Chapter 3

Porifera

Sponges are by far the simplest multicellular animals and are very
different from all the others. They have no fixed body shape, no plane
of symmetry and are covered in holes. All sponges live in water,
nearly all in the sea. The cells are uncoordinated, cell differentiation
is entirely reversible and cells may wander about in the background
jelly. A whole sponge can be regenerated from a few separated cells.
Sponges can almost be regarded not as individuals but as colonies
of separate cells; almost but not quite, as most have a skeleton made
of spicules that supports the body.
These very simple animals are nonetheless very successful and
widespread: since the early Cambrian they have covered most of
the suitable surfaces on the shore and in the shallow sea: the latest
survey found 15 000 living species. How is it that such simple animals
can do so well? What has there been for natural selection to work
on in this phylum? How fundamentally do they differ from other
animals and what are their evolutionary and ecological relationships
with them? To address these questions, we must study the basic
structure and the different kinds of sponges, and indicate the ways
in which they make a living.

3.1 What are the distinguishing characters

of sponges?

Sponges are sessile and immobile, having neither nerves nor

muscles. There may be slight contractility round the larger
pores but it is very restricted.
Sponge cell types are the distinctive collar cells or ‘choanocytes’
(Figure 3.1a), the ‘pinacocytes’ that make an outer layer and the
‘amoebocytes’ wandering through the central jelly or ‘mesohyl’.
This jelly is needed for support; in contrast to other animals,
neighbouring cells are not bound together by a basement
membrane.
24 PORIFERA

Fig. 3.1 The structure of sponges: (a) a choanocyte, component of black regions in (b), (c), (d); (b) basic sponge structure, as in the
late larva; (c) folding of walls to make flagellated chambers, for example in Leucosolenia; (d) part of Grantia, a fully elaborated sponge;
(e) spicules.

Small pores perforate the whole body (the name of the phylum,
Porifera, means ‘pore bearing’). Water carrying food particles
enters the body by many small pores (‘ostia’), moved in by the
beating of the flagella of the internal collar cells. These cells
extract food particles from the water, which flows out through
larger pores, the ‘oscula’ (Figure 3.1b). The structure becomes
elaborated during evolution, as in Figure 3.1c,d. This arrange-
ment, with the principal openings exhalant, is unique to
sponges.
The skeleton is made of spicules (Figure 3.1e) of calcite (a calcium
salt) or silica (a silicon salt) with or without a matrix of horny
collagen-type protein. Such use of silica is a unique feature.
An unusually wide range of skeletal materials occurs among
 WHAT DIFFERENT KINDS OF SPONGE ARE KNOWN? 25

closely related species of sponges. Animal skeletons, whether

hard or hydrostatic (based on the incompressibility of water)
usually translate muscle contraction into movement. Sponges,
however, have no muscles and the supporting skeleton instead
serves to prevent movement from occurring. Spicules have
further important functions in preserving the sponge’s shape,
keeping the pores open and maintaining the internal channels
(as well as making the sponge even nastier to eat than it
probably would be anyway).

3.2 What different kinds of sponge are known?

Calcarea, with calcareous spicules. They occur in shallow waters

(less than 100 metres). Examples are Leucosolenia and Grantia
(Figure 3.2b,d).

Fig. 3.2 Drawings of (a) a hexactinellid, showing spicules fused to form a lattice; (b) Sycon (larger) and Leucosolenia; (c) Halichondria;
(d) Grantia.
26 PORIFERA

Demospongiae, with siliceous spicules. Examples are Halichondria,

the breadcrumb sponge (Figure 3.2c), and Spongia, the once
commonly used bath sponge, in which there are no spicules
but only protein fibres.
Hexactinellida, the ‘glass sponges’ (Figure 3.2a), are very different
from other sponges. They consist of a small group (400
500
species) in the deep sea (below 200 metres); their growth is
orientated to the constant water currents there. The tissues
are 75% syncitial (lacking cell boundaries) with the remaining
cells connected by cytoplasmic bridges; even the choanocytes
are not separate cells. There are no pinacocytes and no cells
with any contractility. The skeleton is a lattice made from
six-rayed spicules of silica. The skeletal network is useful in
distributing light, as spicules have fibreoptic qualities with
refractive differences between the core and the shell of the
fibres. Some authorities believe that the hexactinellids evolved
separately, in parallel with other sponges, but there is increas-
ing evidence that they are a specialised offshoot from cellular
sponges: for example, the embryos resemble other sponges in
their early development, with cell boundaries, but later hatch
as syncitial larvae.
Subdivision of the classes has traditionally depended on spicule
structure, but more recently biochemical or reproductive characters,
or those discovered by electron microscopy, have been preferred.
For a time a further class, the ‘Sclerospongiae’, was proposed for some
15 species of sponges where spicules have become fused to make
massive skeletons, building coral-like reefs, since the Palaeozoic.
This, however, is not a natural group; it includes some Calcarea
and some Demospongiae that have evolved in parallel.

3.3 How do sponges make a living?

3.3.1 Feeding method

Choanocytes both create water currents and trap food. The collar
is a ring of about 30 small folds (‘microvilli’, Figure 3.1a), linked
by cross-bridges. Food is absorbed by food-engulfing cells at the foot
of the collar. Within both the Calcarea and the Demospongiae the
internal pattern has changed (Figure 3.1c,d) to provide canals and
flagellated chambers. This arrangement provides a larger surface
area covered in choanocytes and slows the flow of water past these
cells, allowing more time for food capture.
Sufficient food particles arrive because water pumping is
surprisingly intensive: for example, a specimen of Leuconia
(Demospongiae) growing 10 cm tall with a diameter of 1 cm contained
about 2.2 million flagellated chambers and pumped 22.5 litres of
water per day. The oscular outward jet was 8.5 cm per second.
Amoebocytes assist the circulation of food through the sponge: all
 HOW DO SPONGES MAKE A LIVING? 27

digestion is intracellular. As in many permeable marine animals,

dissolved organic material may be a subsidiary source of food.

3.3.2 Behaviour
The animal is sessile, there are no sense organs, nerves or muscles;
what can there possibly be in the way of behaviour? Starting with
the lowest expectations, we can find quite a bit. For example, the rate
of flagellar beat can be influenced by currents: in one experiment
with Halichondria the flagella beat at 3 cm per second in still water
and this rose to 7 cm per second as the external current was
increased. There is a measure of communication, even coordination,
between cells: for example, dilation of a channel may be propagated
and stimuli such as touch, exposure to air or poisons can result in
the closure of a distant osculum. Although there are no tissue
junctions between cells there may be communicating channels,
rapidly and temporarily formed. Reactions are slow, as is shown by
Hymeniacidon, a common encrusting orange growth, about a centi-
metre thick, on British beaches. Poke it, and about 10 minutes later
the osculum will close. It is not clear how this contractility is
achieved, but occurring round the oscula there are some amoebo-
cytes called ‘myocytes’, that are particularly rich in microfilaments
and microtubules. There is some evidence that myocytes may con-
tain the fibrils (actin and myosin) which are the basis of contraction
in all other animals investigated (and even in some unicells). Nor is
it clear how cells can be re-extended, except by the pull of neigh-
bouring cells, but water pumping must help to retain the shape of
the sponge. After all, a sponge does not need rapid reactions: it needs
only to close up fast enough to avoid desiccation when the tide
goes out. A worse hazard would be to close the exhalant osculum
while the flagella continued to beat: the sponge might burst.
Sponge larvae are motile, using flagella. Their movements are
not under any form of nerve control, but the cells respond to
changes in light intensity, which can alter the direction of swimming.
When first released the larvae swim upwards in the sea, rotating
as they swim: increased intensity of sunlight from above stiffens
the flagella of the rotating larvae, steering them to darker areas
down below.
It is in the very different Hexactinellida that a greater degree of
coordination has been found. Here the diameter of oscula cannot
change, but on mechanical or electrical stimulation the flagella in
all the chambers may stop beating. There are no nerves: electrical
impulses pass along the continuous tissue of the syncitium. In other
sponges no such conducted electrical signals are known.

3.3.3 Reproduction and development

Sponges have remarkable powers of regeneration: they can be
strained through a fine mesh yet the cells will come together,
aggregate and divide to reconstitute the sponge. Cells from different
28 PORIFERA

Fig. 3.3 Sponge development:

(a) amphiblastula larva;
(b) gastrulation after settling.

species will not aggregate. In the sponge body asexual reproduction

by budding occurs readily: it is hard to distinguish from growth.
Some freshwater sponges bud off parts of the body to form
‘gemmules’, stages resistant to adverse environments that regenerate
when conditions are favourable.
In sexual reproduction, gametes are formed in the mesohyl by
dedifferentiation of other cells. Most sponges are hermaphrodites,
but cross-fertilised. When sperm of the same species enters through
an ostium, it is engulfed by a choanocyte which loses its flagellum
and moves through the jelly until it finds an egg, a procedure very
different from that of other Metazoa.
Sessile adult animals always need free-swimming larvae for
dispersal. Sponges have simple flagellated larvae, usually developing
in the parent body and then freed to swim and settle in suitable
sites (Figure 3.3a). Some species achieve further dispersal by asexual
fragmentation followed by release of larvae from the dispersed
fragments. Once the larva has settled the cells move and become
rearranged in a process which corresponds to the gastrulation
(see Glossary) of other Metazoa and is controlled by similar genes.
The larva now has differentiated cells patterned along an axis
(Figure 3.3b); this pattern becomes less clear at metamorphosis.
In some sponges the outer flagellated cells are lost at gastrulation,
but in the Calcaria they dedifferentiate into multipotent cells. The
cells, whatever their derivation, then move, divide and differentiate.

3.4 What changes have evolved during

sponge history?

3.4.1 Morphological change

In these simple animals, with so little connection between cells,
there is not much scope for morphological evolution beyond the
elaboration of flagellated chambers related to increased efficiency
of water circulation (see above). Sponges show an array of growth
forms according to environmental conditions: where it is exposed
to wave action a species usually displays a flat encrusting growth,
but in crevices or still water the same species may grow tall or hang
down, increasing the surface area. Deeper in the sea there is more
variety, including metre-thick spheres and the reef builders already
mentioned. In turbulent water continual disturbance brings in
 WHAT CHANGES HAVE EVOLVED DURING SPONGE HISTORY? 29

plenty of food, but in still water a sponge is in danger of recycling

the water and gaining no more food. Increased body size is then
an advantage, and a raised osculum will achieve a more powerful
exhalant jet. A few of the simpler Calcarea can be recognised by
their shape: for example Figure 3.2 shows the upright tubes of
Leucosolenia and the flat ‘purses’ of Grantia, but usually the general
appearance of a sponge is no certain guide to its identity.

3.4.2 Physiological differences

Such differences may be marked and may be correlated with different
ecological niches. In the Demospongiae, for example, Mycale is an
opportunist generalist readily colonising new sites but never growing
to the maximum possible size; most of its energy is devoted to rapid
growth and reproduction. Tethya by contrast forms permanent
populations of large individuals in less favourable environments;
its energies are channelled to physical resistance and it reproduces
only slowly.
One extraordinary group of sponges has become carnivorous:
living in deep-sea habitats where small particle food is scarce, they
capture small crustaceans with Velcro-like raised hooked spicules.
The crustaceans become entangled, grown over and gradually
digested. These sponges are hydroid-like in form (see Chapter 4) and
have entirely lost choanocytes, ostia, oscula and water channels.
They can be recognised as sponges (Demospongiae) only by their
spicules and by the nature of the outer layer.
However, the general form and function of animals so much
constrained by structural simplicity did not give natural selection
very much to work on. Diversity is shown far more strongly at the
biochemical level.

3.4.3 Sponge biochemistry

Why are sponges so highly coloured? Not all their colours can be
due to the commonly occurring symbiotic algae. On the shore and
in the shallow sea they are often yellow, orange, red, green or
violet. How can animals so openly exposed avoid predation, even if
they do contain spicules? Why are they the only animals to be able
to extract and build with silica? The answers to all these questions
reside in unusual and very varied cell biochemistry. Colours may
be due to pigment granules in amoebocytes and may serve as a
warning of inedibility. Sponges produce an array of biotoxins that
discourage predators; they may extend their use of poisons to
chemical warfare with other sessile invertebrates, to compete for
living space. One tropical encrusting sponge, Terpios, can grow as
much as 23 mm per month as it poisons its neighbours. The family
Clionidae (Demospongiae) includes ‘boring’ sponges with specialised
amoebocytes whose chemical secretions remove calcareous frag-
ments from coral skeletons, clams and scallops. The chips are
collected into the exhalant currents within the sponge and pass out
30 PORIFERA

through the osculum. In coral reefs boring sponges gain protective

shelter, and cause considerable damage.
Cell biologists and pharmacologists are currently very interested
in sponges. They provide model systems for the study of cell junctions
of the simplest and most labile kind, and for the investigation of cell
surface proteins that mediate cell recognition in a very basic immune
system. Sponges produce bioactive compounds, some of which may
directly benefit us: for example a sponge long used by New Zealand
Maoris to promote wound healing has been found to contain high
concentrations of a potent anti-inflammatory agent. Antimicrobial
action is often found; a growing list of such examples emphasises
that we need to conserve the biochemical diversity of sponges.

3.5 How are sponges related to other phyla?

This question has aroused much speculation. For a time sponges

were separated from all other multicellular animals as a subking-
dom, the Parazoa, and stigmatised as a ‘side issue’. This classification
and the ideas behind it have disappeared. Sponges are now known
to be Metazoa of the simplest possible kind, as expected since the
cells are so loosely held together. The only other phylum with equally
simple structure is the Placozoa, very small flat plate-like aggrega-
tions of amoeboid cells (Figure 3.4a). They were once thought to
be some developmental stage of a sponge, until sexually mature
individuals were recorded. Placozoa are now placed in a separate
phylum, with one genus, Trichoplax; recent molecular work compar-
ing specimens from different sources has revealed that this ‘genus’

Fig. 3.4 Non-sponges: (a) Trichoplax, a placozoan (100 mm in length), superficially sponge-like; (b) a choanoflagellate colony,
related to sponge ancestors.
 HOW HAVE SPONGES BECOME SO SUCCESSFUL? 31

in fact contains a great variety of forms. Placozoa are not at all closely
related to sponges: their nearest relatives may be the Ctenophora
(described in Chapter 5).
Sponges are known to be ancient: spicules fossilise readily and
are very common in Cambrian deposits. They have recently also
been found among the very earliest fossil animals in the Precambrian
era (see Chapter 2). Recently discovered deposits in China contain
not only spicules but prints of soft tissue, embryos and larvae from
about 580 million years ago.
Sponges cannot be classed with the Cnidaria as having two cell
layers (nor are they radially symmetrical) and they are even less
like all other Metazoa. Yet molecular evidence strongly supports
a single origin for the Metazoa, with sponges separated from the
rest more than 600 million years ago. Tracing this single origin is
clearly difficult, and no phylum can be derived from present-day
forms, but choanocytes do resemble the unicellular Choanoflagellata,
some of which form colonies (Figure 3.4b). Study of choanoflagellate
proteins reveals the expression of a number of developmental genes
which occur in sponges and other Metazoa but in no other organisms
(see Chapter 20). Accordingly, the dominant hypothesis at present
is that sponges did arise from ancestors shared with those of
choanoflagellates, which are therefore seen as the sister group to all
Metazoa.

3.6 How have sponges become so successful?

Morphological simplicity and lack of coordination have not pre-

vented sponges from being extremely successful animals, if success is
measured by survival, large numbers and very widespread distribu-
tion (in the sea). Sponges remind us that complexity of form is not
the only route to success. Sponge diversity may be limited in
morphological terms but their relatively independent cells have
been able to evolve a variety of unusual biochemical specialisations.
Sponges are ‘alternative’ animals: they can respond to the
environment and behave as functional units, but they do it in ways
unique among multicellular animals.