6 Phylum Porifera: UNCORRECTED PAGE PROOFSsponges make up almost 75 percent of the total benthic biomass at a depth of 100–200 m. Subtidal and deeper water species that do not confront

he phylum Porifera (Latin porus, “pore”; ferre, “to bear”) contains thoseanimals commonly called sponges. Figures 6.1 and 6.2 illustrate a variety

of sponge body forms and some sponge anatomy. Box 6A lists the majorcharacteristics of sponges. Poriferans are sessile, suspension-feeding, multicellu-lar animals that utilize flagellated cells called choanocytes to circulate waterthrough a unique system of water canals. Porifera is the only phylum at the para-zoan grade of body construction (i.e., Metazoa lacking true embryological germlayering). Not only are true tissues absent, but most of the body cells are totipo-tent—they are capable of changing form and function. Despite the fact thatsponges are large-bodied multicellular animals, they function largely like organ-isms at the unicellular grade of complexity. As you will discover in this chapter,their nutrition, cellular organization, gas exchange, and response to environmen-tal stimuli are all very protist-like.

About 5,500 living species of sponges have been described, nearly all of whichare restricted to benthic marine environments. They occur at all depths, but un-polluted littoral and tropical reef habitats harbor especially rich sponge faunas.Most littoral sponges grow as thick or thin layers on hard surfaces. Benthicsponges that live on soft substrata are often upright and tall, thus avoiding burialby the shifting sediments of their environment. Some sponges reach considerablesize (up to 2 m in height on Caribbean reefs, and even larger in the Antarctic) andmay constitute a significant portion of the benthic biomass. In Antarctica,sponges make up almost 75 percent of the total benthic biomass at a depth of100–200 m. Subtidal and deeper water species that do not confront strong tidalcurrents or surge are usually large and exhibit a stable, even symmetrical, exter-nal form. The deeper water hexactinellid sponges often assume unusual shapes,

Phylum Porifera: The Sponges

Sponges have made no progress in theformation of an anterior end or a head.Libbie Hyman,The Invertebrates, Vol. 1, 1940

6

T

UNCORRECTED PAGE PROOFS

many being delicate glasslike structures, others roundand massive, and still others ropelike. A few species inthe class Demispongiae inhabit fresh waters. (Figure 6.1).

Sponges display nearly every color imaginable, in-cluding bright lavenders, blues, yellows, crimsons, andwhite. Many species harbor symbiotic bacteria or uni-cellular algae that may color the sponge’s body.

Taxonomic History and ClassificationThe sessile nature of sponges and their generally amor-phous (asymmetrical) growth form convinced early nat-uralists that they were plants. It was not until 1765,when the nature of their internal water currents was de-scribed, that sponges were recognized as animals. Thegreat naturalists of the late eighteenth and early nine-teenth centuries (Lamarck, Linnaeus, and Cuvier) classi-fied the sponges under Zoophytes or Polypes, regardingthem as allied to anthozoan cnidarians. Throughoutmuch of the nineteenth century they were placed withcnidarians under the name Coelenterata or Radiata. Themorphology and physiology of sponges were first ade-quately understood by R. E. Grant. Grant created forthem the name Porifera, although other names were fre-quently used (e.g., Spongida, Spongiae, Spongiaria).Huxley (1875) and Sollas (1884) first proposed the sepa-ration of sponges from “higher” Metazoa.

Historically, the classes of Porifera have been definedby the nature of their internal skeletons. Until recently,four classes were recognized: Calcarea, Hexactinellida,Demospongiae, and Sclerospongiae. The class Sclero-spongiae included those species that produce a solid,calcareous, rocklike matrix on which the living animalgrows. These poriferans are also known as corallinesponges; about 15 living species have been described.This class, however, was abandoned over a decade ago,and its members relegated to the Calcarea and Demo-spongiae (Vacelet 1985).* Demospongiae is the largestsponge class, comprising about 95 percent of the livingspecies. Because of its size and variability, the Demo-spongiae presents the most problems to taxonomists. Ina series of papers published between 1953 and 1957,Lévi proposed an important reappraisal of the Demo-spongiae, incorporating reproductive characteristics forthe first time.

Although the mainstay of sponge taxonomy has tra-ditionally been the anatomy of the spicules, these skele-tal structures have proven inadequate for developingstable phylogenetic hypotheses and classifications.Indeed, some sponge species lack spicules altogether.Hence specialists are now using embryological, bio-

chemical, histological, and cytological methods to diag-nose sponge taxa. The great variability in sponge mor-phology and the difficulty in precisely setting the limitsof a sponge species have probably driven many potentialporiferologists to frustration (and to other taxa) early intheir careers. Even the great sponge taxonomist ArthurDendy was known to frequently end a species diagnosiswith a question mark. This state of affairs was summa-rized by one student regarding California sponge studies(Ristau 1978):

The study of California sponges has not generated afervor of activity over the years, nor has the literaturebeen saturated with information about this little-stud-ied . . . phylum. Probably the greatest interest generat-ed by sponges occurred recently, when several newsagencies reported that giant, and presumably mutant,sponges were found growing on undersea nuclearwaste storage containers (San Francisco Chronicle, Sep-tember 14, 1976). It has been rumored that the Japaneseare now planning a motion picture in which a sleeze†

of giant sponges rises from the depths of the FarallonIslands and phagocytizes the North Beach area of SanFrancisco. Undoubtedly, when this epic materializes,research and interest in California sponges will in-crease. Until that time, however, those interested in thesponge fauna of this area must be content with thepaucity of scientific literature on this subject.

Recently a host of important bioactive compounds hasbeen discovered in sponges, many having potentialpharmacological significance (e.g., antimicrobial, anti-

180 CHAPTER SIX

*It has also been suggested that the Hexactinellida should beremoved from the Porifera and assigned to a separate phylum,Symplasma, but this proposal has not received much support.

†“Sleeze” is a term coined by Ristau for an aggregation ofsponges; the usage is comparable to other such collective nounsthat define animal groups (e.g., flock, herd, gaggle).

1. Metazoa at the cellular grade of construction,without true tissues; adults asymmetrical orsuperficially radially symmetrical

2. Cells totipotent

3. With unique flagellated cells—choanocytes—thatdrive water through canals and chambers consti-tuting the aquiferous system

4. Adults are sessile suspension feeders; larval stagesare motile and usually lecithotrophic

5. Outer and inner cell layers lack a basement mem-brane (except perhaps in the subclassHomoscleromorpha)

6. Middle layer—the mesohyl—variable, but alwaysincludes motile cells and usually some skeletalmaterial

7. Skeletal elements, when present, composed ofcalcium carbonate or silicon dioxide (typically inthe form of spicules), and/or collagen fibers

BOX 6A Characteristics of the Phylum Porifera


PHYLUM PORIFERA: THE SPONGES 181

Figure 6.1 Representative sponges. (A) Leucetta, a shallow-water cal-careous sponge. (B) Euryspongia, a demosponge. (C) Agelas, a demo-sponge. (D) Three specimens of “stalked” glass sponges (Hexactinellida).(E) Skeleton of Euplectella, a hexactinellid sponge known as Venus’s flowerbasket. (F) A coralline sponge from tropical coral reefs. (G) The commonencrusting demosponge Haliclona. (H) The freshwater demospongeSpongilla.

(C)

FPOLo res-hi res

tocome

This ishi ressize

(A) (B)

(F) (G)

(E)(D)

(H)

e-fileneedsproof

e-fileneedsproof

e-fileneedsproof


inflammatory, antitumor, cytotoxic, and anti-foulingcompounds). The discovery of these natural products insponges has led to a renewed interest in this group and acall for the training of more sponge taxonomists, whosenumbers have dwindled over the past few decades.

PHYLUM PORIFERACLASS CALCAREA: Calcareous sponges (Figure 6.1A). Spiculesof mineral skeleton composed entirely of calcium carbonate laiddown as calcite; skeletal elements often not differentiated intomegascleres and microscleres; spicules usually 1, 3, or 4-rayed;body with asconoid, synconoid, or leuconoid construction; allmarine.

SUBCLASS CALCINEA: Free-living larvae are hollowcoeloblastulae, flagellated, and can become solid parenchy-mula-like structures by cellular ingression; choanocyte nu-clei located basally; flagellum arises independent of nucle-us; with regular triradiate spicules; spicules free, but somespecies (e.g. Murrayona) with massive calcite skeleton.(e.g., Clathrina, Dendya, Leucascus, Leucetta, Soleniscus)

SUBCLASS CALCARONEA: Free-living larvae are partly fla-gellated amphiblastulae; choanocyte nuclei apical; flagel-lum arises directly from nucleus; spicules free or fused. (e.g.,Amphoriscus, Grantia, Leucilla, Leucosolenia, Petrobiona,Scypha [= Sycon])

CLASS HEXACTINELLIDA: Glass sponges (Figure 6.1D,E).Spicules siliceous and basically 6-rayed (hexactinal); bothmegascleres and microscleres always present; body wall cav-ernous, with trabecular network; external pinacoderm absentand replaced by a noncellular dermal membrane; choanocytelayer may be syncytial; exclusively marine; primarily deep-water.

SUBCLASS AMPHIDISCOPHORA: Body never attached toa hard substratum but anchored in soft sediments by a basaltuft or tufts of spicules; megascleres discrete spicules, neverfused into a rigid network; with birotulate microscleres,never hexasters; mostly deep-water. (e.g., Hyalonema,Monorhaphis, Pheronema)

SUBCLASS HEXASTEROPHORA: Usually attached to hardsubstrata, but sometimes attached to sediments by a basalspicule tuft or mat; microscleres are hexasters; megascleressometimes free, but usually fused into a rigid skeletal frame-work, in which case sponge may assume large and elabo-rate morphology. (e.g., Aphrocallistes, Caulophacus, Eu-plectella, Hexactinella, Leptophragmella, Lophocalyx, Rosella,Sympagella)

CLASS DEMOSPONGIAE: Demosponges (Figure 6.1B,C,F–H).With siliceous spicules; spicules not 6-rayed; spicule skeletonmay be supplemented or replaced by an organic collagenousnetwork (“spongin”); marine, brackish, or freshwater sponges,occurring at all depths.

SUBCLASS HOMOSCLEROMORPHA: Embryos incubated,larvae amphiblastulae-like (“cinctoblastula”); differentiationof spicules into mega- and microscleres not evident; allspicules very small (usually less than 100 mm) and distrib-uted in large numbers throughout body, with little region-al organization; with a “pseudobasal membrane” underly-ing the pinacoderm. Usually littoral, but some occuring atshelf and slope depths. (e.g., Corticium, Oscarella, Plakina,Plakortis, Pseudocorticium)

SUBCLASS TETRACTINOMORPHA: Reproduction typical-ly oviparous, but incubation with direct development oc-curs in one order; larvae, when present, typically parenchy-mulae, with distinct megascleres and microscleres;megascleres organized into distinct patterns, either axial orradial; numerous orders and families. (e.g., Asteropus,Chondrilla, Chondrosia, Cliona, Cryptotethya, Geodia, Poly-mastia, Rhabderemia, Stelletta, Suberites, Tethya, Tetilla). Thissubclass now contains some sponges from the recentlyabandoned Sclerospongiae: the merliids (Merlia) and atleast some tabulates (Acanthochaetetes).

SUBCLASS CERACTINOMORPHA: Mostly viviparous, withincubation of parenchymulla larvae; distinct microscleresand megascleres present; spongin present in all but onefamily (Halisarcidae); includes the freshwater familiesSpongillidae and Potamelepidae. (e.g., Adocia, Agelas,Aplysilla, Aplysina [= Verongia], Asbestopluma, Axinella, Axo-ciella, Callyspongia, Clathria, Coelosphaera, Halichondria, Hal-iclona, Halisarca, Hymeniacidon, Ircinia, Lissodendoryx, Mi-crociona, Mycale, Myxilla, Spongia, Spongilla, Tedania). TheCeractinomorpha now contain some sponges previously assigned to the Sclerospongiae, including the stromato-porids (e.g., Astrosclera, Calcifibrospongia), the ceratoporellids(e.g., Ceratoporella, Stromatospongia, Hispidopetra, Goreauiel-la), and the enigmatic Vaceletia crypta.

The Poriferan BauplanIn Chapter 3 we discussed some of the limitations of theparazoan grade of construction, in which true tissuesand organs are absent. Now we discuss the various waysin which sponges have overcome the handicaps im-posed by their primitive level of organization. You willnotice a striking resemblance to the protists in many re-gards. Two unique organizational attributes definesponges and have played major roles in poriferan suc-cess: the water current channels, or aquiferous system(and its choanocytes), and the highly totipotent nature ofsponge cells. The tremendous diversity among spongesin size and shape has occurred both evolutionarily andindividually and is largely derived from these twounique characteristics. Increases in size and surface areaare accomplished by folding of the body wall into a vari-ety of patterns. Furthermore, variation in the overallshapes of sponges results from different growth patternsin various environments. This general plasticity in size,shape, and construction, plus the fact that most individ-ual sponge cells are capable of radically altering theirform and function as needed, compensate in part for theabsence of tissues and organs. The aquiferous systembrings water through the sponge and close to the cells re-sponsible for food gathering and gas exchange. At thesame time, excretory and digestive wastes and reproduc-tive products are expelled by way of the water currents.The volume of water moving through a sponge’s aquif-erous system is remarkable. A 1 × 10-cm individual ofthe complex sponge Leuconia pumps about 22.5 l ofwater through its body daily. Researchers have recorded

182 CHAPTER SIX UNCORRECTED PAGE PROOFS

sponge pumping rates that range from 0.002 to 0.84 ml ofwater per second per cubic centimeter of sponge body. Alarge sponge filters its own volume of water every 10 to20 seconds.

Early workers treated sponges as essentially colonialanimals. The other view, and the one that we prefer,holds that the entirety of a sponge—that is, any and allsponge material bounded by a continuous outer cover-ing—constitutes a single individual. The fact is, a wholesponge grows as a whole body, dictated largely by en-vironmental factors (e.g., water flow dynamics, sub-stratum contours). Changes in body form can arise any-where in or on the organism in response to theseenvironmental pressures. Sponges grow by continuallyadding new cells that differentiate as needed; this is notusually viewed as colonial asexual reproduction. Theexistence of some coordinated behavior in sponges(e.g., cessation of choanocyte pumping, synchronousoscular contractions) further supports the view thateach sponge is, in its entirety, an “individual.”

Body Structure and the Aquiferous SystemThe outer surface cells of a sponge make up the pinaco-derm and are called pinacocytes. Most of the inner sur-faces comprise the choanoderm and are composed offlagellated cells called choanocytes. Both of these layersare a single cell thick. Between these two thin cellularsheets is the mesohyl, which may be very thin in somesimple sponges, or massive and thick in larger species(Figures 6.2). The pinacoderm is perforated by smallholes called dermal pores or ostia (singular, ostium),depending on whether the opening is surrounded byseveral cells or one cell, respectively (Figure 6.3). Wateris pulled through these openings and is driven acrossthe choanoderm by the beating of the choanocyte flagella.The choanocytes pump large volumes of water throughthe sponge body at very low pressures, establishing thewater current (aquiferous) system.

A cuticle, or layer of coherent collagen, may cover (or even replace) the pinacoderm in some species. Thepinacoderm itself can be a simple external sheet, but typ-ically it also lines some of the internal cavities of theaquiferous system where choanocytes do not occur.Pinacoderm cells that line internal canals are called en-dopinacocytes. The choanoderm also can be simple andcontinuous, or folded and subdivided in various ways.The mesohyl varies in thickness and plays vital roles indigestion, gamete production, secretion of the skeleton,and transport of nutrients and waste products by specialameboid cells. The mesohyl includes a noncellular col-loidal mesoglea in which are embedded collagen fibers,spicules, and various cells; as such, it is really a type ofmesenchyme. A great number of cell types may be foundin the mesohyl. Most of these cells are able to changefrom one type to another as required; but some differen-tiate irreversibly, , such as those that commit themselves

to reproduction or to skeleton formation.The mobility of all cells, including pinacocytes and

choanocytes, has been demonstrated by dramatic time-lapse cinematography. The cells of the pinacoderm andchoanoderm are more stable than those of the mesohyl,but in general, the whole structure may be thought of asa continuously mobile system. In fact, recent observa-tions by Bond (1997/1998) confirm a report of nearly halfa century ago that some sponges actually do move fromone place to another. Ameboid cells along the base of thesponge “crawl” as others bring spicules as support forthe leading edge of the sponge. Bond reported thatsome ameobocytes actually broke free from the spongeand moved about on their own for a time, eventually re-turning to the parent sponge body. This locomotion insponges is not sufficient to provide them with a quickescape mechanism from predators, however; in Bond’swords, “The champion speedster regularly moved morethan four millimeters a day.”

During growth, the pinacoderm and choanoderm areeach only one cell thick. By increasing their folding asmesohyl volume increases, these layers maintain a sur-face area-to-volume ratio sufficient to sustain adequatenutrient and waste exchange throughout the whole in-dividual. The one-cell thick choanoderm may remainsimple and continuous (the asconoid condition), or itmay become folded (the syconoid condition), or it maybecome greatly subdivided into separate flagellatedchambers (the leuconoid condition) (Figure 6.3).

The asconoid condition is found in some adult, radiallysymmetrical calcareous sponges (e.g., Clathrina,Leucosolenia) and in the early growth stage (olynthus) of newly settled calcareous sponges (Figure 6.4A).Asconoid sponges rarely exceed 10 cm in height and re-main as simple, vase-shaped, tubular units. The thin wallsenclose a central cavity called the atrium (= spongocoel),which opens to the outside via a single osculum. Thepinacoderm of asconoid and very simple syconoidsponges has specialized cells called porocytes. Duringembryogeny, each porocyte elongates and rolls to form acylindrical tube. The porocyte extends all the way throughthe pinacoderm, the thin mesohyl, and the choanoderminto the atrium, emerging between adjacent choanocytes(Figures 6.4B and 6.7C). The external opening of the poro-cyte canal is called an ostium or incurrent pore. Thechoanoderm is a simple, unfolded layer of choanocyteslining the entire atrium. Water moving through an as-conoid sponge flows through the following structures: os-tium → spongocoel (over the choanoderm) → osculum.

Simple folding of the pinacoderm and choanodermproduces the syconoid condition, within which severallevels of complexity are possible (Figure 6.3B,C). Ascomplexity increases, the mesohyl may thicken and ap-pear to have two layers. The outer “cortical region,” orcortex, often contains skeletal elements that are differentfrom those found in the interior portion of the mesohyl.

PHYLUM PORIFERA: THE SPONGES 183UNCORRECTED PAGE PROOFS

In those sponges with a cortex, the incurrent openingsare lined by several cells (not formed by a single poro-cyte) and are referred to as dermal pores. In the syco-noid condition, choanocytes are restricted to specificchambers or diverticula of the atrium called choanocytechambers (or flagellated chambers, or radial canals).Each choanocyte chamber opens to the atrium by awide aperture called an apopyle. Syconoid spongeswith a thick cortex possess a system of channels or in-current canals that lead from the dermal pores throughthe mesohyl to the choanocyte chambers. The openingsfrom these channels to the choanocyte chambers are

called prosopyles. In such a complex syconoid sponge,water moving from the surface into the body flowsalong the following route: incurrent (dermal) pore → in-current canal → prosopyle → choanocyte chamber →apopyle → atrium → osculum. Syconoid construction isfound in many calcareous sponges (e.g., Scypha, alsoknown as Sycon). Some syconoid sponges appear radial-ly symmetrical, but their complex internal organizationis largely asymmetrical.

The leuconoid condition is produced by additionalfolding of the choanoderm and further thickening of themesohyl by cortical growth. These modifications are ac-

184 CHAPTER SIX

Figure 6.2 Sponge body forms. (A)The unusual demosponge Coelosphaerahatchi (height in life 27 mm). (B) Thecoralline sponge Merlia normani (verticalsection) has a basal calcareous matrixwithin which individual compartmentsare filled by secondary deposition. Thesuperficial soft tissue contains the cho-anocyte chambers and is supported bytracts of siliceous spicules. (C) Thedemosponge Haliclona permollis, asponge with a tubular type of architec-ture; three successive levels of magnifi-cation are shown, from left to right. (D)Microciona prolifera, a demosponge witha more solid type of architecture; threesuccessive levels of magnification areshown, from left to right.

(A) (B)

(D)

(C)


companied by subdivision of the flagellated surfacesinto discrete oval choanocyte chambers (Figure 6.3D). Inthe leuconoid condition, one finds an increase in num-ber and a decrease in size of the choanocyte chambers,which typically cluster in groups in the thickened meso-hyl. The atrium is reduced to a series of excurrent canals(or exhalent canals) that carry water from the cho-

anocyte chambers to the oscula (Figure 6.5). The flow ofwater through a leuconoid sponge is: dermal pore → in-current canal → prosopyle → choanocyte chamber →apopyle → excurrent canal → osculum. Leuconoid or-ganization is typical of most calcareous sponges and allmembers of the Demospongiae.

It is important to realize that the flow rate is not uni-form through the various parts of the aquiferous sys-tem. Functionally, it is critical that water be moved veryslowly over the choanoderm, allowing time for ex-


Figure 6.3 Body complexity in sponges. (Arrows indi-cate flow of water.) (A) The asconoid condition. (B) A sim-ple syconoid condition. (C) A complex syconoid conditionwith cortical growth. (D) A leuconoid condition.

(A) (B)Osculum

Ostium

Spongocoel(= atrium)

Choanocyte

(C) (D)

Apopyle Incurrent canal

Choanocyte canal (chamber)

Prosopyle

Choanocytes

CortexDermal pore

Incurrent canal Excurrent canal

Dermal pore

Incurrent canal

Prosopyle

Osculum

Apopyle

Choanocytechamber

Choanocyte

Choanocytecanal (chamber)

Prosopyle

Ostium

Apopyle

changes of nutrients, gases, and wastes between thewater and the choanocytes. The changes in water flowvelocity through this plumbing are a function of the ef-fective accumulated cross-sectional diameters of thechannels through which the water moves (see Chapter3, or your old physics notes). Water flow velocity de-creases as the cross-sectional diameter increases; thus, ina sponge, velocities are lowest over the choanoderm.Furthermore, water leaving the oscula must be carriedfar enough away to prevent it being recycled by thesponge. In environments of relatively high turbulence,currents, or wave action, this potential recycling ofwastes is not a problem. However, sponges that residein relatively calm water rely on the maintenance of highvelocities of water flow through the oscula (or on modi-fied body shapes) to push the excurrent water farenough away from the sponge to avoid the incoming

currents. In an irregularly shaped leuconoid sponge liv-ing in quiet water, the combined cross-sectional diame-ter of all the incurrent pores is far less than that of all thechoanocyte chambers. But the total oscular diameter iseven less than that of the incurrent pores. Simply put,

186 CHAPTER SIX

Figure 6.4 The asconoid condition. (A) An olynthus,the asconoid form that follows larval settlement in cal-careous sponges. (B) Major cell types in an asconoidsponge. (C) The simple calcareous sponge Leucosoleniashows the asconoid body form and skeleton of CaCO3spicules.

(C)


the water enters at some velocity x, slows to a small frac-tion of x as it passes over the choanoderm, then exits thesponge at a velocity much greater than x. In complexsponges, the differences in velocity are dramatic. Flowrate regulation is also facilitated in some sponges, inpart, by the activity of ameboid cells (called centralcells) that reside near the apopyles of the choanocytechambers. These cells can slow or speed the exit ofwater from the chambers by changing shape and posi-tion across the apopyle (see Figure 6.7I).

The recognition of the various levels of organizationand complexity among poriferans allows one to quicklyand simply describe a sponge’s basic anatomical plan.There is very little evidence, however, that the asconoidplan is necessarily the most primitive, or that all spongelineages have moved through these three levels of com-plexity during their evolution. Nor do all sponges passthrough three such developmental stages. In addition,gradations of and intermediates between the three basicplans are common. Nonetheless, among adult sponges,the simplest organizations (asconoid and syconoid)occur only in the class Calcarea, which is thought to bethe most primitive class of living poriferans. Further-more, calcareous sponges of the leuconoid condition dopass through asconoid and syconoid stages as theygrow, and it is only in this class that all three organiza-tional body plans occur.

The hexactinellid sponges. The hexactinellids differconsiderably from calcareous sponges and demo-sponges (Figure 6.6). The bodies of hexactinellid

sponges display a greater degree of radial, or superficialradial, symmetry than any other group. There is nopinacoderm or its equivalent in hexactinellids. A dermalmembrane is present, but it is extremely thin; no dis-crete or continuous cellular structure supports it. In-current pores are simple holes in this dermal mem-brane. Cellular material is sparsely distributed andforms a trabecular network stretching across intercon-necting internal cavities called subdermal lacunae(Figure 6.6A). The thimble-shaped flagellated chambersare arranged in a single layer and are supported withinthe trabecular network. Both the trabecular networkand the walls of the flagellated chambers appear to besyncytial (i.e., discrete choanocytes do not exist). Waterenters the incurrent pores, passes into the subdermal la-cunae, and from there enters the choanocyte chambersvia the prosopyles.

The unique structure of hexactinellids is so strikingthat some workers (e.g., Bergquist 1985) have even sug-gested the hexactinellids might be regarded as a sepa-rate phylum (the Symplasma). However, as explained


Osculum Excurrent canal

Dermal (incurrent) pores

Figure 6.5 The surface of a living demosponge (Clathria).The complex system of ostia opens into underlying incur-rent canals, and large oscula receive several excurrentcanals.

Figure 6.6 Internal anatomy of Hexactinellida. (A) Thebody wall of Euplectella (transverse section). A dermal layercovers the trabecular network. (B) The choanosyncytiumof Aphrocallistes vastus (vertical section).


in Chapter 2, phylogenetic relationships are best soughtin similarities among groups, not in differences, and bythis reasoning we treat the hexactinellids as poriferans.Also, the syncytial nature of hexactinellids has recentlybeen questioned.

Cell TypesBecause of the nontissue nature of sponges and becausecellular totipotency plays a major role in poriferan biol-ogy, considerable effort has gone into describing andclassifying sponge cell types. Prior to the 1970s, textsgenerally recognized only a few basic kinds of poriferancells. However, subsequent detailed histochemical andultrastructural studies have revealed a host of cell types.These discoveries, combined with the dynamic and toti-potent nature of sponge cells, makes succinct classifica-tion of their cells difficult. We present below an abbrevi-ated version of Bergquist’s (1978) cell classification.

Cells that line surfaces. Pinacoderm forms a continuouslayer on the external surface of sponges and also lines allincurrent and excurrent canals. The pinacocytes thatmake up this layer are usually flattened and often over-lapping (Figure 6.7A,B). Internal, canal-lining pinaco-cytes (endopinacocytes) are usually more fusiform inshape and have less overlap than outer exopinacocytes.Furthermore, ciliated endopinacoderm occurs in thelarge excurrent canals of some leuconoid sponges. Al-though the endopinacoderm is “epithelial” in function,and probably phagocytic as well, the apparent absenceof a basal membrane distinguishes sponge pinacodermfrom the true tissue epithelia of the higher Metazoa.*External cells of the basal or attaching region of a spongesurface are called basopinacocytes. These flattened, T-shaped cells are responsible for secreting a fibrillar colla-gen–polysaccharide complex called the basal lamina,which is the actual attachment structure. In freshwatersponges, the basopinacocytes are active in feeding andextend ameba-like “filopodia” to engulf bacteria. Fresh-water sponge basopinacocytes also play an active role inosmoregulation and contain large numbers of water ex-pulsion vesicles, or contractile vacuoles.

Porocytes are cylindrical, tubelike cells of the pinaco-derm that form the ostia (Figure 6.7C,D). They are con-tractile and can open and close the pore and regulate theostial diameter; however, no microfilaments have beenobserved in them and their precise method of contrac-tion and expansion is unknown. Some can produceacross the ostial opening a diaphragm-like cytoplasmicmembrane that also regulates pore size.

Choanocytes are the flagellated cells that make upthe choanoderm and create the currents that drive waterthrough the aquiferous system (Figure 6.7F–H).Choanocytes are not coordinated in their beating, not

even within a given chamber. However, they are alignedsuch that the flagella are directed toward the apopyleand beat from base to tip. Water is thus drawn into thechamber through the prosopyles, driven across thechoanoderm, and then out the apopyle into the atriumor an excurrent canal. The long flagellum is always sur-rounded by a so-called collar, which is made up of 20 to55 cytoplasmic microvilli (= villi). The villi have micro-filament cores and are connected to one another byanastomosing mucous strands (a mucous reticulum).Choanocytes rest on the mesohyl, held in place by inter-digitation of adjacent basal surfaces. In keeping withtheir central role in phagocytosis and pinocytosis, cho-anocytes are highly vacuolated.

Cells that secrete the skeleton. There are several typesof ameboid cells in the mesohyl, some of which secretethe various elements of sponge skeletons. In almost allsponges, the entire supportive matrix is built on aframework of fibrillar collagen. The cells that secretethis material are called collencytes, lophocytes, andspongocytes. Collencytes are morphologically nearlyindistinguishable from pinacocytes, whereas lopho-cytes are large, highly motile cells that can be recog-nized by a collagen tail they typically trail behind them(Figure 6.8C). The primary function of both cell types isto secrete the dispersed fibrillar collagen found intercel-lularly in virtually all sponges. Spongocytes producethe fibrous supportive collagen referred to as spongin(Figure 6.10A). Spongocytes operate in groups and arealways found wrapped around a spicule or sponginfiber (Figure 6.8D).

Sclerocytes are responsible for the production of cal-careous and siliceous sponge spicules (Figure 6.8A,B).They are active cells that possess abundant mitochon-dria, cytoplasmic microfilaments, and small vacuoles.Numerous types of sclerocytes have been described;these cells always disintegrate after spicule secretion iscomplete.

Contractile cells. Contractile cells in sponges, called my-ocytes, are found in the mesohyl (Figure 6.7E). They are

188 CHAPTER SIX

*Recent work suggests that a basal membrane may be present inthe Homoscleromorpha.

Figure 6.7 Cells that line sponge surfaces. (A) A pinaco-cyte from the surface of the demosponge Halisarca (drawnfrom an electron micrograph). The outer surface is coveredwith a polysaccharide-rich coat. The cell is fusiform andoverlaps adjacent pinacocytes. (B) Pinacoderm from a cal-careous sponge (section). T-shaped pinacocytes alternatewith fusiform pinacocytes. (C,D) A porocyte from the cal-careous asconoid sponge Leucosolenia. (C) Cross section.(D) Side view. (E) Myocytes surrounding a prosopyle. (F) Asection of choanoderm, showing three choanocytes;arrows indicate direction of water current. (G) A choano-cyte. (H) Ultrastructure of a choanocyte (longitudinal sec-tion, drawn from an electron micrograph). (I) A choano-cyte chamber opening into an excurrent canal in a demo-sponge.


PHYLUM PORIFERA: THE SPONGES 189UNCORRECTED PAGE PROOFSPinacocyte

Pinacocytes

usually fusiform and grouped concentrically around os-cula and major canals. Myocytes are distinguished by thegreat numbers of microtubules and microfilaments con-tained in their cytoplasm. Because of the nature of theirfilament arrangement, it has been suggested that my-ocytes are homologous with the smooth muscle cells ofhigher invertebrates. Myocytes are independent effectorswith a slow response time, and, unlike neurons and truemuscle fibers, they are insensitive to electrical stimuli.

Some other cell types. Archaeocytes are ameboid cellsthat are capable of differentiating and giving rise to vir-tually any other cell type. Archaeocytes are large, highlymotile cells that play a major role in digestion and food

transport (Figure 6.9). These cells possess a variety of di-gestive enzymes (e.g., acid phosphatase, protease, amy-lase, lipase) and can accept phagocytized material fromthe choanocytes. They also phagocytize material direct-ly through the pinacoderm of water canals. As the prin-cipal macrophage of a sponge, archaeocytes carry outmuch of the digestive, transport, and excretory activi-ties. As cells of maximum totipotency, archaeocytes areessential to the developmental program of sponges andto various asexual processes (e.g., gemmule formation).

Spherulous cells are large mesohyl cells containingvarious chemical inclusions. These cells often containthe secondary metabolites which are so abundant insponges.

190 CHAPTER SIX

Figure 6.8 Cells that secrete the sponge skeleton. (A)The formation of a triaxon calcareous spicule: (a) sclero-cytes associate to form a triad of three founder cells; (b)nuclear division in each founder cell produces central andperipheral nuclei; (c) the calcite ray is secreted betweeneach pair of nuclei, as thickener cells resulting from thenuclear division gradually move outward along the rays;(d) as spicule formation draws to a close, the founder cellsalso migrate along the rays toward the tips. (B) A sclero-cyte of Mycale (Demospongiae) with a rudimentarysiliceous spicule extending between two vacuoles (drawnfrom an electron micrograph). (C) A lophocyte with its tailof collagen fibers. (D) Spongocytes work in series tosecrete collagen fibrils in a demosponge.


Several other cell types have been identified insponges, but most of these have been characterized onlymorphologically and their functions remain unknown.

Cell AggregationAround the turn of the twentieth century, H. V. Wilsonfirst demonstrated the remarkable ability of sponge cellsto reaggregate after being mechanically dissociated. Al-though this discovery was interesting in itself, lendinginsight into the plasticity and cellular organization ofsponges, it also foreshadowed more far-reaching cyto-logical research. Recent studies on sponges have shedlight on the basic questions of how cells adhere, segre-gate, and specialize. Many sponges that are dissociatedand maintained under proper conditions will form ag-gregates, and some will eventually reconstitute theiraquiferous system. For example, when pieces of the

Atlantic “red beard” sponge(Microciona prolifera) are pressedthrough fine cloth, the separat-ed cells immediately begin toreorganize themselves by ac-tive cell migration. Within 2 to3 weeks, a functional spongere-forms and the original cellsreturn to their respective func-

tions. Furthermore, if cell suspensions of two differentsponge species are mixed, the cells sort themselves outand reconstitute individuals of each separate species—evidence of the ability of self recognition.

SupportThe skeletal elements of sponges are of two types, or-ganic and inorganic. The former is always collagenousand the latter either siliceous (hydrated silicon dioxide)or calcareous (calcium carbonate in the form of calcite oraragonite). Sponges are the only animals that use hydrat-ed silica as a skeletal material.

Collagen is the major structural protein in inverte-brates; it is found in virtually all metazoan connectivetissues. In sponges, it is either dispersed as thin fibrils inthe intercellular matrix or organized as a fibrous frame-work called spongin in the mesohyl. True spongin isfound only in members of the class Demospongiae; dis-persed collagen fibrils are found in all sponges. Theamount of this fibrillar collagen varies greatly fromspecies to species. In hexactinellids it is quite sparse,whereas in demosponges it is abundant and may formdense bands in the cortex.

Traditionally, the sponge organic skeleton has beentermed spongin. This term, however, should be restrict-ed to the form of collagen that constitutes a distinct orga-nized network in the mesohyl of demosponges (Figure6.10A). The network often contains very thick fibers, andmay incorporate siliceous spicules into its structure.Spongin often cements siliceous spicules together at theirpoints of intersection. The encysting coat of the asexualgemmules of freshwater (and some marine) sponges isalso composed largely of spongin.

Mineral skeletons of silica or calcium are found in al-most all sponges, except certain members of the classDemospongiae. Several demosponge genera lack bothspongin and a spicule skeleton (e.g., Chondrosia, Eu-spongia, Halisarca, Oscarella). Sponges lacking mineral

PHYLUM PORIFERA: THE SPONGES 191UNCORRECTED PAGE PROOFSFigure 6.9 Archaeocytes. (A) A typical archaeocyte with alarge nucleus and a prominentnucleolus. (B) Photo of a typicalarchaeocyte. (C) An archaeocyteengages in phagocytosis.

(B)

(C)

Archaeocyte Nucleus of archaeocyte

skeletons possess only fibrous colla-gen networks. They are still used asbath sponges, despite the prevalencenowadays of synthetic “sponges.”

Sponges have been harvested formillennia; Homer and other ancientGreek writers mention an activeMediterranean sponge trade. Prior tothe 1950s, an active natural spongefishery thrived in south Florida, theBahamas, and the Mediterranean. Theindustry peaked in 1938, when theworld’s annual sponge catch (includ-ing cultivated sponges) exceeded 2.6million pounds, 700,000 pounds ofwhich came from the United Statesand the Bahamas. Almost all commer-cial sponges belong to the generaHippospongia and Spongia, but thesesponges have been largely “fishedout” in the traditional sponge huntinggrounds of the Mediterranean andFlorida.

Sponge spicules (Figure 6.10) areproduced by special mesohyl cellscalled sclerocytes, which are capableof accumulating calcium or silicateand depositing it in an organizedway. In some cases, one sclerocyteproduces one spicule; in others, sever-al sclerocytes work together to pro-duce a single spicule, often two cells per spicule ray(Figure 6.8A–D). The construction of a siliceous spiculebegins with the secretion of an organic axial filamentwithin an elongated vacuole in a sclerocyte. As the axialfilament elongates at both ends, hydrated silica is secret-ed into the vacuole and deposited around the filament.Unlike siliceous spicules, calcium carbonate spicules donot have an organic axial structure. Calcareous spiculesare produced extracellularly, in intercellular spacesbounded by a number of sclerocytes. Each spicule is es-sentially a single crystal of calcite or aragonite.

Considerable taxonomic weight has been given tospicule morphology, and an elaborate nomenclature ex-ists to classify these skeletal structures. According totheir morphology, spicules are termed either micro-scleres or megascleres. The former are small to minutereinforcing (or packing) spicules; the latter are largestructural spicules. The demosponges and hexactinellidshave both types; calcareous sponges often have onlymegascleres. Descriptive terms that designate the num-ber of axes in a spicule end in the suffix -axon (e.g., mon-axon, triaxon). Terms that designate the number of raysend in the suffixes -actine or -actinal (e.g., monactinal,hexactinal, tetractinal). In addition, there is a detailednomenclature specifying shape and ornamentation ofvarious spicules (Figure 6.10).

A spicular skeleton may be viewed as a supplemen-tal supporting structure. If the amount of inorganic ma-terial is increased in relation to organic material, thesponge becomes increasingly solid until the texture ap-proaches that of a rock, as it does in members of thedemosponge orders Choristida and Lithistida. In con-trast to discrete spicules, the massive calcareous skele-tons of some species (the coralline sponges and “scle-rosponges”) have a polycrystalline microstructure; theyare composed of needles (“fibers”) of either calcite oraragonite embedded in an organic fibrillar matrix. Theadvantage of incorporating organic matter into the cal-careous framework has been compared to lathe-and-plaster, or reinforced concrete. The mix of organic andinorganic materials probably yields fibrous calcites andaragonites that are less prone to fracture while also pro-ducing substances that are more easily molded by theorganism.

Nutrition, Excretion, and Gas ExchangeAlthough sponges lack the complex organs and organsystems seen in the higher Metazoa, they are neverthe-less a highly successful group of animals. Their successappears to be due largely to their cell totipotency, theaquiferous system, and the general plasticity of theirbody form.

192 CHAPTER SIX

(A)



Figure 6.10 Sponge skeletal systems. (A) Photomicrograph of the superficial der-mal spongin–fiber skeleton typical of thedemosponge family Callyspongiidae. (B)Arrangement of calcareous triaxon spiculesnear the oscular opening in Leucosolenia. (C) Arrangement of monaxon and triaxoncalcareous spicules near the oscular open-ing of Scypha. (D) Cross-section of a sim-ple syconoid calcareous sponge (atrium on right) illustrating placement of triaxonspicules. (E) Some common types ofsiliceous spicules from demosponges. (F) Some siliceous spicules from scle-rosponges. (G) Various spicule types(SEMs).

(G)

(F)

Unlike most Metazoa, nearly all sponges rely on in-tracellular digestion, and thus on phagocytosis and pino-cytosis as means of food capture. The aquiferous systemhas already been described; sponges more or less con-tinuously circulate water through their bodies, bringingwith it the microscopic food particles upon which theyfeed. They are size-selective particle feeders, and thearrangement of the aquiferous system creates a series of“sieves” of decreasing mesh size (e.g., inhalant ostia ordermal pores → canals → prosopyles → choanocyte villi→ intertentacular mucous reticulum). The upper limitof the diameter of incurrent openings is usually around50 µm, so larger particles do not enter the aquiferoussystem. A few species have larger incurrent pores,reaching diameters of 150 to 175 µm, but in most speciesthe incurrent openings range from 5 to 50 µm in diame-ter. Internal particle capture in the 2 to 5 µm range (e.g.,bacteria, small protists, unicellular algae, organic detri-tus) is by phagocytic motile archaeocytes that move tothe lining of the incurrent canals. Then, as water passesover the choanoderm, eddies are formed around thechoanocyte collars. This water passes between the villi,into the collar, and is driven out the collar opening.Particles in the 0.1 to 1.5 µm range (e.g., bacteria, largefree organic molecules) are trapped in the mucous retic-ulum between the collar villi. The distance between ad-jacent villi is consistently 0.1 to 0.2 µm . Undulations ofthe collar move the trapped food particles down to thechoanocyte cell body, where they are ingested byphagocytosis or pinocytosis.

In the case of archaeocyte phagocytosis, digestiontakes place in the food vacuole formed at the time ofcapture. In the case of choanocyte capture, food particlesare partly digested in the choanocytes and then quicklypassed on to a mesohyl archaeocyte (or other wander-ing amebocyte) for final digestion. In both cases, the mo-bility of the mesohyl cells assures transport of nutrientsthroughout the sponge body.

The efficiency of food capture and digestion was dra-matically shown in a study by Schmidt (1970) using flu-orescence-tagged bacteria fed to the freshwater spongeEphydatia fluviatilis. By monitoring the movement of thefluorescent material, Schmidt determined that 30 min-utes elapsed from the onset of feeding until the bacteriahad been captured by choanocytes and moved to thebase of the cells. Transfer of the fluorescent material tothe mesohyl commenced 30 minutes later. Twenty-fourhours later, fluorescent wastes began to be dischargedinto the water, and no fluorescent material remained inthe sponges after 48 hours. Additional studies on thissame species led to an estimate of 7,600 choanocytechambers per cubic millimeter of sponge body, eachchamber pumping approximately 1,200 times its ownvolume of water daily. More complex leuconoidsponges have as many as 18,000 choanocyte chambersper cubic millimeter. In some thin-walled asconoid andsimple syconoid sponges, a distinctive mesohyl is hard-

ly present. In these sponges the choanocytes assumeboth capture and digestive/assimilative functions.

Sponges also take up significant amounts of dissolvedorganic matter (DOM) by pinocytosis from the waterwithin the aquiferous system. Studies by Reiswig in the1970s on Jamaican sponges showed that 80 percent ofthe organic matter taken in by these sponges was of asize below that resolvable by light microscopy. Theother 20 percent comprised primarily bacteria and dino-flagellates.

Recent studies show that at least some sponges formsimple fecal pellets. Experiments on the cosmopolitanspecies Halichondria panicea have revealed that undigest-ed material is expelled as discrete capsules coated witha thin layer of mucus.

Although the phylum Porifera is characterized by fil-ter feeding, members of the demosponge family Clado-rhizidae display an entirely different and unique modeof feeding. Species in this group have lost the character-istic choanocyte-lined aquiferous system and insteadfeed as macrophagous carnivores! They do so by trap-ping small prey on hook-shaped spicules that protrudefrom the surfaces of tentacle-like structures (Figure 6.11).Trapped prey are gradually enveloped by migratingfeeding cells that accomplish digestion and absorption.Although most cladorhizids live at great depths, onespecies of Asbestopluma lives in shallow caves in theMediterranean, where it has been the subject of consid-erable study (Vacelet and Boury-Esnault 1995). Anotherof the remarkable cladorhizid sponges, an undescribedspecies of Cladorhiza, has been discovered to harbormethanotrophic bacterial symbionts in its cells, such asseen in certain animals inhabiting hydrothermal ventsand cold seeps. The sponge thus feeds both by preda-tion and by direct consumption of its microbial sym-bionts (Vacelet et al. 1998).

Excretion (primarily ammonia) and gas exchange areby simple diffusion, much of which occurs across thechoanoderm. We have already seen how folding of thebody, combined with the presence of an aquiferous sys-tem, overcomes the surface-to-volume dilemma posedby an increase in size. The efficiency of the poriferanbauplan is such that diffusion distances never exceedabout 1.0 mm, the distance at which gas exchange bydiffusion becomes inefficient. In addition, water expul-sion vesicles (contractile vacuoles) occur in freshwatersponges and presumably aid in osmoregulation.

Activity and SensitivityThere is no conclusive evidence that sponges possessneurons or discrete sense organs. Furthermore, actionpotentials have never been recorded in sponges, andnothing resembling the synaptic connections of higherMetazoa are known in these animals. However, they arecapable of responding to a variety of environmentalstimuli by closure of the ostia or oscula, canal constric-tion, backflow, and reconstruction of flagellated cham-



Figure 6.11 These remarkable SEMs and color pho-tographs show predation in the carnivorous spongeAsbestopluma. (A–D) Asbestopluma in ambush posture(A), followed by capture of a mysid. (E) Fifteen minutes

after capture of a mysid on its tentacle-like feeding fila-ments. (F–H) The mysid prey has been partly engulfed bythe sponge. (I) The prey is entirely engulfed.

(E) (F)

(G)

(H)

(I)

(A) (B) (C)

(D)

bers. The usual effect of most of these actions is to re-duce or stop the flow of water through the aquiferoussystem. For example, when suspended particulates be-come too large or too concentrated, sponges typicallyrespond by closing the incurrent openings and immobi-lizing the choanocyte flagella. Direct physical stimula-tion will also elicit this reaction, which is easily ob-served by simply running one’s finger across a spongesurface and observing the dermal pore or oscular con-tractions with a hand lens or low-power microscope.

Activity also varies with certain endogenous factors.For example, during a major growth phase, such ascanal or chamber reorganization, activity levels typical-ly fall and pumping rates drop. Periods of reproductiveactivity also cause a substantial decrease in water pump-ing, because many choanocytes are expended in the re-production process (see the next section). Even undernormal conditions, variations in pumping rates occur.Important studies by Reiswig (see references) on Carib-bean sponges documented a number of endogenous ac-tivity patterns. Some sponges cease pumping activity pe-riodically, for a few minutes or for hours at a time; otherscease activity for several days at a time.

The switch from full pumping activity to completecessation requires at least several minutes; consideringthe organism, however, this is a fairly short responsetime. The spread of stimulation and response in spongesappears to be by simple mechanical stimulation fromone cell to the adjacent cells, and perhaps also by diffu-sion of certain chemical messengers associated with theirritability of cytoplasm in general. The contractile my-ocytes of sponges act as independent effectors; they areorganized into a network formed by contacts betweenfilopodial extensions of adjacent myocytes and pinaco-cytes. Response time of myocytes is relatively slow.Latency periods average 0.01 to 0.04 seconds, and con-duction velocities are typically less than 0.04 cm/sec (ex-cept in the hexactinellids, where velocities of 0.30 cm/sechave been recorded). Conduction is always unpolarizedand diffuse. Considerable research once focused on themyocytes in attempts to shed light on the possible pres-ence of a sponge nervous system analogous or homolo-gous to that in higher Metazoa. But, in spite of these ef-forts, there has been no verification of such a system.

One study on sponge activity hypothesized a diffuseconduction system in the hexactinellid sponge Rhab-docalyptus (Lawn et al. 1981). Both mechanical and elec-trical stimulation elicited a diffuse all-or-none responsewherein pumping activity ceased within 20 to 50 sec-onds. Conduction velocities of 0.17 to 0.30 cm/sec wereestimated. Although the authors agreed that this is tooslow for a true neuronal system, they felt it was too fastfor conduction by simple chemical diffusion.

Reproduction and DevelopmentAll sponges appear to be capable of sexual reproduc-tion, and several types of asexual processes are also

common. Many of the details of these processes are un-known, however, largely because sponges lack distinctor localized gonads (gametes and embryos occurthroughout the mesohyl). Furthermore, within any spe-cies and population, there is a marked asynchronyamong individuals in terms of reproductive activity; atany given moment, reproductive activity may be takingplace in only a small number of individuals in any area.

Asexual reproduction. Probably all sponges are capa-ble of regenerating viable adults from fragments. Somebranching species “pinch off” branch ends by a processof cellular reorganization. The dislocated pieces fall offand regenerate into new individuals. This regenerativeability used to be used by Florida commercial spongefarmers, who propagated their sponges by attaching“cuttings” to submerged cement blocks. Additionalasexual processes of poriferans include formation ofgemmules and reduction bodies, budding, and possiblyformation of asexual larvae.

In freshwater sponges of the family Spongillidae,small spherical structures called gemmules are pro-duced at the onset of winter (Figure 6.12). These dor-mant overwintering bodies are invested with a thickcollagenous coat in which supportive siliceous mi-croscleres are embedded. Gemmules are highly resistantto both freezing and drying. The gemmules of somespecies can withstand exposure to –70°C for up to anhour, while others experience mass mortality at –10°C(Ungemach et al. 1997).

The formation and eventual growth of gemmules areremarkable examples of poriferan cell totipotency. Aswinter approaches, archaeocytes aggregate in the meso-hyl and undergo rapid mitosis. “Nurse cells” calledtrophocytes stream to the archaeocyte mass and are en-gulfed by phagocytosis. The result is a mass of archaeo-cytes containing food reserves stored in elaboratevitelline platelets. This entire mass eventually becomessurrounded by a three-layered spongin covering. Dev-eloping amphidisc spicules are transported by their par-ent cells to the growing gemmule and incorporated intothe spongin envelope. The final bit of the gemmule to beenclosed by the spongin case is covered only by a singlelayer of spongin that is devoid of spicules; this single-layered patch is the micropyle. Thus formed, hiber-nation of the gemmule commences, while the parentsponge usually dies and disintegrates.

When environmental conditions are again favorable,the micropyle opens and the first archaeocytes begin toflow out (Figure 6.12C). They immediately flow over thegemmule and onto the substratum, whereupon theybegin to construct a framework of new pinacoderm andchoanoderm. The second wave of archaeocytes to leavethe gemmule colonizes this framework. In the course ofgemmule “hatching,” archaeocytes give rise to every celltype of the adult sponge. Gemmule dormancy appearsto be of two types, a quiescence and a true diapause.


Quiescence is imposed by generally unfavorable condi-tions, including low temperatures, and ends when suit-able conditions return. Diapause, on the other hand, isimposed by a combination of endogenous mechanismsand adverse environmental conditions. The breaking ofa diapause state typically requires exposure to very lowtemperatures for a prescribed number of days.

No other sponge group produces gemmules as com-plex as those of the Spongillidae. However, many ma-rine species produce asexual reproductive bodies (calledreduction bodies) that are roughly similar to freshwatergemmules but incorporate a variety of amebocytes andhave a less complex wall structure.

Many marine sponges produce buds of varioustypes. They appear as squat or elongate club-shapedprotrusions arising on the sponge surface. The buds fallfrom the parent sponge surface and may be carriedabout by water currents for a brief period before theyadhere to the substratum to form a new individual.Some members of the family Clionidae produce uniquearmored buds that are rich in stored foods and can driftin the plankton for extended periods of time.

Some sponges are reported to be capable of produc-ing larvae by asexual means. This little-studied and con-troversial process has been suggested as a means of as-suring production of a free dispersal stage even whenfertilization has failed.

Sexual processes. Most sponges are hermaphroditic,but they produce eggs and sperm at different times.This sequential hermaphroditism may take the form ofprotogyny or protandry, and the sex change may occuronly once, or an individual may repeatedly alternatebetween male and female. In some species individualsappear to be permanently male or female. In still otherspecies, some individuals are permanently gonochoris-tic, whereas some in the same population are hermaph-

roditic. In all cases, cross-fertilization is probably therule.

Sperm appear to arise primarily from choanocytes;eggs arise from choanocytes or archaeocytes. Sper-matogenesis usually occurs in distinct spermatic cysts(= sperm follicles), which form either when all the cellsof a choanocyte chamber are transformed into sper-matogonia or when transformed choanocytes migrateinto the mesohyl and aggregate there (Figure 6.13A).Little is known about oogenesis, although available in-formation suggests that solitary oocytes develop withincysts surrounded by a layer of follicle cells and nursecells (trophocytes). Meiosis commences after an oogoni-um has accumulated a sufficient quantity of food re-serves, presumably supplied by feeding on the tropho-cytes (Figure 6.13B).

There is only one, rather brief, account of the em-bryogeny of a hexactinellid (Okada 1928). Therefore,our discussion is restricted to generalities about theDemospongiae and Calcarea. Mature sperm and oocytesare released into the environment through the aquifer-


Figure 6.12 (A) Reduction bodies forming in a marinesponge. (B) A gemmule (in section) of a freshwater sponge(Spongillidae). (C) A gemmule (in section) of the freshwa-ter sponge Spongilla in the process of hatching.

ous system. The rapid release of sperm from sponge os-cula is dramatic, and such individuals are often referredto as “smoking sponges” (Figure 6.13C). Sperm releasemay be synchronized in a local population or restrictedto certain individuals. Fertilization usually takes placein the water (ovipary) with subsequent planktonic lar-vae. However, some sponges practice vivipary, and inthese species sperm are taken into the aquiferous sys-tem of neighboring oocyte-containing individuals. Theymust then cross the cellular barrier of the choanoderm,enter the mesohyl, locate the oocytes, penetrate the fol-licular barrier, and finally fertilize the egg. In at leastsome species, this impressive feat involves sperm cap-ture by choanocytes and enclosure in an intracellularvesicle (somewhat like the formation of a food vacuoleduring feeding). The choanocyte then loses its collar and

flagellum and migrates through the mesohyl as an ame-boid cell, transporting the sperm to the oocyte (Figure6.14). The migratory choanocyte is called a carrier cell,or transfer choanocyte. Choanocytes no doubt regular-ly consume and digest the unlucky sperm of differentspecies of sponges and other benthic invertebrates but,by some as yet undiscovered recognition mechanism,they respond with a remarkably different behavior tosperm of their own kind.

In viviparous species, embryos are typically releasedas mature swimming larvae. Release of the larva isthrough either the excurrent plumbing of the aquiferoussystem or a rupture in the parent’s body wall. Larvaemay settle directly, they may swim about for severalhours or a few days before settling, or they may simplycrawl about the substratum until ready to attach. In allknown cases, the larvae are lecithotrophic. In general,littoral sponges tend to produce planktonic larvae,whereas subtidal species’ larvae tend to settle directly ormove about on the ocean floor for a few days before be-ginning growth into a new adult individual.

198 CHAPTER SIX

Figure 6.13 Sexual reproduction in sponges. (A) Spermfollicle (in section) containing mature spermatozoa. (B) Anoocyte (in section) of Ephydatia fluviatilis (Demospongiae)is phagocytizing a trophocyte. Inside the oocyte is a tro-phocyte that was recently ingested. (C) Sperm releasefrom a tubular West Indian sponge, Aplysina archeri (Demo-spongiae). The sponge is about 1.5 m tall. (D) Oocyterelease in the sponge Agelas (Demospongiae). The individ-ual in the foreground is covered by cords of yellow mucusthat surround the oocytes during their early development;two specimens in the center show no sign of oocyte re-lease.

(C)

(D)


Three basic larval types have been described insponges: “coeloblastula” larvae (= “blastula” larvae),parenchymula larvae (= parenchymella larvae), andamphiblastula larvae. Most demosponges incubate em-bryos until a late stage, producing a solid parenchymu-la larva with an outer surface of monoflagellated cellsand an inner mesohyl-like core of matrix and cells(Figure 6.15). Parenchymula larvae have a short plank-tonic life, usually just a few days. During this swim-ming phase the larvae of at least some species canchange shape rapidly from elongate to ovoid to flat.

Following settling, the external flagellated cells disap-pear and flagellated choanocytes appear internally, aschoanoderm. This process has long been attributed to aunique embryological inversion process wherein exter-nal cells drop their flagella, migrate to the inner celllayer, then re-form the flagella. However, recent workhas challenged the existence of this inversion processand suggests that the external flagellated cells are sim-ply shed or phagocytized during larval metamorphosis,the internal choanocytes subsequently forming anewfrom archaeocytes. In any case, the result of this postset-tlement metamorphosis is a tiny leuconoid form called arhagon.

Calcareous sponges (and a few demosponges) oftenrelease their embryos early, as free-swimming “coelo-blastula” larvae (Figure 6.16A). These larvae may under-go one of two developmental processes. In the simplest

case, transformation of the larva in-volves an inward migration of surfacecells that have lost their flagella; thesesame cells subsequently regain theirflagella as they metamorphose intochoanocytes.

A more complex embryonic devel-opment produces two distinct cell types


Figure 6.14 Fertilization in the calcareous spongeGrantia. (A) Sperm are trapped by choanocytes; an egg islying in the mesohyl adjacent to the choanoderm. (B) Atransfer choanocyte gives up its sperm to the egg; notethat the egg lies next to the choanoderm and that thechoanocyte has lost its flagellum.

Figure 6.15 Parenchymula larvaeof various demosponges. (A) Larvaof Clathrina. (B) Larva of Spongia. (C) Larva of Tethya. (D) Larva ofLissodendoryx isodictyalis.

(D)

resembling the macromeres and micromeres of sometrue Metazoa. At the 16-cell stage, eight large round cells(“macromeres”) rest at one pole, and eight smaller cells(“micromeres”) form most of the hollow embryo. Thelarger cells are destined to be future pinacoderm andmesohyl, and the smaller cells become the choanoderm.The “micromeres” divide rapidly and develop flagellathat extend into the embryo’s cavity. The “macromeres”remain undivided for some time and never develop fla-gella; in the center of the “macromere” cell cluster is apore to the outside. This stage is called the stomoblastu-la. While still within the mesohyl of the adult sponge,the stomoblastula ingests nutrient-rich amebocytes. Asdevelopment proceeds, a remarkable process of inver-sion takes place, in which the stomoblastula turns insideout through the pore, moving the flagella from the in-side to the outside and producing a hollow, flagellated,amphiblastula larva (Figure 6.16B–D). This larva subse-quently is released from the parent sponge. There is noknown counterpart to this process in any other spongegroup, or in the higher Metazoa.

The initiation of settlement and metamorphosis insponges is poorly understood, especially since they ap-parently lack formal sensory receptors and neurons atany life stage. Recent work by Woollacott and Hadfield

(1996) shows that certain chemicals (KCl and CsCl) in-duce metamorphosis in the larvae of the demospongeAplysilla, but the mechanism of this phenomenon is stilla mystery.

After a free-swimming period, the amphiblastulalarva settles on its flagellated end. Metamorphosis in-volves a rapid proliferation of the “macromeres” to formpinacoderm that overgrows the flagellated hemisphere.The flagellated cells pocket inward to form a chamberlined with cells destined to become choanocytes (Figure6.16F). An osculum breaks through, and the tiny as-conoid-like sponge becomes capable of circulating waterand feeding. This initial functional stage is called anolynthus (Figure 6.4A). After further growth, it will be-come an asconoid, syconoid, or leuconoid adult.

The preceding account of development and larvaltypes is drastically simplified. In fact, sponges show

200 CHAPTER SIX

Figure 6.16 “Coeloblastula” and amphiblastula larvae(in section). (A) Typical “coeloblastula” larva with its poste-rior “macromeres.” (B–D) During the remarkable processof inversion in Scypha, the stomoblastula turns itself insideout to form an amphiblastula larva with externally directedflagella. (E) A typical amphiblastula larva (Scypha). (F) Set-tled young sponge (Scypha) after invagination of flagellat-ed cells.


more variation in embryological development thanmany other animal groups. We recommend Bergquist(1978) as a good starting place if you wish to learn moredetails.

Some Additional Aspects of Sponge BiologySome basic sponge ecology has been presented in theprevious sections of this chapter. However, becausesponges play such important roles in so many marinehabitats, we add here some special aspects of their nat-ural history.

Distribution and EcologyCertain distributional patterns are evident among thethree classes of sponges. Calcareous sponges (andcoralline demosponges) are far more abundant in shal-low waters (less than 200 m), although they are not un-common at slope depths, and a few species (particularlyScypha) have even been reported from depths to 3,800 m.Hexactinellids, which were common in shallow seas ofpast eras, are now largely restricted to depths below 200m, except in extremely cold environments (such asAntarctica), where they occur in shallow waters. Thedemosponges live at all depths. Calcareous sponges areprobably restricted largely to shallow waters becausethey require a firm substratum for attachment. On theother hand, many demosponges and hexactinellidsgrow on soft sediments, attaching by means of rootlikespicule tufts or mats. The coralline sponges, once a pre-dominant group on shallow tropical reefs, are nowlargely restricted to shaded crevices and caves, or tosubreef depths, where their potential competitors (thehermatypic corals) cannot grow. They are thought to berelicts of major reef-constructing groups of Mesozoicand Paleozoic seas.

Although sponges are very sensitive to suspendedsediment in their environment, they seem to be quite re-sistant to hydrocarbon and heavy metal contamination.Many species can actually accumulate these contami-nants without apparent harm. The capacity of certainspecies to accumulate metals at far higher levels thanthat of the environment has been suggested as a possi-ble defense mechanism (antipredation, antifouling).Detergents also do not appear to affect many sponges,and in fact may even serve as a source of nutrition forthese amazingly adaptable animals.

Sponges are the dominant animals in a great manybenthic marine habitats. Most rocky littoral regions har-bor enormous numbers of sponges, and recent work in-dicates that they even occur in large numbers (and larg-er size) around Antarctica. Although many animals preyon sponges, the amount of serious damage they do isusually slight. Some tropical fishes and turtles crop cer-tain kinds of sponges, and small predators (mainly

opisthobranchs) consume limited amounts of sponge“tissue” in both warm and temperate seas. Overall,however, sponges appear to be very stable and long-lived animals, probably in part due to their spicules andtoxic and/or distasteful compounds that discourage po-tential predators.

Biochemical AgentsEven a casual seashore explorer or SCUBA diver willquickly notice that sponges are just about everywhere.Most grow on open rock or occasionally sand/mud sur-faces, where they are obviously exposed to potentialpredation. Clearly, some mechanism(s) must be work-ing to prevent these animals from being cropped exces-sively by predators. The primary defense mechanismsin sponges are mechanical (skeletal structures) and bio-chemical. Studies over the past two decades show thatsponges manufacture a surprisingly broad spectrum ofbiotoxins, some of which are quite potent. A few, such asTedania and Neofibularia, can cause painful skin rashes inhumans.

Research in sponge biochemistry has also revealedthe widespread occurrence of antimicrobial agents insponges. Sponges appear to use “chemical warfare” notonly to reduce predation and prevent infection by mi-crobes, but also to compete for space with other sessileinvertebrates such as ectoprocts, ascidians, and evenother sponges. Different species have evolved chemicals(allelochemicals) that may be species-specific deter-rents or actually lethal weapons for use against compet-ing sessile and encrusting organisms. For example, thecoral-inhabiting sponge Siphonodictyon releases a toxicchemical into the mucus exuded from its oscula, thuspreventing potential crowding by maintaining a zone ofdead coral polyps around each osculum (Figure 6.17).


Figure 6.17 Siphonodictyon coralliphagum infests thehermatypic coral Montastrea cavernosa on a Caribbeanreef. Note the “dead zone” between the oscular chimneysof the sponge and the coral polyps.

Many of the chemicals produced by sponges andother marine invertebrates are being closely studied bynatural products chemists and biologists interested intheir potential as pharmaceutical agents. Compoundswith respiratory, cardiovascular, gastrointestinal, anti-inflammatory, antitumor, and antibiotic activities havebeen identified from many marine sponges. One NewZealand sponge (Halichondria moorei) has long beenused by native Maoris to promote wound healing andwas recently discovered to contain remarkably highconcentrations (10 percent of the sponge dry weight) ofthe potent anti-inflammatory agent potassium fluorosil-icate. Sponge antimicrobial compounds are also of po-tential use to humans. For example, a compound that isactive against the herpes virus (belonging to a class ofchemicals called arabinosides) has been found in thetropical sponge Cryptotethya crypta. Some sponges, including the west Pacific species Luffariella variabilis,produce a remarkable terpenoid compound calledmanoalide that is not only an extremely powerful an-tibacterial compound but also acts as both an analgesicand anti-inflammatory agent. One study (Bergquist andBedford 1978) found that 87 percent of the temperatesponges, and 58 percent of the tropical species exam-ined in New Zealand produced extracts with antibacte-rial activity. Sponges of the genera Halichondria andPandaros are known to produce potent antitumor com-pounds belonging to a group of chemicals called hali-chondrins. The coming decades will undoubtedly wit-ness the emergence of many new pharmacologicalcompounds of poriferan origin.

Growth RatesLittle is known regarding growth rates in sponges, butavailable data suggest that rates vary widely amongspecies. Some species are annuals (especially small-bod-ied calcareous sponges of colder waters); hence theygrow from larvae or gemmules to reproductive adult-hood in a matter of months. Others are perennials andgrow so slowly that almost no change can be seen fromone year to the next; this growth pattern is especiallytrue of tropical and polar demosponges. Age estimatesof perennial species range from 20 to 100 years.

Some sponges are capable of very rapid growth, andthey regularly overgrow neighboring flora and fauna.For example, the tropical encrusting sponge Terpiosgrows over both living and nonliving substrata. InGuam this sponge grows at rates averaging 23 mm permonth over almost every live coral in the area as well asover hydrocorals, molluscs, and many algae. Experi-ments have shown that Terpios is toxic to living corals,and presumably to many other animals. Still anotherphysiological trick of some sponges is the ability torapidly produce copious amounts of mucus when dis-turbed. On the west coast of North America, the beauti-ful red-orange Plocamia karykina covers itself with a thick

layer of mucus when injured or disturbed. Yet the littlered sea slug Rostanga pulchra has evolved the ability tolive and feed inconspicuously on this and other sponges,and even lays its camouflaged red egg masses on thesponge’s exposed surface without eliciting the mucousreaction.

SymbiosesCommensalism is common among sponges of all kinds.It would be difficult to find a sponge that is not utilizedby at least some smaller invertebrates and often by fish-es (e.g., gobies and blennies) as refuge. The porous na-ture of sponges makes them ideally suited for habitationby opportunistic crustaceans, ophiuroids, and variousworms. A single specimen of Spheciospongia vespariafrom Florida was found to have over 16,000 alphaeidshrimps living in it, and a study from the Gulf of Cali-fornia found nearly 100 different species of plants andanimals in a 15 × 15-cm piece of Geodia mesotriaena.

Most symbionts of sponges use their hosts only forspace and protection, but some rely on the sponge’swater current for a supply of suspended food particles.A classic example of this phenomenon is the male–fe-male pair of shrimp (Spongicola) that inhabit hexactinel-lid sponges known as Venus’s flower basket (Euplectella;Figure 6.1E). The shrimp enter the sponge when theyare young, only to become trapped in their host’s glass-like case as they grow too large to escape. Here theyspend their lives as “prisoners of love.” Appropriately,this sponge (with its guests) is a traditional wedding giftin Japan—a symbol of the lifetime bond between twopartners.

Other even more intimate symbiotic relationshipswith sponges are common. Some snails and clams char-acteristically have specific sponges encrusting theirshells, and many species of crabs (hermits and brachy-urans) collect certain sponges and cultivate them ontheir shell or carapace. Demosponges, such as Suberites,are commonly involved in these commensalistic rela-tionships. The sponge serves primarily as protectivecamouflage for its host, and it perhaps benefits by beingcarried about to new areas. And the sponge no doubtfeeds off small bits of animal matter dislodged duringthe feeding activities of its host.

Another spectacular example of poriferan symbiosisare certain sponge–bacteria and sponge–algae associa-tions that appear to be mutualistic. For example, a typi-cal member of the demosponge order Verongida con-tains a mesohyl bacterial population accounting forsome 38 percent of its body’s volume, far exceeding theactual sponge-cell volume of only 21 percent. Presum-ably, the sponge matrix provides a rich medium for bac-terial growth, and the host benefits by being able to con-veniently phagocytize the bacteria for food. Similarrelationships are common between poriferans and vari-ous cyanobacteria. Recent evidence suggests that some


products of normal cyanobacterial metabolism (e.g.,glycerol and certain organic phosphates) are translocat-ed directly to the sponge for nutrition. In many sponges,both regular bacteria and cyanobacteria occur, the for-mer in deeper cellular regions, the latter closer to the sur-face where light is available. In a remarkable study, C. R.Wilkinson (1983) showed that 6 of the 10 most commonsponge species on the forereef slope of Davies Reef(Great Barrier Reef) are actually net primary producers,with three times more oxygen produced by photosyn-thesis (by their symbionts) than consumed by respira-tion. In some areas of the Caribbean and Great BarrierReef, sponges are second only to corals in overall bio-mass, and they appear to owe their rapid growth to thepresence of large populations of symbiotic cyano-bacteria. Most freshwater spongillids maintain similarrelationships with zoochlorellae (symbiotic green algae;Chlorophyta). These sponges grow larger and morerapidly than specimens of the same species that are keptin dark conditions. Some marine sponges (e.g., the boring sponges Cliona and Spheciospongia) harbor com-mensal zooxanthellae similar to those of corals. Com-mensalistic relationships have also been reported be-tween sponges and red algae, filamentous green algae,and diatoms.

Not all sponge symbioses are commensal or mutual-istic. In fact, some are clearly harmful—for example, theboring demosponges that excavate complex galleries incalcareous material such as corals and mollusc shells(Figure 6.18). The phenomenon of boring, known asbioerosion, causes significant damage to commercialoysters as well as to natural coral, clam, and scalloppopulations. The active boring process involves a chem-ical and mechanical removal of fragments or chips ofthe calcareous material by specialized archaeocytes

called etching cells. The use of carbonic anhydrase hasbeen implicated in this process. The chips are expelledin the excurrent canal system and can contribute signifi-cantly to local sediments.

Sponge bioerosion has a significant impact on coralreefs. Perhaps even more important than actual erosionis the weakening of attachment regions of large corals.This action may result in much coral loss during heavytropical storms. Boring sponges do not appear to gainany direct nourishment from their host coral; rather theyuse it as a protective casing in which they reside. If youcarefully examine the shells of dead bivalves along anybeach, you will discover that many of them are perforat-ed with minute holes and galleries of boring sponges.These poriferans are responsible for a major portion ofthe initial breakdown of such calcareous structures, andthus they set the stage for their eventual decompositionand recycling through Earth’s geochemical cycle.

Poriferan Phylogeny

The Origin of Sponges

Sponges are an ancient group, and the important eventsin their origin and early evolution lie hidden in Pre-cambrian time. The unique nature of the poriferan bau-plan is clear, however, and it is strikingly revealed by itsaquiferous system, cellular totipotency and reproduc-tive flexibility, and by the lack of true tissues, reproduc-tive organs, body polarity, and basement membrane.These features, in combination with the prevalence andimportance of flagellated (monociliated) cells insponges, strongly suggest a direct protistan ancestry. Itwould seem that the poriferans share as many similari-


(A)

Figure 6.18 Boring sponges. (A) Surface of a coral (stel-late openings) infected by the sponge Cliona (circularoscula); (B) A close view (SEM) of the surface of a clam

shell, showing six eroded “chips,” two of which have beenentirely removed and four that are only partly etched byCliona.

(B)

e-fileneedsproof

ties with protists than they do with the higher Metazoa(e.g., cellular totipotency; excretory, respiratory, and os-moregulatory strategies; reliance on flagellated cells in avariety of ways, including feeding; strictly intracellulardigestion). At the same time, sponges appear to standapart from all other Metazoa in their possession of theunique aquiferous system, which represents a keysynapomorphy defining this phylum. Sponges thusconstitute a biological grade that contributes to our un-derstanding of the transition from unicellular to multi-cellular life.

Current opinion views the Porifera as originating fromflagellated protist ancestors—either a simple hollow, free-swimming colonial form or a colonial choanoflagellate.The choanoflagellates possess certain features that seem toally them strongly to sponges. For example, the collar cellsof sponges are strikingly similar to the collar cells ofchoanoflagellates. Curiously, similar collar cells have alsobeen found in some widely divergent Metazoa (certainechinoderm larvae, in the oviducts of some sea cucum-bers, and in certain corals). However, these discoveries dolittle to diminish the force of the argument for a choanofla-gellate ancestry to the sponges.

The mesohyl is generally viewed as originating (evo-lutionarily) by simple ingression of surface cells, as seenin the embryogeny of many living sponges. Adoption ofa benthic lifestyle by the earliest sponges could havefostered increased body size. Increase in size led to sur-face-to-volume problems that were overcome by theevolution of the syconoid and leuconoid bauplans, in-creasing the surface area of the choanoderm-lined areasand maintaining small diffusion distances as the aquif-erous system became more complex.

The solutions that poriferans evolved to problems ofsurvival created a group of animals unlike any other.Sponges achieved multicellularity and large body sizewithout such typically metazoan traits as embryologicaltissue layering, neuronal coordination, extracellular di-gestion, excretory structures, or fixed reproductivestructures. Taken together, these and other poriferan at-tributes suggest that the sponges arose very early inmetazoan cladogenesis.

Evolution within the PoriferaSponges are such an ancient and enigmatic phylum thattheir phylogeny has largely eluded biologists. There isno generally agreed-upon phylogenetic hypothesis ofrelationships among the classes. We do know thatsponges probably evolved during the Precambrian(Figure 6.19). Their hard skeletal components have leftgood fossil records for all three extant classes, beginningin the Cambrian and extending to the present. Well over1,000 fossil genera have been described, about 20 per-cent of which are still extant. The early Paleozoic wit-nessed the growth of massive tropical reefs composedlargely of four spongelike groups: the archaeocyathans,stromatoporoids, sphinctozoans, and chaetetids. The

oldest group, Archaeocyatha (Figure 6.20), had a shortlife within the Cambrian (550–500 mya). Sphinctozoansalso appeared in the Cambrian (about 540 mya), whilechaetetids and stromatoporoids first appeared in theOrdovician (about 480 mya). The affinities of these fourgroups have been debated for the past 100 years, andvarious alliances with cyanobacteria, red algae, ecto-procts, cnidarians, and foraminferans have been pro-posed. The discovery of living coralline sponges ledmost workers to conclude that the majority of species inthese four groups were primitive, but true, sponges.

204 CHAPTER SIX

Figure 6.19 Fossil record of the three sponge classes,the coralline sponges, and Archaeocyatha. Dashed linesindicate suggested occurrence, even though fossils havenot yet been found. “R” indicates the times when thegroup in question is known to have been an importantmarine reef builder.

Ter

tiary

Cre

tace

ou

sJu

rass

icT

ria

ssic

Per

mia

nC

arb

on

ifer

ou

sD

evo

nia

nS

ilu

ria

nO

rdo

vic

ian

Cam

bri

an

Pre

cam

bri

an

Ca

lcin

eaC

alc

aro

nea

Str

om

ato

po

roid

a

Oth

erco

rall

ine

spo

ng

es

Hex

act

inel

lid

a

Dem

osp

on

gia

e

Arc

haeo

cyath

a

R

R

R

Calcarea

Corallinesponges

R


Stromatoporoids have a long geological history, ex-tending from late Cambrian times to the present. Thenature of these reef-building invertebrates has beenhotly debated. That the fossil stromatoporoids were truesponges is most strongly suggested by the apparent ho-mology of structures called astrorhizae, found in theircalcareous skeletons, to similar stellate impressions inthe skeletons of recent coralline sponges (“sclero-sponges”). In living sponges these stellate marks are thetraces of the converging exhalent canal systems (be-

neath the oscula). The absence of siliceous spicules inthe calcareous skeletons of fossil stromatoporoids hasbeen cited as evidence against a true sponge relation-ship. However, not all recent stromatoporoids havesiliceous spicules; and in some that do, these spiculesare never incorporated into the calcareous basal skele-ton that would be fossilized anyway.

The coralline sponges have also had a long history,and many distinctive early lines have died out sincetheir origins in Cambrian and Precambrian times. In-terestingly, some of these early calcareous forms wereimportant reef builders during the Permian, but likestromatoporoids, the few surviving genera of those lineshave all retreated to marine cave habitats in moderntropical seas.

Unlike the coralline sponges, which have decreasedin abundance and diversity since the Mesozoic, the re-maining calcareous sponges and the demosponges ap-pear to have increased in diversity throughout their his-tory. Hexactinellids, on the other hand, were mostdiverse and abundant during the Cretaceous. The oldesthexactinellid fossils, of the early Cambrian, were allthin-walled, saclike sponges with a dispersed surfacespicule layer that probably could not support a thickbody wall. During the Paleozoic, hexactinellids werecommon in shallow-water environments. Since then,however, they have become restricted largely to thedeep oceans. The demosponges were well establishedby the mid-Cambrian, from which the earliest fossils areknown, and all known orders of modern demospongesare found in Cretaceous rocks. But only recently havethe complex relationships among the dozen or so ordersof Demospongiae been critically examined. The chal-lenge of unraveling the relationships in this ponderousgroup of sponges is still in an early phase.


Figure 6.20 A typical archaeocyathan. A vertical sectionhas been partly cut away to show the structure betweenthe inner and outer walls (i.e., vertical parieties and hori-zontal tabula).

Selected ReferencesGeneral ReferencesBakus, G. J. 1964. The effects of fish-grazing on invertebrate evolution

in shallow tropical waters. Occ. Pap. Allan Hancock Fndtn. 27:1–29.

Bakus, G. J. 1969. Energetics and feeding in shallow marine waters.Int. Rev. Gen. Exp. Zool. 4: 275–369.

Bakus, G. J., N. M. Targett and B. Schulte. 1986. Chemical ecology ofmarine organisms: An overview. J. Chem. Ecol. 12: 951–987.

Bayer, F. M. and H. B. Owre. 1968. The Free-Living Lower Invertebrates.Macmillan, New York.

Bergquist, P. R. 1978. Sponges. University of California Press, Berkeley.Bergquist, P. 1985. Poriferan relationships. In S. C. Morris et al. (eds.),

The Origin and Relationships of Lower Invertebrates. Syst. Assoc. Spec.Vol. 28, Oxford, pp. 14–27.

Bergquist, P. R. and J. J. Bedford. 1978. The incidence of antibacterialactivity in marine Demospongiae; systematic and geographic con-siderations. Mar. Biol. 46: 215–221.

Bergquist, P. R. and R. J. Wells. 1983. Chemotaxonomy of the Porifera:The development and current status of the field. In Marine NaturalProducts, III. Academic Press, New York, pp. 1–50.

Bond, C. 1997/1998. Keeping up with the sponges. Nat. Hist. 106(11):22–25.

Borojevic, R. 1970. Différentiation cellulaire dans l’embryogenèse et lamorphogenèse chez les spongiaires. Symp. Zool. Soc. Lond. 25:467–490.

Borojevic, R., W. G. Fry, W. C. Jones, C. Lévi, R. Rasmont, M. Sarà andJ. Vacelet. 1967. Mise au point actuelle de la terminologie desÉponges. (A reassessment of the terminology for sponges.) Bull.Mus. Hist. Nat. Paris (2) 39: 1224–1235.

Boury-Esnault, N. 1973. L’exopinacoderme des Spongiaires. Bull. Mus.Nat. D’Hist. Natur. (Zool.) 117: 1193–1206.

Boury-Esnault, N. 1977. A cell type in sponges involved in the metab-olism of glycogen. The gray cells. Cell Tiss. Res. 175: 523–539.

Boury-Esnault, N., L. deVos, C. Donadey and J. Vacelet. 1985. Ultra-structure of choanosome and sponge classification. 3rd Internat.Sponge Conf., 1985, pp. 237–244.

Bowerbank, J. S. 1861, 1862. On the anatomy and physiology of theSpongiidae. Part 1: On the spicula. Philos. Trans. R. Soc. Lond. 148:279–332. Part 2: Proc. R. Soc. Lond. 11: 372–375.

Bowerbank, J. S. 1861. On the anatomy and physiology of theSpongiidae. Philos. Trans. R. Soc. Lond. 152: 747–829, 1087–1138.

Bowerbank, J. S. 1864, 1866, 1874. A monograph of the BritishSpongiidae. Vols. 1, 2, and 3. Ray Society, London.


Brien, P. 1968. The sponges, or Porifera. In M. Florkin and B. T. Scheer(eds.), Chemical Zoology, Vol. 2. Academic Press, New York, pp.1–30.

Brien, P., C. Lévi, M. Sara, O. Tuzet and J. Vacelet. 1973. Spongiaires. InP. Grassé (ed.), Traité de Zoologie 3(1): 716. Masson et Cie, Paris.

Brill, B. 1973. Untersuchungen zur Ultastruktur der Choanocyte vonEphydatia fluviatilis, L. Z. Zellforsch. 144: 231–245.

Burkholder, P. R. 1973. The ecology of marine antibiotics and coralreefs. In O. A. Jones and R. Endean (eds.), Biology and Geology ofCoral Reefs, Vol. 1. Biology, 1. Academic Press, New York, pp.117–182.

Burkholder, P. R. and K. Ruetzler. 1969. Antimicrobial activity of somemarine sponges. Nature Lond. 222: 983–984.

Burton, M. 1932. Sponges. Discovery Rep. 6: 327–392.Carballo, J. L., J. E. Sanchez-Moyano and J. C. Garcia-Gomez. 1994.

Taxonomic and ecological remarks on boring sponges (Clionidae)from the Strait of Gibraltar: Tentative bioindicators? Zool. J. Linn.Soc. 112: 407–424.

Cox, G. and A. W. D. Larkum. 1983. A diatom apparently living insymbiosis with a sponge. Bull. Mar. Sci. 33(4): 943–945.

Curtis, A. 1979. Individuality and graft rejection in sponges: A cellularbasis for individuality in sponges. In G. Larwood and B. Rosen(eds.), Biology and Systematics of Colonial Organisms. AcademicPress, New York, pp. 39–48.

Dayton, P. K. 1979. Observations of growth, dispersal and populationdynamics of some sponges in McMurdo Sound, Antarctica. In C.Lévi and N. Boury-Esnault (eds.), Biologie des Spongiaires, CentreNat. Recherche Scient., Paris, pp. 271–282.

de Laubenfels, M. W. 1955. Porifera. In R. C. Moore (ed.), Treatise onInvertebrate Paleontology. Archaeocyatha and Porifera, E21–E112. Geo-logical Society of America and the University of Kansas Press,Lawrence.

De Vos, L., K. Rützler, N. Boury-Esnault, C. Donadey and J. Vacelet.1991 Atlas of Sponge Morphology. Smithsonian Institution Press,Washington, D.C.

Faulkner, D. J. 1973. Variabilin, an antibiotic from the sponge Irciniavariabilis. Tetrahedron Let. 29: 3821–3822.

Faulkner, D. J. 1977. Interesting aspects of marine natural productschemistry. Tetrahedron Let. 33: 1421–1443.

Fautin, D. G. (ed.). 1988. Biomedical Importance of Marine Organisms.Mem. Calif. Acad. Sci. No. 13.

Fell, P. E. 1974. Porifera. In A. C. Giese and J. S. Pearse (eds.),Reproduction of Marine Invertebrates, Vol. 1. Academic Press, NewYork, pp. 51–132.

Fell, P. E. 1976. Analysis of reproduction in sponge populations: Anoverview with specific information on the reproduction ofHaliclona loosanoffi. In F. W. Harrison and R. R. Cowden (eds.),Aspects of Sponge Biology. Academic Press, New York, pp. 51–67.

Fell, P. E. 1995. Deep diapause and the influence of low temperatureon the hatching of the gemmules of Spongilla lacustris (L.) andEunapius fragilis (Leidy). Invert. Biol. 114(1): 3–8.

Fell, P. E. 1997. Porifera: The sponges. In S. F. Gilbert and A. M. Raunio(eds.), Embryology: Constructing the Organism. Sinauer Associates,Sunderland, MA, pp. 39–54.

Finks, R. M. 1970. The evolution and ecologic history of sponges dur-ing Palaeozoic times. Symp. Zool. Soc. Lond. 25: 3–22.

Frost, T. M. 1976. Sponge feeding: A review with a discussion of con-tinuing research. In F. W. Harrison and R. R. Cowden (eds.),Aspects of Sponge Biology. Academic Press, New York, pp. 283–298.

Fry, W. G. (ed.). 1970. The Biology of the Porifera. Academic Press, NewYork.

Garrone, R. 1978. Phylogenesis of Connective Tissue. S. Karger, Basel.Garrone, R. and J. Pottu. 1973. Collagen biosynthesis in sponges:

Elaboration of spongin by spongocytes. J. Submicrosc. Cytol. 5:199–218.

Goodwin, T. W. 1968. Pigments of Porifera. In M. Florkin and B. T.Scheer (eds), Chemical Zoology, Vol. 2. Academic Press, New York,pp. 53–64.

Harrison, F. W. and R. R. Cowden (eds.). 1976. Aspects of SpongeBiology. Academic Press, New York.

Harrison, F. W. and L. De Vos. 1990. Porifera. In F. W. Harrison and J.A. Westfall (eds.), Microscopic Anatomy of Invertebrates, Vol. 2. AlanR. Liss, New York.

Hartman, W. D. 1958. Natural history of the marine sponges of south-ern New England. Bull. Peabody Mus. Nat. Hist. 12: 1–155.

Hartman, W. D. 1982. Porifera. In S P. Parker (ed.), Synopsis andClassification of Living Organisms, Vol. 1. McGraw-Hill, New York,pp. 641–666.

Hartman, W. D. and T. F. Goreau. 1975. A Pacific tabulate sponge, liv-ing representative of a new order of sclerosponges. Postilla 167:1–21.

Hartman, W. D. and H. M. Reiswig. 1973. The individuality ofsponges. In R. S. Boardman, A. H. Cheetham and W. A. Oliver(eds.), Animal Colonies. Dowden, Hutchinson & Ross, Inc.,Stroudsburg, PA, pp. 567–584.

Hartman, W. D., J. W. Wendt and F. Wiedenmayer. 1980. Living andFossil Sponges. (Compiled by R. N. Ginsburg and P. Reid.)Sedimenta VIII (Comparative Sedimentology Lab, Div. Mar. Geol.and Geophysics, University of Miami).

Hildemann, W. H., I. S. Johnson and P. L. Jobiel. 1979. Immuno-competence in the lowest metazoan phylum: Transplantation im-munity in sponges. Science 204: 420–422.

Hill, D. 1972. Archaeocyatha. In R. C. Moore (ed.), Treatise onInvertebrate Paleontology 1: 158. Geological Society of America andthe University of Kansas Press, Lawrence.

Hill, D. and E. C. Strumm. 1956. Tabulata. In R. C. Moore (ed.), Treatiseon Invertebrate Paleontology, F: 444–477. Geological Society ofAmerica and the University of Kansas Press, Lawrence.

Hooper, J. N. A., R. J. Capon, C. P. Keenan, D. L. Parry, and N. Smit.1992. Chemotaxonomy of marine sponges. Families Micro-cionidae, Raspailiidae and Axinellidae, and their relationshipswith other families in the orders Poecilosclerida and Axinellida(Porifera: Demospongiae). Invertebr. Taxon. 6: 261–01.

Humphreys, T. 1963. Chemical dissolution and in vitro reconstructionof sponge cell adhesions. I. Isolation and functional demonstrationof components involved. Dev. Biol. 8: 27–47.

Humphreys, T. 1970. Species-specific aggregation of dissociatedsponge cells. Nature 228: 685–686.

Hyman, L. H. 1940. The Invertebrates, Vol. 1, Protozoa through Cteno-phora. McGraw-Hill, New York, pp. 284–364.

Jackson, J. B. C. 1977. Competition on marine hard substances: Theadaptive significance of solitary and colonial strategies. Am. Nat.111: 743–767.

Jackson, J. B. C. and L. Buss. 1975. Allelopathy and spatial competitionamong coral reef invertebrates. Proc. Natl. Acad. Sci. USA 72:5160–5163.

Jefford, C. W., K. L. Rinehart, and L. S. Shield (eds). 1987. Pharma-ceuticals and the Sea. Technomic Publ. Co., Basel, Switzerland.

Jones, W. C. 1962. Is there a nervous system in sponges? Biol. Rev. 37:1–50.

Kaye, H. and T. Ortiz. 1981. Strain specificity in a tropical marinesponge. Mar. Biol. 63: 165–173.

Kázmierczak, J. 1984. Favositid tabulates: Evidence for poriferan affin-ity. Science 225: 835–837.

Kázmierczak, J. and S. Kempe. 1990. Modern cyanobacterial analogsof Paleozoic stromatoporoids. Science 250: 1244–1248.

Koltun, V. M. 1968. Spicules of sponges as an element of the bottomsediments of the Antarctic. In Symposium on Antarctic Ocean-ography, Scott Polar Res. Inst., Cambridge, pp. 121–123.

Kuhns, W., G. Weinbaum, R. Turner and M. Burger. 1974. Sponge cellaggregation: A model for studies on cell–cell interactions. Ann.N.Y. Acad. Sci. 234: 58–74.

LaBarbera, M. and G. E. Boyajian. 1991. The function of astrorhizae instromatoporoids: Quantitative tests. Paleobiology 17(2): 121–132.

Lang, J. D., W. D. Hartman and L. S. Land. 1975. Sclerosponges:Primary framework constructors on the Jamaican deep forereef. J.Mar. Res. 33: 223–231.

Lawn, I. D., G. O. Mackie, and G. Silver. 1981. Conduction system in asponge. Science 211: 1169–1171.


Lecompte, M. 1956. Stromatoporoidea. In R. C. Moore (ed.), Treatise onInvertebrate Paleontology, F: F107–F114. Geological Society ofAmerica and the University of Kansas Press, Lawrence.

Lévi, C. 1957. Ontogeny and systematics in sponges. Syst. Zool. 6:174–183.

Lévi, C. and N. Boury-Esnault (eds.). 1979. Sponge Biology. ColloquesInternationaux du Centre National de la Recherche Scientifique.Ed. Cen. Nat. Resch. Sci. No. 291.

Li, C.-W., J.-Y. Chen and T.-E. Hua. 1998. Precambrian sponges withcellular structures. Science 279: 879–882.

Minchin, E. A. 1900. Sponges. In E. R. Lankester (ed.), A Treatise inZoology, Pt. 2. Adam and Charles Black, London, pp. 1–178.

Neigel, J. E. and G. P. Schmahl. 1984. Phenotypic variation within his-tocompatibility-defined clones of marine sponges. Science 224:413–415.

Paine, R. T. 1964. Ash and calorie determinations of sponge andopisthobranch tissue. Ecology 45: 384–387.

Palumbi, S. R. 1984. Tactics of acclimation: Morphological changes ofsponges in an unpredictable environment. Science 225: 1478–1480.

Randall, J. E. and W. D. Hartman. 1968. Sponge-feeding fishes of theWest Indies. Mar. Biol. 1: 216–225.

Reitner, J. and H. Deupp (eds.). 1991. Fossil and Recent Sponges.Springer-Verlag, Berlin.

Reiswig, H. 1975. Bacteria as food for temperate-water marinesponges. Can. J. Zool. 53: 582–589.

Rezvoi, P. D., I. T. Zhuravleva and V. M. Koltun. 1971. PhylumPorifera. In Y. A. Orlov and B. S. Sokolov (eds.), Fundamentals ofPaleontology, Vol. 1, Pt. II. Porifera, Archaeocyatha, Coelenterata,Vermes. Israel Program for Scientific Translations, Jerusalem, pp.5–97.

Rinehart, K. L., Jr., and 25 others. 1981. Marine natural products assources of antiviral, antimicrobial, and antineoplastic agents. PureAppl. Chem. 53: 795–817.

Rützler, K. 1970. Spatial competition among Porifera: Solution by epi-zoism. Oecologia 5: 85–95.

Rützler, K. (ed.) 1990. New Perspectives in Sponge Biology. SmithsonianInstitution Press, Washington, D.C.

Sara, M. 1970. Competition and cooperation in sponge populations.Symp. Zool. Soc. Lond. 25: 273–285.

Sara, M. 1974. Sexuality in the Porifera. Bull. Zool. 41: 327–348.Sara, M. and J. Vacelet. 1973. Ecologie des Demosponges. In P. Grassé

(ed.), Traite de Zoologie 3(1): 462–576. Masson et Cie, Paris.Schwab, D. W. and R. E. Shore. 1971. Fine structure and composition

of a siliceous sponge spicule. Biol. Bull. 140: 125–136.Sharma, G. H. and B. Vig. 1972. Studies on the antimicrobial sub-

stances of sponges. VI. Structure of two antibacterial substancesisolated from the marine sponge Dysidea herbacea. TetrahedronLett. 28: 1715–1718.

Shore, R. E. 1972. Axial filament of siliceous sponge spicules, its organ-ic components and synthesis. Biol. Bull. 143: 689–698.

Simpson, T. L. 1984. The Cell Biology of Sponges. Springer-Verlag, NewYork.

Stearn, C. W. 1975. The stromatoporoid animal. Lethaia 8: 89–100.Stearn, C. W. 1977. Studies of stromatoporoids by scanning electron

microscopy. Mem. Bur. Rech. Geol. Min. 89: 33–40.Sullivan, B., D. J. Faulkner and L. Webb. 1983. Siphonodictidine, a

metabolite of the burrowing sponge Siphonodictyon sp. that inhibitscoral growth. Science 221: 1175–1176.

Tuzet, O. 1963. The phylogeny of sponges according to embryological,histological and serological data, and their affinities with theProtozoa and Cnidaria. In E. C. Dougherty, Z. N. Brown, E. D.Hanson and W. D. Hartman (eds.), The Lower Metazoa: ComparativeBiology and Phylogeny. University of California Press, Berkeley, pp.129–148.

Vacelet, J. 1985. Coralline sponges and the evolution of the Porifera. InS. C. Morris et al. (eds.), The Origins and Relationships of LowerInvertebrates. Syst. Assoc. Spec. Vol. No. 28, Oxford, pp. 1–13.

Vacelet, J. 1988. Indications de profondeur données par les Spongiairesdans les milieux benthiques actuels. Géolog. Médierranéene, 15(1):13–26.

Vacelet, J. and N. Boury-Esnault. 1995. Carnivorous sponges. Nature373: 333–335.

Vacelet, J., N. Boury-Esnault, A. Flala-Medioni and C. R. Fisher. 1998.A methanotrophic carnivorous sponge. Nature 377: 296.

Van de Vyver, G. 1975. Phenomena of cellular recognition in sponges.In A. Moscona and A. Monroy (eds.), Current Topics in Develop-mental Biology, Vol. 4. Academic Press, New York, pp. 123–140.

Van Soest, R. W. M., T. M. G. van Kempen and J. C. Braekman (eds.).1994. Sponges in Time and Space: Biology, Chemistry, Paleontology.Balkema, Rotterdam.

Warburton, F. E. 1966. The behaviour of sponge larvae. Ecology 47:672–674.

Weissenfels, N. 1989. Biologie und mikroskopische Anatomie derSüsswasserschwämme (Spongillidae). Gustav Fischer, Stuttgart.

Westinga, E. and P. C. Hoetjes. 1981. The intrasponge fauna ofSpeciospongia vesparia at Curaçao and Bonaire. Mar. Biol. 26: 139.

Wilkinson, C. R. 1983. Net primary productivity in coral reef sponges.Science 219: 410–412.

Wilkinson, C. R. and J. Vacelet. 1979. Transplantation of marinesponges to different conditions of light and current. J. Exp. Mar.Biol. Ecol. 37: 91–104.

Wilson, H. V. 1891. Notes on the development of some sponges. J.Morphol. 5: 511–519.

Wolfrath, B. and D. Barthel. 1989. Production of fecal pellets by themarine sponge Halichondria panicea Pallas, 1766. J. Exp. Mar. Biol.Ecol. 129: 81–94.

Wood, R. 1990. Reef-building sponges. Am. Sci. 78: 224–235.

CalcareaBorojevic, R. 1979. Evolution des spongiaires Calcarea. In C. Lévi and

N. Boury-Esnault (eds.), Sponge Biology, Colloques Internat.C.N.R.S. 291: 527–530.

Borojevic, R., B.-E. Esnault and J. Vacelet. 1990. A revision of thesupraspecific classification of the subclass Calcinea (Porifera, ClassCalcarea). Bull. Mus. Natn. Hist. Nat. 12: 243–276.

Burton, M. 1963. A revision of the classification of the calcareoussponges. Brit. Mus. Nat. Hist., London.

Hartman, W. D. 1958. A re-examination of Bidder’s classification of theCalcarea. Syst. Zool. 7: 97–110.

Jones, W. C. 1965. The structure of the porocytes in the calcareoussponge Leucosolenia complicata (Montagu). J. R. Microsc. Soc. 85:53–62.

Jones, W. C. 1970. The composition, development, form and orienta-tion of calcareous sponge spicules. Symp. Zool. Soc. Lond. 25:91–123.

Ledger, P. W. and W. C. Jones. 1978. Spicule formation in the calcare-ous sponge Sycon ciliatum. Cell Tiss. Res. 181: 553–567.

Tuzet, O. 1973. Éponges Calcaires. In P. Grassé (ed.), Traité de Zoologie3(1): 27–132. Masson et Cie, Paris.

Ziegler, B. and S. Rietschel. 1970. Phylogenetic relationships of fossilcalcisponges. Symp. Zool. Soc. Lond. 25: 23–40.

HexactinellidaOkada, Y. 1928. On the development of a hexactinellid sponge Farrea

sollasii. Tokyo University Fac. Sci. J., Sect. IV, 2: 1–27.Reiswig, H. M. 1979. Histology of Hexactinellida (Porifera). In C. Lévi

and N. Boury-Esnault (eds.), Sponge Biology, Colloques Internat.C.N.R.S. 291: 173–180.

DemospongiaeAyling, A. L. 1980. Patterns of sexuality, asexual reproduction and re-

cruitment in some subtidal marine Demospongiae. Biol. Bull. 158:271–282.

Ayling, A. L. 1983. Growth and regeneration rates in thinly encrustingDemospongiae from temperate waters. Biol. Bull. 165: 343–352.

Bergquist, P. R. and J. H. Bedford. 1978. The incidence of antibacterialactivity in marine Demospongiae: Systematic and geographic con-siderations. Mar. Biol. 46: 215–221.

Bergquist, P. R. and W. D. Hartman. 1969. Free amino acid patternsand the classification of the Demospongiae. Mar. Biol. 3: 247–268.


Bergquist, P. R. and J. J. Hogg. 1969. Free amino acid patterns inDemospongiae: A biochemical approach to sponge classification.Cah. Biol. Mar. 10: 205–220.

Bergquist, P. R., M. E. Sinclair and J. J. Hogg. 1970. Adaptation to inter-tidal existence: Reproductive cycles and larval behaviour inDemospongiae. Symp. Zool. Soc. Lond. 25: 247–271.

Bryan, P. G. 1973. Growth rate, toxicity and distribution of the encrust-ing sponge Terpios sp. (Hadromerida: Suberitidae) in Guam,Mariana Islands. Micronesica 9: 237–242.

Cimino, G., S. D’Stefano, L. Minale and G. Sodano. 1975. Metabolismin Porifera. III. Chemical patterns and the classification of theDemospongiae. Comp. Biochem. Physiol. 50B: 279–285.

Connes, R., J.-P. Diaz and J. Paris. 1971. Choanocytes et cellule centralechez la Démosponge Suberites massa Nardo. C. R. Hebd. Séanc.Acad. Sci. 273: 1590–1593.

de Laubenfels, M. W. 1948. The order Keratosa of the phylum Porifera:A monographic study. Occ. Pap. Allan Hancock Fndtn. 3: 1–217.

Elvin, D. W. 1976. Seasonal growth and reproduction of an intertidalsponge, Haliclona permollis (Bowerbank). Biol. Bull. 151: 108–125.

Fell, P. E. 1969. The involvement of nurse cells in oogenesis and em-bryonic development in the marine sponge Haliclona ecobasis. J.Morphol. 127: 133–149.

Fell, P. E. 1976. The reproduction of Haliclona loosanoffi and its apparentrelationship to water temperature. Biol. Bull. 150: 200–210.

Fell, P. E. and K. B. Lewandrowski. 1981. Population dynamics of theestuarine sponge Halichondria sp., within a New England eelgrasscommunity. J. Exp. Mar. Biol. Ecol. 55: 49–63.

Finks, R. M. 1967. The structure of Saccospongia laxata Bassler(Ordovician) and the phylogeny of the Demospongiae. J. Paleon-tol. 41: 1137–1149.

Frost, T. M. and C. E. Williamson. 1980. In situ determination of the ef-fect of symbiotic algae on the growth of the freshwater spongeSpongia lacustris. Ecology 61: 1361–1370.

Gerrodette, T. and A. O. Fleschig. 1979. Sediment-induced reductionin the pumping rate of the tropical sponge Verongia lacunosa. Mar.Biol. 55: 103–110.

Guida, V. G. 1976. Sponge predation in the oyster reef community asdemonstrated with Cliona celata Grant. J. Exp. Mar. Biol. Ecol. 25:109–122.

Hatch, W. I. 1980. The implication of carbonic anhydrase in the physi-ological mechanism of penetration of carbonate substrata by themarine burrowing sponge Cliona celata (Demospongiae). Biol. Bull.159: 135–147.

Lévi, C. 1956. Étude des Halisarca de Roscoff. Embryologie et systéma-tiques des Démosponges. Arch. Zool. Exp. Gen. 93: 1–181.

Lévi, C. 1973. Systématique de la classe des Demospongiaria(Démosponges). In P. Grassé (ed.), Traité de Zoologie 3(1): 577–631.Masson et Cie, Paris.

Penney, J. T. 1960. Distribution and bibliography (1892–1957) of thefresh water sponges. University of South Carolina Publ. Biol. 3:1–97.

Penney, J. T. and A. A. Racek. 1968. Comprehensive revision of aworldwide collection of freshwater sponges (Porifera: Spongil-lidae). Bull. U.S. Nat. Mus. No. 272: 1–184.

Pond, D. 1992. Protective-commensal mutualism between the queenscallop Chlamys opercularis (Linnaeus) and the encrusting spongeSuberites. J. Moll. Studies 58: 127-134.

Rasmont, R. 1962. The physiology of gemmulation in fresh watersponges. In D. Rudnick (ed.), Regeneration, 20th Growth Symposium.Ronald Press, New York, pp. 1–25.

Reiswig, H. M. 1970. Porifera: Sudden sperm release by tropicalDemospongiae. Science 170: 538–539.

Reiswig, H. M. 1971a. In situ pumping activities of tropical Demo-spongiae. Mar. Biol. 9: 38–50.

Reiswig, H. M. 1971b. Particle feeding in natural populations of threemarine Demosponges. Biol. Bull. 141: 568–591.

Reiswig, H. M. 1973. Population dynamics of three JamaicanDemospongiae. Bull. Mar. Sci. 23: 191–226.

Reiswig, H. M. 1974. Water transport, respiration, and energetics ofthree tropical sponges. J. Exp. Mar. Biol. Ecol. 14: 231–249.

Reiswig, H. M. 1975. The aquiferous systems of three marineDemospongiae. J. Morphol. 145(4): 493–502.

Ristau, D. A. 1978. Six new species of shallow-water marine demo-sponges from California. Proc. Biol. Soc. Wash. 91: 203–216.

Rützler, K. 1975. The role of burrowing sponges in bioerosion.Oecologia 19: 203–216.

Rützler, K. and G. Reiger. 1973. Sponge burrowing: Fine structure ofCliona lampa penetrating calcareous substrata. Mar. Biol. 21:144–162.

Schmidt, I. 1970. Phagocytose et pinocytose chez les Spongillidae. Z.Vgl. Physiol. 66: 398–420.

Simpson, T. L. and J. J. Gilbert. 1973. Gemmulation, gemmule hatchingand sexual reproduction in freshwater sponges. I. The life cycle ofSpongilla lacustris and Tubella pennsylvanica. Trans. Am. Microsc.Soc. 92: 422–433.

Sollas, W. J. 1884. On the origin of freshwater faunas: A study in evo-lution. Trans. R. Soc. Dublin 2(3): 87–118.

Woollacott, R. M. 1990. Structure and swimming behavior of the larvaof Halichondria melanadocia (Porifera: Demospongiae). J. Morph.205: 135–145.

Woollacott, R. M. and M. G. Hadfield. 1996. Induction of metamor-phosis in larvae of a sponge. Invert. Biol. 115(4): 257–262.

Ungemach, L. F., K. Souza, P. E. Fell and S. H. Loomis. 1997.Possession and loss of cold tolerance by sponge gemmules: A com-parative study. Invert. Biol. 116(1): 1–5.


Documents

6 Phylum Porifera: UNCORRECTED PAGE PROOFSsponges make up almost 75 percent of the total benthic biomass at a depth of 100–200 m. Subtidal and deeper water species that do not confront