A revision of the supraspecific classification of the ...sciencepress.mnhn.fr/sites/default/files/articles/pdf/z2000n2a2.pdfcation of the subclass Calcaronea (Porifera, class Calcarea).

Borojevic R., Boury-Esnault N. & Vacelet J. 2000. — A revision of the supraspecific classifi-cation of the subclass Calcaronea (Porifera, class Calcarea). Zoosystema 22 (2) : 203-263.

ABSTRACTA revision of all the genera of the subclass Calcaronea (Porifera, Calcarea) isgiven. In addition to the two previously described orders, LeucosoleniidaHartman, 1958 emend. and Lithonida Vacelet, 1981, we recognize a thirdone: the Baeriida. The order Leucosoleniida includes nine families, one ofwhich is new (the Jenkinidae), and 42 genera of which four are new(Breitfussia, Leucandrilla, Polejaevia and Syconessa). The order Lithonidaincludes two families and six genera. The order Baeriida includes three fami-lies of which two are new (the Baeriidae and the Trichogypsiidae), and eightgenera. The Leucosoleniida seem to have evolved from the olynthus grade, aform that is probably present in the early stages of ontogenesis of allLeucosoleniida and subsists at the adult stage in Leucosolenia. TheLeucosoleniida comprises a diverse group with several pathways of progres-sing complexity of form, starting with sponges of a simple sycettid organiza-tion and leading to sponges with a complex aquiferous system andskeleton. Increase in size from the sycettid grade of organization may occur bytwo different processes: 1) the growth and elongation of radial tubes increa-sing the thickness of the sponge body (seen in the Sycettidae-Grantiidae lineand the Heteropiidae), or 2) the growth of the central tube containing the

Radovan BOROJEVICDepartamento de Histologia e Embriologia, Instituto de Ciências Biomédicas,

Universidade Federal do Rio de Janeiro, Caixa Postal 68021,21941-970 Rio de Janeiro (Brazil)

[email protected]

Nicole BOURY-ESNAULTJean VACELET

Centre d’Océanologie de Marseille(CNRS-Université de la Méditerranée, UMR 6540 DIMAR),

Station marine d’Endoume, F-13007 Marseille (France)[email protected]@com.univ-mrs.fr

A revision of the supraspecific classification of the subclass Calcaronea (Porifera, class Calcarea)

203ZOOSYSTEMA • 2000 • 22 (2) © Publications Scientifiques du Muséum national d’Histoire naturelle, Paris. www.mnhn.fr/publication/

SYSTEMATIC INDEX

Subclass CALCARONEA Bidder, 1898 Order LEUCOSOLENIIDA Hartman, 1958

Family LEUCOSOLENIIDAE Minchin, 1900Genus Leucosolenia Bowerbank, 1864Genus Ascute Dendy & Row, 1913Genus Ascyssa Haeckel, 1872

Family SYCETTIDAE Dendy, 1892Genus Sycetta Haeckel, 1872Genus Sycon Risso, 1826

Family GRANTIIDAE Dendy, 1892Genus Grantia Fleming, 1828

Genus *Sycandra Haeckel, 1872Genus *Teichonopsis Dendy & Row, 1913Genus Ute Schmidt, 1862Genus *Sycute Dendy & Row, 1913Genus *Synute Dendy, 1892Genus Amphiute Hanitsch, 1894Genus *Sycodorus Haeckel, 1872Genus Leucandra Haeckel, 1872Genus Aphroceras Gray, 1858 Genus Leucandrilla n. gen.Genus *Leucettaga Haeckel, 1872

Family SYCANTHIDAE Lendenfeld, 1891Genus Sycantha Lendenfeld, 1891Genus *Dermatreton Jenkin, 1908

Borojevic R., Boury-Esnault N. & Vacelet J.

204 ZOOSYSTEMA • 2000 • 22 (2)

MOTS CLÉSSpongiaires, Calcaronea,

évolution, définitions génériques,

Baeriida, Jenkinidae.

atrial cavity which increases the length of the sponge body (seen in theJenkinidae and the simple forms of Amphoriscidae). The sponges classified inthe Baeriida and the Lithonida have very divergent forms that are representedby only a few living species. Identification keys and illustrations are providedfor all the valid genera.

RÉSUMÉRévision de la classification supraspécifique de la sous-classe Calcaronea (Porifera,classe Calcarea).Une révision de tous les genres de la sous-classe Calcaronea (Porifera,Calcarea) est faite. Le nouvel ordre Baeriida est proposé en addition aux deuxordres précédemment reconnus, Leucosoleniida Hartman, 1958 emend. etLithonida Vacelet, 1981. L’ordre Leucosoleniida comprend neuf famillesdont une nouvelle (Jenkinidae) et 42 genres, dont quatre nouveaux(Breitfussia, Leucandrilla, Polejaevia et Syconessa). L’ordre Lithonida com-prend deux familles et six genres. L’ordre Baeriida comprend trois familles,dont deux nouvelles (Baeriidae et Trichogypsiidae) et huit genres. LesLeucosoleniida semblent avoir évolué à partir du stade olynthus. Cette formeest probablement présente dans les stades précoces de l’ontogenèse chez toutesles Leucosoleniida et subsiste à l’état adulte chez Leucosolenia. LesLeucosoleniida sont un groupe florissant dans lequel on reconnaît plusieurslignées ayant un système aquifère et un squelette de complexité croissante àpartir de l’organisation de type sycettide. La croissance en taille à partir dustade sycettide peut avoir lieu en suivant deux voies : 1) la croissance et l’élon-gation des tubes radiaires, qui accroissent l’épaisseur du corps de l’éponge etqui sont représentées par la lignée Sycettidae-Grantiidae et les Heteropiidae ;2) la croissance du tube central contenant la cavité atriale qui accroît la lon-gueur du corps, comme chez les Jenkinidae et les formes simples desAmphoriscidae. Au contraire, les Baeriida et les Lithonida sont des groupestrès divergents, représentés seulement par un petit nombre d’espèces. Des clésd’identification et des illustrations sont données pour tous les genres consi-dérés comme valides.

KEY WORDSPorifera,

Calcaronea, evolution,

generic definitions, Baeriida,

Jenkinidae.

Family JENKINIDAE n. fam.Genus Breitfussia n. gen.Genus Jenkina Brøndsted, 1931Genus *Leucascandra Borojevic

& Klautau, 2000Genus *Anamixilla Poléjaeff, 1883Genus *Polejaevia n. gen.Genus *Uteopsis Dendy & Row, 1913

Family HETEROPIIDAE Dendy, 1892Genus *Syconessa n. gen.Genus Sycettusa Haeckel, 1872Genus *Grantilla Row, 1909Genus Grantessa Lendenfeld, 1885Genus Heteropia Carter, 1886Genus *Paraheteropia Borojevic, 1965Genus Vosmaeropsis Dendy, 1892

Family AMPHORISCIDAE Dendy, 1892Genus Amphoriscus Haeckel, 1872Genus Leucilla Haeckel, 1872Genus Paraleucilla Dendy, 1892

Family STAURORRHAPHIDAE Jenkin, 1908Genus Achramorpha Jenkin, 1908Genus Megapogon Jenkin, 1908

Family LELAPIIDAE Dendy & Row, 1913Genus Grantiopsis Dendy, 1892Genus *Kebira Row, 1909Genus *Paralelapia Hôzawa, 1923Genus Lelapia Gray, 1867

Family INCERTAE SEDIS

Genus Sycyssa Haeckel, 1872

Order BAERIIDA n. ord.Family BAERIIDAE n. fam.

Genus Baeria Miklucho-Maclay, 1870Genus *Lamontia Kirk, 1895Genus *Leucopsila Dendy & Row, 1913Genus *Eilhardia Poléjaeff, 1883

Family TRICHOGYPSIIDAE n. fam.Genus Trichogypsia Carter, 1871Genus *Kuarrhaphis Dendy & Row, 1913Genus *Leucyssa Haeckel, 1872

Family LEPIDOLEUCONIDAE Vacelet, 1967Genus *Lepidoleucon Vacelet, 1967

Order LITHONIDA Vacelet, 1981Family MINCHINELLIDAE Dendy & Row,1913

Genus Minchinella Kirkpatrick, 1908Genus Plectroninia Hinde, 1900Genus *Monoplectroninia Pouliquen &Vacelet, 1970Genus *Petrostroma Döderlein, 1892Genus *Tulearinia Vacelet, 1977

Family PETROBIONIDAE Borojevic, 1979Genus *Petrobiona Vacelet & Lévi, 1958

* Genus with only one described species.

INTRODUCTION

In a previous study (Borojevic et al. 1990), werevised the classification of Recent calcareoussponges belonging to the subclass CalcineaBidder, 1898, in an attempt to redefine the cur-rently recognized families and genera and to tracethe possible evolutionary pathways in this sub-class of the Calcarea. The present study is a con-tinuation of this revision, extending it now to theRecent sponges belonging to the subclassCalcaronea Bidder, 1898. The common characteristic of all representativesof the Calcarea is the presence of calcium carbon-ate spicules that have a basal diactine or triactinestructure. Calcareous spicules are secreted into anintercellular space that is delimited by two ormore cells. Although molecular evolutionarystudies have identified a potential early commonorigin of the two subclasses of the Calcarea, thereis no convincing molecular evidence for a closerelationship between Calcarea and other sponges(Lafay et al. 1992; Cavalier-Smith et al. 1996;Borchiellini et al. 1999). To our knowledge, nostudy has been conducted on molecular phyloge-ny within the subclass Calcaronea, and the pre-sent revision is based on morphological andanatomical data.Dendy & Row (1913) conducted the first majorgeneral revision of the Calcaronea in an attemptto classify all the described genera into the evolu-tionary pathways recognized at the time. Sincethen there have been several modifications of that

Taxonomy of Calcaronea

205ZOOSYSTEMA • 2000 • 22 (2)

proposal (Laubenfels 1936; Hartman 1958;Borojevic 1979; Vacelet 1991). But despite therecent description of new species and even highertaxa, and the availability of new informationgathered from cell and developmental biology,genetics, biochemistry, as well as observations ofmorphology at both the microscopic and theultrastructural level, there has been no overviewof the classification that attempts to group all thedescribed calcaronean genera so as to show theputative evolutionary development of the group,with the exception of Burton’s (1963) “Revisionof the Classification of the Calcareous Sponges”.The fundamental rationale of the Burton’s revi-sion was to analyse the supposed great intraspe-cific variability of the Calcarea that has resultedin the merging of species and higher taxa, whichhad been previously recognized as distinct phylo-genetic and taxonomic units, into a small numberof “genera” and “species”. This drastic decrease inlower systematic units has not been universallyaccepted (see discussion in Borojevic et al. 1990).Recent biochemical studies of the Calcarea haveshown that slight morphological differences maycorrespond to large genetic differences, and that afull genetic separation of sympatric or allopatricpopulations is often associated with subtle oreven undetectable differences as inferred by themore conventional morphological criteria (Solé-Cava et al. 1991; Klautau et al. 1994). Thus theclassical taxonomy based on the morphologicalcriteria is overconservative, and many specimensthat had been classified as simple variations of thepreviously described and often cosmopolitanspecies, probably represent distinct taxonomicunits.In the present work we have tried to identify thetaxonomic units that potentially represent mono-phyletic groups of species within the Calcaronea,and we have tried to identify all the possible evo-lutionary pathways, leading from the simplestCalcaronea, such as Leucosolenia, to the mostcomplex, such as Lelapia. We present the morecomplex types of skeletal and tissue organizationas deriving from simpler ones, and follow theconventional view that the simple “ascon” type ofsponge organization is “primitive”. This does notmean that we interpret the progressive increase ofcomplexity as necessarily the true evolutionary

pathway, but this cannot be reconstructed frommorphological data alone. Using this approach, we have confirmed the sepa-ration of sponges that we now group in the orderLeucosoleniida from those belonging to theLithonida, in agreement with the previously pro-posed classifications (Borojevic 1979; Vacelet1991). We were also led to separate a group ofsponges considered as “aberrant” by Dendy &Row (1913) from the Leucosoleniida, and pro-pose the recognition of the Baeriida as a neworder in Calcaronea for this group. The scope of most of the genera is that proposedby Dendy & Row (1913), who provided verydetailed descriptions of the genera, and discussedsynonymy and correspondence with the previ-ously described taxa extensively. The reader isreferred to that revision for a detailed discussionon earlier synonymies. An analysis of the proposed classification willreveal that many points are still uncertain. One ofthe major drawbacks of any attempt to prepare ageneral revision of calcareous sponges is the factthat our knowledge on this group is still veryfragmentary. The largest collection of Calcareaever studied is that described more than a centuryago by Haeckel (1872), who analysed andreviewed most of the specimens collected up tohis time. It is noteworthy that Haeckel proposeda large portion of the presently recognized genera,and in many cases new specimens have not sincebeen found. The Indo-Pacific, Antarctic andJapanese Calcarea have received more attention,mostly during the period between the end of thelast century and the first part of this century, butthe fauna of calcareous sponges in many otherregions remains very poorly known. A review ofany collection, even from regions one wouldexpect to be much studied such as the Europeancoasts of the Atlantic or the Mediterranean, pro-duces many new species that frequently belong tonew higher taxa, indicating that our knowledgeof the diversity of this group of sponges is veryincomplete. The present revision aims at gather-ing and assessing the available data so as to guidethe supraspecific identification of the Calcaronea,and propose a framework for future cellular andmolecular studies, which should give newinsights into the biology of this group. This


206 ZOOSYSTEMA • 2000 • 22 (2)

approach will highlight the taxonomic questionswhich should be addressed in future morphologi-cal, genetic and molecular studies of the Calcarea.For all the terms of sponge morphology we referthe reader to Boury-Esnault & Rützler (1997).

ABBREVIATIONS USED

MNHN Muséum national d’Histoire naturelle, Paris.BMNH Natural History Museum, London.

SYSTEMATICS

Class CALCAREA Bowerbank, 1864

DIAGNOSIS. — Marine Porifera in which the mineralskeleton is composed entirely of calcium carbonate.

The skeleton is composed of free diactine, triactine,tetractine and/or polyactine spicules, to which can beadded a solid basal calcitic skeleton with basal spiculeseither cemented together or completely embedded inan enveloping calcareous cement. The aquiferous sys-tem can be asconoid, syconoid, sylleibid or leuconoid.Members of the Calcarea are viviparous and their lar-vae are blastulae.

Subclass CALCARONEA Bidder, 1898

DIAGNOSIS. — Calcarea with diactines and/or sagittaltriactines and tetractines, rarely also with regularspicules. In addition to the free spicules, there can be anon-spicular basal calcareous skeleton in which basalspicules are cemented together or completely embed-ded in an enveloping calcareous cement. In ontogeny,the first spicules to be produced are diactines in thesettled larva. No information is available for the early


207ZOOSYSTEMA • 2000 • 22 (2)

FIG. 1. —Sycon sycandra; A, SEM view of the choanoderm; B, a choanocyte showing the typical apical nucleus (n) of the Calcaronea(TEM). Scale bars: A, 3 µm; B, 0.7 µm.

A

n

B

stages of postlarval development in the Baeriida andthe Lithonida. Choanocytes are apinucleate and thebasal system of the flagellum is adjacent to the apicalregion of the nucleus. The first stage in embryogenesisis a coeloblastula in which the flagella are internal andface into the central cavity. This blastula passesthrough a complex inversion, turning the flagellatedpole of the blastomeres to the outside, and giving riseto an amphiblastula larva, in which the anterior pole isflagellated and the posterior pole is bare. After settle-ment, the flagellated cells give rise to choanocytes, andthe large posterior aflagellated cells give rise to theother cell categories of the sponge, pinacocytes, poro-cytes, sclerocytes and to the amoeboid cells that arefound between the choanoderm and the pinacoderm.

DESCRIPTION

Like the calcineans, calcaroneans are extremelyvariable in size, form, organization of the aquifer-ous system, and skeleton. Most of the representa-tives are known only from Recent seas. Isolated

spicules, which may belong to calcaroneansponges have been reported from Early Cambrianreefs (James & Klappa 1983) and in Ordovicianstrata (Van Kempen 1978), but no unequivocalcalcaronean fossils have yet been found (Reitner& Mehl 1995). The aquiferous system in the Calcaronea can beasconoid, syconoid, sylleibid or leuconoid.Asconoid, syconoid, and sylleibid systems arefound only in the Leucosoleniida. The leuconoidaquiferous system, such as that seen in theLeucosoleniida, can easily be derived from asyconoid type of organization, as these spongesretain traces of the radial organization of theskeleton and the usually clearly defined centralatrium. However, the leuconoid systems in theBaeriida and the Lithonida bear no trace of anoriginal tubular or radial organization, but areinstead quite similar to the leuconoid aquiferoussystems of the Demospongiae.Calcaronean sponges have choanocytes with anapical, ovoid or pyriform nucleus. The basal fla-gellar roots are always in contact with the nuclearenvelope at the apical pole of the nucleus. In manyspecies, a glycocalyx layer is present between themicrovilli which form the collar (Fig. 1). The inter-pretation of the localization of the nucleus withinthe choanocytes is often hampered by artefactscaused by handling of sponge collections and theirfixation. Since this is one of the most distinctivecharacters distinguishing the Calcinea from theCalcaronea, sections of preserved material must beinterpreted with caution (Vacelet 1964).Despite the observed great diversity of body plan,our present knowledge of calcaronean biology, inparticular their cell and skeletal morphology,indicates that there are a number of homologiesamong currently known species, and stronglysupports the hypothesis of their common originas well as a rather close relationship among all thesponges belonging to this subclass. Most notably,sponges in the Calcaronea have a typical fertiliza-tion process and a very particular pattern ofembryogenesis and larval morphogenesis. During fertilization, the spermatozoa are cap-tured by choanocytes, which transform into aparticular spermatozoon carrier cell containingthe spermiocyst (Duboscq & Tuzet 1937, 1942;Vacelet 1964; Gallissian 1989; Gallissian &


208 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 2. — Fertilization in Leucillla. Transmission electronmicroscopy (TEM). Abbreviations: cc, choanocyte chamber;o, osculum; sp, spicule. Scale bar: 1 µm. (Courtesy Dr M.-F.Gallissian).

ccsp

o

Vacelet 1990). These cells migrate into the sub-choanodermal space where they fertilize largemature oocytes. The entrance point of the carriercell into the oocyte determines the symmetry ofthe future larva in the Leucosoleniida, but appar-ently not in a sponge that we currently classify inthe Baeriida (Duboscq & Tuzet 1937, 1942).Only small differences were observed in the fertil-ization process in different species of the cal-caronean sponges studied until now. There isonly one report of a similar fertilization processin the rest of the Calcarea (Tuzet 1947) (Fig. 2).The amphiblastula larva has large aflagellatedcells at one end and small flagellated cells at theother. There are four “cellules en croix” whichhave the presumed function of photoreceptors(Duboscq & Tuzet 1941; Borojevic 1970;Amano & Hori 1992). At the early blastula stagethe flagella are directed inwards into the primaryblastocoel. Subsequently, in the stage called thestomoblastula, the aflagellated cells form anopening through which the flagellated blastulawall evaginates, inverting the larval wall and turn-

ing the flagella outwards. The larva closes again,delimiting a secondary blastocoel (Fig. 3). At thisstage the larva is a typical amphiblastula withclearly marked poles: the flagellated pole corre-sponds to the anterior pole of the free-swimminglarva, while the large aflagellated cells are restrict-ed to the posterior pole (Fig. 4). After settlement,the large aflagellated cells give rise to pinacocytes,sclerocytes and to other amoeboid cells, while theflagellated cells differentiate into choanocytes(Amano & Hori 1993). The inversion of theearly larva is unique and specific to the subclassCalcaronea, and is reminiscent of the morpho-genesis of Volvox (Ivanov 1971). The amphiblas-tula larvae of the Calcaronea differ from all othersponge larvae. They are only superficially similarto the larvae of the Homoscleromorpha, whichhave been described as amphiblastula, but whichare now termed “cinctoblastula” in order tounderline these differences (Boury-Esnault et al.1995). In the order Leucosoleniida, after the set-tlement of the larva, an asconoid tubular spongeis formed, which can remain at this stage of orga-


209ZOOSYSTEMA • 2000 • 22 (2)

FIG. 3. — Diagram of the stomoblastula showing the phenomenon of inversion; A, fertilization; B, stomoblastula with the flagellum ofthe flagellated cells inside the blastocoel; C, inversion; D, amphiblatula with flagellae outside; E, mature free-swimming amphiblastula;F, young rhagon after metamorphosis.

A B C

D E

F

nization (e.g. Leucosolenia) or form radial out-growths which give rise to the radial tubes of thesyconoid grade of organization (Fig. 5) (Schulze1875). The postlarval development of the othertwo orders of the Calcaronea is not known.

Order LEUCOSOLENIIDA Hartman, 1958 emend.

DIAGNOSIS. — Calcaronea with a skeleton composed ofexclusively free spicules, without calcified non-spicularreinforcements. The aquiferous system is asconoid,syconoid, sylleibid or leuconoid. In the latter case, theradial organization around a central atrium can gener-ally be detected by a well-formed atrial skeleton tan-gential to the atrial wall, and/or a subatrial skeletonconsisting of subatrial tri- or tetractines with the pairedactines tangential to the atrial wall and the unpairedactine perpendicular to it. The post-larval developmentpasses (presumably always) through an olynthus stage.

DESCRIPTION

Like the calcinean order Clathrinida (Borojevic etal., 1990), the Leucosoleniida represents a

homogenous group of sponges, in which arefound all the possible modifications of the funda-mental pattern of the sponge body organization,from asconoid to leuconoid, and including mostof the intermediate stages of the progressive mod-ifications of the associated skeleton. Conse-quently, we consider that the Leucosoleniidarepresents a single taxonomic unit that cannot bedivided into two groups, according to homocoelor heterocoel grade of organization as proposedby Hartman (1958). The simplest forms correspond to the olynthusgrade of organization, with a single tubular cen-tral cavity lined by choanocytes (familyLeucosoleniidae) (Fig. 6). However, whereas inthe Clathrinida, the olynthus form has given riseto several independent evolutionary lineages (seeBorojevic et al. 1990) in the Leucosoleniida, themajor and, as far as we are aware, sole evolution-ary line from the homocoel to the heterocoelgrade of organization passes through a sycettidgrade of organization. Sycetta is characterized by asingle central tube devoid of choanocytes, which


210 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 5. — Young Sycon sycandra in the sycettid stage. Scalebar: 1.7 mm.

FIG. 4. — An amphiblastula larva (a) in the parental sponge.Scanning electron microscopy (SEM). Scale bar: 8 µm.

a

corresponds to the atrium, from which tubeswith a choanoderm radiate (Fig. 7). This spongehas only an exhalant aquiferous system; theincurrent water flows directly into the radialtubes through inhalant pores (Figs 7; 8).The first group of morphological characters usedto define the Leucosoleniida is the overall shapeand the underlying skeletal support of the tubes.Increase in size from the sycettid grade of organi-zation may occur by two different processes: thegrowth and elongation of radial tubes whichincreases the thickness of the sponge body, or thegrowth of the central tube containing the atrialcavity, which increases the length of the spongebody. Both processes can be observed in theLeucosoleniida: A) the first process has given rise to two evolu-tionary pathways. The first is well-depicted byvery young specimens of Sycon, as well as adultSycetta, where the radial tubes are short and sepa-

rate. At this stage they have an inarticulatechoanoskeleton, i.e. the central atrial tube has adistinctive tangential skeleton, but the radialtubes perpendicular to the atrium are primarilysupported by subatrial triactines whose pairedactines are adjacent to the atrial skeleton, and theunpaired ones support the radial tube walls. Thedistal cones have peculiar small triactines. The evolutionary lineages of the Sycettidae-Grantiidae and Heteropiidae bifurcate from thispoint. Whereas in the Sycettidae, this type ofinarticulated organization is found only in veryyoung specimens of Sycon and Sycetta, in theHeteropiidae, the inarticulate choanoskeleton isfound in Syconessa and in Sycettusa, which alsohas distal cones that are fused into a continuoustangential cortical layer. In both families theelongation of radial tubes and their progressivecoalescence result in a compact body that has astrictly radial organization, such as is found in


211ZOOSYSTEMA • 2000 • 22 (2)

FIG. 6. — Diagram of an asconoid aquiferous system, such as isfound in Leucosolenia. Abbreviations: ps, pinacoderm and theskeletogenous layer; ch, choanoderm. The arrow shows thedirection of water flow.

FIG. 7. — Diagram of the Sycetta type of organization of thesponge wall. Abbreviations: ps, pinacoderm and the skeletoge-nous layer; a, atrium; o, osculum. The arrow shows the directionof water flow.

ps ps

a

o

ch

typical adult representatives of Sycon (Fig. 8) andGrantessa. The radial tubes are intercalated withnarrow inhalant canals with an inhalant pore-bearing membrane devoid of skeleton.Progressively, a common cortex covers the externalpart of the radial tubes and the openings of inhalantcanals, i.e. the inhalant pores move to the corticalsurface, and the cortex becomes supported by a spe-cific skeleton. Such corticalization has given rise toa wide range of sponges with a solid body and withelaborate skeleton. The aquiferous system changesfrom long choanocyte chambers arranged radiallyaround the central atrium, characteristic of thesyconoid system (Fig. 8), to shorter elongate orovoid choanocyte chambers arranged aroundradial exhalant cavities, such as observed in thesylleibid aquiferous system (Fig. 9), and to ovoidor spherical choanocyte chambers arrangedbetween the inhalant and exhalant canals, such asobserved in sponges with a typical leuconoid aquif-erous system (Fig. 10). Sponges belonging to thelatter evolutionary pathway usually have a typicalarticulate choanoskeleton, i.e. several rows of sim-

ilar triactine spicules. In the first subatrial row, thepaired actines are adjacent to the atrial skeleton andthe unpaired actine is perpendicular to it, lying inthe wall of the radial tube. This is the most com-mon form of subatrial skeleton, and is easily rec-ognized in all heterocoel Leucosoleniida. Althoughthe spicules of the choanoskeleton can be irregu-larly scattered in massive sponges with a leuconoidaquiferous system such as Leucandra, the originalorientation of many triactines with the unpairedangle turned to the atrium and the unpaired actinepointing distally, is frequently preserved. As indi-cated earlier, this evolutionary line has bifurcatedquite early into two pathways that are distin-guished by the presence or absence of pseu-dosagittal spicules in the distal part of the radialtubes. In both pathways, there is both increasedcorticalization, and progressive evolution of thesyconoid organization into the leuconoid one (seedescriptions of the families Grantiidae andHeteropiidae).B) The second process has also given rise to twoevolutionary pathways. The first one is analogous


212 ZOOSYSTEMA • 2000 • 22 (2)

ps

dc

FIG. 8. — Diagram of Sycon type of sponge wall organization.Abbreviations: ps, pinacoderm and the skeletogenous layer; ch,choanoderm; a, atrium; dc, distal cones; ic, inhalant canals. Thearrow shows the direction of water flow.

ch

ic

a

FIG. 9. — Diagram of the sylleibid type of aquiferous systemorganization, such as observed in Polejaevia, Paralelapia, andLeucilla. Abbreviations: cx, cortex; ch, choanoderm; a, atrium;ic, inhalant cavities; ec, exhalant cavities.

cx

ic

ec a

ch

to that of the Levinellidae in Calcinea (see Boro-jevic & Boury-Esnault 1986). During the longi-tudinal growth of the central tube, the radialoutgrowths of the sycettid type of organizationdo not increase in length, but multiply. They canbecome grouped around the common cavities,each of them opening into the atrium. Thesegroups of outgrowths are intercalated by shallowinhalant spaces. This organization, classified as“aberrant” by Dendy & Row (1913) in compari-son with the typical Sycon form of growth, wasdescribed by Lendenfeld (1891) for the genusSycantha (Fig. 11) and by Jenkin (1908a) for hisgenera Tenthrenodes, Hypodictyon and Derma-treton. A partial corticalization can occur in this

evolutionary line by an increase of tangential tri-actine spicules in the distal parts of the fusedradial tubes between the inhalant cavities so as toform a loose network such as observed inDermatreton. Since this network does not providesufficient mechanical support, the atrial skeletontakes over this function, and is thickened in orderto provide the required rigidity. This evolutionarypathway has only given rise to a few sponges,which we group in the family Sycanthidae. In the second evolutionary pathway, the shortradial tubes retain their regular distribution onthe central atrial tube and become covered by atrue continuous cortex. The result is a spongewith a thin body surrounding a large atrial cavity,


213ZOOSYSTEMA • 2000 • 22 (2)

cx

ic

ic

ch

ch

ec

ps

o

aa

FIG. 10. — Diagram of the leuconoid type of aquiferous systemorganization, such as observed in many Calcaronea. Instead ofa central atrial cavity there is a network of aquiferous exhalantcanals that increase in size from the distal regions to the oscula.Abbreviations: cx, cortex; ch, choanoderm; ic, inhalant cavities;ec, exhalant cavities; a, atrium.

FIG. 11. — Diagram of the Sycantha type of sponge wall organi-zation. Abbreviations: ps, pinacoderm and the skeletogenouslayer; ch, choanoderm; a, atrium; o, osculum; ic, inhalant cavi-ties. The arrow shows the direction of water flow.

with a rigid and well-developed atrial and corticalskeletons (Fig. 12). This morphology is found inseveral independent evolutionary lineages. In twofamilies, the Jenkinidae and the simple forms ofAmphoriscidae, a continuous and dense cortex isassociated with an inarticulate choanoskeletoncomposed of only the unpaired actines of thesubatrial spicules, and occasionally the actines ofcortical or subcortical spicules (e.g. Ampho-riscidae). In the basal region of these sponges,where the wall can be thicker, a number of suba-trial or cortical spicules may be found at somedistance from respectively the atrial or the corti-cal plane. Nonetheless, they clearly retain themorphology of cortical or subatrial spicules andnever form an articulate choanoskeleton. Amongthe sponges with a thick wall and an articulateskeleton, the original syconoid organization ofthe thin-wall sponges can also result in a moreelaborate sylleibid or an irregular alveolar leu-conoid aquiferous system (Figs 9; 10).

The growth of the Jenkinidae is longitudinal. Asthese long tubular structures become fragile, largespecies form a complex cormus of branched andoccasionally anastomosed tubes (e.g. Anamixilla,Uteopsis, Leucascandra), quite similar to the largecormi of Leucosolenia in the Calcaronea, andLevinella or Leucaltis in the Calcinea. Conversely,in the Amphoriscidae, large specimens ofParaleucilla can secondarily form massive bodiesby a secondary thickening of the body wall ratherthan by the distal elongation of the original radialtubes, such as occurs in Sycon. The original inar-ticulate organization of the choanoskeleton is stillclearly visible at least in the external part of thesponge. The thickening of the sponge wall maybe caused by the insertion of new layers betweenthe atrial and the subatrial skeletons, forming asubatrial area with a specific skeleton derivedfrom subatrial, atrial or both types of spicules.Alternatively, the thickened body can be a conse-quence of multiple folding and coalescence of theoriginally thin sponge body wall.It is conceivable that an inarticulate choanoskele-ton can also be derived from an articulate one bythe secondary reduction of the choanosome wallthickness. Dendy & Row (1913) favoured thispossibility, concluding that both inarticulate andarticulate types of choanoskeleton can coexist inthe same genus, the former one being derivedfrom the latter one. We find, like Brøndsted(1931), that there is a relative morphological andgeographical homogeneity of sponges with aninarticulate choanoskeleton, and consequentlyconsider this form of skeleton is a primary mor-phological characteristic. We have now tried togroup those sponges with an inarticulate skeletonin separate taxa. However, we are aware that somespecies may be difficult to fit into the proposedsystem, and that the thickening of the originallythin choanosome may be a natural consequenceof the growth of the sponge body.The second group of morphological characterscorresponds to the different patterns of spiculesthat participate in the composition of the skele-ton in specific regions of the sponge wall. Thesecharacteristics can be used to subdivide theLeucosoleniida, into the families Grantiidae,Heteropiidae, Staurorrhaphidae and Lelapiidae.Since all these families derive from the sycettid


214 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 12. — Diagram of the Jenkinidae type of the sponge wallorganization. Abbreviations: cx, cortex; ch, choanoderm; a, atri-um; o, osculum; ic, inhalant cavities.

ic

cx

a

o

ch

grade of organization mainly through the distalincrease of their radial tube length and subse-quent corticalization, they all still bear clear tracesof the radial organization, in the tubes growingout from the central atrium. This is generallyquite easily noted in the proximal subatrial skele-ton, which is perpendicular to the atrial one, andindicates the original position of radial tubes(Fig. 13). Within each of these families, the gen-era are generally defined by the presence orabsence of certain types of spicules (e.g. largediactines) in defined regions of the sponge. Thisdivision may be rather artificial, but it is conve-nient for classification of many sponges whichbelong to the heterocoel Leucosoleniida.In typical species of this order and in fully-grownspecimens, the main characteristic of each familyis quite easily recognized. For example, giant cor-tical tetractines typify the Amphoriscidae andsubcortical pseudosagittal spicules are typical ofthe Heteropiidae. However, this is not the case inyoung specimens, and their identification can bequite difficult. Furthermore, a particular spiculetype can be rare in some specimens, or in certainregions of a sponge (Borojevic 1966). Newspicule types can appear in families where theyare not originally found, representing a sec-

ondary, rather than a primary and diagnostic,morphological characteristic (e.g. the corticaltetractines in Leucandrilla, which does not belongto Amphoriscidae). In a similar way, the normallythick sponge wall can be thin and supported onlyby a reduced choanoskeleton in young sponges,or in the suboscular region, causing a spongebelonging to the Grantiidae to appear similar tothose in the Jenkinidae. It is always difficult toresolve such cases, and we can only agree withDendy & Row (1913) in stating that “it must befrankly admitted that the boundary line… is byno means sharply defined”. We hope that furtherstudies will shed more light on the problematiccases.

Family LEUCOSOLENIIDAE Minchin, 1900

TYPE GENUS. — Leucosolenia Bowerbank, 1864 byoriginal designation.

DIAGNOSIS. — Leucosoleniida with a cormuscomposed of frequently branched, but rarely anasto-mosed, asconoid tubes, and with a continuous choan-oderm that lines all the internal cavities of the sponge.There is neither a common cortex covering the cor-mus, nor a delimited inhalant or exhalant aquiferoussystem.


215ZOOSYSTEMA • 2000 • 22 (2)

FIG. 13. — Diagram of the subatrial region of heterocoel Leucosoleniida. Abbreviations: a, atrium; as, atrial skeleton, composed oftriactines and tetractines tangential to the atrial surface; ap, apopyle of the radial choanocyte chamber; ss, subatrial spicules; ar,articulate choanosomal skeleton; ch, choanoderm.

as

ap

a

ar

ch

ss

DESCRIPTION

The family Leucosoleniidae includes all the cal-caronean homocoel sponges. In contrast to theClathrinidae, which frequently form large mas-sive cormi, the Leucosoleniidae are most oftensmall and creeping tubular sponges that onlyrarely form cormi several centimetres large, suchas Leucosolenia complicata (Montagu, 1818) orLeucosolenia eleanor Urban, 1905.

Genus Leucosolenia Bowerbank, 1864

TYPE SPECIES. — Spongia botryoides Ellis & Solander,1786 by original designation.

DIAGNOSIS. — Leucosoleniidae in which the skeletoncan consist of diactines, triactines and/or tetractines.There is no reinforced external layer on the tubes.

DESCRIPTION

While the genus Leucosolenia is morphologicallyvery homogenous, it is nonetheless cosmopolitanand includes numerous species. The asconoidtubes may be creeping and only rarely branched,or be copiously ramified but not anastomosed;they may form a large arborescent cormus such asseen in L. complicata (Montagu, 1818). The cor-mus of Leucosolenia is always simple, withoutsubdivisions or differentiations into regions withdistinct functions, although in larger specimensthe central and proximal tubes are usually widerthan the distal ones (Fig. 14).

Genus Ascute Dendy & Row, 1913

TYPE SPECIES. — Leucosolenia uteoides Dendy, 1892 byoriginal designation.


216 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 14. — Leucosolenia complicata from the Channel Sea(Roscoff) MNHN C.1968,341. Scale bar: 1 cm.

FIG. 15. — Section of the wall of Ascute. Specimen from theWilson collection collected near Port Philips Heads (Australia),BMNH 1983.6.9.33. Scale bar: 160 µm.

DIAGNOSIS. — Leucosoleniidae in which the skeletoncan be composed of diactines, triactines and/ortetractines, with an outer layer that is supported bygiant longitudinal diactines.

DESCRIPTION

Dendy & Row (1913) placed the genus Ascuteamong sponges with basinucleate choanocytes.We have examined the type specimen of Leuco-solenia uteoides Dendy, 1892 (BMNH 1893.6.9.33) and found that it only has typical sagittaltriactine and tetractine spicules that are organizedin a pattern very similar to the skeleton ofLeucosolenia (Fig. 15), and quite different fromthe Clathrinidae, which are characterised by reg-ular spicules. Since the appearance of thechoanocytes may be considered altered by fixa-tion (Vacelet 1964), we prefer to place this genusclose to Leucosolenia until examination of newspecimens and a revision of their cytology is pos-sible. Only two species were described in thisgenus; both are from Australia: A. asconoides(Carter, 1886) and A. uteoides (Dendy, 1892).

Genus Ascyssa Haeckel, 1872

TYPE SPECIES. — Ascyssa troglodytes Haeckel, 1872 bysubsequent designation (Dendy & Row 1913).

DIAGNOSIS. — Leucosoleniidae with a skeleton com-posed entirely of diactines.

DESCRIPTION

Haeckel (1872) described the two species of thegenus Ascyssa from very few small specimens; rep-resentatives of this genus have not been foundsince. Since in the Calcaronea the first spicules tobe secreted are diactines, these specimens maysimply represent very young Leucosolenia, as sug-gested by Dendy & Row (1913). However,because Haeckel (1872) indicated that the speci-men of A. acufera Haeckel, 1872 was sexuallyreproductive, this hypothesis is unlikely.

Family SYCETTIDAE Dendy, 1892

TYPE GENUS. — Sycetta Haeckel, 1872 by original des-ignation.

DIAGNOSIS. — Leucosoleniida with a central atrialtube and perpendicular regularly arranged radial tubeslined by choanoderm. The distal cones of the radialtubes, which may be decorated with tufts of diactines,are clearly noticeable on the sponge surface. They arenever covered by a cortex supported by tangential tri-actines and/or tetractines. The proximal skeleton ofthe radial tubes is composed of a row of subatrial tri-actines and/or tetractines, which are usually followedby only a few or several rows of triactines and/ortetractines. Distal pseudosagittal spicules are absent. Atangential layer of triactines and/or tetractines sup-ports the atrial wall.

DESCRIPTION

In the Leucosoleniida, the transition from homo-coel to heterocoel grade of organization apparent-ly passes only through the sycettid-grade oforganization (Dendy & Row 1913). The sycettidorganization is essentially a sponge likeLeucosolenia in which the median region of thesingle central tube is decorated with regularlyarranged short and unbranched radial tubes. It isstructurally analogous to the organization of sim-ple Levinellidae from the Calcinea (Fig. 7). Thetransition from the homocoel to the heterocoelorganization involves the progressive restrictionof choanocytes to the radial tubes, while the cen-tral tube acquires the sole function of an exhalantatrium. In the Sycettidae the elongation of theradial tubes is concurrent with their partial or fulllongitudinal coalescence around the radialinhalant canals. This organization, typical in thegenus Sycon, gives compactness to the sponge,simultaneously maintaining an efficient water cir-culation.

Genus Sycetta Haeckel, 1872

TYPE SPECIES. — Sycetta sagittifera Haeckel, 1872 bysubsequent designation (this work).

DIAGNOSIS. — Sycettidae with a central atrial tubedecorated with short, completely separate radial tubes.There is no defined inhalant aquiferous system. Theskeleton of the radial tubes is composed of triactinesand tetractines, and diactines may be found in the dis-tal cones.

DESCRIPTION

The genus Sycetta, as defined by Dendy & Row(1913), comprised three species described under


217ZOOSYSTEMA • 2000 • 22 (2)

ar

dc

ss

as

a

the names Sycetta primitiva Haeckel, 1872,S. sagittifera Haeckel, 1872 and Sycaltis coniferaHaeckel, 1872. Dendy & Row 1913 designatedSycetta primitiva as the type species. Haeckel(1872) characterized this species by the presenceof regular, equiangular and equiradiate spicules,which are clearly described and represented assuch in the corresponding figure. In the same fig-ure, Haeckel (1872: vol. III, pl. 41) shows thatthe choanocytes are closer to the basinucleatethan to the apinucleate type. Although Haeckel’sdescriptions may be taken with some reservation,and S. primitiva has not been observed since that

time, the original description indicates quiteclearly that this is a calcinean sponge, and shouldbe classified as a typical member of the familyLevinellidae (Borojevic & Boury-Esnault 1986).Haeckel (1872) classified Sycetta primitiva in thesubgenus Sycettaga, and we propose to transfer itas a genus to the family Levinellidae, with a singlespecies Sycettaga (Sycetta) primitiva Haeckel,1872. Sycetta sagittifera being an originallyincluded nominal species is designated here as thetype species of Sycetta. This species displays allthe characteristics of the genus as understood byDendy (1893), Dendy & Row (1913), and sub-sequent authors.Brøndsted (1931) described two sponges fromthe Deutsche Südpolar Expedition collection,Sycetta antarctica and Tenthrenodes primitivus.Whilst the former one is a typical Sycetta, the lat-ter is characterized by the presence of diactinesand the occasional coalescence of the radial tubes,which, however, are not fused. We have nowplaced the genus Tenthrenodes Jenkin, 1908 insynonymy with Sycantha Lendenfeld, 1891.Tenthrenodes primitivus Brøndsted, 1931 is how-ever much closer to a typical Sycetta and we pro-pose to transfer this species to the genus Sycetta.Sycetta (Tenthrenodes) primitiva (Brøndsted,1931) should be distinguished from Sycettaga(Sycetta) primitiva Haeckel, 1872, which belongsnow to the family Levinellidae. Dendy & Row(1913) specified that sponges in the genus Sycettahave no diactines, as all the sponges described inthe genus up to their time were devoid of them.Their presence in Sycetta primitiva (Brøndsted,1931) leads us to modify this point accordinglyin the definition of the genus Sycetta.

Genus Sycon Risso, 1826

TYPE SPECIES. — Sycon humboldtii Risso, 1826 by sub-sequent designation (Dendy & Row 1913).

DIAGNOSIS. — Sycettidae with radial tubes partially orfully coalescent; distal cones are decorated by tufts ofdiactines. The inhalant canals are generally well-defined between the radial tubes and are often closedat the distal end by a membrane that is perforated byan ostium, devoid of a skeleton. There is no continu-ous cortex covering the distal ends of the radial tubes.Skeleton of the atrium and of the tubes composed oftriactines and/or tetractines.


218 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 16. — Diagram of a transverse section through the wall ofSycon natalense Borojevic, 1967. Abbreviations: a, atrium; as,atrial skeleton composed of tangential triactines and tetractines;ss, subatrial spicules; ar, articulate choanosomal skeleton; dc,distal cone, with a short tuft of diactines (from Borojevic 1967b).Scale bar: 100 µm.


219ZOOSYSTEMA • 2000 • 22 (2)

FIG. 17. — The organization of the aquiferous system in Sycon sycandra (SEM); A, tubes (t); B, distal chones (d); C, detail of tubes (t)and inhalant openings (i); D, apopyles (a). Scale bars: A, 160 µm; B, 65 µm; C, 13 µm; D, 43 µm.

A B

C D

t

t

a

i

d

DESCRIPTION

The genus Sycon is cosmopolitan, and it is oftenconsidered to be a perfect example of the calcare-ous sponges (Figs 16; 17). Many representativeshave a simple radially organized body with a sin-gle osculum, occasionally with a short peduncle.Species that grow larger may be arborescent,with a peduncle and ramified body, each branchrepresenting a complete syconoid organiza-tion. The radial tubes are generally simple, butin large specimens they can also be ramified. Inthis case, the branches remain parallel, and eachbranch ends by a peculiar terminal cone. In somespecies (e.g. Sycon elegans Bowerbank, 1845;Sycon gelatinosum Blainville, 1837), the distalcones have dense tufts of diactines, which termi-nate all at the same level, giving the external sur-face a smooth, tabulate appearance. Thisorganization should not be misinterpreted as acortex, which is always characterized by tangen-tial triactine spicules.A group of small representatives of the genus arisefrom solid or tubular creeping stolons. Thestolons can produce terminal hollow sphericalbuds (e.g. Sycon sycandra Lendenfeld, 1885),which detach, and form propagules with a pecu-liar skeleton. They are usually hispid due to thepresence of long diactines, which act as flotationdevices and promote their subsequent anchorage,attachment to the substrate, and formation of theyoung sponge. These propagules can live for along time in the water column, and are quite fre-quently collected in the mesopsammon.However, they cannot be identified as Sycon untilthey attach to the solid substrate and grow intothe typical adult sponge. Alternatively, sphericalpropagules can be formed from the distal parts ofthe radial tubes through the constriction and sub-sequent detachment of the region just under thedistal cones (e.g. Sycon frustulosum Borojevic &Peixinho, 1976).Most species of Sycon are attached to hard sub-strates, but occasionally they can live on a softbottom [e.g. S. villosum (Haeckel, 1872) ,S. raphanus (Schmidt, 1862)]. In this case, theyare anchored by long and ornate spicules, whichare not found on the specimens that are attachedto hard substrates. Since all the other morpholog-ical characteristics are identical, and similar

sponges can grow in close proximity on differentsubstrata, we interpret the presence of thesespicules to be a secondary adaptation to the typeof substratum.Sycon is a common genus which has been exten-sively studied. Following Laubenfels (1936),Burton (1963) revived the name Scypha Grant,1821 which was described in the Flora of theBritish Plants, and which has recently been usedby non-taxonomists. Since 1899, no taxonomistexcept Laubenfels and Burton has used this nameand more than 25 works, and 10 authors haveused the generic name Sycon since 1950. We wishto maintain the commonly used younger syn-onym to avoid confusion. As the rule 23.9.1.2 ofreversal of precedence cannot be applied strictly,we refer the case to the Commission with anappropriate recommendation for a ruling underthe plenary power (Art. 81). The use of thejunior name is to be maintained while the case isunder consideration (Art. 82).

Family GRANTIIDAE Dendy, 1892

TYPE GENUS. — Grantia Fleming, 1828 by originaldesignation.

DIAGNOSIS. — Leucosoleniida in which there is alwaysa cortex, supported by a skeleton of tangential spiculesthat can be diactines, triactines, tetractines, or anycombination of them. The aquiferous system is eithersyconoid with radial and elongate choanocyte cham-bers, or sylleibid or leuconoid with elongate or spheri-cal, scattered choanocyte chambers. The inhalant andexhalant aquiferous systems are always fully developed.The choanoskeleton is articulate, tubular in syconoidspecies, and contains few to several rows of triactinesand/or tetractines, or is, in leuconoid species, arrangedwithout apparent order. In the latter case, the choa-noskeleton always preserves traces of the radial organi-zation, particularly at the level of the subatrialtriactines and/or tetractines. The atrial skeleton con-sisting of tangential triactines and/or tetractines iswell-developed.

DESCRIPTION

The family Grantiidae has a central positionamong the Leucosoleniida. Its major characteris-tic is the development of a distinct cortex. Thedevelopment of a cortex is quite progressive inthe Grantiidae, and simple forms, such as G. com-pressa (Fabricius, 1780), clearly indicate their


220 ZOOSYSTEMA • 2000 • 22 (2)

proximity to Sycon by the presence of tufts ofdiactines at the end of their radial tubes. Indeed,in large Sycon species, the inhalant canals are par-tially closed by a membrane devoid of spicules.Formation of a specific skeleton in this mem-brane, with the production of tangential spiculeswhich do not derive from those of the tubes, is anew feature which marks a major evolutionarystep, and the boundary between the familiesSycettidae and Grantiidae (Fig. 18).The family Grantiidae is very large. While some ofthe genera were designated to include sponges witha very particular type of growth or skeleton, andconsequently include only a single species (e.g.Teichonopsis, Sycute, Synute), others have a rather

basic type of organization and skeleton, and includenumerous species that are present throughout allthe oceans (e.g. Grantia, Leucandra).

Genus Grantia Fleming, 1828

TYPE SPECIES. — Spongia compressa Fabricius, 1780 byoriginal designation.

DIAGNOSIS. — Grantiidae with a syconoid organiza-tion. The cortex is composed of tangential triactinesand/or tetractines, occasionally with small perpendicu-lar diactines. Longitudinal diactines, if present, are notfound exclusively in the cortex, but cross obliquely, atleast a part of the choanosome and protrude from theexternal surface.


221ZOOSYSTEMA • 2000 • 22 (2)

FIG. 18. — Diagram of a transverse section through the wall of Grantia socialis Borojevic, 1967. Abbreviations: a, atrium; as, atrialskeleton composed of tangential triactines and tetractines; ss, subatrial spicules; ar, articulate choanosomal skeleton; cx, cortex(from Borojevic 1967a). Scale bar: 100 µm.

ar

a

as

ss

cx

DESCRIPTION

Typical species of Grantia have long and regularradial tubes, which may be branched distally, anda relatively thin atrial and cortical skeletons.Diactines frequently protrude from the externalsurface of the sponge. Many species of Grantiathat form small solitary tubes or large bushysponges have been described from all oceans.

Genus Sycandra Haeckel, 1872

TYPE SPECIES. — Ute utriculus Schmidt, 1870 by sub-sequent designation (Dendy & Row 1913).

DIAGNOSIS. — Grantiidae with a large flattened body;the atrial cavity with a complex network of tissuetracts, supported by parallel diactines.

DESCRIPTION

Dendy & Row (1913) retained Haeckel’s genusSycandra for a single species S. utriculus that wascharacterized by a complex network of tissuetracts in the atrial cavity, supported by bundlesof parallel diactines. Similar structures can beseen inside the atrial cavity of several large andflattened Grantiidae and Amphoriscidae. Thesurface of the opposite sides of the central atrialcavity can be close, touch and become coalescent(e.g. Leucilla saccharata Haeckel, 1872; Amphiutelepadiformis Borojevic, 1967). These regionsthus become connected by tissue tracts support-ed by a skeleton that is derived from the atrialone, and which can contain modified atrialspicules. However, in other Sycettidae diactinesare not normally present in the atrial skeleton,and the presence of an internal atrial networkwith a specific skeleton is a new morphologicalcharacteristic. Here we follow the opinion ofDendy & Row (1913) and consider that thischaracter is sufficient to separate the genus fromother Grantiidae.

Genus Teichonopsis Dendy & Row, 1913

TYPE SPECIES. — Teichonella labyrinthica Carter, 1878by monotypy.

DIAGNOSIS. — Pedunculate calyciform Grantiidaewith a syconoid organization and an expanded atrium.The thin wall is highly folded and the convoluted edgecorresponds to the oscular margin.

DESCRIPTION

The separation of this genus from Grantia is jus-tified because of its particular pattern of growth,through which the atrial cavity becomes wideopen. As it grows the sponge wall becomes ahighly folded asymmetric leaf, freely traversed bythe water current that runs from the lower corti-cal to the upper atrial surface.

Genus Ute Schmidt, 1862

TYPE SPECIES. — Ute glabra Schmidt, 1864 by mono-typy.

DIAGNOSIS. — Grantiidae with a syconoid organiza-tion. The cortex is supported by giant longitudinaldiactines, and the choanoskeleton is articulate, com-posed of several rows of triactines with occasionaltetractines. There are no radial fascicles of diactines.

DESCRIPTION

Calcarea belonging to the genus Ute are amongthe most beautiful calcareous sponges. They havea regular tubular form with a vitreous, smoothand shiny surface due to many longitudinal, par-allel diactines (Fig. 19).The relationship between the genera Ute andAphroceras has been discussed previously(Borojevic 1966).

Genus Sycute Dendy & Row, 1913

TYPE SPECIES. — Sycon dendyi Kirk, 1895 by mono-typy.

DIAGNOSIS. — Grantiidae with a syconoid organiza-tion. The cortex is supported by giant longitudinaldiactines. The distal part of the choanocyte chambersis crowned by fascicles of radial diactines locatedbetween the longitudinal diactines.

DESCRIPTION

This genus has a single species. Like Sycon, it ischaracterized by tufts of diactines that decoratethe distal cones of the radial tubes, and like Ute ithas longitudinal cortical giant diactines.

Genus Synute Dendy, 1892

TYPE SPECIES. — Synute pulchella Dendy, 1892 bymonotypy.


222 ZOOSYSTEMA • 2000 • 22 (2)

DIAGNOSIS. — Grantiidae with a cormus entirelymade of fused syconoid units and surrounded by acommon cortex with a special skeleton containinggiant longitudinal diactines (Dendy 1892a).

DESCRIPTION

This genus is monospecific and only known fromthe southern Australian coasts. Its organization isreminiscent of colonial ascidians such asBotryllus.

Genus Amphiute Hanitsch, 1894

TYPE SPECIES. — Amphiute paulini Hanitsch, 1894 bymonotypy.

DIAGNOSIS. — Grantiidae with a syconoid organiza-tion. Both cortical and atrial skeletons are supportedby giant longitudinal diactines.

DESCRIPTION

These are syconoid sponges that form largecormi, which are supported in both the atrial andcortical surfaces by giant longitudinal diactines(Fig. 20). Their relationship with the familyHeteropiidae has been discussed previously(Borojevic 1965).

Genus Sycodorus Haeckel, 1872

TYPE SPECIES. — Sycandra (Sycodorus) hystrix Haeckel,1872 by subsequent designation (Dendy & Row 1913).


223ZOOSYSTEMA • 2000 • 22 (2)

FIG. 19. — Diagram of a longitudinal radial section through the wall of Ute gladiata Borojevic, 1966. Abbreviations: a, atrium; as, atrial skeleton; ss, subatrial spicules; ar, articulate choanosomal skeleton; cx, cortex (from Borojevic 1966). Scale bar: 100 µm.

a

cx

ar

ss

as

DIAGNOSIS. — Grantiidae with a syconoid organiza-tion. Only the atrial skeleton is supported by giantlongitudinal diactines.

DESCRIPTION

Sycodorus is a variation of the type of spongesbelonging to the “group” Ute, whose skeleton isprovided with longitudinal diactines. They arecharacterized by the presence of longitudinal

diactines that are restricted to the atrial tangentialskeleton.

Genus Leucandra Haeckel, 1872

TYPE SPECIES. — Sycinula egedii Schmidt, 1870 bysubsequent designation (Dendy & Row 1913).

DIAGNOSIS. — Grantiidae with a sylleibid or leu-conoid organization. Longitudinal large diactines, ifpresent, are not restricted to the cortex, but lie


224 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 20. — Diagram of a transverse section through the wall of Amphiute lepadiformis Borojevic, 1967. Abbreviations: a, atrium; as,atrial skeleton; ss, subatrial spicules; ar, articulate choanosomal skeleton; cx, cortex (from Borojevic 1967b). Scale bar: 100 µm.

a

as

as

ss

ar

cx

obliquely across the external part of the sponge walland protrude from the surface of the sponge.

DESCRIPTION

This is a very large genus containing many species(Fig. 21). Initially it was defined primarily by neg-ative characters, and included most of the leu-conoid Calcarea. Dendy & Row (1913) narrowedthe definition of the genus, and succeeded in giv-ing it a more positive and circumscribed defini-tion. In particular, they clearly perceived thedifference between Leucetta and Leucandra, thefirst step required to separate a large group of leu-conoid Calcinea from the genus Leucandra. In thepresent study we separate another group of leu-conoid sponges, which have a particular skeletalorganization and had been classified as Leucandra,into the order Baeriida. At the same time, follow-ing Jenkin (1908a) and Brøndsted (1931), we sep-arate the sponges with thin walls and an inarticulatetype of choanoskeleton into the family Jenkinidae,and following Dendy (1913), we separate grantiidsponges with tetractines in the cortical skeletoninto the new genus Leucandrilla.As pointed out by Dendy & Row (1913),Leucandra can be derived from grantiid sponges bya progressive substitution of the syconoid aquifer-ous system by a sylleibid or a leuconoid organiza-tion, and a concomitant replacement of thechoanoskeleton of the tubes with a scattered one.Nonetheless, traces of the original radial organiza-tion are clearly preserved in the subatrial skeleton.Several authors have considered Leuconia Grant,1841 as a senior synonym of Leucandra. As shownby Vosmaer (1887) and Dendy (1893), Leuconiahas to be rejected, being previously used for a genusof mollusks. Leucandra Haeckel, 1872 being a validsynonym is the valid name of the taxon.Leucandra has numerous representatives in alloceans.

Genus Aphroceras Gray, 1858

TYPE SPECIES. — Aphroceras alcicornis Gray, 1858 bymonotypy.

DIAGNOSIS. — Grantiidae with a leuconoid organiza-tion. The cortex is supported, at least in part, by giantlongitudinal diactines.

DESCRIPTION

Aphroceras is differentiated from Leucandra bythe presence of internal longitudinal diactines inthe cortex (Fig. 22). In a previous study(Borojevic 1966), it was shown that the numberof these spicules can be quite variable, rangingfrom a continuous dense layer to only very rarespicules, or even absence. In the latter case, it isnot possible to distinguish this sponge from atypical Leucandra. However, we retain this genusat present, as we feel that the typical Aphrocerasare easy to identify. Dendy & Row (1913)pointed out that Aphroceras probably derivesdirectly from Leucandra by a secondary acquisi-tion of longitudinal internal diactines, and notfrom Ute by a modification of the syconoid


225ZOOSYSTEMA • 2000 • 22 (2)

FIG. 21. — A transverse section through the wall of Leucandraaspera (Schmidt, 1862) (light micrograph). Abbreviations: a,atrium; as, atrial skeleton; ch, choanosome; cx, cortex. Scalebar: 230 µm.

cx

ch

as

a

aquiferous system into the leuconoid one. Ourstudies on A. ensata (Bowerbank, 1858)(Borojevic 1966), however, point to a close rela-tionship between Ute and Aphroceras.

Genus Leucandrilla n. gen.

TYPE SPECIES.—Leucilla wasinensis Jenkin, 1908 BMNH

1908.9.25.59 by original designation. Not Leuconia wasi-nensis BMNH 1936.3.4.537, in Burton (1959).

DIAGNOSIS. — Grantiidae with a leuconoid organiza-tion. In addition to triactines the cortex containstetractines, with the apical actines turned into thechoanoderm. The articulate choanoskeleton is sup-ported by subatrial triactine spicules, and numerousrows of choanosomal triactines and/or tetractines, withapical actines of cortical tetractines in the distal region.


226 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 22. — Diagram of a transverse section through the wall of Aphroceras ensata (Bowerbank, 1858). Abbreviations: a, atrium; as,atrial skeleton; ss, subatrial spicules; ch, choanosome; cx, cortex (from Borojevic 1966). Scale bar: 100 µm.

cx

ch

ss

as

a

DESCRIPTION

Leucandrilla is differentiated from Leucandra bythe presence of tetractines in the cortical skeleton(Fig. 23). Like Aphroceras, which is distinguishedfrom Leucandra by the presence of corticaldiactines, the separation between Leucandrilla andLeucandra is not clear-cut. Dendy (1913), Dendy& Row (1913) and Borojevic & Boury-Esnault(1987) have already pointed out that some of thesponges with cortical tetractines that had beenclassified among Amphoriscidae are not very dif-ferent from a typical Leucandra, but are quite dis-tinct from Leucilla, and should consequently beclassified in the Grantiidae. In particular they havea complete articulate choanoskeleton, reminiscentof the grantiid organization, which is absent inAmphoriscidae. While Dendy (1913) proposedthat these sponges should be included inLeucandra, we now propose to isolate them in aseparate genus in the family Grantiidae, analogouswith the recognized separation of Aphroceras. Itshould be noted that we consider the corticaltetractines in this genus to be a secondary charac-ter, corresponding to a modification of normal cor-tical triactines, while cortical tetractines of thefamily Amphoriscidae are a primary character,marking an independent evolutionary line.In addition to L. wasinensis (Jenkin, 1908b),which we propose to be the type species of thegenus Leucandrilla, other sponges that had beenclassified in the genus Leucandra, such as L. inter-media (Row, 1909) and L. lanceolata (Row &Hôzawa, 1931), also belong to this genus.

Genus Leucettaga Haeckel, 1872

TYPE SPECIES. — Leucetta (Leucettaga) loculiferaHaeckel, 1872 by subsequent designation (Dendy &Row 1913).

DIAGNOSIS. — Grantiidae (?) with a leuconoid organi-zation. The skeleton is composed of only triactines,arranged without apparent order in the cortex and inthe choanosome. The atrium is crossed by numeroussepta, which possess a special skeleton containingminute triactines.

DESCRIPTION

Dendy & Row (1913) retained the genusLeucettaga with a single species, Leucetta

(Leucettaga) loculifera Haeckel, 1872 for thesponge described as one of the subspecies ofL. pandora Haeckel, 1872, based on the presenceof spicular tracts in the atrium whose skeleton isquite different from that of the sponge wall.Sponges described by Haeckel (1872) underL. pandora are quite heterogeneous, containingthe most divergent forms of spicules. The draw-ing representing the sponge wall organization(Haeckel 1872: table 22, 3b and c) shows quitean unusual structure, and is difficult to interpret.We have considered the presence of atrial tractsthat have a specific skeleton as a distinctive char-acter for the genus Sycandra, and consequentlywe retain the genus Leucettaga in the scope pro-posed by Dendy & Row (1913). However, this


227ZOOSYSTEMA • 2000 • 22 (2)

t

FIG. 23. — Leucandrilla organization. Section made on the typespecimen from Jenkin (1908b) (BMNH 1908.9.25.59). The cortexcontains tetractines (t), with the apical actines turned into thechoanoderm. Scale bar: 150 µm.

sponge is so incompletely described that it isquite doubtful whether it belongs to the familyGrantiidae, in which a skeleton composed of onlytriactines is quite unusual. The precise classifica-tion of this genus will only be possible after theexamination of new specimens.

Family SYCANTHIDAE Lendenfeld, 1891

TYPE GENUS. — Sycantha Lendenfeld, 1891 by origi-nal designation.

DIAGNOSIS. — Leucosoleniida with an irregularsyconoid organization, and the skeleton primarily sup-ported by triactine spicules, with occasionally diactinesin the distal cones. The large central atrium bears numer-ous short radial tubes lined by choanoderm. Radial tubesare grouped and fused proximally, each group commu-nicating through a wide opening with the atrial cavity.The distal free or coalescent cones are intercalated bylarge inhalant spaces, which often reach the external sur-face of the atrial skeleton. When coalescent, distal conescan have tangential triactines, but there is no continu-ous cortex covering the choanosome and delimiting theinhalant cavities externally.

DESCRIPTION

We propose to include a small group of Leuco-soleniida, which are derived from sponges with asycettid type of organization and have a particulartype of growth, in the family Sycanthidae, in a sim-ilar scope to the subfamily Sycanthinae proposedby Lendenfeld (1891). In these sponges, a thin wallsurrounds a large atrial cavity that has numerousshort radial tubes, which are not regularly distrib-uted on the central atrium but form groups whichcommunicate with the central atrial cavity by a largeopening. Distally, the grouped radial tubes bearindividual cones (e.g. Sycantha), which maybecome coalescent and protected by tangentialspicules similar to those present in the radial tubes(e.g. Dermatreton). Despite the presence of thesespicules, a continuous cortex is not formed, butrather a loose cortical network perforated by largeopenings of the inhalant cavities covers the distalregions of the radial tubes. The inhalant spaces leftbetween the groups of radial tubes are quite large,and can reach the external face of the atrial wall,giving the external side of the sponge a honey-combed appearance. Lendenfeld (1891) observedthat the radial tubes communicate among them-

selves in the proximal region, and that the waterflow passes from one tube to another throughpores, before reaching the atrial cavity. Dendy(1892b), and subsequently Jenkin (1908a) whohad the opportunity to examine the type specimendescribed by Lendenfeld (1891), refuted thisinterpretation. Both Jenkin (1908a) and Dendy &Row (1913) considered Sycantha tenellaLendenfeld, 1891 as an aberrant species of Sycon.However, having observed sponges with a similartype of growth in the National AntarcticExpedition collections, Jenkin (1908a) proposedthe genera Tenthrenodes, Hypodictyon andDermatreton for sponges with chambers that arefused in the proximal region, in an almost identi-cal manner to that in the genus Sycantha (Dendy& Row, 1913). We thus consider that Tenthrenodesand Hypodictyon are synonyms of Sycantha, whilewe retain the genus Dermatreton for sponges withlinked choanocyte chambers that have developedan external tangential meshwork that is supportedby tangential spicules, corresponding topologicallyto a cortex, but differing from it by the fact that itdoes not delimit an inhalant aquiferous systemexternally.

Genus Sycantha Lendenfeld, 1891

TYPE SPECIES. — Sycantha tenella Lendenfeld, 1891 bymonotypy.

DIAGNOSIS. — Sycanthidae that have fused radialtubes with free distal cones decorated by diactinespicules.

DESCRIPTION

Only Lendenfeld (1891) observed Sycantha tenel-la from a specimen collected in the northern partof the Adriatic Sea, and he gave quite a detaileddescription of this species. Tenthrenodes antarc-ticum (Jenkin, 1908) is similar to pedunculatesmall Sycon species; the description of the linkedtype of radial tubes is not fully convincing, andfollowing Dendy & Row (1913) we propose tokeep it in the genus Sycon. As pointed out byDendy & Row (1913), the sponge described asTenthrenodes scotti Jenkin, 1908 has tangentialspicules at the distal parts of the radial tubes. Thisspecies has the organization typical of the Sycan-


228 ZOOSYSTEMA • 2000 • 22 (2)

thidae and belongs to the genus Dermatreton aswe understand it now. Sycantha (Hypodictyon)longstaffi (Jenkin, 1908) is apparently one of thetypical representatives of the genus. As discussedunder the family Staurorrhaphidae, the presenceof the subatrial spicules with a lone centrallydirected apical actine, is common in manyLeucosoleniida, and does not merit the separa-tion of the genus Hypodictyon from Sycantha.

Genus Dermatreton Jenkin, 1908

TYPE SPECIES. — Dermatreton hodgsoni Jenkin, 1908by subsequent designation (this work).

DIAGNOSIS. — Sycanthidae with coalescent radialtubes whose distal parts are supported by tangentialtriactines that form a loose meshwork perforated bylarge inhalant cavities.

DESCRIPTION

We use the genus Dermatreton in the mannerproposed by Jenkin (1908a). The loose cortex,which covers the distal parts of fused radial tubes,is in the form of a meshwork with broad open-ings formed by the inhalant spaces. As such itcannot give sufficient mechanical rigidity to thesponge, and consequently the atrial skeleton isthickened and rigid. Jenkin (1908a) has not des-ignated the type species of the genus. Among theoriginally included species we designate D. hodg-soni as the type species. Similar morphology isobserved in Dermatreton (Tenthrenodes) scotti(Fig. 24). The description and illustrations ofDermatreton chartaceum suggest that it should beincluded in the genus Breitfussia.

Family JENKINIDAE n. fam.

TYPE GENUS. — Jenkina Brøndsted, 1931 by originaldesignation.

DIAGNOSIS. — Leucosoleniida with a syconoid,sylleibid or leuconoid organization. The thin wall sur-rounding the large atrial cavity is supported by tangen-tial atrial and cortical skeletons, and essentially aninarticulate choanoskeleton consisting of unpairedactines of the subatrial triactines and/or tetractines,and occasionally with small radial diactines. The prox-imal part of the large radial diactines that protrudefrom the external surface, or the tangential triactines

scattered irregularly in the cortex, may also form thechoanoderm. Large cortical tetractines or subcorticalpseudosagittal triactines are not present.

DESCRIPTION

We propose the family Jenkinidae for a group ofsponges characterized by an inarticulate choanos-keleton (Fig. 25). Dendy & Row (1913) consid-ered this character not to be relevant at the genericlevel, and only Brøndsted (1931) proposed sepa-rating the leuconoid sponges with an inarticulateskeleton into the genus Jenkina. However, a pri-mary inarticulate type of choanoskeleton is a char-acteristic of the family Amphoriscidae, in which itis always associated with the presence of large cor-tical tetractines (Borojevic & Boury-Esnault1987). Dendy (1913) and Dendy & Row (1913)underlined the difference between the sponges ofthe genus Leucilla that have an inarticulate type ofchoanoskeleton and which derive fromAmphoriscus, and those with an articulate skeleton,by transferring the former group to the genus


229ZOOSYSTEMA • 2000 • 22 (2)

FIG. 24. — Diagram of the Dermatreton type of sponge wallorganization. Abbreviations: ch, choanoderm; ic, inhalant cavi-ties; cx, cortex.

ch

ic

cx

Leucandra (placed now into the new genusLeucandrilla). They thus implied that the inartic-ulate type of the choanoskeleton, and not the cor-tical tetractines, is the primary characteristic of thefamily Amphoriscidae. We now consider that theinarticulate type of sponge wall organization, witha thin choanoderm and well-defined atrial and cor-tical skeletons, is a consequence of a particular typeof growth, and is not a secondary reduction of thesponge wall thickness during evolution.Consequently, sponges with this organizationshould be separated from those with a massive typeof growth as is observed in the Grantiidae. Whilein the Amphoriscidae the cortical skeleton is alwayssupported by large tetractines, in the Jenkinidae itcan be thin (e.g. Jenkina, Leucascandra), or rein-forced by large diactines or triactines (e.g. Uteopsisand Anamixilla, respectively). It should be empha-sized that young specimens of Grantiidae, and thesuboscular region of adult Grantiidae in which thesponge grows longitudinally, can have an inartic-ulate skeleton that becomes an articulate one when

sponge is fully grown. Conversely, the Jenkinidaeare characterized by an inarticulate skeleton in thefully-grown sponges. As mentioned previously thegrowth of these long tubular sponges into a largebranched cormus such as observed in Leucascandra,Anamixilla and Uteopsis is a consequence of therestriction of their radial growth, and this is uniquein the Leucosoleniida.

Genus Breitfussia n. gen.

TYPE SPECIES. — Ebnerella schulzei Breitfuss, 1896.

DIAGNOSIS. — Jenkinidae with a simple tubular bodyand syconoid organization. The choanoskeleton isreduced to the unpaired actines of the subatrial tri-actines, and occasionally contains the proximal part ofradial diactines.

DESCRIPTION

In the system proposed by Lendenfeld (1891),and followed by Breitfuss (1896), the genusEbnerella of the subfamily Amphoriscinae wascharacterized by an inarticulate skeleton, contain-ing diactines, triactines and/or tetractines. Thespecies included by Lendenfeld (1891) in thisgenus belongs now to the genus Amphoriscus andEbnerella is thus a junior synonym of Ampho-riscus. Among the species described by Breitfuss(1896) in Ebnerella, E. kuekenthali belongs to thefamily Heteropiidae (Sycettusa), but a new nameis needed for E. schulzei. In addition, species withan inarticulate choanoskeleton, described underthe genus Grantia or Dermatreton, should beincluded in the new genus Breitfussia as nowdefined, such as Breitfussia (Grantia) vitiosa(Brøndsted, 1931) and Breitfussia (Dermatreton)chartacea (Jenkin, 1908). Breitfussia is knownonly from cold Arctic or Antarctic waters.

Genus Jenkina Brøndsted, 1931

TYPE SPECIES. — Leucandra hiberna Jenkin, 1908 bysubsequent designation (Laubenfels 1936).

DIAGNOSIS. — Jenkinidae with a simple tubular bodyand a sylleibid or leuconoid organization of the aquif-erous system. The choanoskeleton is composed of theunpaired actine of subatrial spicules and, occasionally,the proximal part of radial diactines that cross thesponge wall.


230 ZOOSYSTEMA • 2000 • 22 (2)

cx ss as

a

FIG. 25. — Diagram of the Jenkinidae-type of skeleton.Abbreviations: a, atrium; as, atrial spicules; ss, subatrialspicules; cx, cortex.

DESCRIPTION

Brøndsted (1931) proposed the genus Jenkina fora group of sponges described by Jenkin (1908a) andby himself from Antarctica. These sponges are well-characterized by an inarticulate choanoskeletonthat contains unpaired actines of subatrial spiculesand occasionally radial diactines, which cross thethin choanoderm and protrude from the externalsurface of the sponge. The thin choanosome fre-quently lacks a typical leuconoid aquiferous system,and some doubt remains about the divisionbetween the genera Breitfussia and Jenkina.

Genus Leucascandra Borojevic & Klautau, 2000

TYPE SPECIES. — Leucascandra caveolata Borojevic &Klautau, 2000 by monotypy.

DIAGNOSIS. — Jenkinidae with a complex cormuscomposed of copiously branched and anastomosed

tubes. Each tube has a thin wall with a rather irregularalveolar type of leuconoid aquiferous system, and aninarticulate choanoskeleton that is supported only byunpaired actines of subatrial triactines. Both corticaland atrial skeletons consists of a thin layer of tangentialtriactines and/or tetractines.

DESCRIPTION

In the Clathrinida, tubular sponges frequently forma large cormus composed of ramified and anasto-mosed tubes (e.g. Clathrina, Ascandra, Ascaltis,Leucascus, Leucaltis). In the Leucosoleniida, thisgrowth form is quite rare. While the Jenkinidaefrom cold Antarctic or Arctic waters grow as smallsolitary tubes, those from warmer waters can formlarge complex cormi. The genus Leucascandra isthus characterized by a tendency to form a large cor-mus composed of extensively branched and anas-tomosed tubes (Fig. 26), an inarticulatechoanoskeleton, and a thin cortex (Fig. 27).


231ZOOSYSTEMA • 2000 • 22 (2)

FIG. 26. — Specimen of Leucascandra caveolata Borojevic &Klautau from New Caledonia (Poindimié, 30 m) (PhotoP. Laboute) (from Borojevic & Klautau 2000). Scale bar: 1.4 cm.

FIG. 27. — Leucascandra organization. Transverse sectionthrough the sponge wall; the choanoskeleton is composed ofonly subatrial triactines and the atrial skeleton contains tri-actines and tetractines. Abbreviations: a, atrial skeleton; ch,choanoskeleton. Scale bar: 40 µm.

a

ch

Genus Anamixilla Poléjaeff, 1883

TYPE SPECIES. — Anamixilla torresi Poléjaeff, 1883 bymonotypy.

DIAGNOSIS. — Jenkinidae with a syconoid organiza-tion. The thick cortex is supported by many layers oftriactines. The choanoskeleton is composed of theunpaired actine of the subatrial spicules, and giant tan-gential triactines similar to those in the cortex butlying scattered in the choanosome.

DESCRIPTION

In Anamixilla, large triactines form a thick cortexand apparently invade the choanosome (Fig. 28).A similar phenomenon is observed in theLelapiidae, in which the reduction of the classicalchoanoskeleton is concomitant with its partialsubstitution by large cortical diactines. Largediactines have been observed in the oscular regionof A. irregularis Burton, 1930. We have examinedthe specimen deposited in the British Museum(BMNH 1929.8.30.6) and found that it does notbelong to the genus Anamixilla.

Genus Polejaevia n. gen.

TYPE SPECIES. — Polejna telum Lendenfeld, 1891 bymonotypy.

DIAGNOSIS. — Jenkinidae with a sylleibid organiza-tion. The cortex is supported by a layer of large tan-gential triactines. The choanoskeleton is composed of

the unpaired actines of the subatrial triactines, and ofrare small scattered triactines.

DESCRIPTION

In the system proposed by Lendenfeld (1891),the genus Polejna Lendenfeld, 1885 was used forsylleibid sponges with triactines and tetractines.The type species of Polejna, described originallyas Leucilla uter Poléjaeff, 1884 is in fact a goodspecies of Leucilla (Borojevic & Boury-Esnault1987). Polejna is thus a junior synonym ofLeucilla. Subsequently, Lendenfeld (1891)described in Adriatic a new species in the genus,Polejna telum Lendenfeld, 1891, that we considerto be different from Amphoriscidae and placenow in the family Jenkinidae. A new name is thusrequired for the Jenkinidae with a sylleibid orga-nization and triactines and tetractines such asPolejna telum, for which we propose Polejaevia.The position of Polejaevia in the family Jenkinidaeis somewhat dubious, since small choanosomaltriactines have been described in it and are repre-sented in the illustration of the type species, dis-tinguishing it from typical Jenkinidae. The sizeand distribution of the triactines is quite unusual,and as they are not reminiscent of the articulatechoanoskeleton of the tubes of the Grantiidae,classification of P. telum in the genus Leucandra isimpossible. Lendenfeld (1891) suggested that thetriactines might be young cortical triactines.Secondary spicules may be found in thechoanosome in the absence of any other skeleton,as seen in the genus Leucettusa Haeckel, 1872(Borojevic et al. 1990). On the other hand, the or-ganization of Polejaevia can be understood to bequite similar to Anamixilla: while in the formerthe additional triactines in the choanosome arenew spicules, in Anamixilla the cortical spiculesapparently invade the choanoskeleton. We pro-pose that the genus should be maintained in theJenkinidae until new specimens are examined.The description of Leucandra mawsoni Dendy,1918 suggests that it might belong to Polejaevia.We have examined the specimens deposited inthe British Museum (BMNH 20.12.9.95) andfound that this species is a calcinean sponge,belonging to the genus Leucascus. Hence Polejaeviatelum is the only known species belonging to thisgenus.


232 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 28. — Diagram of a transverse section through the wall ofAnamixilla torresi Poléjaeff, 1883. Abbreviations: a, atrium; as,atrial skeleton; ss, subatrial spicule; ch, choanosome; cx, cor-tex (from Poléjaeff 1883).

cx

ch

assas

Genus Uteopsis Dendy & Row, 1913

TYPE SPECIES. — Ute argentea Poléjaeff, 1883 bymonotypy.

DIAGNOSIS. — Jenkinidae with a syconoid organiza-tion. The cortex is thick and is supported by giant lon-gitudinal diactines. The choanoskeleton is reduced tothe unpaired actines of the subatrial spicules andsmaller distal radial diactines.

DESCRIPTION

Uteopsis is well-described and illustrated byPoléjaeff (1883). It is characterized by an inartic-ulate choanoskeleton and a thick cortex com-posed of giant longitudinal diactines andtriactines.

Family HETEROPIIDAE Dendy, 1892

TYPE GENUS. — Heteropia Carter, 1886 by originaldesignation.

DIAGNOSIS. — Leucosoleniida with a syconoid or leu-conoid organization. The choanoskeleton is composedof a proximal layer of subatrial triactines and a distinctdistal layer of pseudosagittal triactines and/or pseu-dosagittal tetractines, often separated by an intermedi-ate layer that is supported by several rows of triactinesand/or tetractines. The atrial skeleton is well-devel-oped.

DESCRIPTION

The family Heteropiidae is characterized by thepresence of a layer of subcortical pseudosagittalspicules. At a first glance, these spicules may seemto be sagittal triactines with the paired actinesadjacent to the cortex, and the unpaired actineturned inwards in a position symmetrical to thatof the subatrial spicules. However, as indicated byPoléjaeff (1883) and Dendy & Row (1913), boththe length and the form of the paired actines inthese spicules are unequal. The longer pairedactine is perpendicular to the cortex, while theshorter one as well as the unpaired actine areadjacent to the cortex (Fig. 29). This position isclearly observed for the distal triactines of theradial tubes in sycon-like sponges that we nowplace in the genus Syconessa. This indicates thatthe formation of pseudosagittal spicules precedesthe formation of a cortex and appears early in theevolution of the Heteropiidae, immediately after

the acquisition of the sycettid type of organiza-tion. The evolutionary pathway of the Hetero-piidae, well-represented by the genus Syconessa,diverges from sponges that are very similar toSycon. In the type species Syconessa syconiformis(Borojevic, 1967a), the choanoskeleton is inartic-ulate, or has only a few spicules in the proximalpart of the choanoskeleton. The corticalization ofsuch a sponge can lead both to the genus


233ZOOSYSTEMA • 2000 • 22 (2)

FIG. 29. — Diagram of one parasagittal (p) and two sagittal (s)spicules in Heteropiidae. The arrow indicates the unpairedangle, and u indicates the unpaired actine. The double line rep-resents the cortex (cx). Note that the two paired actines of theparasagittal spicules are of quite different sizes, and they arealways in a subcortical position with the longer paired actinedirected inwards.

cx

u

s

p

u

u

Sycettusa, which is characterized by a thin bodywall with the choanosome devoid of its ownskeleton, and to Grantessa , in which thechoanoskeleton is articulate. It should be notedthat several species of Grantessa have a very thinand/or poorly defined cortex, and the distal conesof the radial tubes are still decorated by tufts ofdiactines (e.g. Grantessa ramosa Haeckel, 1872),clearly showing their relationships to spongeswith a sycon-like organization.Corticalization has apparently arisen several timesin the Leucosoleniida. In the family Hetero-piidae, corticalization associated with the mainte-nance of the choanoskeleton of the tubes that isreduced to subatrial and subcortical spicules hasgiven rise to the genus Sycettusa, whilst the corti-calization associated with the thickening of thechoanosome has produced Grantessa. The subse-quent transition to the leuconoid type of organi-

zation is seen in the genus Vosmaeropsis. Similarprogression is observed among the Leucoso-leniida that lack pseudosagittal spicules, in whichthe first route has given rise to the Jenkinidae andthe Amphoriscidae, and the second the Gran-tiidae.The family Heteropiidae contains a series of gen-era that are analogous to those of the familyGrantiidae, the sole difference being the presenceof subcortical pseudosagittal spicules. In any cal-caronean sponge with a strong cortex, some sub-cortical spicules may be in the position and havethe shape of pseudosagittal spicules, due to therestriction of their growth by the rigidity of the cor-tical skeleton. They should not be interpreted asan indication that the sponge belongs to the fam-ily Heteropiidae (see discussion on Amphiutepaulini Hanitsch, 1894 in Borojevic 1965: 665-670). Consequently, the regular presence of a dis-


234 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 30. — Diagram of a transverse section through the wall of Syconessa syconiformis (Borojevic, 1967a). Abbreviations: a, atrium;as, atrial skeleton; ss, subatrial spicule; ps, parasagittal spicules; dc, distal cones (from Borojevic 1967b). Scale bar: 50 µm.

a

as

ss

ps

dc

tinct layer of subcortical pseudosagittal spiculesresembling the continuous layer composed exclu-sively of these spicules found in Grantessa, shouldbe interpreted as an indication that the spongebelongs to the Heteropiidae, whilst isolated pseu-dosagittal spicules should be understood to be thesecondary modification of subcortical spicules thatis found in some representatives of the Grantiidae.We are aware that this distinction is often unclear,and a search for complementary cytological or bio-chemical criteria should be undertaken to identifythose features that could distinguish the Grantiidaefrom the Heteropiidae, to allow the correct classi-fication of some of the problematic cases.

Genus Syconessa n. gen.

TYPE SPECIES. — Grantessa syconiformis Borojevic,1967 by original designation.

DIAGNOSIS. — Heteropiidae with short radial tubessupported by a skeleton composed of a proximal rowof subatrial triactines and distal pseudosagittal tri-actines, with occasional tube spicules. The pairedactines of proximal subatrial spicules are adjacent tothe atrial skeleton, while the unpaired actine is in thewall of the radial tube. The shorter paired actine andthe unpaired actine of the distal pseudosagittal spiculessupport the distal cones, while the longer paired actineis inside the wall of the radial tube.

DESCRIPTION

As discussed by Dendy & Row (1913) and byBorojevic (1965, 1967a), the pseudosagittalspicules are derived from the spicules of the distaltubes rather than from the cortical spicules. InSyconessa, distinct pseudosagittal spicules can beseen at the distal part of the radial choanocytechambers, where their unpaired actine and theshorter paired actine participate in the skeleton ofthe distal cones, while the centripetal longer pairedactine of pseudosagittal triactines support theexternal part of the radial tubes (Fig. 30). Thisgenus is the starting point of the evolutionary lineof the Heteropiidae. We now propose the newgenus Syconessa for sponges with a syconoid typeof organization and a layer of pseudosagittalspicules in the distal part of their choanocyte cham-bers, but without a cortex. Dendy & Row (1913)pointed out that Sycon ensiferum Dendy, 1892 alsohas triactines that have a typical form of pseu-

dosagittal spicules in the distal part of radial tubes,rendering the species almost indistinguishablefrom Grantessa. We have examined the slides of thisspecies in the British Museum (BMNH 93.6.9.6a,25.11.1.1746/47), and found that indeed thissponge belongs to the genus Sycon, despite the occa-sional presence of spicules of the pseudosagittal type.

Genus Sycettusa Haeckel, 1872 emend.

TYPE SPECIES. — Sycetta (Sycettusa) stauridia Haeckel,1872 by monotypy.

DIAGNOSIS. — Heteropiidae with a syconoid organi-zation. Atrial and cortical skeletons are formed by tangential triactines and/or tetractines. The choanos-keleton is inarticulate, and is composed of unpairedactines of the subatrial triactines, and of centripetalactines of the pseudosagittal subcortical triactines.


235ZOOSYSTEMA • 2000 • 22 (2)

cx ps ss as

a

FIG. 31. — Diagram of the Sycettusa inarticulate skeleton.Abbreviations: a, atrium; as, atrial spicules; ss, subatrialspicules; ps, parasagittal subcortical spicules; cx, cortex.

DESCRIPTION

We propose to divide the genus Grantessa (asdefined by Dendy & Row 1913) into two groups:one with an inarticulate choanoskeleton, to becalled Sycettusa, and the other with an articulatechoanoskeleton, to be called Grantessa. As statedearlier, we consider that the former genus evolvedby the corticalization of sponges with an inarticu-late skeleton similar to Syconessa syconiformis(Borojevic, 1967), thereby maintaining this char-acteristic of the choanoskeleton (Fig. 31). Thesponges assembled in the genus Sycettusa can bedivided into two groups, one common in the Arcticregion and the other in the Indo-Pacific. The rela-tionship between these two groups remains to beestablished. The Arctic group includes Sycettusa(Sycaltis) glacialis (Haeckel, 1872), S. (Ebnerella)kuekenthali (Breitfuss, 1896), S. (Ebnerella) lance-olata (Breitfuss, 1898), S. (Amphoriscus) murma-nensis (Breitfuss, 1898), S. (Amphoriscus) thompsoni(Lambe, 1900) and S. (Ebnerella) nitida (Arnesen,1901). The Indo-Pacific group includes S. stau-ridia Haeckel, 1872, S. (Sycortis) sycilloides (Schuf-fner, 1877), S. (Amphoriscus) poculum (Poléjaeff,1883), S. (Grantessa) simplex (Jenkin, 1908b),S. (Grantessa) glabra (Row, 1909) and S. (Grantessa)hastifera (Row, 1909).Haeckel (1872) proposed the subgenus Sycettusafor the single species S. stauridia from the RedSea, which is a typical syconoid Heteropiidaewith an inarticulate choanoskeleton, and weretain this genus and species name in the samecombination.

Genus Grantilla Row, 1909

TYPE SPECIES. — Grantilla quadriradiata Row, 1909by monotypy.

DIAGNOSIS. — Heteropiidae with a syconoid organi-zation. The skeleton of the tubes is inarticulate, composed of subatrial triactines, and subcorticalpseudosagittal triactines and tetractines.

DESCRIPTION

The vast majority of Heteropiidae have only pseu-dosagittal triactines in the subcorticalskeleton. The genus Grantilla has been proposedfor the single species G. quadriradiata Row, 1909,

with pseudosagittal tetractines. Other mor-phological characteristics are quite similar toSycettusa.

Genus Grantessa Lendenfeld, 1885

TYPE SPECIES. — Grantessa sacca Lendenfeld, 1885 bymonotypy.

DIAGNOSIS. — Heteropiidae with a syconoid organiza-tion and an articulate choanoskeleton. A thin cortex isformed by triactines but lacks longitudinal largediactines. The distal part of the radial tubes is frequentlydecorated by tufts of radially arranged diactines, indi-cating a close relationship to the genus Syconessa.

DESCRIPTION

We include in the genus Grantessa s.s. thesyconoid Heteropiidae with articulate choanos-keletons (Fig. 32). We consider that they arederived from sponges similar to Syconessa, inwhich the increase of radial tubes had generatedthe articulate choanoskeleton. Grantessa are com-mon in warm seas, and often grow as largearborescent or bushy cormus.

Genus Heteropia Carter, 1886

TYPE SPECIES. — Aphroceras ramosa Carter, 1886 bymonotypy.

DIAGNOSIS. — Heteropiidae with a syconoid organiza-tion, an articulate choanoskeleton, and where the cor-tical skeleton consists of longitudinal large diactines,with occasionally tangential triactines and perpendicu-lar small diactines.

DESCRIPTION

Heteropia in the family Heteropiidae correspondsto the same grade of skeletal complexity than Utein the family Grantiidae.

Genus Paraheteropia Borojevic, 1965

TYPE SPECIES. — Amphiute ijimai Hôzawa, 1916 bymonotypy.

DIAGNOSIS. — Heteropiidae with a syconoid organiza-tion, an articulate choanoskeleton, and with both cor-tical and atrial skeletons containing longitudinaldiactines.


236 ZOOSYSTEMA • 2000 • 22 (2)

DESCRIPTION

Paraheteropia in the family Heteropiidae corre-sponds to the same grade of skeletal complexitythan Amphiute in the family Grantiidae.

Genus Vosmaeropsis Dendy, 1892

TYPE SPECIES. — Heteropia macera Carter, 1886 bysubsequent designation (Dendy & Row 1913).

DIAGNOSIS. — Heteropiidae with a sylleibid or leu-conoid organization. The choanoskeleton is composedof proximal subatrial triactine spicules and an irregularlayer of scattered triactines and tetractines.

DESCRIPTION

The genus Vosmaeropsis most often has a typicalleuconoid grade of organization, and correspondsclosely to the genus Leucandra in the familyGrantiidae. When the sponge wall is thick, theskeleton has a tendency to be irregular, with theprogressive loss of traces of the radial organiza-tion. While the layer of subatrial triactines ortetractines in general is well-preserved, and theproximal part of the choanoskeleton is supportedexclusively by unpaired actines of these spicules,the distal layer of the choanoskeleton, that is sup-ported by centripetal rays of pseudosagittal sub-cortical spicules, becomes blurred by the invasionby other choanosomal spicules. Alternatively, thepseudosagittal subcortical spicules and the facingsubatrial spicules can retain their original relation-ship, and the thickening of the wall can be obtainedby insertion of new spicules between the subatrialspicules and the atrial skeleton, such as observedin Vosmaeropsis hozawai Borojevic & Klautau,2000. This is similar to the secondary thickeningof the sponge wall in the genus Paraleucilla. In thesecases, the distinction between Vosmaeropsis andLeucandra is very difficult.

Family AMPHORISCIDAE Dendy, 1892

TYPE GENUS. — Amphoriscus Haeckel, 1870 by origi-nal designation.

DIAGNOSIS. — Leucosoleniida with a syconoid,sylleibid or leuconoid organization, and a distinct cor-tex supported by tangential tetractines whose cen-tripetal apical actines cross the outer part of or thewhole of the choanosome. Tangential triactines and

small tetractines may be also present in the cortex. Thechoanoskeleton typically is inarticulate, composed ofthe apical actines of cortical tetractines and theunpaired actines of subatrial spicules. In species with athick wall scattered triactines and/or tetractines may bealso present, either among the spicules of the inarticu-late choanoskeleton, or forming a distinct subatriallayer. An atrial skeleton is always present.

DESCRIPTION

The family Amphoriscidae is well-characterizedby a distinct subcortical layer exclusively support-


237ZOOSYSTEMA • 2000 • 22 (2)

FIG. 32. — Diagram of a transverse section through the wall ofGrantessa ramosa (Haeckel, 1872). Abbreviations: a, atrium; as,atrial skeleton; ar, articulate choanosomal skeleton; ss, subatrialspicule; ps, parasagittal spicules; cx, cortex (from Borojevic1967a). Scale bar: 100 µm.

cx

ps

ar

ss

asa

ed by the apical actines of giant corticaltetractines. In species of Leucilla that have a leu-conoid organization and a thick wall, thechoanoskeleton is disorganized, unlike the sim-pler syconoid genus Amphoriscus (Fig. 33). Weunderstand this to indicate that the simple inar-ticulate choanoskeleton is a primitive condi-tion. This is an argument against the derivationof the Amphoriscidae from the typicalGrantiidae, by reduction of the choanoskeletonand the secondary presence of the apical actinesof cortical triactines. The regular presence of sub-atrial triactines in the Amphoriscidae clearly indi-cates that they derive from a Sycetta type oforganization through the precocious develop-ment of a cortical skeleton reinforced by gianttetractines. Since the articulate skeleton of thetubes, typical of adult specimens of Sycon andGrantia, is not found in the Amphoriscidae, theorigin of this family lies closer to the Jenkinidaethan to Grantiidae. However, it is impossible to

decide whether these two families have a com-mon origin or have evolved independently.In most Leucosoleniida, the thickening of thechoanosome is associated with the growth of newspicules in the central part of the choanoskeleton,and with the increasing distance between the cor-tical skeleton on one side, and the atrial and sub-atrial skeletons on the other (e.g. Grantessaramosa, Fig. 32). In the Amphoriscidae, thesponge wall can thicken through the addition of anew layer between the atrial and subatrial skele-tons, and the subatrial and cortical skeletonsstrictly maintain their close primary relationship(Fig. 33). The invasion of the choanoskeleton byspicules derived from the inner atrial skeleton inParaleucilla is analogous to the invasion of thechoanoskeleton from the outer cortical skeletonin the Jenkinidae (e.g. Anamixilla and Uteopsis).In both families, the primary inarticulatechoanoskeleton is preserved, despite the progres-sive thickening of the sponge wall and the neces-sity to introduce new skeletal structures tosupport it.

Genus Amphoriscus Haeckel, 1870

TYPE SPECIES. — Ute chrysalis Schmidt, 1864 by subse-quent designation (Dendy & Row 1913).

DIAGNOSIS. — Amphoriscidae with a syconoid organi-zation of the aquiferous system. Scattered spicules inthe choanosome are always absent.

DESCRIPTION

Amphoriscus is a well-characterized genus, andcontains several species of solitary sponges withbeautiful vitreous transparent walls. The genus isfound in all the oceans.Dendy & Row (1913) kept the genus SyculmisHaeckel, 1872 for Amphoriscidae with a root-tuft of diactines and anchoring tetractines. Wehad the opportunity to observe similar anchoringstructures in the families Sycettidae andJenkinidae, and we do not feel that this charactercalls for the creation of a special genus; we pro-pose the inclusion of the sponge described asSyculmis synapta by Haeckel (1872), in the genusAmphoriscus.


238 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 33. — Diagram of the Amphoriscidae type of inarticulatechoanoskeleton (Amphoriscus or Leucilla). Abbreviations: a,atrium; as, atrial spicules; ss, subatrial spicules; st, subcorticaltetractines; cx, cortex.

cx

ss as

a

st

Genus Leucilla Haeckel, 1872

TYPE SPECIES. — Leucilla amphora Haeckel, 1872 bysubsequent designation (Dendy & Row 1913).

DIAGNOSIS. — Amphoriscidae with a sylleibid or leu-conoid organization. The choanoskeleton is formedprimarily by the apical actines of giant cortical tri-actines and the unpaired actines of subatrial triactinesor tetractines. It may contain dispersed spicules, but atypical articulate choanoskeleton is always absent.

DESCRIPTION

The genus Leucilla is quite close to Amphoriscus,and most species have a sylleibid aquiferous sys-tem. The simple species of Leucilla, such asL. amphora Haeckel, 1872, always have a thinsponge wall and an inarticulate choanoskeleton,which is reduced to the apical actines of corticaltetractines and to the unpaired actines of subatri-al triactines or tetractines (Fig. 34). In specimensof Leucilla that build a thicker wall, scattered tri-actines or tetractines can be found in thechoanosome, but they clearly derive from the cor-tical or the subatrial skeleton. There is no prima-ry choanoskeleton derived from radially arrangedspicules, and Leucilla species have never anystructures reminiscent of the articulate arrange-ment of the choanoskeleton.A group of sponges that have the organizationtypical of Leucandra, has been described underthe genus Leucilla. Tetractines are present in theircortical skeleton, but their apical actines do notrepresent the main support of the choanoskele-ton, which is typically articulate and clearly remi-niscent of a grantiid organization with many rowsof choanosomal triactines. Dendy & Row (1913)placed them in the genus Leucandra and we nowinclude these sponges in the genus Leucandrilla.

Genus Paraleucilla Dendy, 1892

TYPE SPECIES. — Leucandra cucumis Haeckel, 1872 bymonotypy.

DIAGNOSIS. — Amphoriscidae with a leuconoid orga-nization. The thick wall is divided into two regions.The outer region is supported by the skeleton whichremains essentially inarticulate, with the apical actinesof cortical tetractines pointed inwards, and a layer oftriactines and/or tetractines with the unpaired actine

pointed outwards. The inner region of the choanos-keleton is intercalated between the original subatrialskeleton and the atrial one, and it is supported by largetriactines and/or tetractines, that are scattered in disar-ray, and whose form is similar to the spicules found inthe outer layer of the choanoskeleton, or inside theatrial skeleton. Since the original subatrial layer stillremains in the outer part of the choanosome, facingthe cortical tetractines, there are no typical subatrialspicules adjacent to the atrial skeleton.

DESCRIPTION

Some leuconoid Amphoriscidae are massivesponges, with a folded, and thickened body. Inthese cases, the inarticulate organization isretained only in the outermost layer of the choa-nosome, which has a typical inarticulate skeletonconsisting of the apical actines of cortical tetrac-tines. Subatrial triactines or tetractines maintaintheir original position with their unpaired angledirected towards the atrium and their unpairedactine pointed towards the cortex (Fig. 35). The


239ZOOSYSTEMA • 2000 • 22 (2)

FIG. 34. — Diagram of a transverse section through the wall ofLeucilla. Abbreviations: a, atrium; as, atrial skeleton; ss, subatri-al spicule; st, subcortical tetractines; cx, cortex.

cx

ss as

a

st

latter spicules, however, are far from the surfaceof the atrium or larger exhalant canals, since athick layer supported by numerous scatteredirregular triactines and/or tetractines is intercalat-ed in between. This inner layer never has anytraces of a radial structure, and is clearly a newacquisition due to the intense growth of thesponge in this region. This structure had beenwell-described in Leucandra cucumis Haeckel,1872, but the outermost inarticulate layer waserroneously interpreted as containing onlyinhalant cavities. In specimens that we haveobserved, the outer layer contains the choano-some, although the lack of scattered spicules, thatare present in the inner part of the wall gives animpression of loose cavities.Dendy (1892b) proposed the genus Paraleucillafor Haeckel’s species Leucandra cucumis. He after-wards abandoned this idea and included thespecies in the genus Leucilla (Dendy, 1893), butsubsequently returned to use the genus (Dendy

& Row 1913) in order to underline the particularorganization of the subcortical region. Afterexamination of the material listed below, we havefound that other sponges classified in the genusLeucilla showed the organization typical ofParaleucilla in which we now classify them:Paraleucilla (Leucilla) saccharata Haeckel, 1872,(material studied MNHN-LBIM-C1968-681;BMNH 86.6.7.64 and 25.11.1.690a), Para-leucilla (Leucilla) crosslandi Row, 1909 (materialstudied BMNH 1954.2.24.25), Paraleucilla(Leucilla) proteus Dendy, 1913 (material studiedBMNH 20.12.9.60a), Paraleucilla (Leucilla) prin-ceps Row & Hôzawa, 1931 (material studiedBMNH 25.11.1.90a).

Family STAURORRHAPHIDAE Jenkin, 1908

TYPE GENUS. — Achramorpha Jenkins, 1908 by origi-nal designation.

DIAGNOSIS. — Leucosoleniida with a continuous cor-tex covering all the choanosome. Cortical tetractinesare absent. The organization of the aquiferous systemis syconoid, sylleibid or leuconoid. A tangential atrialskeleton is present only in the oscular region. In theatrial cavity, only the paired actines of subatrial chi-actines support the atrial surface, while the apicalactine is bent and points into the atrial cavity, makingits surface hispid.

DESCRIPTION

The Staurorrhaphidae have been proposed byJenkin (1908a) to include sponges with chi-actines (“cruciform” spicules), which aretetractines whose apical actine is bent so that itfollows the same line as the unpaired actine, butin the opposite direction (Fig. 36). These spiculesare found in the atrial wall: the paired actines areadjacent to the atrial surface, the apical actine isfree in the atrial cavity giving the atrial surface ahispid appearance, and the unpaired actine lies inthe wall of the radial tubes. Since the samesponges have no tangential spicules in the atrialskeleton, Jenkin (1908a) proposed that chiactinesoriginate from the atrial tetractines. However, aspointed out by Dendy & Row (1913), chiactineshave a typical subatrial origin and position. Inseveral other genera of Leucosoleniida, subatrialtetractines are found with the apical actine curved


240 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 35. — Diagram of a transverse section through the wall ofParaleucilla. Abbreviations: a, atrium; as, atrial skeleton; sl, sec-ondary subatrial layer inserted between the atrial skeleton andthe primary subatrial spicules, derived from the primary subatrialskeleton; pt, primary tetractines indicating the original positionof the subatrial skeleton; st, subcortical tetractines; cx, cortex.

a

cx

st

sl

pt

as

either in the direction of the unpaired actine,thus pointing towards the distal end of the radialtube, or in the opposite direction, pointing intothe atrium, such as the case of chiactines. Solelythe presence of chiactines would not justify thecreation of the family Staurorrhaphidae.However, in sponges included in this family theatrial skeleton is reduced to the region immedi-ately adjacent to the osculum, and in the spongebody there are no atrial tangential tri- ortetractines. As indicated by Dendy & Row(1913), this is an unusual modification of theskeletal organization among the Leucosoleniidaand justifies the separation of the two genera,Achramorpha and Megapogon, from otherLeucosoleniida. Since in most sponges the atrialcavity is echinated or hispid, probably as a pro-tection from invading organisms, in theStaurorrhaphidae the subatrial tetractines haveapparently taken over this function, forming along apical actine bent towards the atrial cavity.The family Staurorrhaphidae is thus characterizedsimultaneously by the absence of the atrial tan-gential skeleton and the presence of subatrialtetractines that are chiactines. All the knownspecies in the Staurorrhaphidae have rather a thinwall with either an inarticulate skeleton, or only afew scattered spicules in the choanoskeleton. Bothgenera of Staurorrhaphidae are known only fromAntarctica.

Genus Achramorpha Jenkin, 1908

TYPE SPECIES. — Achramorpha nivalis Jenkin, 1908 bysubsequent designation (Dendy & Row 1913).

DIAGNOSIS. — Staurorrhaphidae with a syconoidorganization.

Genus Megapogon Jenkin, 1908

TYPE SPECIES. — Leuconia crucifera Poléjaeff, 1883 bysubsequent designation (Dendy & Row 1913).

DIAGNOSIS. — Staurorrhaphidae with a sylleibid orleuconoid organization.

Family LELAPIIDAE Dendy & Row, 1913

TYPE GENUS. — Lelapia Gray, 1867 by original desig-nation.

DIAGNOSIS. — Leucosoleniida with a syconoid, sylleibidor leuconoid organization. The choanoskeleton con-tains typical subatrial spicules in the proximal region,associated with spicular tracts, consisting of modifiedtriactines arranged in parallel, which traverse either radi-ally or obliquely the choanosome. The cortex containstangential triactines and occasionally large longitudinaldiactines and/or small perpendicular diactines.

DESCRIPTION

The family Lelapiidae is characterized by spicularfibres or tracts, that are not found in otherLeucosoleniida. These tracts contain triactineswith reduced paired actines (nail-spicules) or dia-pasons (tuning-fork spicules), which replace thetypical choanoskeleton. This feature attracted a lotof attention in the past, since it is similar to thetype of skeleton found in “the Pharetronida”, andwas understood to be an indication of the relation-ship between the Lelapiidae and the fossilCalcarea. Two lines of evolution can be distin-guished in this family. In the Grantiopsis-Kebiraline the fibres are formed by the “nail-spicules”,while in the Paralelapia-Lelapia line they areformed by diapasons. In both lines, the simplestsponges are quite reminiscent of the Grantiidae,while in the most complex ones the cortical skele-


241ZOOSYSTEMA • 2000 • 22 (2)

FIG. 36. — Diagram of the subatrial skeleton of the Stauror-rhaphidae. Note the absence of the atrial skeleton, and the pres-ence of the apical actines of the subatrial tetractines that pointinwards into the atrial cavity. Abbreviations: a, atrium; ss, suba-trial spicules.

ss

a

ton participates progressively in the reinforcementof the choanoskeleton, replacing the typical mesh-work of choanosomal triactine spicules. The twotypes of modifications of the typical tangential tri-actines that participate in the spicule tracts are ap-parently related to the mechanical and spatialconstraints of these linear structures. Similar mod-ifications are found in the genus Guancha in theCalcinea, where the same constraints inside thepeduncle induce either the reduction of the pairedactines or their curvature into the diapason form(Borojevic et al. 1990). Consequently, we considerthat the Lelapiidae belong to the Leucosoleniida,where they represent a rather specialized and well-de-limited family, but that they do not have a close rela-tionship with other calcareous sponges related to thefossil groups, such as the Lithonida andMurrayonida, which have diapasons. The Lelapiidaeare characteristic of the Indo-Pacific region.

Genus Grantiopsis Dendy, 1892

TYPE SPECIES. — Grantiopsis cylindrica Dendy, 1892by monotypy.

DIAGNOSIS. — Lelapiidae with a syconoid or sylleibidorganization. The cortex is composed of tangential tri-actines, and occasionally has an external layer of smalldiactines perpendicular to the surface. The proximallayer of the choanoskeleton is composed of subatrialtriactines and/or tetractines, whose unpaired actinesare associated with modified triactines that have veryreduced paired actines. These modified triactines areeither isolated or form short bundles joined by anorganic material, and support the external part of thechoanosome.

DESCRIPTION

Grantiopsis has a particular skeleton that is char-acterized by triactines with reduced pairedactines in the wall of the tubes. As is typical fortriactines of the tubes, they are associated proxi-mally with the unpaired actines of subatrial tri-actines. In Grantiopsis species that have a thinwall, such as young G. fruticosa Dendy &Frederick, 1924, the triactines occasionally occursingly; but in sponges with a thicker wall, suchas G. cylindrica, they form distinct bundles(Fig. 37). Their reduced paired actines give thema form similar to diactines. They are parallel andtightly bound by an organic material that ismore resistant to dissolution with sodiumhypochlorite (which is used usually to dissociatecalcareous spicules) than other parts of the skele-ton. The nature of this material is unknown. Theorganization of their choanoskeleton is quitesimilar to a typical Grantia in which the articu-late skeleton of radial tubes is substituted byspicular tracts formed of nail-shaped triactines.These tracts are much more developed and con-spicuous in Kebira, clearly pointing to the originof the Lelapiidae from sponges like theGrantiidae, in which Grantiopsis had previouslybeen classified.

Genus Kebira Row, 1909

TYPE SPECIES. — Kebira uteoides Row, 1909 by mono-typy.

DIAGNOSIS. — Lelapiidae with a leuconoid organiza-tion. The choanoskeleton has large diactines andspicule tracts consisting of triactines with rudimentarypaired actines. The atrial and cortical skeletons arecomposed of triactines and diactines.


242 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 37. — Diagram of a transverse section through the wall ofGrantiopsis. Abbreviations: a, atrium; as, atrial skeleton;ss, subatrial spicules; ct, choanosomal spicule tracts composedof nail-like triactines; cx, cortex.

cx

as

a

ctss

DESCRIPTION

Kebira is clearly related to Grantiopsis with whichit shares the nail-form triactines, with reducedpaired actines, that are bundled in long tracts. Ithas a thick wall and a leuconoid organization. Thechoanoskeleton is supported by giant diactines andmultispicular tracts, which most often maintainthe radial organization. The atrial surface is sup-ported by tangential triactines (Fig. 38). AlthoughRow (1909) and Ilan & Vacelet (1993) do notspecifically mention subatrial spicules, we haveexamined the specimens studied by the latterauthors, and found that each choanosomal tract ofspicules is anchored at the atrial surface or at thesurface of larger exhalant canals, in a single suba-trial spicule, just as is observed in Grantiopsis(Fig. 37). Both Grantiopsis and Kebira lack the clas-sical articulate choanoskeleton composed of triac-tine or tetractine spicules. In the thick-walledKebira, large diactines, apparently derived from thecortical ones, participate in the formation of theskeleton of the choanosome, in addition to thespicular tracts.

Genus Paralelapia Hôzawa, 1923

TYPE SPECIES. — Lelapia nipponica Hara, 1894 bymonotypy.


243ZOOSYSTEMA • 2000 • 22 (2)

FIG. 38. — Diagram of the subatrial region of the Kebira skele-ton. Abbreviations: a, atrium; as, atrial skeleton; ss, subatrialspicules; ct, choanosomal spicule tracts composed of nail-liketriactines; cd, choanosomal diactines.

ct

cd

ssa

as

FIG. 39. — Diagram of a cross-section through the wall ofParalelapia nipponica Hara, 1894. Abbreviations: a, atrium;as, atrial skeleton; ct, choanosomal spicule tracts formed oftuning-fork-shaped triactines; cx, cortex; ec, exhalant canal;ic, inhalant canal. Scale bar: 60 µm (from Hôzawa 1923).

a

as

ct

ct

cx

ec

ic

as

DIAGNOSIS. — Lelapiidae with a sylleibid organizationof the aquiferous system. The thick cortex is composedof an external layer of triactines and an internal layerof giant longitudinal diactines. The choanoskeleton iscomposed of radially arranged loose tracts of diapa-sons, originating proximally from unpaired actines oftypical subatrial triactines. A well-developed atrialskeleton consists of tangential tri- and tetractines.

DESCRIPTION

The relationship of Paralelapia to Lelapia is quitesimilar to that of Grantiopsis and Kebira. InParalelapia, the sylleibid aquiferous system andthe organization of the choanoskeleton clearlyhave a radial organization, reminiscent of theGrantiidae. The loose spicular tracts are associat-ed proximally with the subatrial spicules, and thecortical skeleton is well-separated from thechoanoskeleton (Fig. 39).

Genus Lelapia Gray, 1867

TYPE SPECIES. — Lelapia australis Gray, 1867 bymonotypy.

DIAGNOSIS. — Lelapiidae with a leuconoid organiza-tion. The cortex is formed by external layers of tri-actines; it may also have an internal layer of largelongitudinal diactines, as well as radial thin diactines

or microdiactines. The choanoskeleton has radially orsubradially arranged spicule tracts consisting of diapa-sons, and large scattered diactines that are similar tothe cortical ones. The atrial skeleton is composed oftangential triactine spicules.

DESCRIPTION

Lelapia antiqua Dendy & Frederick, 1924 andL. australis represent a series of modificationsfrom Paralelapia (Dendy & Frederick 1924). Asin Kebira, the choanoskeleton is reduced tospicule tracts, that are progressively invaded bythe cortical diactines, which lead to the forma-tion of a thick and rigid sponge wall (Fig. 40).

Family INCERTAE SEDIS

Genus Sycyssa Haeckel, 1872

TYPE SPECIES. — Sycyssa huxleyi Haeckel, 1872 bymonotypy.DIAGNOSIS. — Leucosoleniida (?) with a syconoidorganization. The skeleton consists of diactines only.

DESCRIPTION

This species has been described using two speci-mens collected by Haeckel in the Adriatic; it wasnever found since. The absence of all the radiatespicules is quite remarkable. This condition is


244 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 40. — Skeletal arrangement in a longitudinal section of Lelapia australis Gray, 1867. Abbreviations: a, atrium; dc, dermal cortex;dt, dermal tuft of triactines and slender diactines; fi, spicular fibres formed of tuning-fork-shaped triactines; gc, atrial cortex. Scalebar: 350 µm (from Dendy 1894).

dt

dc

fi fi

fi

fi

a

a a

a

gc

KEY OF GENERA OF LEUCOSOLENIIDA

1. Homocoel organization: all the internal cavities are lined by the choanoderm ...... 2

— Heterocoel organization; with separate choanocyte chambers and an exhalant aquife-rous system devoid of choanocytes, with or without a separate inhalant system ...... 4

2. Skeleton contains only diactines .................................................................... Ascyssa

— Skeleton composed of diactines, triactines and/or tetractines ................................ 3

3. Giant longitudinal diactines forming a continuous layer on the external surface.... Ascute

— No giant longitudinal diactines ............................................................ Leucosolenia

4. Skeleton composed exclusively of diactines .................................................. Sycyssa

— Skeleton composed of diactines, triactines and/or tetractines ................................ 5

5. Sponge body composed of a central atrial tube bearing, at least in its median region,radial tubes lined with choanoderm and ending in conspicuous distal cones that arefrequently crowned by tufts of radial diactines ...................................................... 6

— Sponge body covered by a cortex supported by tangential spicules ...................... 10

6. Elongate radial tubes regularly arranged around the central tube are completely sepa-rate from one another; no inhalant aquiferous system, the incurrent water entersdirectly through pores into the radial choanocyte chambers .......................... Sycetta

— Radial tubes are coalescent at least in their proximal region .................................. 7

7. Radial tubes are parallel and regularly arranged and are occasionally ramified in theirdistal part. Radial tubes are coalescent for most of their length, with inhalant canalsin between that open to the outer surface between the distal cones through ostia .. 8

— A thin sponge wall consisting of the atrial skeleton supporting irregularly groupedshort radial tubes that are coalescent or fused proximally, with free distal cones; eachgroup of tubes communicates through a common opening with the atrial cavity;large irregular inhalant cavities are left between the groups of radial tubes, and fre-quently reach the outer surface of the central atrial tube ............................ Sycantha

8. The atrial cavity contains an internal tissue network supported by a skeleton ofparallel bundles of diactines ...................................................................... Sycandra

— No network inside the atrial cavity ........................................................................ 9

9. Pseudosagittal spicules are present in the distal cones of the radial tubes .... Syconessa

— Absence of pseudosagittal spicules in the distal cones ...................................... Sycon


245ZOOSYSTEMA • 2000 • 22 (2)

10. The whole sponge is covered by a continuous cortex supported by tangential spi-cules .................................................................................................................. 11

— Only the grouped distal cones are covered by a cortical network supported by tan-gential triactines, leaving large openings to the inhalant cavities, and giving to theexternal surface a honeycombed aspect ................................................ Dermatreton

11. Presence of a distinct layer of subcortical pseudosagittal spicules ........................ 12

— Absence of a distinct layer of subcortical pseudosagittal spicules ........................ 17

12. Inarticulate choanoskeleton composed of the unpaired actines of subatrial spiculesand the centripetal paired actines of subcortical pseudosagittal spicules .............. 13

— Articulate choanoskeleton, containing few to several rows of tube spicules, whichare more or less scattered between the subatrial and subcortical spicules ............ 14

13. Pseudosagittal spicules are triactines only .................................................. Sycettusa

— Pseudosagittal spicules are tetractines and triactines ................................ Grantilla

14. Sylleibid or leuconoid organization .................................................... Vosmaeropsis

— Syconoid organization ...................................................................................... 15

15. Without large longitudinal diactines in the cortical or atrial skeleton ...... Grantessa

— With large longitudinal diactines in the cortical or atrial skeleton ...................... 16

16. Longitudinal diactines in the cortical skeleton only ................................ Heteropia

— Longitudinal diactines in the cortical and atrial skeleton .................... Paraheteropia

17. Articulate choanoskeleton, with at least some trace of the tube organization; themajority of spicules have the unpaired actine pointing towards the outer surface ofthe sponge ........................................................................................................ 18

— Inarticulate choanoskeleton, or choanoskeleton composed of an external inarticula-te layer supported by apical actines of cortical tetractines, and an internal layer ofscattered triactines and/or tetractines, without any apparent order .................... 34

18. Without a tangential atrial skeleton substituted by subatrial chiactines .............................................................................................................. (Staurorrhaphidae) 19

— With tangential atrial skeleton composed of triactines and/or tetractines .......... 20

19. Syconoid organization ........................................................................ Achramorpha

— Sylleibid or leuconoid organization ...................................................... Megapogon


246 ZOOSYSTEMA • 2000 • 22 (2)

20. Spicular tracts in the choanoskeleton .............................................. (Lelapiidae) 21

— No spicular tracts .......................................................................... (Grantiidae) 24

21. Spicular tracts made of “nail-shaped” triactines, with highly reduced paired actines .. 22

— Spicular tracts made of diapasons ...................................................................... 23

22. Tubular sponge, syconoid or sylleibid organization, choanoskeleton with short,radially arranged tracts between the distal parts of the radial tubes ........ Grantiopsis

— Massive sponge with a thick wall containing large diactines, choanoskeleton withlong tracts in an approximately radial arrangement ...................................... Kebira

23. Cortex composed of giant longitudinal diactines, which do not invade the choanos-keleton .................................................................................................. Paralelapia

— Giant diactines in the choanoskeleton ........................................................ Lelapia

24. Absence of longitudinal diactine in the atrial and/or cortical skeleton ................ 25

— Presence of longitudinal diactines in the atrial and/or cortical skeleton .............. 29

25. Syconoid organization ...................................................................................... 26

— Sylleibid or leuconoid organization .................................................................... 27

26. Diameter of the osculum smaller than that of the atrium .......................... Grantia

— Diameter of the osculum larger than that of the atrium: pedunculate calyciformsponge with a thin folded wall ............................................................ Teichonopsis

27. Septa with a specific skeleton of minute triactines within the atrial cavity ... Leucettaga

— No septa within the atrial cavity ........................................................................ 28

28. Cortex composed of triactines and possibly diactines, which protrude from the cor-tex making it hispid ................................................................................ Leucandra

— Cortex with triactines and tetractines .................................................. Leucandrilla

29. Leuconoid organization .......................................................................... Aphroceras

— Syconoid organization ...................................................................................... 30

30. Sponge with individual syconoid tubes or an arborescent cormus composed of sepa-rate tubes .......................................................................................................... 31

— Massive cormus composed of coalescent syconoid units, covered by a common cortex............................................................................................................ Synute


247ZOOSYSTEMA • 2000 • 22 (2)

31. Longitudinal diactines present only in the cortex .............................................. 32

— Longitudinal diactines present in the atrial skeleton .......................................... 33

32. Tufts of thin radial diactines decorate the distal parts of the radial tubes, and crossthe cortex between the longitudinal diactines ................................................ Sycute

— No tufts of radial diactines .............................................................................. Ute

33. Longitudinal diactines present only in the atrial skeleton.......................... Sycodorus

— Longitudinal diactines present both in the atrial and the cortical skeleton ... Amphiute

— No giant cortical tetractines .......................................................... (Jenkinidae) 37

35. Syconoid organization ........................................................................ Amphoriscus


36. Inarticulate choanoskeleton; scattered spicules occasionally between subcortical andsubatrial layers ............................................................................................ Leucilla

— Choanoskeleton divided in two parts: the external part has an inarticulate organiza-tion, while the internal one is intercalated between the subatrial spicules and the atrialskeleton, and is supported by scattered triactines and/or tetractines .......... Paraleucilla

37. Syconoid organization ...................................................................................... 38


38. Thin cortical skeleton, composed of one to several layers of triactines .... Breitfussia

— Reinforced cortical skeleton .............................................................................. 39

39. Skeleton reinforced with giant tangential triactines, which are also scattered in thechoanoskeleton .................................................................................... Anamixilla

— Reinforced skeleton with longitudinal diactines ........................................ Uteopsis

40. Without scattered spicules in the choanoskeleton .............................................. 41

— With scattered spicules in the choanoskeleton, smaller than those of the cortex, andwithout defined position ........................................................................ Polejaevia

41. Sponge growing as small individual tubes .................................................. Jenkina

— Sponge forming a large cormus composed of copiously anastomosed and ramifiedtubes .................................................................................................. Leucascandra


248 ZOOSYSTEMA • 2000 • 22 (2)

similar to that seen in the family Trichogypsiidae inthe order Baeriida. However, no other characteris-tics of the Baeriida are present, and the organiza-tion of S. huxleyi is otherwise quite similar to othergrantiid sponges with a syconoid choanosome, inparticular to Sycodorus hystrix Haeckel, 1872, aspointed out by the author. At present we are unableto decide what the relationship of this sponge is toother Calcaronea. However, we believe that it iscloser to the Leucosoleniida than to the Baeriida.

Order BAERIIDA n. ord.

DIAGNOSIS. — Leuconoid Calcaronea with the skele-ton either composed exclusively of microdiactines, orin which microdiactines constitute exclusively or pre-dominantly a specific sector of the skeleton, such aschoanoskeleton or atrial skeleton. Large or giant

spicules are frequently present in the cortical skeleton,from which they can partially or fully invade thechoanoderm. In sponges with a reinforced cortex, theinhalant pores can be restricted to a sieve-like ostia-bearing region. Dagger-shaped small tetractines (pugi-oles) are frequently the sole skeleton of the exhalantaquiferous system. Although the skeleton may be high-ly reinforced by the presence of dense layers of micro-diactines in a specific region, an aspicular calcareousskeleton is not present.

DESCRIPTION

Dendy & Row (1913) already wrote that “aber-rant genera [such] as Leucopsila, Baeria,Kuarraphis, Leucyssa and Trichogypsia can only beincluded in the Grantiidae provisionally. The dif-ficulty of arranging the genera probably arisesfrom the fact that great gaps exist in the familyowing to extinction of intermediate forms”. This


249ZOOSYSTEMA • 2000 • 22 (2)

FIG. 41. — Transverse sections through the wall of Baeria john-stoni (Carter, 1871) (light micrograph). Abbreviations: ex, largeexhalant canals; ch, choanosome; cx, cortex. Scale bar:230 µm.

FIG. 42. — Transverse sections through the wall of Baeria nivea(Johnston, 1842) (light micrograph). Abbreviations: ex, largeexhalant canals; ch, choanosome; cx, cortex. Scale bar:230 µm.

cxcx

ex

ch

ex

ch

statement clearly outlines the major argumentsfor separating the above-named sponges from theLeucosoleniida. We now propose to create a neworder, the Baeriida in the Calcaronea, for a groupof sponges with quite a distinct type of organiza-tion, in which no traces of radial symmetry canbe observed, and which apparently has not fol-lowed the sycettid pathway of evolution.The aquiferous system of the Baeriida is alwaysleuconoid, with choanocyte chambers distributed

irregularly throughout the sponge wall and oftenarranged in groups around large exhalant canals(Figs 41; 42). No true atrial skeleton, reminiscentof the central tube, is found in the exhalant aquif-erous system, nor does a clear subatrial skeletonindicate the original position of the radial tubes.We found no sponges with an asconoid orsyconoid type of aquiferous system, which couldbe included in this order. The postlarval develop-ment of the Baeriida is not known, and the pres-ence of an early olynthus stage in theirdevelopment has not been established. It is note-worthy that a comparison of the morphogenesisof diamorphs, obtained after dissociation andreaggregation of Sycon vigilans and Grantia com-pressa (representing the Leucosoleniida) andBaeria (Leuconia) nivea (representing theBaeriida) (Sarà et al. 1974; Peixinho 1980) havedisclosed clear differences. During morphogene-sis both Sycon and Grantia pass through the olyn-thus stage, acquiring a sycettid grade oforganization by the subsequent formation ofradial tubes, similar to the postlarval develop-ment of the Leucosoleniida. Conversely, Baeriadoes not pass through olynthus and sycettidstages during morphogenesis, but develops a leu-conoid type of aquiferous system by the forma-tion of a rhagon, similar to that described for theDemospongiae by Lévi (1956) (Fig. 43).Consequently, we assume that in the Baeriida, asin the Demospongiae, the development of theaquiferous system involves the formation ofspherical choanocyte chambers, simultaneouswith the formation of the inhalant and exhalantaquiferous systems.Apart from these aspects of the aquiferous systemand the associated skeleton, the Baeriida are char-acterized by having two distinct categories ofspicules, which correspond to the megascleresand microscleres in Demospongiae. Smallspicules (most frequently microdiactines) arefound throughout the sponge and giant spiculesare limited to the cortical region (e.g. Baeria john-stoni and Lamontia zona), or invade thechoanosome from the cortex and form a scatteredskeleton throughout the body (e.g. Baeria nivea). One or several types of very small spicules may bepresent. A very particular type of tetractine termed“unicorvo-cruciform” by Bowerbank (1864),


250 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 43. — Diagram of the formation of choanocyte chambers ofdiamorphies during the process of reconstitution from dissociat-ed cells; drawing from original micrographs by Peixinho (1980);A, pre-sycon stage of Grantia compressa; B, pre-leucon stage inBaeria nivea. Scale bar: A, 26 µm; B, 45 µm.

A

B

“kreuzförmigen Vierstrahlern” by Haeckel (1872),“dagger-shaped tetracts” by Grant (1826), Dendy(1892b) and Kirk (1895), or harpoon-liketetractines by more recent authors is found in theskeleton of exhalant surfaces. We propose to namethese spicules “pugioles” (pugiolus, in Latin smalldagger). Typical pugioles are found in Baeria andLamontia. They constitute the exhalant canalskeleton, in which the paired actines are adjacentto the canal surface, the unpaired actine is per-pendicular to it lying inside the adjacent tissue, andthe apical actine is free in the canal lumen (Figs 44;45). Consequently, they have the main axis (theone passing through the unpaired angle of the basaltriactine system) perpendicular to the exhalantcanal surface, as opposed to atrial spicules in theLeucosoleniida where this axis is parallel to theatrial surface oriented longitudinally with theunpaired angle in most of the cases turned towardsthe osculum. The position and function of pugi-oles are not unlike those of the equally “cruciform”large spicules named chiactines (Jenkin 1908a)that are character i s t ic of the fami lyStaurorrhaphidae. We consider that this is a con-sequence of the same need for reinforcement andprotection of the surface of the exhalant system. Inboth cases, the atrial surface is devoid of skeleton,and consequently is bald and exposed to invasion.Chiactines are modified subatrial spicules and thusparticipate both in forming the proximal part ofthe choanoskeleton and in the protection of theatrial cavity through long apical actines bent cen-tripetally across the atrial surface of the sponge.However, pugioles are present only in theBaeriidae, where they point their apical actinetowards the inside of the atrium. Apparently, theyare derived from spicules that were tangential tothe surface of exhalant canals, but which have sub-sequently acquired the particular position and ori-entation observed in Baeria johnstoni and B. nivea.The second type of small spicules are microdi-actines, often termed “Stäbchen-Mörtel” (Haeckel1872) or “mortar spicules” (Dendy 1892b)(Fig. 45). These spicules apparently derive fromsmall triactines, in which one of the paired actinesis rudimentary, giving the spicule a lanceolate-likeshape. Similar spicules can be found amongsponges in the Leucosoleniida, but in the Baeriidathey may be the sole spicule type present, or can

constitute a specific part of the skeleton, eitheralone or as its major component, such aschoanoskeleton (e.g. Baeria, Lamontia, Eilhardia,Lepidoleucon) or atrial skeleton (e.g. Leucopsila). There may also be small triactines with two shortor rudimentary paired actines, bent together toform a club-shaped spicule, often with a more orless pronounced hole adjacent to the centre of thespicule; these are the “needle-eye” spicules seen inthe genus Kuarraphis and in Baeria ochotensis.Baeria gladiator Dendy, 1892 also has typical tri-chodragmas. To our knowledge, this is the onlycalcareous sponge with this type of microdiactine. The external skeleton of the Baeriida may have athick and continuous layer of reinforced corticalspicules that obstructs the free flow of inhalantwater and causes the inhalant pores to be restrict-ed to a specific cribriform region (e.g. Lamontia,Lepidoleucon).In the subclass Calcaronea, continuous evolu-tionary lineages with all the intermediate formsare observed in Leucosoleniida, suggesting arecent evolutionary radiation. In contrast, theBaeriida and Lithonida are represented by well-characterized and very distinct genera, with nointermediate forms, suggestive of long-term evo-lution in which a small number of only the mostspecialized forms are conserved. Similarly, in thesubclass Calcinea, all the transitional forms arefound in Clathrinida but not in Murrayonida(Borojevic et al. 1990).

Family BAERIIDAE n. fam.

TYPE GENUS. — Baeria Miklucho-Maclay, 1870 byoriginal designation.

DIAGNOSIS. — Baeriida with a choanoskeleton consist-ing of giant triactines, and/or of tetractines in no partic-ular order, and/or of very numerous microdiactines. Notraces of radial organization can be seen in thechoanoskeleton. The cortical skeleton consists of tri-actines, giant diactines, and/or numerous microdiactines,and occasionally the basal actines of cortical gianttetractines. The choanoskeleton consists of scatteredspicules similar to those observed in the cortex, to whichnumerous microdiactines can be added, or which can beentirely replaced by microdiactines. The exhalant aquif-erous system is formed by ramified canals that have notangential skeleton, being loosely or densely covered byharpoon-shaped pugioles and/or microdiactines.


251ZOOSYSTEMA • 2000 • 22 (2)

Genus Baeria Miklucho-Maclay, 1870

TYPE SPECIES. — Baeria ochotensis Miklucho-Maclay,1870 by monotypy.

DIAGNOSIS. — Baeriidae in which the choanoskeletonconsists of giant triactines and/or tetractines, lyingwithout apparent order, and of very numerous micro-diactines. A cavity equivalent to the atrium, localizedonly under the oscula, has a skeleton supported by tan-gential triactines. All the other exhalant canals have askeleton composed of harpoon-shaped pugioles.

DESCRIPTION

In the genus Baeria we include the hithertodescribed species B. nivea (Grant, 1826), B. john-stoni (Carter, 1871), B. ochotensis Miklucho-Maclay, 1870, B. gladiator (Dendy, 1892) andB. prava (Breitfuss, 1898), which bear a number

of similarities. They all have a choanoskeletoncontaining numerous microdiactines (“mortarspicules”), and very large triactines or tetractinesthat lie scattered without any apparent order. Thecortical skeleton consists of triactines and micro-diactines, which are different in size and formfrom those in the choanosome. The atrium isalways limited to the space immediately belowthe osculum, and the oscular area contains adense layer of sagittal triactines. However, theexhalant canals of all sizes are devoid of a tangen-tial skeleton and contain only pugioles (Fig. 44).In Baeria johnstoni an additional type of verysmall tetractine is found in the choanosome.Although Haeckel (1872) has represented themicrodiactines of all these species as smooth andequally acerate on both ends, we have observed


252 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 44. — Pugioles (p) in an exhalant canal (ex) of Baeria nivea(Johnston, 1842). Scale bar: 90 µm.

ex

p

FIG. 45. — Diagram of pugioles and microdiactines; A, Baeria glad-iator; B, Lamontia zona; C, Baeria nivea; D, Baeria ochotensis.

A B

C D

that in B. nivea and B. johnstoni the microdi-actines are clearly hastate in form, with one longand one short actine, the shorter one having anangular twist close to the centre of the spicule.Dendy & Row (1913) reported that after re-examination of Haeckel’s preparations ofB. ochotensis, many of microdiactines were foundto have a “needle-eye” form. These correspond totiny triactines, with two very much reducedpaired actines bent to lie approximately parallel,which can be slightly swollen and fused at theend (Fig. 45); such spicules can be observed inthe Lithonida. Some of these spicules have theirdistal ends free, thus corresponding exactly todiapasons. Similar spicules are found in Kuar-raphis Dendy & Row, 1913, which we place alsoin the Baeriida. In the original description ofKuarraphis (Leucyssa) cretacea, Haeckel (1872)indicated that the “needle-eye” spicules werefound in Baeria ochotensis although they were notfigured in its description.B. johnstoni and B. nivea have previously beenincluded in the genera Spongia, Grantia, Leucandraand Leuconia. The first three genera are now usedin a different context, and the last one has beenshown to be invalid (Dendy 1893; Dendy & Row1913). The genus Baeria has been proposed previ-ously by Dendy & Row (1913) for B. ochotensis,and the present definition of the genus is quite closeto the one proposed by those authors.

Genus Lamontia Kirk, 1895

TYPE SPECIES. — Lamontia zona Kirk, 1895 by mono-typy.

DIAGNOSIS. — Baeriidae in which the choanoskeletonconsists of microdiactines. Cortical and atrial skeletonshave triactines and tetractines, and the cortex ispierced by large diactines. A specialized ostia-bearingzone, located below the osculum, leads the incurrentwater flow to inhalant cavities.

DESCRIPTION

Lamontia is a particular sponge, which bears anumber of similarities with Baeria. The choanos-keleton is composed of microdiactines, and thereare pugioles in the skeleton of the exhalant aquif-erous system. It is distinguished by a special ostia-bearing zone and large diactines, which make thesponge hispid.

Genus Leucopsila Dendy & Row, 1913

TYPE SPECIES. — Leuconia stilifera Schmidt, 1870 bymonotypy.

DIAGNOSIS. — Baeriidae in which the cortex is formedby tangential triactines and microdiactines. Thechoanoskeleton is composed almost exclusively ofirregularly scattered giant tetractines, and numerousmicrodiactines. Both the cortical and atrial surfaces arecovered by a dense layer of microdiactines. While inthe cortex microdiactines overlay the continuous layerof tangential triactines, they are the sole skeleton of theexhalant aquiferous system.

DESCRIPTION

Like Baeria, Leucopsila has a massive body, withan irregular leuconoid aquiferous system orga-nized around exhalant canals that are distributedin the choanosome in the form of an anastomos-ing network. The organization of the skeleton issimilar in Leucopsila stilifera (Schmidt, 1870) andBaeria johnstoni (Carter, 1871); the major dis-tinction between the two being the replacementof pugioles in the skeleton of the exhalant systemof Leucopsila by microdiactines.Leucopsila is a large sponge that has been reportedonly from arctic and subarctic waters, both fromthe Atlantic and Pacific regions (Schmidt 1870;Haeckel 1872; Hôzawa 1919).

Genus Eilhardia Poléjaeff, 1883

TYPE SPECIES. — Eilhardia schulzei Poléjaeff,1883 by monotypy.

DIAGNOSIS. — Calyciform Baeriidae with inhalantostia on the inner surfaces, and oscula on the outersurfaces. The ostia-bearing surface is supported by athin layer of tangential triactines and scattered micro-diactines. The skeleton of the exhalant system and ofthe choanoskeleton is composed of large triactines andmicrodiactines. The cortical skeleton consists of giantlongitudinal and small diactines as well as tangentialtriactines.

DESCRIPTION

Eilhardia schulzei Poléjaeff, 1883 is quite anunusual calcareous sponge, with a calyciformbody in which the inhalant surface is theinner one, and exhalant surface the outer one.Its internal organization is similar to otherBaeriidae. Poléjaeff (1883) has given a very


253ZOOSYSTEMA • 2000 • 22 (2)

detailed description and beautiful illustration ofthis sponge.

Family TRICHOGYPSIIDAE n. fam.

TYPE GENUS. — Trichogypsia Carter, 1871 by originaldesignation.

DIAGNOSIS. — Baeriida with a skeleton entirelyformed by diactine spicules.

DESCRIPTION

We propose to put sponges that have affinitieswith the Baeriidae, but that have only diactinespicules in the family Trichogypsiidae. As dis-cussed above, one of the characteristics of theBaeriida is the presence of small “mortar-shaped”diactines that make up either all of, or a large partof a specific portion of the skeleton. TheTrichogypsiidae have large diactines, which areprobably not homologous with “mortar-shaped”diactines. It is difficult to establish whether theabsence of triactine spicules is a primitive condi-tion or is a consequence of a secondary reduction

of the skeleton. In the Leucosoleniida, diactinesare the first spicules to be secreted, but it is notknown if this is also true for the Baeriida. All theTrichogypsiidae are very poorly known, havingbeen described in early studies from a small num-ber of specimens; there are no recent studies thatprovide a detailed description of their cytology orbiology. Up to now, the Trichogypsiidae havebeen only described from boreal or arctic regions.

Genus Trichogypsia Carter, 1871

TYPE SPECIES. — Trichogypsia villosa Carter, 1871 bysubsequent designation (Dendy & Row 1913).

DIAGNOSIS. — Trichogypsiidae with a skeleton com-posed of spined diactines.

DESCRIPTION

Haeckel (1872) described two subspecies thatwere subsequently raised by Dendy & Row(1913) to the species level: Trichogypsia (Leucyssa)incrustans Haeckel, 1872 and T. villosa Carter,1871. He pointed out the similarity of both theexternal form and the organization of the aquifer-ous system to Baeria nivea, but we have no infor-mation on the internal organization of thechoanoderm or the choanoskeleton. We haveexamined the slide prepared from the originalspecimen, described by Carter (1871) (BMNH1870.10.1.9). It contains a fragment of anencrusting sponge with a thick cortex elevated inconules. The cortex is supported by very denselypacked parallel diactines, arranged in bunches inthe conules. A large, deep cavity present at thesurface is surrounded by a particular skeletoncomposed of bent diactines. A canal parallel tothe surface is connected to this cavity. Both thecanal surface and the internal skeleton consist ofstraight or slightly curved spiny diactines thatform a rather loose network (Fig. 46).

Genus Kuarrhaphis Dendy & Row, 1913

TYPE SPECIES. — Leucyssa cretacea Haeckel, 1872 bymonotypy.

DIAGNOSIS. — Trichogypsiidae with a skeleton com-posed exclusively of small perforated needle-eyediactines.


254 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 46. — Trichogypsia villosa (Carter, 1871); A, a diagram of atransverse section of the body wall, showing a pit-like cavity anda canal parallel to the external surface that bears conules con-taining very densely arranged parallel diactines; B, diactines ofthe cortical conules; C, diactines of the internal reticular skele-ton; D, diactines that line the pit-like cavity.

A B

C

D

DESCRIPTION

Dendy & Row (1913) proposed the genusKuarraphis for a sponge described by Haeckel(1872) that is characterized by the presence ofonly needle-eye spicules. The sponge is encrust-ing, has a leuconoid organization without anatrium, and has irregular exhalant canals.Haeckel (1872) pointed out the similarity of thechoanosome and the skeleton of this sponge withthose described by him in the “subgenusLeucomalthe”, which comprises the species thatwe place now in the Baeriidae. As indicated byDendy & Row (1913), the “needle-eye” spiculesare similar to diapasons, and are also found inBaeria ochotensis.

Genus Leucyssa Haeckel, 1872

TYPE SPECIES. — Leucyssa spongilla Haeckel, 1872 bymonotypy.

DIAGNOSIS. — Trichogypsiidae (?) with a skeletoncomposed only of smooth diactines.

DESCRIPTION

The external shape of this sponge, which has apedunculate cormus with a large clathrate bodyof anastomosed tubes, is quite different fromboth the Baeriida and Leucosoleniida, and israther common among the Clathrinida.However, Eilhardia also has quite an unusualshape, and it is possible that our knowledge ofonly a very small number of sponges in theBaeriida gives the impression of a great diver-gence of the observed forms. We place thissponge among the Baeriida, following Haeckel’s(1872) description of the rather irregular alveolarchoanosome supported by a dense skeleton ofsmall diactines, scattered without order.

Family LEPIDOLEUCONIDAE Vacelet, 1967

TYPE GENUS. — Lepidoleucon Vacelet, 1967 by mono-typy.

DIAGNOSIS. — Baeriida with a leuconoid organization andwith an irregular outer layer of scales derived from tri-actines. The choanoskeleton is exclusively composed ofscattered microdiactines. The ostia are localized in a spe-cial area where the triactines are not transformed into scales.

DESCRIPTION

The Lepidoleuconidae is characterized by the for-mation of triangular scales in the cortex. Haeckel(1872) reported similar spicules in Leucetta trigo-na described from a single dried specimen fromSouth Africa. As the description of this species isquite incomplete it is possible that it couldbelong to the family Lepidoleuconidae. The orga-nization of the skeleton in the Lepidoleuconidaeis similar to that of other Baeriida, and in particu-lar to Baeria johnstoni and Eilhardia schulzei in


255ZOOSYSTEMA • 2000 • 22 (2)

FIG. 47. — Diagram of the organization of Lepidoleucon inflatumVacelet, 1967; A, osculum; B, inhalant area. Scale bar: 75 µm(from Vacelet 1967a).

A

B

KEY OF GENERA OF BAERIIDA

1. Skeleton composed of only diactines ..........................................(Trichogypsiidae) 2

— Skeleton composed of diactines, triactines and/or tetractines.................................. 4

2. Microdiactines are the “needle-eye” type................................................ Kuarrhaphis

— Lanceolate diactines with spines on one or both ends, or smooth diactines ............ 3

3. Lanceolate spiny diactines .................................................................... Trichogypsia

— Smooth diactines ........................................................................................ Leucyssa

4. Sponge with a specialized inhalant zone bearing pores .......................................... 5

— Sponge without a specialized inhalant zone ........................................ (Baeriidae) 6

5. Cortex composed of scales derived from triactines .............................. Lepidoleucon

— Cortex composed of triactines and tetractines and sometimes diactines; pugioles arepresent in the atrial skeleton .................................................................... Lamontia

6. Calyciform sponges with inhalant ostia localized on the inner surface and oscula onthe outer surface ........................................................................................ Eilhardia

— not calyciform sponges .......................................................................................... 7

7. Atrial skeleton with pugioles .......................................................................... Baeria

— Atrial skeleton with microdiactines .......................................................... Leucopsila


256 ZOOSYSTEMA • 2000 • 22 (2)

which the choanoskeleton consists solely ofmicrodiactines, and to Lamontia, which also hasthe ostia-bearing inhalant area.

Genus Lepidoleucon Vacelet, 1967

TYPE SPECIES. — Lepidoleucon inflatum Vacelet, 1967by monotypy.

DIAGNOSIS. — Same definition as the family (Fig. 47).The single osculum has a circlet of modifiedtetractines.

Order LITHONIDA Vacelet, 1981

DIAGNOSIS. — Calcaronea with reinforced skeletonconsisting either of linked or cemented basal actines of

tetractines, or of a rigid basal mass of calcite. Diapasonspicules are generally present and the canal system isleuconoid.

DESCRIPTION

The families Lelapiellidae (Calcinea), Lepidoleu-conidae (Baeriida) and Lelapiidae (Leucoso-leniida) are now excluded from this order.

Family MINCHINELLIDAE Dendy & Row, 1913

TYPE GENUS. — Minchinella Kirkpatrick, 1908 byoriginal designation.

DIAGNOSIS. — Lithonida with a choanoskeleton con-sisting of a primary network of tetractines cemented orlinked together in a variety of ways.

DESCRIPTION

Vacelet (1981) provided an identification key ofthe known Recent species. The family includesseveral fossil genera, whose definitions have to berevised.

Genus Minchinella Kirkpatrick, 1908

TYPE SPECIES. — Minchinella lamellosa Kirkpatrick,1908 by monotypy.

DIAGNOSIS. — Minchinellidae in which thechoanoskeleton consists of tetractines cementedtogether into a rigid network by their basal actines andsubsequently embedded in an enveloping cement(Fig. 48). The cortical skeleton is composed of tri-actines, diapasons, and diactines.

Genus Plectroninia Hinde, 1900

TYPE SPECIES. — Plectroninia halli Hinde, 1900 bymonotypy.

DIAGNOSIS. — Minchinellidae with a choanoskeletoncomposed of tetractines, the basal actines of which arefused with the basal actines of adjacent spicules, while

the apical actines remain free and point outward. Thischoanoskeleton is made up of two layers, an outerlayer of large tetractines and a basal layer of smalltetractines. The cortical skeleton consists of freespicules arranged tangentially. A perioscular circlet oftetractines, and rarely of triactines, is usually present.

DESCRIPTION

The genus Plectroninia, which was erected for fos-sil sponges, is difficult to distinguish from severalother fossil genera, such as Bactronella Hinde,1884; Porosphaera Steinmann, 1878; TretocaliaHinde, 1900; Porosphaerella Welter, 1910;Sagittularia Welter, 1910. The type species ofPlectroninia, P. halli Hinde, 1900 from the mid-Miocene, was turbinate in shape and was probablyfree-living on a muddy sand bottom (Pickett, pers.comm.). It was therefore quite different from theRecent species first allocated to this genus byKirkpatrick (1900) (Figs 49-51). Kirkpatrick con-cluded that his specimen of P. hindei was a juvenilethat had not reached its ultimate shape. The rela-tionships of the Recent species, which are encrust-


257ZOOSYSTEMA • 2000 • 22 (2)

FIG. 48. — Surface view of the skeleton in Minchinella lamellosaKirkpatrick, 1908, showing the progressive embedding of thetetractine framework (t) in a calcareous cement (c) (SEM). Scalebar: 31 µm (from Vacelet 1991).

ct

FIG. 49. — Transverse section of Plectroninia hindei Kirkpatrick,1900 (light micrograph). Abbreviations: p, papilla; o, osculum;c, choanosome; cs, cortical skeleton; lt, large fused tetractinesof the choanoskeleton; st, small fused tetractines of thechoanoskeleton. Scale bar: 330 µm (from Pouliquen & Vacelet1970).

o

p

cs

lt

st

c

ing, with the massive Petrostroma Döderlein, 1892are also unclear at present. This genus contains 12 known Recent speciesthat have an unusually large depth distribution,being known from shallow water caves down to1600 m in depth.

Genus Monoplectroninia Pouliquen & Vacelet, 1970

TYPE SPECIES. — Monoplectroninia hispida Pouliquen& Vacelet, 1970 by monotypy.

DIAGNOSIS. — Minchinellidae in which thechoanoskeleton is composed of a basal layer made ofone category of small tetractines cemented together bytheir basal actines, while their apical actine remainsfree and points outward. The cortical skeleton consistsof free spicules.

Genus Petrostroma Döderlein, 1892

TYPE SPECIES. — Petrostroma schulzei Döderlein, 1892by monotypy.

DIAGNOSIS. — Minchinellidae with tetractines fused bytheir basal actines forming radial lines that are linked bysmaller tetractines, which are fused by their basal actines.The cortical skeleton consists of free spicules.

DESCRIPTION

The genus Petrostroma has not been found sinceit was originally described by Döderlein (1892)

and then redescribed again by Döderlein (1898).It is difficult to distinguish this genus fromPlectroninia. The existence of ascending radiallines in the main framework of large tetractines ofthe single known species may be a function of itsmassive shape as compared with encrustingRecent Plectroninia. Although these may repre-sent different growth forms, we continue to sepa-rate the two genera, pending a revision thatincludes the fossil Minchinellidae. For illustra-tion, see Döderlein 1898.

Genus Tulearinia Vacelet, 1977

TYPE SPECIES. — Tulearinia stylifera Vacelet, 1977 bymonotypy.

DIAGNOSIS. — Minchinellidae (?) with a basal skele-ton consisting of tetractines with basal actines that areinterwoven but are not cemented, and with underlyinglayers of triactines linked in the same way.

DESCRIPTION

This genus may represent the first step in theprocess of the spicule linkage that is characteristicof Minchinellidae (Fig. 52). However, its inclu-sion in this family is questionable because theactine tips are only slightly modified, and truezygosis and diapasons are absent.


258 ZOOSYSTEMA • 2000 • 22 (2)

FIG. 51. — Cortical skeleton and osculum of Plectroninia vas-seuri Vacelet, 1967b. Scale bar: 66 µm (from Vacelet 1967b).

FIG. 50. — View of the choanoskeleton of Plectroninia sp., show-ing two layers of tetractines fused by their basal actines (SEM).Scale bar: 50 µm (from Vacelet 1991).


259ZOOSYSTEMA • 2000 • 22 (2)

FIG. 52. — Diagram of dissociated spicules of Tulearinia styliferaVacelet, 1977 (light micrograph); A, superficial diactines;B, microdiactines; C, perioscular triactines; D, triactines;E, tetractines from the basal network; F, tetractine from canals.Scale bars: A, D, E, 50 µm; B, C, 15 µm; F, 20 µm (from Vacelet1977).

Acx

ch

ct

d

e

B

C

DE

FFIG. 53. — Diagram of the organization of Petrobiona massilianaVacelet & Lévi, 1958. Abbreviations: cx, cortical skeleton; ch,choanosome; ct, tracts of storage cells; d, solid calcareousskeleton; e, parasitic excavating sponge (from Vacelet 1964).

Family PETROBIONIDAE Borojevic, 1979

TYPE GENUS. — Petrobiona Vacelet & Lévi, 1958 bymonotypy.

DIAGNOSIS. — Lithonida in which the basal skeletonis a solid mass. The living tissue is located between thecrests and spines of the basal skeleton. Thechoanoskeleton and cortical skeleton consist of freespicules that may be trapped within the rigid skeleton.

Genus Petrobiona Vacelet & Lévi, 1958

TYPE SPECIES. — Petrobiona massiliana Vacelet & Lévi,1958 by monotypy.

DIAGNOSIS. — Same definition as the family.

DESCRIPTION

The genus is known by a single species, which isthe only member of the Calcarea that is providedwith survival structures (“pseudogemmules”)

FIG. 54. — Surface view of the solid calcareous skeleton ofPetrobiona massiliana, showing terminal spines of the scleroder-mites (sc) and partially entrapped spicules (sp) (SEM). Scale bar:50 µm (from Vacelet 1991).

sc

sp

enclosed within the calcareous skeleton (Vacelet1964, 1990) (Fig. 53). The skeleton is formed bya solid mass of calcite consisting of elongated scle-rodermites that form a series of crests betweenwhich lies the living tissue. Spicules trapped withinthe massive skeleton do not dissolve (Fig. 54). Themassive skeleton differs in microstructure fromthat of Murrayona in the Calcinea. The species isknown from both living and fossil specimens fromPleistocene strata in the Mediterranean.

REFERENCES

Amano S. & Hori I. 1992. — Metamorphosis of cal-careous sponges. 1: Ultrastructure of free-swimminglarvae. Invertebrate Reproduction and Development21 (2): 81-90.

Amano S. & Hori I. 1993. — Metamorphosis ofcalcareous sponges. II: Cell rearrangement and dif-ferentiation in metamorphosis. Invertebrate Repro-duction and Development 24 (1): 13-26.

Borchiellini C., Chombard C., Lafay B. & Boury-Esnault N. 2000. — Molecular systematics ofsponges (Porifera). Hydrobiologia 420: 15-27.

Borojevic R. 1965. — Éponges calcaires des côtes deFrance. I: Amphiute paulini Hanitsch; les genres

Amphiute Hanitsch et Paraheteropia n. g. Archives deZoologie expérimentale et générale 106: 665-670.

Borojevic R. 1966. — Éponges calcaires des côtes deFrance. III: Discussion sur la taxonomie deséponges calcaires: Aphroceras ensata (Bowerbank) etUte gladiata sp. n. Archives de Zoologie expérimentaleet générale 107 (4): 703-724.

Borojevic R. 1967a. — Éponges calcaires recueillies enNouvelle-Calédonie par la mission Singer-Polignac:1-10, in Salvat B. (ed.), Mission Singer-Polignac:Pacifique. Volume 2. Fondation Singer-Polignac,Paris.

Borojevic R. 1967b. — Spongiaires d’Afrique du Sud.2: Calcarea. Transactions of the Royal Society of SouthAfrica 37: 183-227.

Borojevic R. 1970. — Différentiation cellulaire dansl’embryogenèse et la morphogenèse chez les spon-giaires, in Fry W. G. (ed.), The Biology of thePorifera. Symposium Zoological Society 25: 467-490.

Borojevic R. 1979. — Évolution des épongesCalcarea: 527-530, in Lévi C. & Boury-Esnault N.(eds), Biologie des Spongiaires. CNRS, Colloquesinternationaux du CNRS 291, Paris.

Borojevic R. & Boury-Esnault N. 1986. — Une nou-velle voie d’évolution chez les éponges Calcinea:Description des genres Burtonulla n. g. et Levinellan. g. Bulletin du Muséum national d’Histoire natu-relle 8 (3): 443-455.

Borojevic R. & Boury-Esnault N. 1987. — Revisionof the genus Leucilla Haeckel, 1872, with a re-

KEY OF GENERA OF LITHONIDA

1. Skeleton including a solid mass of calcite ................................................ Petrobiona

— Skeleton without a solid mass of calcite ................................................................ 2

2. Basal skeleton made of uncemented tri- and tetractines interlaced by their basalactines .................................................................................................... Tulearinia

— Basal skeleton made of tetractines cemented in a rigid network ............................ 3

3. Rigid network of tetractines embedded in a cement .............................. Minchinella

— Rigid network of tetractines not embedded in a cement ........................................ 4

4. Basal skeleton made of a single category of tetractines .................. Monoplectroninia

— Basal skeleton made of two categories of tetractines .............................................. 5

5. Basal skeleton forming a thin basal layer .............................................. Plectroninia

— Basal skeleton with ascending radial lines .............................................. Petrostroma


260 ZOOSYSTEMA • 2000 • 22 (2)

description of the type species Leucilla amphoraHaeckel,1872: 29-40, in Jones W. C. (ed.), Euro-pean Contributions to the Taxonomy of Sponges.Volume 1. Publication of the Sherkin IslandMarine Station, Litho Press Co, Middleton.

Borojevic R. & Klautau M. 2000. — Calcareoussponges from New Caledonia. Zoosystema 22 (2):187-201.

Borojevic R., Boury-Esnault N. & Vacelet J. 1990. —A revision of the supraspecific classification of thesubclass Calcinea (Porifera, class Calcarea). Bulletindu Muséum national d’Histoire naturelle, 4e sér., 12(2): 243-245.

Boury-Esnault N. & Rützler K. (eds) 1997. —Thesaurus of sponge morphology. SmithsonianContributions to Zoology 596: 1-55.

Boury-Esnault N., Muricy G., Gallissian M.-F. &Vacelet J. 1995. — Sponges without skeleton: anew Mediterranean genus of Homoscleromorpha(Porifera, Demospongiae). Ophelia 43 (1): 25-43.

Bowerbank J. S. 1864. — A Monograph of the BritishSpongiadae. Volume 1. Robert Hardwicke, London,289 p.

Breitfuss L. 1896. — Kalschwämme des Bremer-Expedition nach Ost-Spitzbergen. ZoologischerAnzeiger 19: 426-432.

Brøndsted H. V. 1931. — Die Kalkschwämme: vol-ume 20: 1-47, in Drugalski E. von (ed.), DeutschenSüdpolar Expedition.

Burton M. 1959. — Sponges. British Museum, JohnMurray Expedition, London, 281 p.

Burton M. 1963. — A Revision of the Classification ofthe Calcareous Sponges, with a Catalogue of theSpecimens in the British Museum. British Museum(Natural History), London, 653 p.

Carter H. J. 1871. — A description of two newCalcispongiae (Trichogypsia, Leuconia). Annals andMagazine of Natural History 8: 1-28.

Cavalier-Smith T., Allsopp M. T. E. P., Chao E. E.,Boury-Esnault N. & Vacelet J. 1996. — Spongephylogeny, animal monophyly, and the origin ofthe nervous system: 18S rRNA evidence. CanadianJournal of Zoology 74: 2031-2045.

Dendy A. 1892a. — Preliminary account of Synutepulchella, a new genus and species of CalcareousSponges. Proceedings of the Royal Society of Victoria4: 1-6.

Dendy A. 1892b. — Synopsis of the AustralianCalcarea Heterocoela, with proposed classificationof the group, and description of some new generaand species. Proceedings of the Royal Society ofVictoria 5: 69-116.

Dendy A. 1893. — Studies on the comparative anato-my of sponges. V: Observations on the structureand classification of the Calcarea heterocoela.Quarterly Journal of Microscopical Science 35: 159-257.

Dendy A. 1913. — Report on the calcareous spongescollected by H. M. S. Sealark in the Indian

Ocean. Transactions of the Linnean Society ofLondon, Zoology 16: 1-29.

Dendy A. & Frederick L. M. 1924. — On a collectionof sponges from the Abrolhos Islands, WesternAustralia. Journal of the Linnean Society of London,Zoology 35: 477-519.

Dendy A. & Row R. W. 1913. — The classificationand phylogeny of the calcareous sponges, with a ref-erence list of all the described species, systematicallyarranged. Proceedings of the Zoological Society ofLondon 47: 704-813.

Döderlein L. 1892. — Description of Petrostromaschulzei, n. g. et sp. of Calcarea, representing a neworder Lithones. Verhandlungen der DeutschenZoologischen Gesellschaft in Heidelberg 2: 143-145.

Döderlein L. 1898. — Über die Lithonina, eine neueGruppe von Kalkschwämmen. Zoologische Jahr-bücher 10: 15-32.

Duboscq O. & Tuzet O. 1937. — L’ovogenèse, lafécondation et les premiers stades du développe-ment des éponges calcaires. Archives de Zoologieexpérimentale et générale 79: 157-316.

Duboscq O. & Tuzet O. 1941. — Sur les cellules encroix des Sycon (Sycon ciliatum Fabricius, Sycon coro-natum Ellis et Solander, Sycon elegans Bowerbank)et leur signification. Archives de Zoologie expérimen-tale et générale 81: 151-163.

Duboscq O. & Tuzet O. 1942. — Recherches com-plémentaires sur l’ovogenèse, la fécondation et lespremiers stades du développement des éponges cal-caires. Archives de Zoologie expérimentale et générale81: 395-466.

Gallissian M. F. 1989. — Le spermiokyste de Syconsycandra (Porifera, Calcarea): étude ultrastructurale.Comptes Rendus de l’Académie des Sciences 309 (7):251-258.

Gallissian M. F. & Vacelet J. 1990. — Fertilizationand nutrition of the oocyte in the calcified spongePetrobiona massiliana: 175-181, in Rützler K. (ed.),New Perspectives in Sponge Biology. SmithsonianInstitution Press, Washington.

Grant R. E. 1826. — Observations and experimentson the structure and functions of the sponge.Edinburgh Philosophical Journal 14: 336-341.

Haeckel E. 1872. — System der Kalkschwämme(Calciospongien oder Grantien). Verlag von GeorgReimer, Berlin, 418 p.

Hartman W. D. 1958. — A re-examination ofBidder’s classification of the Calcarea. SystematicZoology 7: 55-109.

Hôzawa S. 1919. — Report on the calcareous spongescollected during 1906 by the United StatesFisheries Steamer Albatross in the NorthwesternPacific. Proceedings of the United States NationalMuseum 54: 525-556.

Hôzawa S. 1923. — On a new genus of calcareoussponge. Annotationes Zoologicae Japonenses 10: 183-190.


261ZOOSYSTEMA • 2000 • 22 (2)

Ilan M. & Vacelet J. 1993. — Kebira uteoides(Porifera, Calcarea) a recent “Pharetronid” spongefrom coral reefs. Ophelia 38: 107-116.

Ivanov A. V. 1971. — On the reasons of excurvationof the embryo in the colonial Phytomonadina andcalcareous sponges. Monitore Zoologico Italiano 5: 1-10.

James N. P. & Klappa C. F. 1983. — Petrogenesis ofearly cambrian reef limestones, Labrador, Canada.Journal of Sedimentary Petrology 53 (4): 1051-1096.

Jenkin C. F. 1908a. — Porifera. III: Calcarea: 1-52, inNational Antarctic Expedition. Natural History.Volume 4. BMNH, London.

Jenkin C. F. 1908b. — The marine fauna of Zanzibarand British east Africa, from collections made byCyril Crossland, M.A. in the years 1901 & 1902.The calcareous sponges. Proceedings of the ZoologicalSociety of London 29: 434-456.

Kirk H. B. 1895. — Further contribution to a knowl-edge of New Zealand sponges. Transactions of theNew Zealand Institute 27: 287-292.

Kirkpatrick R. 1900. — Description of sponges fromFunafuti. The Annals and Magazine of NaturalHistory 34: 345-363.

Klautau M., Solé-Cava A. M. & Borojevic R. 1994. —Biochemical systematics of sibling sympatric speciesof Clathrina (Porifera: Calcarea). BiochemicalSystematics and Ecology 22 (4): 367-375.

Lafay B., Boury-Esnault N., Vacelet J. & Christen R.1992. — An analysis of partial 28S ribosomal RNAsequences suggests early radiations of sponges.Biosystems 28: 139-151.

Lambe L. M. 1900. — Sponges from the coasts ofnorth-eastern Canada and Greenland. Transactionsof the Royal Society of Canada 6: 19-38.

Laubenfels M. W. de 1936. — A Discussion of theSponge Fauna of the Dry Tortugas in particular andthe West Indies in general, with Material for aRevision of the Families and Orders of the Porifera.Publications of the Carnegie Institution ofWashington, Papers from Tortugas Laboratory 30,Washington, 225 p.

Lendenfeld R. von 1891. — Die Spongien der Adria. I:Die Kalkschwämme. Wilhelm Engelmann, Leipzig,212 p.

Lévi C. 1956. — Étude des Halisarca de Roscoff.Embryologie et systématique des démosponges.Archives de Zoologie expérimentale et générale 93: 1-184.

Peixinho S. 1980. — Étude cytologique de la reconstitu-tion après dissociation cellulaire de Leuconia nivea(Grant, 1826) et Grantia compressa (Fabricius,1780) (Porifera, Calcarea). Thèse de 3e cycle,Université Pierre et Marie Curie, Paris, France.

Poléjaeff N. 1883. — Report on the Calcarea dredgedby H. M. S. Challenger during the years 1873-1876:1-76, in Thomson C. W. & Murray J. (eds), Reportof the Scientific Results of the Voyage of

H. M. S. Challenger. MacMillan & Co., Zoology24, London.

Pouliquen L. & Vacelet J. 1970. — Nouvelles obser-vations sur des Éponges Pharétronides Minchi-nellidae de Méditerranée. Téthys 2 (2): 437-442.

Reitner J. & Mehl D. 1995. — Early Paleozoic diver-sification of sponges: new data and evidences.Geologische Paläontologische Mitteilungen Innsbruck20: 335-347.

Row R. W. H. 1909. — Reports on the marine biolo-gy of the Sudanese Red Sea. XIX: Report on theSponges collected by Mr. Cyril Crossland in 1904-1905. Journal of the Linnean Society of London,Zoology 31: 182-214.

Sarà M., Gaino E. & Valentini F. 1974. — Olynthusformation by cell aggregation in Sycon vigilans(Porifera, Calcispongiae). Vie et Milieu 24: 225-234.

Schmidt O. 1870. — Grundzüge einer Spongien Faunades Atlantischen Gebietes. Wilhelm Engelmann,Leipzig, 85 p.

Schulze F. E. 1875. — Über den Bau und dieEntwicklung von Sycandra raphanus Haeckel.Zeitschrift für wissenschaftliche Zoologie 25: 247-280.

Solé-Cava A. M., Klautau M., Boury-Esnault N.,Borojevic R. & Thorpe J. P. 1991. — Genetic evi-dence for cryptic speciation in allopatric popula-tions of two cosmopolitan species of the calcareoussponge genus Clathrina. Marine Biology 111 (3):381-386.

Tuzet O. 1947. — L’ovogenèse et la fécondation del’éponge calcaire Leucosolenia coriacea et de l’épongesiliceuse Reniera elegans. Archives de Zoologie expéri-mentale et générale 85: 127-148.

Vacelet J. 1964. — Étude monographique de l’épongecalcaire Pharétronide de Méditerranée, Petrobionamassiliana Vacelet et Lévi. Les Pharétronidesactuelles et fossiles. Recueil des Travaux de la Stationmarine d’Endoume 34: 1-125.

Vacelet J. 1967a. — Descriptions d’épongesPharétronides actuelles des tunnels obscurs sous-récifaux de Tuléar (Madagascar). Recueil des Travauxde la Station marine d’Endoume suppl. 6: 37-62.

Vacelet J. 1967b. — Quelques Éponges Pharétronideset « Silico-calcaires » des grottes sous-marinesobscures. Recueil des Travaux de la Station marined’Endoume 42: 121-132.

Vacelet J. 1977. — Éponges Pharétronides actuelles etSclérosponges de Polynésie Française, Madagascaret de la Réunion. Bulletin du Muséum nationald’Histoire naturelle 444: 345-366.

Vacelet J. 1981. — Éponges hypercalcifiées(« Pharétronides », « Sclérosponges ») des cavités desrécifs coralliens de Nouvelle-Calédonie. Bulletin duMuséum national d’Histoire naturelle 3 (A): 313-351.


262 ZOOSYSTEMA • 2000 • 22 (2)

Vacelet J. 1990. — Storage cells of calcified relictsponges. Chap. 19: 144-152, in Ruetzler K. (ed.),New Perspectives in Sponge Biology. SmithsonianInstitution Press, Washington.

Vacelet J. 1991. — Recent Calcarea with a reinforcedskeleton (“Pharetronids”): 252-265, in Reitner J. &

Keupp H. (eds), Fossil and Recent Sponges. Springer-Verlag, Berlin.

Van Kempen T. M. G. 1978. — Anthaspidellidsponges from the early paleozoic of Europe andAustralia. Neues Jahrbuch für Geologie undPaläontologie Abhandlungen 156: 305-337.

Submitted on 4 May 1999;accepted on 6 October 1999.


263ZOOSYSTEMA • 2000 • 22 (2)

A revision of the supraspecific classification of the ...sciencepress.mnhn.fr/sites/default/files/articles/pdf/z2000n2a2.pdfcation of the subclass Calcaronea (Porifera, class Calcarea).

Documents