuscript. We thank K.P. Smith (Dept. of extemal stimuli is a common feature in Calcarea Invertebrate Zoology, National Museum of and Demospongiae. It results in a decrease in Natural History, Washington, D.C.) and M.C. body size and concurrent increase in number of Diaz (University of Califomia, Santa Cruz) for cel! contacts (review in Simpson, 1984; help with fieldwork in Belize. The staff of the Weissenfels, 1989). One can speculate whether Department of Ultrastructure and Evolutionary the increased number ofcell contacts is only the Biology of the University of Innsbruck are result of a reduced body volume or possibly thanked for the use of facilities and technical serves the intensification of'signal transduction' advice (W. Salvenmoser, in particular). This in the network ofactinocytes. Frequent cell--eell paper was presented at the 5th International contacts between myocyte-like cells were also Sponge Symposium in Brisbane, 1998, and we observed in H communis (Pavans de Ceccatty et wish to acknowledge British Airways for the al., 1970). According to Payans de Ceccatty donation of guest tickets to the Smithsonian (1974a), the microfilament-colftaining Institution which allowed K. Rützler and K.P. pinacocytes play an important role in this process, Smith to attend. Research in Belize, Innsbruck, both for cell contraction and cell communication. and Washington was supported by grants from Owing to the dynamic, 'loose' organisation of Fond Wissenschaftlicher Forschung (Austria) cellular features (Pavans de Ceccatty, 1974b; and funds from the Caribbean Coral Reef Bond, 1992) there are no nervous cells evident in Ecosystems Program (CCRE), National Museum sponges, but one can expect temporary, fixed of Natural History, Smithsonian Institution pathways through connected mesohyl cells at the (contribution no. 484). points of stable intercellular connections. Lentz (1966) reported acetylcholinesterase, monoamine LITERATURE CITED oxidase, epinephrine, norepinephrin, and 5HT (serotonin) in 'myocyte-like' cells of Sycon AMANO, S. & HORI, 1. 1994. Metamorphosis of a demosponge. 1. Cells and structure ofswimming ciliatum. These observations along with the larval. Invertebrate Reproduction and association of cholinesterase and myofilaments Development 25: 193-204. in myocyte-like cells and the report of actin AX, P. 1989. Basic phylogenic systematization of the filaments in pinacocytes (Pavans de Ceccatty, Metazoa. Pp. 229-245. In Fernholm, B., Bremer, 1989) may signif)r myoid and neuroid elements K. & Joernvall, H. (eds) The Hierarchy of Life. from a common integrative system that is (Elsevier: Amsterdam.) coordinating 'tissue' contractions in sponges 1995. Das System der Metazoa 1. (Gustav Fischer although electrophysiological evidence of a Verlag: Stuttgart, Jena, New York.) conducting mechanism is still lacking (Lawn, BAGBY, R.M. 1966. The fine structure ofmyocytes in 1982). the sponges Microciona prolifera (Ellis and Solander) and Tedania ignis (Durchassaing and In conclusion, we believe our findings of Michelotti). Journal ofMorphology 118: 167-182. serotonergic cells in the Parazoa suggest an 1970. The fine structure of pinacocytes in the evolutionary specialisation ofserotonin, separate marine sponge Microciona prolifera (Ellis and from its function in Protists. We interpret our Solander). Zeitschrift fur Zellforschung und observations as supporting the recently mikroskopische Anatomie 105: 579-594. emphasised sister-group relationship with the BERGQUIST, P.R. 1978. Sponges. (Hutchinson: Eumetazoa (Morris, 1993; Müller, 1995; Ax, London.) BERGQUIST, P.R., SINCLAIR, M.E., GREEN, C.R. 1995) by exhibiting the very first steps in the & SILYN-ROBERTS, H. 1979. Comparative evolutionary development of the integrative morphology and behaviour of larvae of system of the Metazoa. Further research Demospongiae. Pp. 103-111. ln Lévi, C. & involving additional species and immuno­ Boury-Esnault, N. (eds) Biologie des cytochemical, electrophysiological, and other Spongiaires, Colloques Internationaux de CNRS approaches is clearly needed. Number 291. (Centre National de la Recherche Scientifique: Paris.) BOURY-ESNAULT, N. & RÜTZLER, K. (eds) 1997. ACKNOWLEDGMENTS Thesaurus of Sponge Morphology. Smithsonian Contributions to Zoology 596: 1-55. We are indebted to G. Rieger, D. Reiter, P. BOND, C. 1992. Continuous cell movements rearrange Ladurner (University ofInnsbruck), and W.E.G. anatomical structures in intact sponges. Journal of Müller (University of Mainz, Germany) for Experimental Zoology 263: 284-302. 664 MEMOIRS OF THE QUEENSLAND MUSEUM

BURLANDO, RE., GAINO. E. & MARCHISIO, P.C. LENTZ, T.L. 1966. Histochemical localisation of 1984. Actin and tubulin in dissociated sponge neurohumors in a sponge. Journal of cells. Evidence for peculiar actin containing Experimental Zoology 162: 171-180. microextensions. European Journal of Cell LETHIAS, c., GARRONE, R. & MAZZORANA, M. Biology 35: 317-321. 1983. Fine structure of sponge cell membranes: CARR, W.E.S., GLEESON, R.A. & Comparative study with freeze-fracture and TRAPIDO-ROSENTHAL, H.G. 1989. conventional thin section methods. Tissue and Chemosensory systems in lower organisms: Cell 15: 523-535. Correlations with internaI receptor systems for MACKIE, G.O. 1990. The elementary nervous system neurotransmitters and hormones. Advances in revisited. American Zoologist 30: 907-920. Comparative Environmental Physiology 5: 25-52. MlYAKE, A 1984. Blepharismone: A conjugation GARRONE, R. & LETHIAS, C. 1990. Freeze-fracture inducting trypotophan derivate in the ciliate study ofsponge cells. Pp. 121-128. In Rützler, K. Blepharisma. Pp. 807-813. ln Schlossberger, H. (ed.) New Perspectives in Sponge Biology. G., Hocken, W, Linzen, B. & Steinhart, H. (eds) (Smithsonian Institution Press: Washington, Progress in Tryptophan and Serotonin Research. D.c.) (De Gruyter: Berlin.) GREEN, C.R. 1978. Variations in septate junction MOGAMI, Y., WATANABE, K., OOSHIMA, C., structure in the invertebrates. Pp. 338-339. In KAWANO, A & BABA, S.A. 1992. Regulation Sturgess, J.M. (ed.) Electron Microscopy II. of ciliary movement in sea urchin embryos: (Microscopical Society ofCanada: Toronto.) Dopamine and 5-HT change the swimming GREEN, C.R. & BERGQUIST, P.R. 1979. Cell behavior. Comparative Biochemistry and membrane specializations in the Porifera. Pp. Physiology 101C: 251-254. 153-158. In Lévi, C. & Boury-Esnault, N. (eds) MORRIS, P.J. 1993. The developmental role of the Biologie des Spongiaires, Colloques extracellular matrix suggests a monophyletic Internationaux du CNRS Number 291. (Centre origin of the Kingdom Animalia. Evolution 47: National de la Recherche Scientifique: Paris.) 152-165. 1982. Phylogenetic relationships within the MüLLER, WE.G. 1982. Cell membranes in sponges. Invertebrata in relation to the structure ofseptate International Revue ofCytology 77: 129-181. junctions and the development of "occluding" 1995. Molecular phylogeny ofMetazoa (Animais): junctional types. Journal of Cell Science 53: Monophyletic origin. Naturwissenschaften 82: 279-305. 321-329. GRELL, K.G. 1974. Vom Einzeller zum Vielzeller. PAVANS DE CECCATTY, M. 1974a. The origin ofthe Hundert Jahre Gastrea-Theorie. Biologie in integrative system: A change in view derived unserer Zeit 4: 65-71. from research on Coelenterates and Sponges. GRELL, K.G. & RUTHMANN, A. 1991. A new Perspectives in Biology and Medicine 17: culture method for Trichoplax adherens F. E. 379-390. Schulze. Cytobiology 4: 216-240. 1974b. Coordination in sponges: The foundation of HALDANE, J.B. 1954. La signalisation animale. integration. American Zoologist 14: 895-903. Année Biologique 58: 89-98. 1989. Les éponges, à l'aube des communications HARRISON, F.W & DE VOS, L. 1991. Porifera. Pp. cellulaires. Pour la Science 142: 64-72. 29-89. In Harrison, F.W & Westfall, JA (eds) PAVANS DE CECCATTY, M., THINEY, Y. & Microscopic Anatomy of Invertebrates, 2: GARRONE, R. 1970. Les bases ultrastructurales Placozoa, Porifera, Cnidaria, and Ctenophora. des communications intercellulaires dans les (Wiley-Liss: New York.) oscules des quelques éponges. Pp. 449-466. In HYMAN, L.H. 1951. The Invertebrates. II. Fry, WG. (ed.) The Biology of the Porifera. Plathyhelminthes and Rhynchocoela. The (Academic Press: London.) acoelomate Bilateria. (McGraw Hill: New York.) PROSSER, C.L. 1967. Problems in the comparative IVANOV, AV. 1988. On the early evolution of the physiology ofnon-striated muscles. Pp.133-149. Bilateria. Fortschritte der Zoologie 36: 349-352. In Wiersma, C.AG. (ed.) Invertebrate Nervous KIMMEL, H.D. & CARLYON, WD. 1990. Persistent Systems: Their Significance for Mammalian effects of a serotonin depletor (p-chloro­ Neurophysiology. (University of Chicago Press: phenylalanine) in regenerated planaria (Dugesia Chicago.) dorotocephala). Behavioral Neuroscience 104: REVEL, J.P. 1988. The oldest multicellular animal and 127-34. itsjunctions. Pp. 135-149. In Hertzberg, E.L.(ed.) LAWN, LD. 1982. Porifera. Pp. 49-72. In Shelton, Gap Junctions. (AR. Liss: New York.) G.AB. (ed.) Electrical Conduction and Behaviour RIEGER, R.M., HASZPRUNAR, G. & SCHUCHERT, in 'Simple' Invertebrates. (Clarendon Press: P. 1991. On the origin ofthe Bilateria: Traditional Oxford.) views and new concepts. Pp. 107-113. In LEDGER, P. W. 1975. Septate j unctions in the Sirnonetta, AM. & Conway Morris, S. (eds.) The calcareous sponge Sycon ciliatum. Tissue and Cell Early Evolution ofMetazoa and the Significance 7: 13-18. ofProblematic Taxa. (Claredon Press: Oxford.) SEROTONIN IN PORIFERA ? 665

RÜTZLER, K. & FELLER, C. 1996. Caribbean epithelial cells ofthe freshwater sponge Spongilla mangrove swamps. Scientific American 274(3): lacustris. Cell and Tissue Research 216: 145-154. 94-99. WEISSENFELS, N. 1989. Biologie und SCHUCHERT, P. 1993. Trichoplax adherens (phylum mikroskopische Anatomie der Placozoa) has cells that react with antibodies Süsswasserschwamme (Spongillidae). (Gustav against the neuropeptide RF-amide. Acta Fischer Verlag: Stuttgart & New York.) Zoologica 74: 1I5-1I7. 1990. Condensation rhythm offresh-water sponges SIMPSON, T.L. 1984. The cell biology of sponges. (Spongillidae, Porifera). European Journal of (Springer-Verlag: New York.) Cell Biology 53: 373-383. WELSH, J.H. 1953. Excitation of the heart of Venus TWAROG, B.M. 1988. Serotonin: History of a mercenaria. Archives ofExperimental Pathology discovery. Comparative Biochemistry and Physiology 91: 21-24. and Pharmacology 219: 23-29. WOOLLACOTT, R.M. 1990. Structure and swimming UMBRlACO, D., ANCTIL, M. & DESCARRlES, L. behaviour of the larva of Halichondria 1990. Serotonin-immunoreactive neurons in the melanadocia (porifera: Demospongiae). Journal cnidarian Renilla koellikeri. Journal of ofMorphology 205: 135-145. Comparative Neurology 291: 167-78. . 1993. Structure and swimming behaviour of the VAN HOUTEN, 1. 1990. Chemoreception in larva of Haliclona tubi/era (Porifera: unicellular eucaryotes. Pp 343-356. In Anderson, Demospongiae). Journal of Morphology 218: P.Y.A. (ed.) Evolution of the First Nervous 301-321. Systems. (Plenum: New York.) WOOLLACOTT, R.M. & PINTO, R.L. 1995. Flagellar WACHTMANN, D., STOCKEM, W. & basal apparatus and its utility in phylogenetic WEISSENFELS, N. 1990. Cytoskeletal analyses of the Porifera. Journal ofMorphology organization and cell organelle transport in basal 226: 247-265. 666 MEMOIRS OF THE QUEENSLAND MUSEUM

BIOCALCIFICATION IN THE INDO-PACIFIC biocalcification processes. It is hypothesized that CORALLINE DEMOSPONGE ASTROSCLERA Astrosclera is able to control the rate ofcalcification by WILLEYANA LISTER - THE ROLE OF the regulation of its bacterial population. The mean BASOPINACODERM. Memoirs ofthe Queensland growth rate ofAstrosclera was measured at 230)lm per Museum 44: 666. 1999:- The aragonitic calcareous year. A detailed description ofsoft tissue ultrastructure basal skeleton of Astrosclera is composed of and its cellular composition has recently been 20-60)lm-sized aragonitic spherulites, produced by a published by Worheide (1998). 0 Porifera, combination of three processes. First, the spherulites Astrosclera, skeletal development, calcification are formed in large vesicle cells (LVC's) inside large regulation, ultrastructure. vesjcles in the ectosome. In a second process, after Literature cited. release from LVC's, basopinacocytes transport the spherulites to the tips ofthe skeletal pillars, where they WORHEIDE, G. 1998. The reef cave dwelling fuse together by epitaxial growth; and in a third ultraconservative coralline demosponge process, during upward growth, the soft tissue is Astrosclera willeyana Lister from the slowly rejected from the lowermost-parts of the Indo-Pacific - Micromorphology, Ultrastructure, skeletal cavities and the remaining spaces are Biocalcification, Taxonomy, Biogeography, subsequently filled by epitaxially-grpwing aragonite Phylogeny, Isotope Record. Facies 38: 1-88. fibers. In the second and third process, Gert Worheide* (email: [email protected]) & Joachim basopinacocytes produce either the insoluble Reitner, Institut und Museum für Geologie und intracrystalline organic matrix, which does not consist Paliiontologie (IMGP), University of Goettingen, of collagen, as weil as the soluble intracrystalline 2 Goldschmidtstr 3, D-370n Goettingen, Germany. matrix, which consists of highly acidic Ca +-binding *Present address: Marine Biology Laboratory, mucus substances. Basopinacocytes control speed and QueenslandMuseum, Po.Box 3300, South Brisbane, Qld, direction of epitaxial growth in both ofthe latter two 4101, Australia; 1 June 1998.