Renewed Study of the Type Material of Pa/aeospongil/a chubutensis Ott and Volkheimer (1972) C. VOLKMER-RlBEIRO and J. REITNER Introduction The first fossilized gemmule-bearing freshwater sponge was described by Ott and Volkheimer (1972) from the Patagonian Lower Cretaceous (Aptian) at the Chubut River valley, in Argentina. However, an accurate reading ofthe original description shows that the photograph ofthe gemmules presented in Fig. 3 does not show the thick gemmular wall and the particular arrangement of gemmos- cleres shown in Fig. 5. Also the drawings ofthe gemmoscleres shown in Fig. 6 do not resemble the gemmoscleres photographed in Fig. 5. In a second paper (Volkheimer and Ott 1973), the authors presented a study ofthe fauna and flora found associated with P. chubutensis and concluded that the paleohabitat was a lacustrine one subjected to strong seasonal droughts. This second paper proposed a phyleticrelationship of P. chubutensiswith the extantgenus Spongilla and, more particularly, with Spongilla alba (Carter 1849). Soon after, Racek and Harrison (1975) proposed a strong relationship of Palaeospongilla with the extant genus Radiospongilla. Only drawings were presented and reference was made to a deposition of the preparations in the Australian Museum (Sydney) and the Smithsonian Institute (Washington). Based on these discrepancies between the illustrations and the phylogenetic ideas given by the original authors and by Racek and Harrison, the present authors decided to undertake a renewed study of these very important fossil materials, which represent the only entirely pre- served freshwater sponges. Material and Methods The holotype material of P. chubutensis and two para types are deposited at the Museo Argentino de Ciencias Naturales Bernardino Rivadavia (Buenos Aires, Argentina). Three pieces of a further paratype are housed with the collection of the Bayerische Staatssammlung für historische Geologie und Paläontologie (Munich, West Germany). One thin section of a paratype (no. Z 4748) is depo- sited at the Australian Museum (Sydney, Australia). The thin sections, which were made from a topotype (Racek and Harrison 1975) could not be found at the Smithsonian Institute or at the Australian Museum. New thin sections and SEM preparations were made from the para type deposited in Munich. The J. Reitner and H. Keupp (Eds.) Fossil and Recent Sponges © Springer-Verlag Berlin Heidelberg 1991 122 C. Volkmer-Ribeiro and J. Reitner thin section loaned from the Australian Museum was throughly restudied and photographed. The SEM studies were carried out by the junior author with a Cambridge 360. New preparation methods were used to figure out the spicular skeleton and preserved soft parts as weIl as preserved gemmules. The probes were etched for 3 h with 5% Titriplex-III-solution (ClOH14N2018 *2H20) (Re"itner and Engeser 1987). This method permits etching very fine and sensitive stnictures, e.g., smaIl gemmoscleres and acanthose microscleres, which are preserved in mono- or paucicrystalline calcite. Under the SEM, backscatter technique (secondary electrons) was used . This method makes it possible to distinguish irregularities on the surface structures caused by special contrast and therefore gives a much better view of spiny structures. Description of the Speeres 1. Palaespongilla chubutensis Ott and Volkheimer The renewed study ofthese fossil specimens revealed that the original description of Ott and Volkheimer and also the redescription presented by Racek and H1;lrrison (1975) app1y to an association of two genera. The occurrence of such associations is a common fact among ex ta nt freshwater sporiges (Volkmer- Ribeiro and de Rosa-Barbosa 1972; Racek 1969). Part of the material described as Palaeospongilla chubutensis is now syn- onymized under Spongilla patagonica n.sp. The main part ofthe sponge material is Palaeospongilla chubutensis sensu strictu. Palaeospongillidae new fami1y Type genus : Palaeospongilla Ott and Volkheimer 1972, as presently redefined. Diagnosis: Fossil freshwater sponges with gemmules devoid of gemmoscleres but with a thin inner coat and a thin pneumatic coat. Gemmules in groups or isolated and protected by an outer armature of tangentially arranged megascleres. Palaeospongilla Ott and Volkheimer 1972, redefined Palaeospongilla Ott and Volkheimer 1972, p. 53 (partim) Type species : Palaeospongilla chubutensis Ott and Vo1kheimer 1972, as presently redescribed. Diagnosis: Sponge forming thick crusts. Gemmu1es without gemmoscleres, with a conspicuous inner coat and thin pneumatic coat. Foramen devoid of a porous tube. Gemmules single but more usually packed in groups oftwo or several, each package or each single gemmule protected by an outer envelope of tangentially arranged megascleres. Megascleres sparsely spined and curved oxea. Micros- cleres not detected. Palaeospongilla chubutensis Ott and Volkheimer 1972 (Figs. 1; 2a,b; 3a,b) Renewed Study of the Type Material of Palaeospongilla chubulensis 123 Fig.l. Para type of Palaeospongilla chubutensis deposited a t Museo Bernardino Rivadavia , Buenos Aires. Arrow points to the sponge fossilized crust. Scale = I cm Palaeospongilla chubutensis (OU and Volkheimer 1972; p. 53 partim and Figs. 2,3,4) Holotype: As originally designated (Ott and Volkheimer 1972, Fig. 2) Paratypes: As originally designated (Ou and Volkheimer 1972, p. 53) Redescription Sponge forming overlaying crusts around the sterns of aquatic plants and plant remains (Fig. 1) (OU and Volkheimer 1972, Fig. 2). Skeleton consists of main ascending skeletal fibers and of secondary thick fibers placed a t right angles to main fibers thus composing a reticulum ofsquare meshes (OU and Volkheimer 1972, Fig. 3). MegascJeres : Stout, straight to feebly curved, sparsely, abruptly pointed oxea (length, 230-260 /Lm ; width, 15- 18 /Lm) . MicroscJeres: not detected; GemmoscJeres: absent Gemmules: As described in the generic definition. Diameters : 483- 644/Lm Distribution: Hitherto known only from type loca lity (valley of the Chubut River, 16 km from locality ofCerro Condor, Chubut Province, Argentina) Age: Patagonian lower Cretaceous (Aptian), Chubut Group (OU and Volk- heimer 1972, p. 53) Preserved Reproductive Characters This exceptionally well-preserved fossil allows study first of the reproductive process of a Cretaceous freshwater sponge. The preserved oocytes, oocyte for- 124 C. Volkmer-Ribeiro and J. Reitner Fig.2. a Gemmule formation in P. chubUlensis photographed from the thin section deposited at the Austra lian Museum. To the right coa ts a lready formed , to the left inner coat in th e form ing, a rea of the foramina l tu be last to be formed. Notice spongocytes groupped around the coats. Scale = 250 ,.,.m . b Deta il of right gemmule of a . Notice inner gemula r coa t, pneumatic coat and the tangentia lly arranged megascleres of the common envelope. Scale = 125 ,.,.m Renewed Study of the Type Material of Pa/aeospongilla chubutensis 125 mation, and gemmule formation are comparable with extant freshwater sponges. The study of the thin section of Palaeospongilla chubutensis deposited in the Australian Museum (Sydney) revealed that the sponge was forming gemmules after aperiod ofoocyte production (Fig. 2a,b). One mature oocyte (Fig. 3a,b) can be seen side by side with several empty follicles. The cytological characteristics of this oocyte and of the gemmules of Palaeospongilla chubutensis are listed in the legends for the figures (Fig. 2a,b ; 3a,b) and generally conform with the literature on this subject (Simpson 1984). The oocytes of Palaeospongilla chubutensis share a large number ofidentities with the oocytes of Tetractinomorpha according to the descriptions and illus- trations of Liaci et al. (1971). The oocyte was fossilized at the moment when phagocytosis was finished and a follicle epithelium was already isolated from the mesohyle cells (synvivodiagenetic). At this moment nuclear extrusion were taking place and a second nutrition al phase to the oocyte was starting with forming of a crown of nurse cells. The nuclei of such nurse cell are conspicuously placed close to the follicle wall and the cell processes are already orientated towards the oocyte surface (Fig.3b). Diameter of the oocyte plus follicle: 330 /Lm; diameter of the nucleus: 133 /Lm. The great number of empty follicles and young gemmules, suggest an alternation of reproductive processes with drastic reduction of the time interval which elapses between seXual reproduction and gemmulation in extant fresh- water sponges of temperate regions (Frost et al. 1982). Three hypotheses can tentatively explain this fact: 1. The sponge was living in a temperate region but in a habitat seasonally submerged for short periods. 2. A sudden drought occurred which induced precocious gemmulation. 3. The sponge was living in a tropical habitat subjected to seasonally long periods of drought. Discussion Palaeospongilla chubutensis displays characteristics not seen in any other pre- sently known extant freshwater sponges. For this reason, a new family is proposed to contain this monotypic fossil genus. Among the extant freshwater sponges only the Spongilla aspinosa/ Spongilla inarmata-Spongilla spoliata group of species (Volkmer-Ribeiro and Maciel1983) have the same common character as seen in Palaeospongilla chubutensis, that gemmules have no gemmoscleres and are protected by a capsule of normal megascleres. However, in the three extant species each gemmule has its own capsule, wh ich is in extreme close contact with the inner gemmule wall, so that some megascleres could be embedded in the gemmule wall. These capsules also exhibit no distinctive particular arrangement of megascleres which are woven in disorderly fashion around the gemmule (cf. Fig. 2 in Volkmer-Ribeiro and Traveset 1987). The inner gemmular coat ofthese extant species is quite thin and distinguishing three layers, including a pneumatic Renewed Study ofthe Type Material of Palaeospongilla chubutensis 127 coat is difficult. A very conspicuous foraminal collar is also present in the gemmules of the three extant species. The fossil freshwater sponge has no microscleres, which are present in Spongilla. The only microscleres within the described sponges are related to a new species, Spongilla patagonica n.sp. (Fig. 5a,b). Also remarkable is the localization of the gemmules of Palaeospongilla chubutensis from base to the top layer ofthe sponge, whilst in the extant Spongilla species the gemmules are usually located close to the sponge surface. The three Spongilla species form thin skeletal fibers wh ich support waving branches and pierce the sponge surface as hispid projections. Palaeospongilla chubutensis, on the contrary, seems to have ftat, thick, folia ted crusts with lobose contour and a smooth surface (Fig. 2 in Ott and Volkheimer 1972) 2. Spongilla patagonica n.sp. Family Spongillidae Gray (1867), sensu Volkmer-Ribeiro (in press) Spongilla Lamarck (1816) sensu Penney and Racek (1968) Palaeospongilla Ott and Volkheimer 1972, p. 53 (partim) Spongilla patagonica n.sp. (Figs. 4a,b; 5a,b) Palaeospongilla chubutensis Ott and Volkheimer 1972; p. 53 (partim, Fig. 5) Palaeospongilla chubutensis Racek and Harrison 1975; p. 158 (partim) Derivatio nominis: After Patagonia, alandscape in Argentina Holotype: The holotype is located on the para type specimen of Palaeospongilla chubutensis which is deposited within the Bayerische Staatssammlung für His- torische Geologie und Paläontologie, Munich. SEM preparation deposited in the Institut für Paläontologie der Freien Uni- versität Berlin IPFUB/JR/89; (Figs. 4a,b; 5a,b). Description Megascleres slim, gradually sharp pointed, slightly curved, smooth oxea (length 210-260 /Lm, width 7-10 /Lm) . Microscleres: usually straight, spined oxea; some slightly curved oxea occur also (length 100- 110 /Lm, width 9-10 /Lm) . Gemmoscleres: short, slightly curved to strongly bent, rarely straight, strongly ... Fig.3. a Maturingoocyte of P. chubutensis(thin section deposited at the Australian Museum). Notice surrounding folicle ephitelium and the collagen matrix starting to form outside the epithelium. Nucleolus displaced from the sectioning plane. Scale = 250 /Lm . b Detailed magnification ofa. Notice dense material being extruded from the nucleus and the lastly phagocytized microgranular cells aligned midway between the follicle epithelium and the nuclear extrusions. The collar of nurse (spherulous) cells·starting to differentiate, their nuclei are seen bounding the epithelium cells and their processes are already oriented towards the egg surface. Scale = 125 /Lm 128 C. Volkmer-Ribe iro a nd J . Reitner Fig.4. SEM micrographs o f Spongil/a paragonica n.sp. aBollom arrolV points to one gemmule with gemmoscleres in situ : left arrow points to microscleres : righr arrow indica tes the thinner magascleres of S. palagonica placed in close contact with the stout megascleres of P. chubulensis. b Same gemmule as a a t la rger magnifica tio n: th e arrow indica tes the place occupied by the long fo ra mina l nipple ; the gemmule is seen to have had a conspicuous inner coa t a nd a la rge pneuma ticcoa t where gemmoscleres we re ra ndomly embedd ed Renewed Study of the Type Material of Palaeospongilla chubulensis 129 Fig. 5. a Same gemmule as Fig. 4a and b, showing details of the gemmular caats and of the gemmoscleres' shape; the two arrows indicate the place occupied by the inner gemmular coat. b Spongilla palagonica n.sp.; Detailed magnification of the area indicated by the left arrow in Fig. 4a shows the megascleres of P. chuburensis in the upper pari and tho e of S. palagonica n.sp. in the middle and bollom parIs ofthe photograph ; alleft grouping of microscleres of S. palagonica n.sp. 130 C. Volkmer-Ribeiro and J. Reitner spined, cylindrical amphistrongyla, some spines larger and recurved, extremities of the scleres with a starlike profile (length 70-80 1Lffi, width 8 p.m). Gemmules: spherical, with a thick inner- and a large pneumatic coat. The gemmoscleres are within the pneumatic layer randomly orientated and loosely packed, the tips ofthe spicules sometimes protruding beyond the outer surface of the gemmule coat. Foramen is single, nipple-like, extending to the surface ofthe pneumatic coat. The gemmule ofFig. 4b has a diameter of 469 p.m. The inner coat is 20 lLffi and the pneumatic coat is 100 p.m thick. Distribution and geological age: Same as for Palaeospongilla chubutensis. Discussion On exhaustive microscopic scrutiny and SEM preparation of material isolated from one para type (Munich specimen) of Palaeospongilla chubutensis, two gemmules could be detected wh ich are completely different from those of Palaeospongilla chubutensis. The gemmule coatings and gemmoscleres (Figs. 4a,b; 5a,b) are almost identical to those ofthe extant Spongilla alba Carter 1849. The gemmoscleres are identical with the ones illustrated by Ott and Volkheimer (1972, p. 58, Fig. 5), and by Racek and Harrison (1975, p. 160, Fig. 6). SEM preparations were made of gemmules of the extant species Spongilla alba (Fig. 6a-c) from material collected by E.H. Cordero from Jacarepagua, coastal lagoonal area of Rio de Janeiro, Brazil, in June 31, 1940. [The specimen is deposited in the Porifera collection ofMuseu de Ciencias Naturais ofFundacao Zoobotanica do Rio Grande do Sul (MCN), no. 1271]. The geinmules ofthe fossil and the extant species exhibit close affinities ofthe inner coat, the pneumatic coat, the foraminal area, the gemmoscleres, and the gemmosclere embedding within the pneumatic coat. A further characteristic feature of the two species is the possession of microscleres which are longer andthinner than the gemmoscleres (cf. Spongilla alba in Volkmer-Ribeiro and Traveset 1987, p. 227, Fig. 3). The description of a new species is justified by the fact that the fossil material does not allow for a precise perception of the spines covering the microscleres. The fossil species shows dimensions which are sm aller than those shown by Penney and Racek (1968) for Spongilla alba. The observed dimensions agree with those presented by the same authors for Spongilla wagneri, a species placed by Poirrier 1976 into synonymy with Spongilla alba. Spongilla alba is thus phylogenetically closely related to Spongilla patagQnica n.sp: Fig.6. SEM micrograph ofa gemmule ofthe extant Spongilla alba Carter 1849 (from specimen MCN no. 1271). a Cross-section of gemmule; notice inner coat and thick pneumatic coat; gemmoscleres randomly embedded in the pneumatic coat. b Magnification ofthe area indicated by the arrow in a shows scanty spongin gluing the gemmosc1eres together and the thin outer coat. c Detail of the gemmular walls of S. alba; notice gemmosc1eres protruding from the pneumatic coat and the extremely variable shape of the gemmosc1eres' extremities Renewed Study of the Type Material of Palaeospol1f!, i/la chublilel1sis 131 \32 C. Volkmer-Ribeiro and J. Reitner General Remarks Racek and Harrison (1975) rejected the possibility of a spicular mixture of two distinct species. Both fossil species were silicified synvivodiagenetically and exhibit no overlapping growth . Only one kind of gemmule type was recognized. Epizoism is a common feature among extant marine (Ruetzler 1970) and freshwater sponges Racek 1969; Volkmer-Ribeiro and de Rosa-Barbosa 1972). The degree of skeletal development in the associated species depends strongly on which species is the dominant one. In the present case Palaeospongilla chubutensis was the dominant species and the skeletal fibers described by Ott and Volkheimer belong to this species. Spongilla patagonica must have formed ti ny patches of isolated gemmules covered by skeletal fibers soon overgrown by Palaeospongilla chubutensis. The taxonomic review of Palaeospongilla chubutensis proposed by Volk- heimer and Ott (1973) was caused by Racek's advice (p. 459) to the authors as to the dose relationship between their species and Spongilla alba Carter 1849. However, Racek and Harrison (1975) proposed a dose relationship of Palaeo- spongilla chubutensis with the extant genus Radiospongilla Penny and Racek (1968). They considered Palaeospongilla chubutensis astern group from wh ich the genera Radiospongilla and Spongilla evolved, each ofthem inheriting part ofthe double spicular set of Palaeospongilla. The fossil sponge supported Penny and Racek's (1968 p. 63) proposal of a monophyletic origin of all gemmule-producing freshwater sponges, grouped by the authors (op.cit) in the Spongillidae. Radio- spongilla would lead the rodlike structures up to the birotulated gemmosderes on account of the incipient rotules found in the gemmosderes of some species of Radiospongilla. After Marshall's (1883) and Brien's (1967) ideas of a polyphyletic origin for freshwater sponges, Volkmer-Ribeiro (1986, in press) restricted the family Spongillidae sensu Penney and Racek (1968). A large nu mb er of genera con- sidered as spongillids by these authors have been transferred into the families Potamolepidae Brien and Metaniidae Volkmer-Ribeiro (1986). However, the restricted family Spongillidae is still considered by Volkmer-Ribeiro as an artificial grouping of genera deserving further studies. The dose morphological similarity betwen Spongilla alba and Spongilla patagonica n.sp. is linked with the distribution and ecology ofthe taxon Spongilla alba, and Spongilla cenota (Harrison 1974). According to Harrison (1974), Spongilla alba has a circumlitoral distribution in tropical and subtropical areas with a strong tolerance to brackish waters. This ecological preference if extended to Spongilla patagonica fits weU into the paleohabitats inferred by Ott and Volkheirner (.1972) and Volkheimer and Ott (1973) for the fossil fauna and flora associated with Palaeospongilla chubutensis, i.e. , a lacustrine habitat in a Mediterranean climate. Also the fact that both sponge species were forming gemmules at the moment of fossilization agrees with the proposition of a habitat subjected to periodical droughts. Renewed Study of the Type Material of Palaeospongilla chubutensis 133 Acknowledgments. The authors are indebted to Dr. Juan Boto, Invertebrate Department, Museo Bernadino Rivadavia, Buenos Aires, for loan ofthe holotype and the two para types of Palaeospongilla chubutensis. They acknowledge the kind assistance ofDr. F .W.E. Rowe, Porifera and Echinodermata Section, Australian Museum, Sydney, with the loan ofthe thin section of Palaeospongilla chubutensis. We thank Dr. F. J. Collier, Dept. ofPaleobiology, Smithsonian Institute (NMNH), for his interest in looking for the material of Palaeospongilla chubutensis within his museum. They acknowledge the encouragement received from Dr. Nicole Boury-Esnault and Dr. Jean Vacelet for presenting the descriptions ofthe oocyte. The senior author is grateful for funds from the National Research Council of Brazil to support the project (Fellowship no. 306134/76 and Grant no. 409734-88.8). She is also indebted to Dr. H. Keupp and Dr. J. Reitner for funding, wh ich allowed her to participate of the Conference on Fossil and Recent Sponges. References Brien P (1967) Eponges du Luapula et du lac Moero. Result Scient Explor hydrobiol Bangwelo- Luapula Miss JJ Symoens 11, 58: I-52 ) , Frost TM, De Nagy GS, Gilbert JJ (1982) Population dynamics and standing biomass of the freshwater sponge Spongilla lacustris. Ecology 63, 5: 1203-1210 Harrison FW (1974) Sponges (Porifera: Spongillidae) In: Hart CW, Fuller S (eds) Pollution ecology offreshwater invertebrates. Academic Press, New York, pp 29-66 Liaci L, Scisioli M, Matarrese A (1971) La riproduzine sessuale di akuni tetractinomorpha (Porifera). Atti Soc Peloritana Sci Fis Mat Nat 17:235-245 Marshall W (1883) On some new siliceous sponges collected by Mr. Pechuel-Loesche in the Congo. Ann Mag Nat Hist 12 :301- 412 ' Ott E, Volkheimer W (1972) Palaeospongilla chubutensis n.g. et n.sp. Ein Süsswasserschwamm aus der Kreide Patagoniens. N Jb Geol Paläont Abh 140:49-63 . Penny JT, Racek AA (1968) Comprehensive revision ofa worldwide collection offreshwater sponges (Porifera: Spongillidae). Bull Smith Inst US Nat Mus 272: 1-184 Poirrier MA (1976) A taxonomic study of the Spongilla alba, S. cenota, S. wagneri species group (Porifera : Spongillidae) with ecological observations of S. alba. In : Harrison FW, Cowden RR (eds) Aspects of Sponge Biology. Academic Press, New York, pp 203-213 Racek AA (1969) The freshwater sponges of Australia (Porifera: Spongillidae). AustJ Mar Freshwat Res 20:267-310 Racek AA, Harrison FW (1975) The systematic and phylogenetic position of Palaeospongilla chubutensis (Porifera: Spongillidae). Proc Linn Soc NS Wales 99, 3: 157-165 Reitner J, Engeser T (1987) Skeletal structures and habitats of Recent and fossil Acanthochaetetes (subclass Tetractinomorpha, Demospongiae, Porifera). Coral Reefs 6: 13-18 Ruetzler K (1970) Spaiial competition amcing Porifera: solution by epizoism. Oecologia 5:85-95 Simpson TL (1984) The cell biology of sponges. Springer, Berlin Heidelberg New York Tokyo, pp 662 Volkheimer W, Ott E (1973) Esponjas de agua duke delcretacico de la Patagonia. Nuevos datos acerca . de su posicion sistmatica y su importancia paleobiografica y paleoclimatica. Actas Quinto Congr Geol Argent 3:455-461 Volkmer-Ribeiro C (1986) Evolutionary Study ofthe freshwater sponge Genus Metania Gray, 1867: III Metaniidae, New family. Amazoniana IX (Kiel), 493-509 Volkmer-Ribeiro C (in press) A new insight inta the systematics, evolution .and taxonomy of freshwater sponges. In: Rützler K, Hartmann WD (eds) New Perspectives in Sponge Biology. Smithsonian Institution Press, Washington Volkmer-Ribeiro C, De Rosa-Barbosa R (1972) On Acalle recurvata and an associated fauna ofother freshwater sponges. Rev Bras Biol 32, 3 :303-317 Volkmer-Ribeiro C, Maciel SB (1983) N ew freshwater sponges from Amazonian waters . Amazoniana 8, 2:255- 264 Volkmer-Ribeiro C, Traveset A (1987) Annotated catalog of the type specimens of Potts' species of freshwater sponges. Proc Acad Nat Sci Phila 139:223-242