INSTITUTE FOR SYSTEMATICS AND POPULATION BIOLOGY UNIVERSITY OF AMSTERDAM. from the Tropical-Western Atlantic. Margriet Kielman!***& SolangePeixinho

Transcription

1 * Instituto Beaufortia INSTITUTE FOR SYSTEMATICS AND POPULATION BIOLOGY UNIVERSITY OF AMSTERDAM Vol. 45, no. 1 March 2, 1995 Mycale escarlatei n. sp. and Mycale unguifera n.sp. (Demospongiae) from the Tropical-Western Atlantic Eduardo HajduJ*, Sven Zea **? Margriet Kielman!***& SolangePeixinho **** *Institute ofsystematks and Population Biology, University ofamsterdam, P.O. Box 94766, 1090-GTAmsterdam, TheNetherlands. **UniversidadNacional de Colombia, Deparlamento de Biologia, INVEMAR - ***LNVEMAR - de Investigaciones Marinas de Punta de Betin, AA 1016, Santa Marta, Colombia. Instituto de Investigaciones Marinas de Punta de Betin, AA 1016, Santa Marta, Colombia. ****Departamento de apologia, Instituto de Biologia, Universidade Federal da Bahia, Campus de Ondina, , Salvador, Ba, Brazil. Key words: Porifera, Poecilosclerida, Mycalidae, Tropical-western Atlantic, Phylogeny Abstract Mycale escarlatei n.sp. is described from the coasts of Bahia and Rio de Janeiro in Brazil and from the Santa Marta area in Colombia. It belongs to the subgenus Aegogropila and is distinguished from other congeners in the area, by slender anisochelae-ii and possession of micracanthoxeas. Mycale unguifera is found n.sp. in the Islas del Rosario and Santa Marta area in Colombia. It belongs to the subgenus Grapelia and is distinguished within this taxon, by its large and abundant sigmas. The specimen described by Carter (1886) as Pseudoesperia enigmatica, is here proposed as neotype of Grapelia australis. Furthermore, we provide a phylogenetic discussion of centrally-thickened, centrotylote-like toxas, micracanthoxeas, absence of an ectosomal skeleton and anisochelae-iii with a basal spur-like projection. INTRODUCTION Eighteen species of Mycale are currently known along the Brazilian coast (three endemics), and from the tropical-western Atlantic (Hajdu & six along the Colombian Caribbean coast (one Desqueyroux-Faundez, 1994; Hajdu & Rutzler, endemic). Our article describes two additional in prep., key included). Of these, seven occur species, both occurring along the Colombian 1

2 Laborel (1969)], while the open coastline is frequently washed by either the cold Falkland Caribbean coast, and only one along the Brazilian coast, thus raising to eight the number of species known from each of both subareas quoted above. The frequent finding of new species of Mycale in areas like the Caribbean (e.g. Van Soest, 1984; Current or by local upwelling waters (e.g. Valentin, 1984; Matsuura, 1986), which bring a subtropical component to its marine fauna and flora (= Patagonian affinity, sensu Mothes et al., Hajdu & Riitzler, in prep.; present article), where 1993). This complex scenario is suggestive of the the acquisition of knowledge relative to the existence of an, as yet poorly known diverse marine life has been a priority for a long time, is fauna of Porifera. Extensive collections of mostly indicative that the diversity of the genus may be unidentified material, including numerous new markedly underestimated worldwide. This is corroborated by the observation of several unpub- species, are housed in several Universities in Brazil, such as those in Rio de Janeiro, Salvador, lished species in the Natural History Museum of Recife andjoâo Pessoa. London, and the Zoôlogisch Museum in Amsterdam. Furthermore, Hooper & Lévi (1994) observed enormous amounts of unpublished / MATERIALS AND METHODS unidentified material in museums all over Australia, which is also bound to contain new species of Mycale. Brazilian specimens of Mycale were collected by EH in Angra dos Reis (Rio de Janeiro), and by New species of sponges are still being found in SP during ecological and systematic studies of the poorly studied Southern Caribbean and the sponges of Salvador (Bahia). The Colombian Brazilian areas where the geological history and specimens were collected by MK and SZ in environmental conditions differ markedly from course of ecological and systematic studies of those of the oceanic Antillean arc, source of most Colombian sponges. Specimens of the monographed faunal collections from the were all collected by SCUBA, except for the Bahian one which Central Atlantic area. The rocky and coral reefs, was found by wading at low tide. Preparation and associated ecosystems located in the conti- techniques followed those outlined by Riitzler nental coast of Colombia in the Southern Caribbean, for example, are scattered among the (1978), with additions by Hajdu & Riitzler (in prep.). mouths of the Atrato and Magdalena, the greatest rivers that open into the Caribbean basin. Abbreviations used are: Moreover, the Santa Marta area in NE Colombia is the scenario of an upwelling of relatively cold waters (ca. 21 C), seasonally alternated with outwelling of continental runoff (Zea, 1993). AMS BMNH DES (Australian Museum, Sidney), (Natural History Museum, London), of Earth (Department Sciences, University of Cambridge), These ecological conditions, together with the INV-POR (INVEMAR, Porifera collection), relatively recent connection to the Eastern Paci- ISP-UvA (Institute of Systematics and Popula- fic before the rise of the Panama isthmus, make tion Biology, University of Amster- the Southern Caribbean a likely source of new dam), faunal elements. This situation is evident along the extensive Brazilian coastline, where tropical coral-reef ecosystems, like those in the Caribbean, are gradually replaced as we move south by a system of alternating bays and sandy beaches in open coastline, interrupted here and there by a few rocky pontoons or mouths of rivers. The bays have a tropical affinity ["oasis coralliennes", MCN NMV NNM SME SMF (Museu de Ciências Naturais, Fundaçâo Zoobotânica do Rio Grande do Sul, Porto Alegre), (Museum of Victoria, Melbourne), Natuurhistorisch (Nationaal Museum, Leiden), (Station Marine d'endoume, Marseille), (Senckenberg Museum, Frankfurt), 2

3 Figs Mycale Janeiro State, 2. Marta, Colombia. Scale UFBaPOR 1 1. n.sp., da Bahia, Po- collection), do Rio 1-17, Table Material Museum of (University Zoology, Cambridge), Museum of Natural His- (Angra 16.X.1992, Museum, University Porifera Museum (Zoological of Hamburg) of de , El Santa m 14.IV. Collection, reef. Topsent 4 cm in Anisochelae varians regular, Gray, tangen- spicule bundles which serrated palmate. Type species: Punta 448, coll. S. deep, de Zea, com- on dead coral Betin 12.XII. INV-POR Marta, Colombia), holotype deposited (Figs. (Figs. naked-eye (Figs. firm but no m Santa coll. S. fringing on under ZMA POR sigmas. Aegogropila 9, 15-17); with Specimens 5 mm visible. diameter. Few, Both (ca. bright-orangy-red, specimens the material or yellow. aquiferous oscula from to consistency Colour of live scattered by 4 thick), slightly 1-2). Surface reticulated No subectosomal channels of the tem micracan- small 1-2), microhispid; compressible. dull-pinkish-red, n.sp possesses very with head 75% of the area), encrusting (< tuberculate 1867 peels off, and in MCN 2829, metamorphic boulder; morphology. External Lundbeck Mycale Gray easily from length DESCRIPTION Demospongiae Sollas often boulder State, Brazil), coll. S. Peixinho, anisochelae-ii, Class an, under Paratypes: Diagnosis: Mycale (A.) escarlatei thoxeas. skeleton of do shaft's ectosomal deep, INV-POR de , SYSTEMATIC DESCRIPTIONS Diagnosis: Mycale Ilha 4451, National Zoology, often under (Bahia Schizotype slender with n.sp. and University of Singapore). Subgenus Aegogropila Gray, m Marta, Colombia), deep, Morro 2 Sargassum. Bahia Itapoâ (Salvador, 449, collection), Reference Department tial, ca. dominated by ington D.C.), (Zoôlogisch (Zoologisch Genus de Reis, Rio de Janeiro State, Brazil), coll. (Bahia Family Mycalidae Rio Reis, Betín, Bahía de Santa Holotype: UFRJPOR dos 5.XI.1991, intertidal; Order Poecilosclerida de 1. Washtory, Smithsonian Institution, Zea, ZRC studied. Hajdu, munity Amsterdam, ZMH dos Bomfim, Angra situ, Punta de Ja- E. ZMA POR do Mycale (Aegogropila) escarlatei Figs. (Universidade Federal (National in cm. Bomfim USNM situ, Ilha paratype (INV-POR 448), neiro, Porifera collection), UMZC in holotype (UFRJPOR 4451), (Aegogropila) escarlatei Mycale = n.sp. (Universidade Federal rifera UFRJPOR escarlatei (Aegogropila) Brazil. 1-2 sys- mm in Angra dos Reis (UFRJPOR 4451, holotype) and from Punta de 3

4 n.sp., holotype (UFRJPOR 4451). 3. Tangential view of Figs Mycale (Aegogropila) escarlatei surface showing reticulation composed partly by bundles of Pores subtylostyles, partly by single subtylostyles. of variable sizes are clearly visible in the meshes. 4. Enlarged central lower portion (rotated upside-down) of figure 3, showing protruding terminations of a few subectosomal brushes of subtylostyles. Abundant anisochelae-iv, and a few anisochelae-ii can be seen to lying tangentially the surface membrane. 5. Detail of surface showing rosettes of anisochelae-i ocurring immediately below the surface. 6. Terminationsof megascleres. 7. Face view ofanisochelae-i. 8. Profile view ofanisochelae-i. Scales: 3 = 500 µm, 4-5 = 100 µm, 6 = 20 µm, 7-8 = 10 µm. 4

5 Betîn (INV POR 448) inhabited by ophiuroid. single small sharpened terminations; toxas-ii (Fig. 13), Skeleton. Ectosomal skeleton, tangential reticulation of paucispicular bundles of subtylostyles (3-10 spicules across, to 70 up (Jm thick, Fig. 3). Meshes very irregular, up to 350 pm wide. Rosettes of anisochelae-i (up to 140 (im in diameter) scattered in nodes of the reticulation (Fig. 5), or, around more rarely, bundles (like a whorl). visible Sigmas-I rare, around bundles, with hooks stouter than -I, centrally-thickened, centrotylotelike, markedly curved centrally, straight prolongations, sharp terminations; micracanthoxeas (Figs. 11, 13-14), straight, large spines, ca. 20 spines per spicule. Etymology The species is named after Sergio Escarlate (formerly with Colégio Santo Inâcio de Rio Loyola, de Janeiro) who introduced EH to marine biology, and also because specimens can directed to the bundle or away be scarlet-red. from it. Anisochelae-II, and -IV scattered in the meshes, the latter predominantly among the pores (< 50 (im in diameter) on the surface membrane (Fig. 4). REMARKS Choanosomal skeleton, plumo-reticulated Mycale escarlatei is easily distinguished from other architecture with much divergence and little species from the tropical-western Atlantic by its anastomosing of tracts (< 100 Jm thick). Tracts possession of both very slender, "duck's bill" like, run perpendicularly to surface, diverging in a anisochelae-ii with a head long 75% of the profusion of acute-angled length of the shaft, and micracanthoxeas. Species brushes in the subectosomal area. The latter support the tangential ectosomal skeleton, only slightly piercing it (< from other biogeographic one or the other character. realms may possess 100 im, Fig. 4). Rosettes of anisochelae-i occur Mycale (Carmia) micracanthoxea Buizer & Van as deep as 500 in (0m the choanosome. Soest, 1977 (NE Atlantic), and M. (Aegogropila) bamfieldense Reiswig & Kaiser, 1989 (NE Pacific) Spicules (measurements in Table 1). Megascleres: subtylostyles (Fig. 6), fusiform, straight, slightly bent, or slightly sinuous, thicker at middle portion, ellyptical head, apex narrowing gradually. possess micracanthoxeas, but differ from M. escarlatei by a collection of features. M. micracanthoxea does not have third a category of anisochelae, and both its sigmas-i and micracanthoxeas Microscleres: anisochelae-i (Figs. 7-8), head are larger. Additionally the colour in life varies about 50% of total length of the shaft, shaft from orange-yellow to brownish-grey, according almost straight on profile view, frontal ala of foot to type of substrate and time of the year (Buizer markedly bent forward (ca. 30 angle with shaft on profile view); anisochelae-ii (Figs. 9, 15-17), & Van Soest, 1977), never the reaching intense bright colours of the new M. species. bamfieldense head about 75% of total length of the shaft, shaft gently curved all over on profile view, general differs from M. escarlatei in the absence of a second category of toxas. Its greyish-yellow colour aspect on profile and face views very narrow, on in life also sets it apart. Both northern species possess sigmas-ii which are considerably smaller (14-25 Jm) than those found in the new species (20-39 im). Their smaller toxas were not reported as centrally-thickened, and none of them possess anisochelae-ii of the peculiarly narrow shape found in M. (A.) escarlatei sp.n. face view microsclere wider on top portion of head [like a duck's bill, sensu Thomas (1968)]; anisochelae-iv (Fig. 10), head about 55% of total length of the shaft, shaft gently curved all over on profile view, foot with markedly reduced alae, the frontal one extends on in top a long, thin digitiform process; sigmas-i, stout, hooks abruptly Comparison with over 60 species and varieties bent and sharp; sigmas-ii (Figs ), slender, hooks markedly bent and sharp; toxas-i (Fig. 12), slender, markedly curved inward centrally, gently curved outward on both prolongations, gradually of Mycale occurring in the Indo-west Pacific (data derived from database by Van Soest; Van Soest, 1994) revealed two species which are likely to be related to M. escarlatei, viz. M. (A.) sulevoidea (Sol- 5

6 Table 1. Spicular micrometric data the on holotype and paratypes of Mycale (Aegogropila) escarlatei sp.n, two species of Mycale for which micracanthoxeas were reported, viz. M. (Carmia) micracanthoxea Buizer & Van Soest, 1977, and M (A.) bamfieldense Reiswig & Kaiser, 1989, and two species which possess similarly shaped duck s bill (sensu Thomas, 1968) anisochelae-ii, viz. M. (A.) sulevoidea (Sollas, 1902) and M. (A.) cavernosa Bergquist, Unless stated otherwise, N = 20 for the holotype and other species studied, and N = 10 for the paratypes. subtylostyles anisochelae sigmas toxas micracanthoxeas length by width of head by width of shaft escarlatei sp.n / I: I: I: holotype by II: II: (N = 2) UFRJPOR 4451 by IV: 13-/ II: escarlatei sp.n I: I: I: n. found paratype by II: (N = 8) II: MCN 2829 by 4-(>.9-8 IV: II: (52?) escarlatei sp.n I: I: I: n. found paratype by II: II: II: INV POR 448 by IV: escarlatei sp.n I: I: I: paratype by II: (N = 8) (N = 5) (N = 8) INV POR 449 by IV: II: II: (N = 4) (N = 2) (N = 5) micracanthoxea I: I: I: Buizer & Van Soest by II: II: II: bamfieldense I: I: I: Reiswig & Kaiser by II: II: III: sulevoidea (Sollas) 360 I: I: I: 50 (?)- 140 by 6-12 II: 33 by 6 II: n. rep III: n. rep. II: n. rep IV: 12 holotype I: I: I: remeasured by II: by (N= 10) by 8-/ III: not found II: not found II: IV: (N = 4) sulevoidea (Sollas) I: I: I: sensu Hentschel' by II: II: II: n. rep. III: n. rep. IV:

7 Table 1. continued subtylostyles anisochelae sigmas toxas micracanthoxeas length by width of head by width of shaft SMF 1031 remeasured by by : : : (N 8) = by (N = 7) II: N.Found II: II: (N 14) = III: (N = 11) IV: (N= 75) cavernosa ^-351 1: : Bergquist by II: by II: holotype remeasured by by I: I: II: by : : ' Perhaps new species. las, 1902) and M. (A.) cavernosa Bergquist, new species. Other specimens differed in several respects to the type of M. sulevoidea, and from the Both species possess anisochelae-ii of comparable dimensions, which on face view resemble a new species described here. duck's bill, and also a reticulated ectosomal skeleton. Nevertheless, detailed scrutiny under We studied the holotype of M. cavernosa SEM failed to reveal the occurrence of micracanthoxeas in either species. (USNM 23703) and found it to possess three categories of anisochelae (anisochelae-i being quite rare), instead of two as originally reported Mycale sulevoidea (holotype studied, UMZC Reg. (Bergquist, Nov ) is distinct from M. escarlatei by its 1965). This aspect agrees with M. escarlatei but M. cavernosa differs by its possession of slightly larger megascleres, anisochelae-i and -IV, and toxas-i; and absence of sigmas-ii (Table 1). thinner megascleres and larger sigmas-i (Table 1), absence of toxas (a single toxa, 26 im long Another five specimens originally assigned to M. sulevoidea were studied for comparative reasons, was seen, but its autochtonous nature is uncertain) and erect shape. two identified by Hentschel (1912, viz. SMF 1031 and 1032), and three identified by Burton Subgenus Grapelia Gray, 1867 (ZMA 1612 and 1613, from the 'Siboga' Exp., unpubl. res; BMNH a, Great Diagnosis: Mycale with three categories of aniso- Barrier Reef Exp.). Only Hentschel's SMF 1031, chelae. Categories-I and -II markedly curved on from the Strait of Malacca, actually resembles profile view and organized the holotype, although possessing larger megas- in rosettes. Anisochelae-I mosdy unguiferate. Anisochelae-II acan- cleres, a fourth category of anisochelae, second those, frequently bipocilla-like. Anisochelae-III category of sigmas and only one category of toxas, which sets it apart (Table 1). It may be a palmate with basal spur-like projection. species: Grapelia australis Gray, Type 7

8 Figs Mycale n.sp., holotype (UFRJPOR 4451). 9. Profile-view of anisochelae-ii. 10. (Aegogropila) escarlatei Profile-view of anisochelae-iv. 11. Sigmas-II. Note anisochelae-iv on the left, and abundant micracanthoxeas all over the picture. 12. Toxa-I (smooth, the preparation is dirty) and sigma-ii. 13. Toxas-II with a thickened (centrotylote-like) central portion. Note micracanthoxeas onthe left. 14. Micracanthoxeas. Scales: 9-11, 13 = 5µm, 12 = 10 µm, 14 = 1 µm. 8

9 while 1885 Carter by of with lost Phillip be sure of the [sensu Gray (1867), and E. parasitica P. (G.) parasitica; here for P. enigmatica), As it is absolute- Mycale (G.) austra- (1885), and BMNH two the G. aus- accordingly, must, names we formally available viz. neotype of the oldest name, tralis. The other M. = junior synonym], conspecificity, consider to propose enigmatica their S (Melbourne). Carter = a validity of both G. = [sensu is conspefrom originates (as to as P. type there too, Heads large dried specimen. a ly impossible australis also the Although is by Carter (1886) Port BMNI I which is (1886)]. which it, viz. Australia, This is opt improperly renamed Pseudoesperia identified specimen (1885) Carter, Esperia parasitica [the type species, specimen lis of description enigmatica cific is known from Carter's Pseudoesperia original be considered junior synonyms Mycale australis Hentschel, a valid species 18-30, Material Rosario studied. low-orange, on 74 08'W, Figs Mycale (Aegogropila) anisochelae-ii. 15. (INV-POR 448). escarlatei n.sp. Profile Paratype (MCN 2829). 17. Paratype (INV-POR 449). view of Scales = 5 coll. of the coral under coll. Comparative Kielman, 28.V.1990, Schizotype ZMA material: Porites porites. and and M. Kielman, slides Santa 10 m from the UFRJPOR Mycale (G.) australis, (as Pseudoesperia Marta, deep, blue, holotype neotype, BMNH enigmatica); P. enigmatica [sensu Dendy (1896)], NMV-F65700, NMV-F65701, REMARKS NMV-F65702,NMV-F65703, NMV-F65704; Esperella (Including the naming of a neotype for Grapelia australis.) The junior rina ca type-specimens of Grapelia synonym both assumed to Pseudoesperia be lost and its Carter, possible 1885 are Whitelegge, 1906, var. carteri (Valentine, in litt). Grape- is known from the single spicule of a hand-drawing by Bowerbank (1864, Fig. specimen from Fremantle, SW 135; of arenosa a out Australia), P. BMNH M. Durban, S. lost, 1911, holotype, ZMH-S1666; 1924, trichophora Dendy and 150; M. Africa, unpublished), slide holotype, & only); M. (G.) Vacelet BMNH (det. 1924, microscopic et al., parasitica (det. sp. P. BMNH Frederick, (original 149 anco- AMS-Z1440; M. parasiti- (G.) parasitica (sensu MNHN D.WL (microscopic is Frederick, & ; only); type Hentschel, Dendy holotype, slides lia australis del deep, yel- m substrate; INV.POR deep, blue, rocky M , Chengue Bay (11 20'N- Marta, Colombia), m substrate. deposited µm. 10 Islas INV-POR 450, 430, Chengue Bay (11 20'N-74 08'W, rocky status n.sp. coll. S. Zea, , dead bottom parts Santa Colombia), Paratype 16. unguifera Holotype: INV.POR 4.VII.1991, aus- Table 2. (Colombia), Paratypes: (Grapeha) until its is confirmed. Mycale (Grapelia) Figs. it renaming M. macilenta (as of M. junior homonym a We refrain from tralis. as is a.) var. 1976), Burton, Van Soest, Sey- 9

10 Fig. 18. Mycale (Grapelia) unguifera n.sp., holotype in situ. Scale = 1 cm. (INV-POR 450), Figs Mycale (Grapelia) unguifera n.sp., holotype (INV-POR 450). 19. Transverse section showing general architecture, with very dense and confused choanosomal skeleton (cho), and abundant subectosomal divergent brushes (br). 20. Tangential view ofsurface showing low density of subtylostyles strewn at random, and abundant anisochelae-i and sigmas. 21. Subectosomal area showing abundance of rosettes of anisochelae-i and patches of sigmas. 22. Detail of rosette of anisochelae-i. Scales: 19 = 500 µm, 20 = 100µm, 21 = 50 µm, 22 = 20 µm. 10

11 Microscleres: Table 2. Spicular micrometric data on the holotype and paratypes of Mycale (Grapelia) unguifera sp.n.. N = 20. subtylostyles anisochelae sigmas length by width of head length by width width by of shaft INV-POR : Islas del Rosario by II: by by III: INV-POR ^392 I: ^61 Santa Marta by II: by III: by INV-POR I: Santa Marta by II: by by III: chelles, unpublished), ZMA 10711; M. parasitica (det. anisochelae each) are very rare. Carpay, Tasmania, unpublished), ZMA Spicules (Measurements in Table 2). Megascleres: subtylostyles (Fig. 23), fusiform, thicker at Diagnosis: Mycale (Grapelia) unguifera n.sp. possesses sigmas longer than 50 ±m, anisochelae-ii mar- central portion, elliptic head, apex sharpening kedly larger than anisochelae-iii; megascleres gradually. anisochelae-i (Figs. 24- with mean length over 350 pm. 25), unguiferate head < 20% of shaft's total length, lateral alae of head very narrow and DESCRIPTION Holotype fragile, slimy crust, sharp, biphid frontal ala, with deep notch between two narrow and sharp projections, shaft with smooth surface (Fig. 18). Colour in life yellow-orange. Paratypes blue in life. INV-POR 429 like a bulgy bladder, composed of two main fragments (ca. 15 markedly curved on profile view, palmate foot with spines on top of alae; anisochelae-ii (Figs ), acanthose head ca. 15% of shaft's total by 9 by 4 and 4 by 4 by 2 mm). INV-POR 430 length, alae of head not recognizable, head with composed of four main fragments, the largest 4 larger and smaller spines, slender shaft markedly by 4 by 2 mm. Both are firm, but compressible. curved on profile view, palmate foot with large Ectosomal skeleton a thin tangential layer of spines on top of alae; anisochelae-iii (Fig. 29), subtylostyles intermingled with abundant head 40% of shaft's total length, lateral alae of anisochelae-i (single or in rosettes, ca. 105 (lm in diameter, anisochelae each) and sigmas (Fig. 20). Choanosomal skeleton markedly confused, with dense spiculation and tracts of megascleres head longer than frontal one, slender shaft slightly curved on profile view, foot with reduced alae and large basal spur-like projection; sigmas (Fig. 30), stout, bent at middle portion, markedly bent on hooks, sharp hooks. not easily discernible (Fig. 19). Anisochelae-I, strewn at random or organized in rosettes, and sigmas abound in subectosomal area (Figs. 21- Etymology. The species is named after the unguiferate shape of its anisochelae-i. 22) among brushes of subtylostyles, perpendicular to, and slightly piercing the surface. Rosettes REMARKS of anisochelae-ii (ca. 60 flm in diameter, The occurrence of slightly larger and thicker 11

12 Figs Mycale (Grapelia) unguifera n.sp., holotype (INV-POR 450). 23. Terminations of subtylostyles. 24. Profile view of anisochela-i with a curved shaft, unguiferate head, and base which is microspined on its top portion. 25. Detail of base of the anisochela-i of Fig. 24 showing spines on its top portion. 26. Profile view of anisochela-ii with a curved shaft, unguiferate head, and base which is acanthose on its top portion. 27. Detail of unguiferate head of the anisochela-ii of Fig. 26 showing larger-spines (= alae?), with smaller ones in between. 29. Detailof spined base ofthe anisochela-ii of Fig. 26. Note size-variation of spines. Profile view of palmate anisochela-iii, with reduced base and basal spur-like projection (sp). 30. Sigma. Scales: 23-24, 30 = 10 µm, 25-26, 29 = 5 µm, = 2 µm. 12

13 spicules in the holotype from the Islas del Ro- pled with the rarity of centrotylote-sigmas and sario on the southern Colombian coast, as compared to both paratypes from the Santa Marta area, conforms to the pattern described by Zea & Van Soest (1986), and Zea (1987). According to these authors more than 50% of the sponge species occurring on the southern Colombian Caribbean coast possess stouter siliceous spicules /or -toxas in the Poecilosclerida makes it likely that the scattered occurrence of this character is independently achieved, thus constituting several distinct apomorphies. Small micracanthoxeas with disproportionately large spines as found in the, as yet, few species of Mycale discussed above also seem unrelated to when compared to specimens from other more other similar microscleres occurring in the Poeci- northerly localities along the Colombian coast, and the Caribbean in general. Of the comparative material of Grapelia studied here, only M. (G) carteri (Dendy & Frederick, losclerida. Some Cladorhizidae De Laubenfels, 1936 may possess microspined (sub)(tylo)styles (see Lundbeck, which 1905), are larger and generally contorted. These are markedly distinct in 1924), the specimen described by Vacelet et al. overall shape, and function (they form a surface (1976), as M. (G.) parasitica, and two undescribed specimens from the western Indian Ocean, from Durban (South Africa) and from the Seychelles, possess sigmas. These are generally smaller (26-45 im long, partly new micrometries; in Hajdu, prep.) than the ones reported here for M. (G.) crust), from the micracanthoxeas of Mycale. The genera of uncertain affinities Phlyctaenopora Topsent, 1904 [synonym Barbozia Dendy, 1922; sensu Lévi & Lévi (1983)], and Discorhabdella Dendy, 1924 may possess aster-like microxea with two semiterminal whorls of large spines, very much unguifera n.sp. (50-69 (lm long). The specimen unlike the pattern observed in Mycale. Neofibularia Hechtel, 1965 possesses microspined microxea, but these are larger, and spines are smaller. The from the Seychelles was blue alive, and has sigmas which are im long, thus coming closer to the new species (both paratypes were blue well known onychaetes of Tedania Gray, 1867 also alive). Nevertheless, it shows at least two additional distinctive characters. The first, a trait differ markedly from the micracanthoxeas reported here because of their much larger size, shared with some other specimens from the smaller spines, and occasional centrotylote Indo-west Pacific-anisochelae-III which are larger than the anisochelae-ii. The second, anisochelae-i almost two times as as those long in M. (G.) unguifera n.sp.. The new is species thus shape. the lack of affinities to other known Despite microscleres mentioned microspined above, the occurrence of micracanthoxeas in Mycale could be uniquely derived, although probably on a considered well distinguished from other congeners. more inclusive level of universality. This occurrence could be an underlying synapomorphy for distinct subgenera or species-groups, rather than PHYLOGENETIC DISCUSSION be interpreted as suggestive of the close affinities of the three species possessing The phylogenetic significance of centrally-thick- them alone. Indeed, despite their likely artificial assignment to ened, centrotylote-like toxas and micracanthox- distinct subgenera, M. micracanthoxea and M. bam- eas as found here in M. escarlatei is uncertain. fieldense seem very close with respect Hajdu (1994) described centrotylote-sigmas from to anatomical characters other than their ectosomal skele- Hamacantha popana (De Laubenfels, 1935), but tons (Table 1). On the other hand, M. escarlatei these are probably only analogous to the centrally-thickened toxas reported here, since both spicule-types (sigmas and toxas) diverged very early. These spicules are synapomorphic for seems far from these species as suggested by anatomical characters other than the micracanthoxeas. The assessment of the true level of universality of the occurrence of micracanthoxeas Haplosclerida and Poecilosclerida (e.g. Van will ultimately depend on a taxonomic revision Soest, 1991; Hajdu et al., 1994b). This fact, cou- of known species of Mycale, since it is quite possi- 13

14 50% uncertain affinities, for the moment considered best assigned to the Mycalidae (Van Soest & Stentoft, To make 1988). the of appearance anisochelae with a basal spur-like projection a single event in mycalid evolution would imply accepting a close relationship of Phlyctaenopora to subgenera Mycale and The halichondroid choanosomal pattern Grapelia. and the ectosomal ble that the character has been overlooked in the past. The observation that M. micracanthoxea and M. bamfieldense, despite distinct with to respect the pattern of their ectosomal skeletons (the first without, the latter do share a with), comparable set of spicules is suggestive of further pitfalls in the use of ectosomal patterns as the sole diagnostic character for Mycale subgenera. Supporting specialization in the form of a second category of evidence comes from Hajdu & Desqueyroux- diactinal megascleres make it very unlikely that Faundez (1994) who noted that although agree- Phlyctaenopora could be so closely related to My- ing in most characters to sponges assigned to cale. It seems pointless to say that Iophon and subgenus Mycale, M. (M.?) lapidiformis (Ridley & Acanthorhabdus also do not share additional fea- Dendy, 1886) did not possess the characteristic tures with Mycale, other than the shape of their tangential confused ectosomal skeleton of other species assigned to the subgenus (sensu Van anisochelae. The conclusion that follows from Soest, 1984). Additionally, M. (A.?) n. sp. (Hajdu the rationale above is that despite their morphological similarity [primary homology, Pinna & Riitzler, in prep.) is described with an unstable (1991)], distinct occurrences of anisochelae with a basal spur-like projection may well constitute tangential ectosomal reticulation of spicule bundles - of reported specimens had it (2 out of several independent achievements by variably 4), the others did not (being Carmia-like). All distantly related poecilosclerid lineages. these observations when taken in conjunction are interpreted here as suggestive that patterns of the ectosomal skeleton may have been often lost. ACKNOWLEDGEMENTS This points towards the unnatural status of Carmia as an artificial assemblage This article benefited from the comments of N. of several convergent evolutionary lines. Begle (1991) stated that absences (such as the ectosomal skeletal pattern of Carmia ) are as useful as characters as presences. However, such features are, according to him, even more dependent of an overall phyloge- Boury-Esnault (SME), K. Riitzler (USNM), R.W.M. van Soest and F.R. Schram (ISP-UvA). Authors are thankful to L.B. Holthuis (NNM) for clarifying doubts concerning the status of Grapelia and Pseudoesperia, and to P.B. Berents (AMS), M. netic analysis before they can be evaluated as Dzwillo (ZMH), M. Grasshoff (SMF), K. Joysey synapomorphic, since their 'astructural' nature (UMZC), Y.C. Man (ZRC), T. Stranks (NMV), prevents the formulation of any hypothesis concerning primary homology (Pinna, 1991). Anisochelae-III with a basal spur-like projection as found here in M. unguifera, occur in other species of Mycale, subgenera Mycale and Grapelia C. Valentine (BMNH) and R. Wood (DES) for the loan of, and/or search for specimens. EH is thankful to G. Muricy and S. Bonnecker (UFRJ) for providing laboratory facilities during field activities in Brazil which led to the collection of (Hajdu & Boury-Esnault, 1991; Hajdu & Des- the holotype of M. (A.) escarlatei sp.n., and to queyroux-faûndez, 1994). They are also charac- Conselho Nacional de Desenvolvimento Cienti- teristic of Iophon Gray, 1867, Phlyctaenopora and fico e Tecnolôgico (CNPq, Brazil) for providing a fellowship in support of his PhD studies in the Acanthorhabdus Burton, As far as morphological similarity goes, hardly any distinction can University of Amsterdam, and funding of field be made between these anisochelae, except per- activities in Brazil in INVEMAR provided haps that those in Phlyctaenopora seem stouter, and field and laboratory facilities to SZ and MK. those in Iophon, more regular. But this view could COLCIENCIAS (Colombia) and WOTRO (The be biased by our greater familiarity with the ones occurring in Mycale. Phlyctaenopora is a genus of Netherlands) funded, respectively, the research of SZ and MK. 14

15 REFERENCES suisse Zool., 101 (3): BEGLE, D.P., Relationships of the osmeroid fishes HAJDU, E. & K. RUTZLER, in prep.. Sponges, Genus Mycale(Poecilosclerida) from a Caribbean mangrove and the use of reductive characters in phylogenetic analysis. Syst. Zool., 40 (1): BERGQUIST, P R., The sponges of Micronesia, and critique of subgeneric classification. HAJDU, E., SOEST, R.W.M. VAN &J.N.A. HOOPER, Part 1. The Palau Archipelago Pacific Sci, 19 (2): Proposal for a phylogenetic subordinal classification of poecilosclerid sponges (Demospongiae, Porifera). In: R.W.M. van Soest, Th.M.G. van Kempen &J.C. BOWERBANK, J.S., A monograph of the British Braekman (eds.), Sponges in time and space, Proc. 4th Spongiadae. Vol I. Ray Society, London, i-xx, 1-290, pis. I-XXXVII. BUIZER, B. & R.W.M. VAN SOEST, Mycale micracanthoxea nov. spec. (Porifera, Poecilosclerida) from the Netherlands. Neth.J. Sea Res, 11 (3/4): BURTON, M., Porifera. Part 2. Antarctic sponges. Nat. Hist. Repts. British Antarctic 'Terra Nova' Exped. 6 (4): int. Porifera Congr.: Balkema, Rotterdam. 515 pp. HAJDU, E., WEERDT, W.H. DE & R.W.M. VAN SOEST, Affinities of the "Mermaid's Glove" sponge Isodictya palmata with a discussion on the synapomorphic value of chelae microscleres. In: R.W.M. van Soest, Th.M.G. van Kempen &J.C. Braekman (eds.), Sponges in time and space, Proc. 4th int. Porifera CARTER, H.J., Descriptions of sponges from the Congr.: Balkema, Rotterdam. 515 pp. neighbourhood ofport Phillip Heads, S. Australia. Ann. HALLMANN, E.F., A revision of the monaxonid Mag. nat. Hist., (5) 15 (X): , pl. IV. species described as new in Lendenfeld's "Catalogue of CARTER, HJ., Supplement to the descriptions of the sponges in the Australian Museum." Part III. Proc. Mr. J. Bracebridge Wilson's Australian sponges. Ann. Mag. nat. Hist., (5) 18 (XLII): , pl. X. DENDY, A., Catalogue of non-calcareous sponges collected by J. Bracebridge Wilson, Esq., M.A., in the neighbourhood ofport Phillip Heads. Part 2. Proc. roy. Soc. Vic., (n.s.), 9: DENDY, A., Report the on Sigmatotetraxonida collected by H.M.S. Sealark in the Indian Ocean. In: Re- Linn. Soc. N.S.W., 39: HECHTEL, G., A systematic study of the sponges of Port Royal, Jamaica. Bull. Peabody Mus. nat. Hist., 20: i-iv, HENTSCHEL, E., Tetraxonida. 2 Teil. In: W. Michaelsen & R. Hartmeyer (eds.), Die Fauna Sudwest- Australiens, 3 (10): Gustav Fischer, Jena. HENTSCHEL, E., Kiesel- und Hornschwamme der ports of the Percy Sladen Trust Expedition to the Aru und Kei-Inseln. Abh. senckenb. naturforsch. Ges., Indian Ocean in Vol. 7. Trans. Linn. Soc. Lond., 1912: Zool., 18: 1-164,pis HOOPER, J.N.A. & C. LÉVI, Biogeography of DENDY, A., Porifera. I. Non-Antarctic sponges. Brit Indo-west Pacific sponges: Microcionidae, Raspailiidae, Antarctic ('Terra Nova') Exped Nat. Hist. Axinellidae. In: R.W.M. van Soest, Th.M.G. van Repts., Zool., 6 (3): , pis. Kempen &J.C. Braekman (eds.), Sponges in time and DENDY, A. & L.M. FREDERICK, On a collection space, Proc. 4th int. Porifera Congr.: of sponges from Abrolhos Islands, western Australia. J. Linn. Soc. (Zool.), 35: , pis Balkema, Rotterdam. 515 pp. LABOREL, J., Les peuplements de madréporaires GRAY, J.E., Notes on the arrangement of sponges, des côtes tropicales du Brésil. Ann. Univ. Abidjan, (E), with the description of some new genera. Proc. zool. 2 (3): Soc. Lond., 1867: , pis. XXVII-XXVIII. LAUBENFELS, M.W. DE, Some sponges of lower HAJDU, E., California (Mexico). Am. Mus. Novit., 779: A phylogenetic interpretation of hamacanthids (Demospongiae, LAUBENFELS, M.W. DE, 1936 A discussion of the Porifera), with the redescription ofhamacanthapopana. J. Zool., 232: HAJDU, E. & N. BOURY-ESNAULT, Marine Pori- sponge fauna of the Dry Tortugas in particular and the West Indies in general, with material for a revision of fera of Cabo Frio (Rio de Janeiro - Brazil). The family the families and orders of the Porifera. Pap. Tortugas Mycalidae Lundbeck, 1905, with the description of a new species. Revta. Brasil. Biol., 51 (3): Lab, 30: LÉVI, C. & P. LÉVI, Démosponges bathyales HAJDU, E. & R. DESQUEYROUX-FAUNDEZ, A récoltées par le N/O 'Vaubarv au sud de la Nouvelle- Calédonie. Bull. Mus. natn. Hist. nat. Paris, (4) 5 (A, 4): synopsis of South American Mycale (Mycale) (Poecilosclerida, Demospongiae), with the description of three new species and a cladistic analysis of Mycalidae. Rev. LUNDBECK, W., Porifera (Part 2). Desma- 15

16 cidonidae. Dan. IngolfExped. 6: SOLLAS, On the sponges collected during the MATSUURA, Y., Contribuiçào "Skeat Expedition" to the Malay Peninsula, ao estudo da estrutura oceanografica da Regiâo Sudeste entre Cabo Frio Proc. zool. Soc. London, 11: , pis (RJ) e Cabo de Santa Marta Grande (SC). Ciência e Cultura, 38 (8): THOMAS, P.A., Studies on Indian sponges - III. Two species of silicious sponges of the family MOTHES, B., SOEST, R.W.M. VAN & G.S. CARVA- LHO, Biogeography ofbrazilian marine sponges (Porifera, Demospongiae). In Book of Abstracts. 4th Int. Ophlitaspongiidae De Laubenfels (Class: Demospongiae Sollas, Order: Poecilosclerida Topsent). J. mar. biol. Ass. India, 10 (2): Porifera Congr. University of Amsterdam, Amsterdam. TOPSENT, E., Spongiaires des Açores. Résuit. PINNA, M.C.C. DE, Concepts and tests ofhomology in the cladistic paradigm. Cladistics, 7: REISWIG, H.M. & H. KAISER, Description of Mycale bamfieldense Poecilosclerida) from Vancouver Island, British Columbia. Can J. Zool., 67: n.sp. (Porifera, Demospongiae, Camp, scient. Prince Albert I, 25: 1-280, pis. I-XVIII, [ ], VACELET, J., P. VASSEUR & C. LÉVI, Spongiaires de la pente externe des récifs coralliens de Tuléar (sud-ouest de Mém. Mus. natn. Madagascar). Hist. nat. Paris, (A, Zool.) 99: 1-116, pis. I-X. RIDLEY, S.O. & A. DENDY, Preliminary report on VALENTIN, J.L., the Monaxonida collected by H.M.S. 'Challenger'. Analyse des paramétres hydrobiologiques dans la remontée de Cabo Frio, Brésil. Mar. Ann. Mag. nat. Hist., (5) 18: , Biol., 82: RÛTZLER, K., Sponges in coral reefs. In: D.R. WHITELEGGE, T., Sponges. Part I. In: Scientific Stoddart & R.E. Johannes (eds.), Coral reefs: research Results of the Trawling Expedition of H.M.C.S. methods. Monogr. oceanogr. Meth. (UNESCO), 5: "Thetis" off the coast of New South Wales in February and March Mem. Austral. Mus. IV (9): , SOEST, R.W.M. VAN, Marine sponges from Curaçao and other Caribbean localities. Part III. Poecilosclerida. Stud. Fauna Curaçao Caribb. Isls., 66 (199): pis. XLIII-XLTV. ZEA, S., Esponjas del Caribe Colombiano. Catâlogo Cientifico, Bogota: ZEA, S., Cover of sponges and other sessile organ- SOEST, R.W.M. VAN, Demosponge higher taxa isms in rocky and coral reef habitats of Santa Marta, classification re-examined. In: J. Reitner & H. Keupp Colombian Caribbean Sea. Car. J. Sci., 29 (1/2): 75- (eds.), Fossil and recent sponges: Springer-Verlag, Berlin and Heidelberg. 595 pp. ZEA, S. & R.W.M. VAN SOEST, Threenew species SOEST, R.W.M. VAN, Demosponge distribution of sponges from the Colombian Caribbean. Bull. mar. patterns. In: R.W.M. van Soest, Th.M.G. vankempen Sci., 38 (2): & J.C. Braekman (eds.), Sponges in time and space, Proc. 4th int. Porifera Congr.: Balkema, Rotterdam. 515 pp. SOEST, R.W.M. VAN & N. STENTOFT, Barbados deep-water sponges. Stud. Fauna Curaçao Caribb. Isls., 70(215): Received: September,