Forgot your password?
Document Actions
Monteiro et al 02
Journal of Experimental Marine Biology and Ecology
270 (2002) 171 – 189
www.elsevier.com/locate/jembe
Patches of the ascidian Pyura stolonifera
(Heller, 1878): structure of habitat and associated
intertidal assemblages
S.M. Monteiro*, M.G. Chapman, A.J. Underwood
Centre for Research on Ecological Impacts of Coastal Cities, Marine Ecology Laboratories A11,
University of Sydney, Sydney NSW 2006, Australia
Received 5 July 2001; received in revised form 15 January 2002; accepted 17 January 2002
Abstract
Beds of the ascidian Pyura stolonifera on rocky shores around Sydney, Australia, provide an
important habitat for many organisms, such as algae, chitons, limpets and snails. Fishermen collect
Pyura for bait, which can change the structure of those beds. This may, in turn, cause changes in their
associated biota. To predict the effects of such disturbances, it is necessary to understand the
relationship between the structure of the habitat provided by beds of Pyura and their associated
assemblages. Beds of Pyura can provide two types of habitat: Clumped, with > 50% cover of closely
packed individuals or Sparse, with < 50% cover, where individuals are mostly isolated from contact
with others. The spaces amongst the Pyura can be bare rock or rock covered by numerous species of
algae and sessile animals. Nineteen algal species and 45 species of animals were identified in the field
in Clumped and Sparse habitats. Assemblages differed significantly between patches of Clumped and
Sparse habitat and much of the difference was attributable to relatively few species. In addition, a
number of species of algae and animals live on the surface of the Pyura themselves. Eighty-four taxa
were found at the scale of individual Pyura, many of which were relatively small and cryptic.
Organisms on Pyura differed between individual ascidians that were isolated from other Pyura (i.e. a
more common situation in Sparse habitat) and those that were surrounded by and in contact with other
Pyura (i.e. more common in Clumped habitat). These differences were mostly attributed to a few
families of small molluscs. It appears that changes in the structure of beds of Pyura have the potential
to make significant alterations to intertidal assemblages at the scale of individual Pyura and at the
scale of the habitats formed by beds of Pyura. D 2002 Elsevier Science B.V. All rights reserved.
Keywords: Patches; Pyura stolonifera; Habitat; Molluscs
*
Corresponding author. Tel.: +61-2-9351-4062; fax: +61-2-9351-6713.
E-mail address: smontess@bio.usyd.edu.au (S.M. Monteiro).
0022-0981/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 2 - 0 9 8 1 ( 0 2 ) 0 0 0 2 0 - 5
172 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
1. Introduction
Patch dynamics has long been a focus of ecological studies in different types of habitat
and geographical areas (Dayton, 1971; Paine and Levin, 1981; Connell and Keough, 1985;
Veblen, 1985; Ramage and Schiel, 1999), but no consistent definition of a ‘‘patch’’ can be
found in the literature. According to White and Pickett (1985), any definition of a patch
should always make reference to a defined spatial scale and structure and should relate to
the system being studied. In forestry, for instance, patches are usually defined as areas
dominated by different species of trees or shrubs (Veblen, 1985; Wu and Levin, 1994). On
the other hand, in intertidal and subtidal areas, patches are often described as areas of open
space that were created by some natural or man-made event (e.g. Connell and Keough,
1985; Sousa, 1985). This makes it impossible to have a general definition of patch that can
be applied to all habitats.
Nevertheless, patches have been divided into two general groups: Type I and Type II
patches. Type I patches are areas of cleared space in the middle of a matrix of a habitat, i.e.
they are surrounded totally or partially by the original habitat. For example, when clumps
of mussels are removed from beds of mussels by a disturbance, the patches created are of
Type I. Type II patches, on the other hand, are isolated from surrounding assemblages and
are the outcome of creation of new substratum. For example, Type II patches are created
by organisms such as barnacles, oysters or coralline algae, that can provide additional
surfaces for settlement of other organisms (Connell and Keough, 1985; Sousa, 1985).
Most landscapes are composed of Type I and/or Type II patches that differ in age, size,
structure, etc. (Veblen, 1985; Mauchamp et al., 1994). As a result, landscapes are often
˜
seen as mosaics of different phases (i.e. patches). For example, Montana (1992) described
the alternating densely vegetated patches and bare areas as a two-phase mosaic, character-
istic of scrublands in the Southern Chihuhuan Desert.
In rocky intertidal systems, primary substratum is often in short supply (Dayton, 1971;
Connell, 1972; Gaines and Roughgarden, 1985; Sousa, 1985) and many organisms live
among mussels and other organisms, which can provide additional substratum. Such
biologically generated habitat is important in enhancing biodiversity on rocky shores
(Lohse, 1993; Seed, 1996; Thompson et al., 1996).
The process of creation or maintenance of habitat, not involving direct trophic
interactions among species, was defined by Jones et al. (1994) as ecosystem engineering.
Ecosystem engineers are organisms that directly or indirectly control the availability of
resources (other than themselves) to other species, by causing changes in biotic or abiotic
materials (Jones et al., 1994). Habitats created by living organisms on intertidal shores are
structurally more complex than the surrounding rock surface. In addition, the amount of
available substratum can be greatly increased by their presence (Lohse, 1993). The
structure of habitat influences the composition of species associated with it and different
habitats can therefore have different associated assemblages (Ojeda and Dearborn, 1989;
Brown, 1991; Lintas and Seed, 1994; Seed, 1996; Thompson et al., 1996).
The structure of habitat can be defined in terms of heterogeneity (the relative abundance
per unit area or per unit volume of different structural components; McCoy and Bell,
1991) and complexity (the absolute abundance per unit area or per unit volume of
individual structural components; McCoy and Bell, 1991). Each of these elements can
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 173
strongly influence the small-scale distribution of species (Bourget et al., 1994; Lemire and
Bourget, 1996) and play an important role in structuring marine assemblages (Kohn and
Leviten, 1976; Bergeron and Bourget, 1986; Chapman and Underwood, 1994; Lemire and
Bourget, 1996). For example, for intertidal organisms, increased structural complexity can
alleviate environmental stresses during low tide (Thompson et al., 1996). Invertebrates are
protected from dislodgment by strong wave-action, insolation, etc., in mussel beds (Seed,
1996), and clumps of Fucus provide shelter for a variety of macro-invertebrates
(Thompson et al., 1996). In the early stages of development of a marine epibenthic
community in Canada, the small-scale distribution of species was strongly influenced by
the heterogeneity (panels with a smooth surface or with crevices of different depths) of the
substratum (Bourget et al., 1994).
The spatial and temporal patchiness of the habitat may influence the behaviour of
intertidal snails (Underwood and Chapman, 1989), competition (Dayton, 1971), the number
of species in assemblages (e.g. Kohn and Leviten, 1976) or structure of populations (e.g. the
snail Littorina unifasciata, Underwood and Chapman, 1992). The ways in which organisms
respond to patchiness can be complex and varied and, among other things, depend on
factors, such as the time of creation of the patch (Sousa, 1985), its area (Keough, 1984;
Sousa, 1984; Butler, 1991), its shape (Hamazaki, 1996) and the types of organisms present in
the patch and surroundings (Keough, 1984; Sousa, 1984; Eggleston et al., 1999). The almost
endless number of possible combinations of these elements at a certain point in time and
space results in the great patchiness and variability that characterize intertidal rocky shores
around the world. It is therefore of utmost importance to be able to describe natural
variability as a preliminary to trying to understand what factor(s) cause it.
Pyura stolonifera (Heller, 1878) (hereafter referred to as Pyura) is a large solitary
ascidian that occurs in dense beds on rocky intertidal reefs, from low intertidal habitats to a
depth of 10 to 12 m (Edgar, 1997). They are often more abundant in areas of strong wave-
action (Paine and Suchanek, 1983; Fielding et al., 1994). In Australia, they can be found
from Western Australia to Queensland and in Tasmania (Edgar, 1997). They are also found
along the coast of southern Africa (Fielding et al., 1994) and in Chile (Paine and
Suchanek, 1983; Edgar, 1997).
Like beds of mussels or kelps, beds of Pyura can greatly increase the range and amount
of habitat available for other organisms in low-shore areas where they dominate (Fielding
et al., 1994). The crevices and interstices among individuals in beds of Pyura provide a
sheltered environment for a wide variety of organisms. These range from macro-in-
vertebrates, such as chitons (in Australia, Onithochiton quercinus and Plaxiphora albida
tend to be more abundant on Pyura than on the surrounding rock), whelks and limpets
(Van Driel and Steyl, 1976; Fielding et al., 1994), to micro-invertebrates such as mic-
rogastropods, limpets and several species of algae (Van Driel and Steyl, 1976; Otway,
1989; Fielding et al., 1994). Living on and amongst Pyura may allow animals and plants
to get the benefits of a high energy environment, such as oxygenated water or a continuous
flow of food, without being subjected to such potentially destructive variables as strong
wave-action or currents (Fielding et al., 1994).
Across their distribution, Pyura are used as food and bait for fishing (Otway, 1989;
Fairweather, 1991; Chapman and Underwood, 1994). Otway (1989) described beds of
Pyura (areas of the shore where Pyura are dominant) as persistent and conspicuous
174 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
elements of low-shore assemblages in NSW, which provided much of the 3D complexity
at that level. Sixty-five percent of the loss of Pyura observed by Otway (1989) was due to
the animals being killed by fishermen. Preliminary observations by Fairweather (1991)
around Sydney suggested that collection by fishermen could potentially eliminate or, at
least, greatly reduce populations of Pyura. The loss of individuals from beds of Pyura may
lead to changes in the structure provided by those beds. Because the removal of Pyura
varies in time and space, patches with different types of structure may be present in
different areas at any time. At Cape Banks, NSW, Australia, the removal of Pyura by
fishermen or storms resulted in marked changes to the low-shore assemblage, possibly due
to the loss of the additional surface area provided by these animals (Otway, 1989) or to the
effects of the storm itself. Patches of bare space and/or patches of basal encrusting
coralline algae developed in spaces where Pyura were removed (Otway, 1989).
Unquantified observations suggested that beds of Pyura around Sydney were com-
prised of two different structural habitats:
Clumped: areas with large cover of Pyura ( > 50%) and with most Pyura in contact or
very close together to form large continuous patches of Pyura (Fig. 1a).
Sparse: areas with a small cover of Pyura ( < 50%) and with most Pyura not in contact
with others and therefore forming small patches of Pyura (Fig. 1b).
To test the model that these observations reflected natural patterns of structure of the
habitats, it was predicted that sampling random patches of Pyura on intertidal shores
around Sydney (Fig. 2) would show that all patches fall into one or the other of the above
categories (i.e. it was predicted that no areas with many isolated Pyura covering more than
50% and no areas with very few scattered patches of dense Pyura would be found).
Organisms associated with Pyura can vary in size from a few millimetres to several
centimetres. Individual Pyura can form clumps and be surrounded by other Pyura or be
found standing by themselves separated from other individuals. Animals and plants living
on the Pyura themselves may vary according to the proximity of surrounding Pyura.
Therefore, the prediction that different assemblages were associated with different types of
Fig. 1. (a) Clumped habitat; (b) Sparse habitat (circles represent individual Pyura and are placed in a way that
represents how Pyura individuals are organised in the field); (c) example of how the structure of habitats was
measured (— substratum not covered by Pyura; — substratum covered by Pyura).
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 175
Fig. 2. Map showing all locations and sites sampled.
habitat was tested at two spatial scales: the scale of patches of Pyura and that of individual
Pyura, because when examining the diversity associated with any habitat or assemblage, it
is important to use a range of resolutions (e.g. Thompson et al., 1996).
2. Materials and methods
Field work was done at four shores (Cape Banks, Long Bay, Little Congwong and
Congwong; Fig. 2) near Sydney, Australia. The experiments relating to the associated
assemblages were only done in the Cape Banks Scientific Marine Research Area in Botany
Bay, NSW, Australia (Underwood et al., 1983; Fig. 2).
176 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
2.1. Patches of Pyura
Low intertidal areas of approximately 3 m  3 m, where Pyura were numerous were
randomly chosen from those available at each site and defined a priori to be either
Clumped or Sparse habitat, according to the qualitative descriptions given above. In each
of these areas, four replicate 1 m  1 m quadrats were randomly placed. Using a 100-cm
tape measure, the substratum covered by Pyura and the substratum not covered by Pyura
were measured along four transects set 20 cm from each other (i.e. 20, 40, 60 and 80 cm
from the top of the frame of the quadrat; Fig. 1c). When either category extended beyond
the frame of the quadrat, the length outside the quadrat was also measured, up to a
maximal distance of 1 m away from the frame (distances greater than 1 m were considered
to be outside the patch being measured).
These measurements were used to estimate three variables to characterize the structure
of each patch of habitat: (i) percentage cover of Pyura; (ii) size of patches of substratum
covered by Pyura and (iii) size of patches of substratum without Pyura. Percentage cover
was calculated based on the total length of Pyura measured in each quadrat and the total
possible area (i.e. 400 cm).
To sample the associated assemblages at the scale of patch (i.e. 1 m  1 m quadrat), the
cover and diversity of the different species of animals and plants were measured by
recording the percentage cover of algae and sessile animals and numbers of mobile
animals in four, randomly placed 30 cm  30 cm sub-quadrats. Percentage covers were
recorded under 100 points in each sub-quadrat. Any sessile organism present, but not
recorded under a point of intersection, was given a cover of 0.5%. Any organism not
identified in the field was taken to the laboratory for identification. Organisms were
generally identified to species.
2.2. Individual Pyura
In order to sample the assemblages at the scale of individual Pyura, five Isolated
(standing by themselves separated from other individuals) and five Clumped Pyura (part
of a clump and surrounded by other Pyura) were collected from each of two sites (CB and
HF; Fig. 2). The individuals collected were randomly selected from those available at each
site. Clumped individuals were collected from Sparse or Clumped habitats, whenever
possible and from different clumps, so that the widest possible range of individuals was
sampled. Isolated individuals were only collected from Sparse habitat (there were no
Isolated individuals in Clumped habitat). Each individual was removed from the rock by
wedging a knife between the bottom of its test and the substratum and using it as a lever.
To avoid the possible loss of any associated animals during collection, a plastic bag was
put around each individual prior to removal. The samples were then taken to the laboratory
and preserved in 7% formalin.
The covers of algae and sessile animals on the tests of each Pyura were estimated
using a 3 cm  3 cm quadrat with 10 random points (five replicate quadrats were
sampled per Pyura; the average area of a typical Pyura is 180 cm2). Mobile animals
were removed by placing the Pyura in a 500-Am sieve and washing it under running
water. When no more animals were found in the sieve, the Pyura was examined under
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 177
a microscope. Any animals still present were removed with forceps. All animals were
then identified and counted. Algae and sessile animals were identified to species, but
mobile animals were only identified to family. This taxonomic resolution allowed
spatial patterns of these taxa to be identified and saved considerable time during
sorting.
3. Results
3.1. Two habitats
Clumped and Sparse habitats were compared, using analysis of variance, in terms of
percentage cover of Pyura and sizes of patches covered by Pyura or covered by other
Table 1
Summary of analyses comparing (a) the average percentage cover of Pyura at three sites, at Cape Banks;
variances were homogenous (Cochran’s test, p > 0.05); (b) the mean size of patch of substratum not covered by
Pyura at three sites, at Cape Banks. Variances were heterogeneous, so data were transformed to Ln(x + 1); (c) the
mean size of patch of Pyura at three sites, at Cape Banks. Variances were heterogeneous, so data were
transformed to Ln(x + 1)
Source of variation df Mean squares F-ratio P
(a)
S 2 45.55 1.68 ns
H 1 2375.06 13.39 ns
Q(S Â H) 18 27.05 1.43 ns
SÂH 2 177.43 6.56 *
Residual 72 18.86
SNK tests (‘‘ = ’’ not significant at P = 0.05); Site 1: C p S; Site 2: C p S; Site 3: C p S
(b)
S 2 2.92 2.74 ns
H 1 25.35 10.33 ns
Q(S Â H) 18 1.07 2.50 **
SÂH 2 2.45 2.30 ns
Residual 72 0.43
(c)
S 2 2.19 2.81 ns
H 1 1.85 0.08 ns
Q(S Â H) 18 0.78 1.55 ns
SÂH 2 22.24 28.51 ***
Residual 72 0.50
SNK tests (‘‘ = ’’ not significant at P = 0.05); Site 1: C = S; Site 2: C p S; Site 3: C p S
S = Site (three levels; random), H = Habitat, i.e. Clumped versus Sparse (two levels; fixed), Q(S Â H) = Quadrats
(four levels in each habitat at each site; random). n = 4 replicate transects.
ns, p > 0.05; *p < 0.05; **p > 0.01; ***p > 0.001.
178 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
substratum. In cases where Cochran’s tests were significant, data were transformed
(Tables 1 and 2). The hypothesis that patches of Pyura would fall into two structural
groups was tested at three sites in Cape Banks (HF, CB and P, see Fig. 2). The
generality of the findings was then tested at the three other locations around Cape Banks
(Fig. 2).
Even though the difference varied among sites (S Â H interactions in Tables 1a and 2a),
the cover of Pyura was greater in Clumped than in Sparse habitat at all sites sampled at
Cape Banks and at all other locations. Across all locations, the cover of Pyura was always
greater in Clumped habitat (Fig. 3a,e). The size of patch of Pyura in Clumped habitat
varied more from site to site and from location to location. At two out of three sites at Cape
Banks, patches of Pyura were larger in Clumped habitat (Fig. 3b,e), but this was not
significant (Table 1b). At all other locations sampled, there were significant differences
between habitats (Table 2b). The size of patches of substratum other than Pyura was
consistently greater in Sparse habitat (Fig. 3c,f ). At Cape Banks, there was a site  habitat
interaction. Patches without Pyura were smaller in Clumped than in Sparse habitats at two
Table 2
Summary of the analyses comparing (a) the average percentage cover of Pyura at three locations; variances were
homogenous (Cochran’s test, p > 0.05); (b) the mean size of patch of substratum not covered by Pyura at three
locations. Variances were heterogeneous, but data could not be transformed; (c) the mean size of patch of Pyura at
three locations. Variances were heterogeneous, so data were transformed to Mx
Source of variation df Mean squares F-ratio P
(a)
L 2 622.17 1.49 ns
H 1 22 330.89 53.5 ***
Q(L Â H) 12 417.42 1.81 ns
LÂH 2 1764.39 4.23 *
Residual 54 230.13
SNK tests (‘‘ = ’’ not significant at P = 0.05); Loc 1: C = S; Loc 2: C p S; Loc 3: C p S
(b)
L 2 6.05 1.72 ns
H 1 78.12 22.17 **
Q(L Â H) 12 3.52 2.42 *
LÂH 2 10.39 2.95 ns
Residual 54 1.45
(c)
L 2 1217.06 2.10 ns
H 1 8975.95 15.52 **
Q(L Â H) 12 578.44 2.41 *
LÂH 2 1463.03 2.53 ns
Residual 54 240.12
L = Locations (three levels; random), H = Habitat, i.e. Clumped versus Sparse (two levels; fixed),
Q(L Â H) = Quadrats (three levels; random). n = 4 replicate transects.
ns, p > 0.05; *p < 0.05; **p > 0.01; ***p > 0.001.
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 179
Fig. 3. Mean ( + SE): (a,d) percentage cover of Pyura; (b,e) size of patches of substratum covered by Pyura; (c,f )
size of patches of substratum not covered by Pyura, (n = Clumped, 5 = Sparse). HF, CB and P are three sites at
Cape Banks (a – c); LB (Long Bay), LC (Little Congwong) and C (Congwong) are three other shores.
out of three sites (SNK tests and Table 1c). At all remaining locations, the patches of
substratum not covered by Pyura were smaller. Using just one of these variables to classify
habitats would not be useful, but combining all three variables makes the distinction
between Clumped and Sparse habitats easier. The cover of Pyura and the sizes of patches
of Pyura were greater in Clumped habitat, while the sizes of patches of other substratum
were greater in Sparse habitat.
180 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
Fig. 4. nMDS plot of assemblages associated with Pyura stolonifera at the scale of patch, from the three sites at
Cape Banks combined (circles, Clumped habitat; triangles, Sparse habitat). There were four quadrats in three
patches of each type of habitat (different shading of symbols represents the three sites).
3.2. Assemblages in the two habitats
Assemblages associated with the two habitats were different at all sites at Cape Banks,
but this difference was only significant in ANOSIM tests (on ranked Bray – Curtis
dissimilarities, Clarke, 1993; P < 0.06) at two of the sites (HF: ANOSIM R = 0.08,
P < 0.06; CB: R = 0.27, P < 0.0005; P: R = 0.53, P < 0.0005). To test the generality of these
Table 3
The contribution towards Bray – Curtis measures of dissimilarity for the most important species in each site,
between assemblages in Clumped versus Sparse habitats
Dissimilarity
Clumped versus Sparse habitat
Site 1 Site 2 Site 3 All sites
Bare % (Rank) 2 (13) 0.3 (10) 7 (5) 3 (6)
Corallina officinalis % (Rank) 21 (1=) 26 (1) 42 (1) 29 (1)
Non-geniculate Corallina % (Rank) 6 (3=) 20 (2) 4 (7) 11 ( = 3)
Sargassum sp. % (Rank) 4 (5) 2 (7=) 1 (9) 2 (7=)
Hildenbrandia rubra % (Rank) 21 (1=) 7 (4) 9 (3) 13 (2)
Ralfsia verrucosa % (Rank) 13 (2) 16 (3) 8 (4) 11( = 3)
Laurencia sp. % (Rank) 1 (14=) 4 (5=) 0.1 (27) 2 (7=)
Tesseropora rosea % (Rank) 1 (14=) 2 (7=) 12 (2) 5 (5)
Montfortula rugosa % (Rank) 6 (3=) 4 (5=) 6 (6) 6 (4)
Small black siphonarians % (Rank) 4 (4) 3 (6) 0 (36) 2 (7=)
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 181
patterns across Cape Banks, the data from all three sites were combined. Assemblages
associated with Clumped and Sparse habitats were significantly different (R = 0.19,
P < 0.0005; Fig. 4). Cover of Pyura was not included in these analyses because Pyura
was considered the provider of habitat. Percentage covers of algae and sessile animals were
corrected after the removal of the cover of Pyura. Where significant patterns were detected,
SimDiss (an equivalent procedure to SIMPER, Primer package, Clarke and Warwick, 1994)
was used to determine the contribution of different species to those patterns.
Fig. 5. Mean ( + SE): (a) number of mobile species; (b) mean number of sessile species; (c) mean total number of
species associated with habitats of Pyura, at the three sites sampled at Cape Banks (n = Clumped, 5 = Sparse).
182 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
Nineteen species of algae and 45 species of animals were found in the habitats provided
by Pyura, across the three sites. Molluscs were the dominant group. Others included
sponges, echinoderms, anemones and crustaceans. Of the 64 taxa, only a small group was
responsible for the majority of the dissimilarities between assemblages (Table 3). Across
all sites, Corallina contributed most to the differences. Others included the encrusting
algae Hildenbrandia and Ralfsia. All these taxa are generally associated with the space
among Pyura and tended to be more common in Sparse habitat. The fissurellid limpet
Montfortula rugosa was the only mobile animal that contributed more than 5% to the
Fig. 6. Mean ( + SE): (a) percent cover of Corallina officinalis; (b) number of Cellana tramoserica; (c) number of
Thais orbita; (d) number of Cabestana spengleri found in habitats of Pyura, at the three sites sampled at Cape
Banks (n = Clumped, 5 = Sparse).
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 183
differences between assemblages. This species was present in both habitats and found on
Pyura and on the surrounding area.
Some species varied from one habitat to the other, but the numbers of taxa in each
habitat appeared to be similar. In light of this, the hypotheses that the number of mobile,
sessile and the total number of taxa were similar between the habitat were tested. The
numbers of mobile and sessile species and the total number of species were similar in each
habitat (Fig. 5a– c). Some species, such as C. officinalis and the limpet Cellana tramoserica
were usually associated with the spaces amongst Pyura and were more abundant in Sparse
habitat (Fig. 6a,b). The whelks Thais orbita and Cabestana spengleri were generally
found attached to Pyura. Thais was found in both habitats and had no specific pattern
of abundance (Fig. 6c). Cabestana, on the other hand, were mostly found in Clumped
habitat (Fig. 6d).
3.3. Assemblages on individual Pyura
Eighty-four taxa were found living on the tests of individual Pyura. Again, molluscs
were the dominant group, although several species of algae, polychaetes and isopods were
also present. The differences (Fig. 7) between the types of Pyura (for HF and CB,
respectively, ANOSIM R = 0.60, P < 0.0008; R = 0.59, P < 0.008) were mainly due to
limpets of the families Siphonariidae and Acmaeidae, bivalves of the family Erycinidae
and micro-gastropods of the family Eatoniellidae. Analysis of variance was initially used
to test the hypothesis that abundances of these taxa differed between Isolated and Clumped
Fig. 7. nMDS plot of assemblages on individual Pyura. Five individuals were collected from Clumped or Sparse
habitat at each of the two sites (C = Clumped Pyura; I = isolated Pyura) to close square.
184
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
Table 4
Analyses of numbers and frequencies of occurrence of types of gastropods on Pyura
Taxon Site Individual Pyura Clumped Pyura v2 (a)
(1 df ) P v2 (b)
(1 df ) P Fi Fc
No. occupied No. individuals No. occupied No. individuals
Siphonaridae 1 4 99 1 13 1.8 ns 66.0 ***
2 5 144 0 0 5.0 144.0 ***
1+2 0.5 1.00
Acmaeidae 1 4 77 2 5 0.7 ns 63.2 ***
2 5 88 3 5 0.5 ns 74.1 *
1+2 0.66 0.5
Erycinidae 1 5 557 5 127 0.0 ns 270.3 ***
2 5 173 5 142 0.0 ns 3.0 ns
1+2 0.67 0.5
Eatoniellidae 1 5 280 5 102 0.0 ns 82.9 ***
2 4 54 5 22 0.1 ns 13.5 ***
1+2 0.59 0.67
For each family of gastropods, at each site, v2 (a) tested the null hypothesis that the frequencies of Isolated and Clumped Pyura occupied by that family were equal; v2 (b)
tested the null hypothesis that the total number of that family on 10 Isolated Pyura equals the total number on 10 Clumped Pyura. Fi is the probability from Fisher’s exact
test of the null hypothesis that the proportion of Isolated Pyura occupied by that family in site 1 equals the proportion occupied in site 2; Fc is the same test for Clumped
individuals.
*p < 0.05; ***p < 0.001.
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 185
Pyura, but the heterogeneity of the data did not allow conclusive results and the
transformation of data did not solve this problem. To overcome this difficulty, three
different hypotheses were tested: (i) the proportion of Isolated Pyura occupied by taxon
i p proportion of Clumped Pyura occupied by taxon i; (ii) the proportion of taxon i on
Isolated Pyura p proportion of taxon i on Clumped Pyura; (iii) the proportion of Clumped
or Isolated Pyura with taxon i at site 1 = proportion of Clumped or Isolated Pyura with
taxon i at site 2. In the cases, where the expected values (under the null hypothesis of no
difference) were smaller than 5, Fisher’s exact tests were used. The tests were done for the
taxa (chosen on the basis of SIMDISS results) that accounted for most of the differences
between the types of Pyura.
For all four taxa tested, with the exception of Siphonariidae at CB, the proportions of
Isolated and Clumped Pyura occupied by taxon were similar (Table 4). At each site,
however, more individuals were found on Isolated than on Clumped Pyura. The only
exception to this trend was Erycinidae at CB. The proportions of Isolated and Clumped
Pyura occupied by each taxon were similar at the two sites.
4. Discussion
Pyura are a conspicuous and abundant component of rocky shores across its area of
distribution in Australia (Kott, 1985; Otway, 1989; Edgar, 1997). Their presence provides
added surface area for the settlement of other organisms, and by ameliorating environ-
mental conditions they may provide different microhabitats for those organisms. Living on
and amongst P. stolonifera may allow animals and plants to get the benefits of a high
energy environment, such as oxygenated water or continuous flow of food, without being
subjected to potentially destructive variables as strong wave-action or currents (Fielding et
al., 1994). Pyura may therefore be considered an example of an intertidal ecosystem
engineer (Jones et al., 1994).
Patches of other substratum (generally occupied by different species of foliose and/or
encrusting algae) in habitats of Pyura fall under the category of Type I patches — areas of
habitat surrounded by a matrix of some other habitat (Connell and Keough, 1985), in this
case Pyura. Patches of Pyura can, on the other hand, be considered Type II patches
because they are a different substratum and provide an additional area for the settlement of
other organisms (Connell and Keough, 1985).
As originally predicted, habitats in beds of Pyura separated into two different structural
types based on a combination of percentage cover and relative sizes of patches of Pyura
and patches of other substratum. Samples were randomly chosen, at all sites and locations
so that the widest possible range of patches was sampled. This also assured that if there
were any overlapping between the two types of habitat, it would be identified in the
analyses. There was only a handful of cases where a patch could have possibly been
considered intermediate but in every such case statistical analyses showed them to be
either Clumped or Sparse and not intermediate.
Clumped habitats were characterized by a large cover of Pyura and by larger patches of
Pyura in relation to those of other substratum. Sparse habitats, on the other hand, were
characterized by a small cover of Pyura, with patches of other substratum being on
186 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
average larger than those of Pyura. These patterns were consistent across all locations
sampled. The fact that no intermediate areas were found and the generality of the patterns
observed support the original idea that areas of the shore where Pyura is found can be
classified either as Clumped or Sparse habitat. The differences between Clumped and
Sparse habitats are presumably the result of combinations of differences in dispersal, rates
of settlement or post-settlement mortality, including differences due to disturbances.
Assemblages can vary at different spatial scales and the type of patterns observed at one
scale may or may not be present at a different one emphasizing the need to sample at
different spatial scales in order to reach a better assessment of the diversity in a given area
(Thompson et al., 1996). In the present study, the scales chosen were patches of Pyura and
individual Pyura.
Various studies have shown that different intertidal habitats are associated with different
assemblages (e.g. Hawkins, 1981; Lohse, 1993; Thompson et al., 1996). This was also the
case for the habitats of Pyura at Cape Banks. Assemblages were different at all three sites.
When the data from all sites were combined, the assemblages were significantly different.
Species normally associated with bare spaces, such as the algae Corallina and Hilden-
brandia, were responsible for most of the differences between the assemblages in the two
habitats. The algae Corallina and the limpet Cellana were more abundant in Sparse habitat,
which can possibly be explained by the greater availability of substratum not covered by
Pyura. Corallina may settle and grow better on rock (at all sites Corallina was mostly
found on the space between Pyura rather than on the Pyura themselves). C. tramoserica
are grazers and the algae they tend to feed on are more common on the substratum that
surrounds the Pyura. Species like the whelks T. orbita and C. spengleri were generally
found attached to Pyura and were more frequent in Clumped habitat.
As was the case at the scale of patch, the assemblages were different at the scale of
individual Pyura, supporting the idea that two different habitats were present in the beds of
Pyura. Differences at this scale were mostly due to differences in the abundances of
organisms rather than the number or types of taxa. What varied between Isolated and
Clumped Pyura were the numbers of animals present and the proportions of Pyura
occupied by those organisms rather than the types of animals. In this case, contrary to what
happened at the scale of patch, where differences were mainly due to algae, the taxa
responsible for most of the differences were mobile molluscs (three families of gastropods
and one family of bivalves).
The results of this study suggest that the variables used, percentage cover of Pyura, the
size of the patches of substratum covered and not covered by Pyura, and whether Pyura
are isolated or in clumps were appropriate for identifying patterns in the structure of
habitat and assemblages associated with the beds of Pyura around Sydney.
Pyura is commonly killed and used as bait by recreational fishermen (Underwood and
Kennelly, 1990) in Australia. Fishermen usually cut off the top of the test and use the
insides as bait. The portion still left attached to the rock usually disappears after about two
weeks. This type of collection can be selective; larger or more readily accessible
individuals are preferred. The numbers of Pyura can be seriously reduced by this type
of collection (Fairweather, 1991) and the structure of the patches of Pyura can also be
affected. A reduction in the number and cover of Pyura may lead to changes in the
structure of the habitat available for other organisms. Removal of Pyura at Cape Banks, by
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 187
fishermen or storms, has been shown to result in marked changes to the assemblages in
low-shore areas (Otway, pers. comm.).
Changes in the structure of a habitat may affect the associated assemblages in different
ways, so, in order to be able to predict the possible effects of such changes, it is important
to know the underlying natural patterns in those assemblages. Possible responses of the
assemblages to changes in the structure of the habitat can include decreases or increases in
abundances of some or all species and/or increases or decreases in the number of species.
Some species can disappear and new ones can replace them or be added to an assemblage.
In the case of habitats provided by Pyura, there are differences in the associated
assemblages, depending on whether the ascidians are clumped or sparsely scattered on
the shore. To determine the extent to which changing the arrangement of the Pyura
themselves may lead to changes in associated assemblages requires manipulative experi-
ments. In order to be able to plan and design such experiments, the natural patterns of
occurrence of the assemblages associated with Pyura must be known. Describing those
patterns quantitatively is the important first step in that understanding.
Acknowledgements
This study was supported by funds from the Australian Research Council through the
Centre for Research on Ecological Impacts of Coastal Cities. S.M. Monteiro was sup-
ߘ ˆ
ported by a scholarship from the Programa Praxis XXI of the Fundacao para a Ciencia e
Tecnologia, Lisbon, Portugal. We thank Dr. P. Archambault and two anonymous referees
for comments on earlier versions of this manuscript and numerous colleagues in the Centre
who helped with fieldwork. [RW]
References
Bergeron, P., Bourget, E., 1986. Shore topography and spatial partitioning of crevice refuges by sessile epibenthos
in an ice disturbed environment. Mar. Ecol. Prog. Ser. 1 – 2 (28), 129 – 145.
Bourget, E., DeGuise, J., Daigle, G., 1994. Scales of substratum heterogeneity, structural complexity, and the
early establishment of a marine epibenthic community. J. Exp. Mar. Biol. Ecol. 181 (1), 31 – 51.
Brown, V.K., 1991. In: Bell, S.S., McCoy, E.D., Mushinsky, H.R. (Eds.), The Effects of Changes in Habitat
Structure During Succession in Terrestrial Communities. Chapman & Hall, New York, USA, pp. 141 –
168.
Butler, A.J., 1991. Effect of patch size on communities of sessile invertebrates in Gulf St. Vincent, South
Australia. J. Exp. Mar. Biol. Ecol. 153, 255 – 280.
Chapman, M.G., Underwood, A.J., 1994. Dispersal of the intertidal snail, Nodilitorina pyramidalis, in response to
the topographic complexity of the substratum. J. Exp. Mar. Biol. Ecol. 179 (2), 145 – 169.
Clarke, K.R., 1993. Non-parametric multivariate analysis of changes in community structure. Austr. J. Ecol. 18,
117 – 143.
Clarke, K.R., Warwick, R.M., 1994. Change in Marine Communities: An Approach to Statistical Analysis and
Interpretation. Plymouth Marine Laboratory, Plymouth, 144 pp.
Connell, J.H., 1972. Community interactions on marine rocky intertidal shores. Annu. Rev. Ecol. Syst. 3, 169 –
192.
Connell, J.H., Keough, M.J., 1985. In: White, P.S., Pickett, S.T.A. (Eds.), Disturbance and Patch Dynamics of
Subtidal Marine Animals on Hard Substratum. Academic Press, New York, USA, pp. 125 – 151.
188 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
Dayton, P.K., 1971. Competition, disturbance and community organisation: the provision and subsequent uti-
lisation of space in a rocky intertidal community. Ecol. Monogr. 41 (4), 351 – 389.
Edgar, G.J., 1997. Australian marine life — the plants and animals of temperate waters. Australia, Reed Books,
544 pp.
Eggleston, D.B., Elis, W.E., Etherington, L.L., Dahlgren, C.P., Posey, M.H., 1999. Organism responses to habitat
fragmentation and diversity: habitat colonization by estuarine macrofauna. J. Exp. Mar. Biol. Ecol. 236 (1),
107 – 132.
Fairweather, P.G., 1991. A conceptual framework for ecological studies of coastal resources: an example of a
tunicate collected for bait on Australian seashores. Ocean Shoreline Manage. 15, 125 – 142.
Fielding, P.J., Weerts, K.A., Forbes, A.T., 1994. Macroinvertebrate communities associated with intertidal and
subtidal beds of Pyura stolonifera (Heller) (Tunicata: Ascideacea) on the Natal coast. S. Afr. J. Zool. 29 (1),
46 – 53.
Gaines, S., Roughgarden, J., 1985. Larval settlement rate: a leading determinant of structure in an ecological
community of the marine intertidal zone. Proc. Natl. Acad. Sci. U. S. A. 82, 3707 – 3711.
Hamazaki, T., 1996. Effects of patch shape on the number of organisms. Landscape Ecol. 11 (5), 299 – 306.
Hawkins, S.J., 1981. The influence of Patella grazing on fucoid-barnacle mosaic on moderately exposed rocky
shores. Proceedings of the 15th European Marine Biology Symposium: Lower Organisms and Their Impor-
tance in the Food Web. Kiel. Meeresforsch., Sonderh. vol. 5, pp. 537 – 543.
Jones, C.G., Lawton, J.H., Shachak, M., 1994. Organisms as ecosystem engineers. Oikos 69 (3), 373 – 386.
Keough, M.J., 1984. Effects of patch size on the abundance of sessile marine invertebrates. Ecology 65 (2), 423 –
437.
Kohn, A.J., Leviten, P.J., 1976. Effect of habitat complexity on population density and species richness in a
tropical intertidal predatory gastropod assemblage. Oecologia 25 (3), 199 – 210.
Kott, P., 1985. The Australian ascidiacea: Part I. Phlebobranchia and stolidobranchia. Mem. Queensl. Mus. 23,
328 – 331.
Lemire, M., Bourget, E., 1996. Substratum heterogeneity and complexity influence microhabitat selection of
Balanus sp. and Tubularia crocea larvae. Mar. Ecol. Progr. Ser. 135, 77 – 87.
Lintas, C., Seed, R., 1994. Spatial variation in the fauna associated with Mytilus edulis on a wave exposed rocky
shore. J. Molluscan, 165 – 174.
Lohse, D.P., 1993. The importance of secondary substratum in a rocky intertidal community. J. Exp. Mar. Biol.
Ecol. 166, 1 – 17.
Mauchamp, A., Rambal, S., Lepart, J., 1994. Simulating the dynamics of a vegetation mosaic — a spatialized
functional model. Ecol. Modell. 1 – 3, 107 – 130.
McCoy, E.D., Bell, S.S., 1991. Habitat structure: the evolution and diversification of a complex topic. In: Bell,
S.S., McCoy, E.D., Mushinsky, H.R. (Eds.), Habitat Structure: The Physical Arrangement of Objects in
Space. Chapman & Hall, London, pp. 3 – 27.
˜
Montana, C., 1992. The colonization of bare areas in two phase mosaics of an arid ecosystem. J. Ecol. 80, 315 – 327.
Ojeda, F.P., Dearborn, J.H., 1989. Community structure of macroinvertebrates inhabiting the rocky subtidal zone
in the Gulf of Maine: seasonal and bathymetric distribution. Mar. Ecol. Progr. Ser. 57 (3), 147 – 161.
Otway, N.M., 1989. The effects of grazing by chitons on mid and low shore intertidal communities. PhD Thesis,
University of Sydney.
Paine, R.T., Levin, S.A., 1981. Intertidal landscapes: disturbance and the dynamics of pattern. Ecological Mono-
graphs 51 (2), 145 – 178.
Paine, R.T., Suchanek, T.H., 1983. Convergence of ecological processes between independently evolved com-
petitive dominants: a tunicate – mussel comparison. Evolution 37 (4), 821 – 831.
Ramage, D.L., Schiel, D.R., 1999. Patch dynamics and response to disturbance of the seagrass Zostera nova-
zealandica on intertidal platforms in southern New Zealand. Mar. Ecol. Prog. Ser. 189, 275 – 288.
Seed, R., 1996. Patterns of biodiversity in the macro invertebrate fauna associated with mussel patches on rocky
shores. J. Mar. Biol. Assoc. U. K. 76, 203 – 210.
Sousa, W.P., 1984. Intertidal mosaics: patch size, propagule availability and spatially variable patterns of suc-
cession. Ecolology 65 (6), 1918 – 1935.
Sousa, W.P., 1985. In: White, P.S., Pickett, S.T.A. (Eds.), Disturbance and Patch Dynamics on Rocky Intertidal
Shores. Academic Press, New York, USA, pp. 101 – 124.
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 189
Thompson, R.C., Wilson, B.J., Tobin, M.L., Hill, A.S., Hawkins, S.J., 1996. Biologically generated habitat
provision and diversity of rocky shore organisms at a hierarchy of spatial scales. J. Exp. Mar. Biol. Ecol.
202, 73 – 84.
Underwood, A.J., Chapman, M.G., 1989. Experimental analyses of the influences of topography of the sub-
stratum on movements and density of an intertidal snail, Littorina unifasciata. J. Exp. Mar. Biol. Ecol. 134,
175 – 196.
Underwood, A.J., Chapman, M.G., 1992. Experiments on topographic influences on density and dispersion of
Littorina unifasciata in New South Wales. In: Grahame, J., Mill, P.J., Reid, D.G. (Eds.), Proceedings of the
Third International Symposium on Littorinid Biology. The Malacological Society of London, London, pp.
181 – 195.
Underwood, A.J., Kennelly, S.J., 1990. Pilot studies for design of surveys of human disturbance of intertidal
habitats in New South Wales. Aust. J. Freshwater Res. 41, 165 – 173.
Underwood, A.J., Denley, E.J., Moran, M.J., 1983. Experimental analyses of the structure and dynamics of mid
shore intertidal communities in New South Wales. Oecologia 56, 202 – 219.
Van Driel, C.D., Steyl, C.D., 1976. Proceedings of the first interdisciplinary conference on marine and freshwater
research in Southern Africa. Port Elizabeth, 5 – 10 July.
Veblen, T.T., 1985. Stand dynamics in Chilean Nothofagus forests. In: White, P.S., Pickett, S.T.A. (Eds.), The
Ecology of Natural Disturbance and Patch Dynamics. Academic Press, New York, USA, pp. 35 – 51.
White, P.S., Pickett, S.T.A., 1985. In: White, P.S., Pickett, S.T.A. (Eds.), Natural Disturbance and Patch Dynam-
ics: An Introduction. Academic Press, New York, USA, pp. 3 – 13.
Wu, J.G., Levin, S.A., 1994. A spatial patch dynamics modelling approach to pattern and process in an annual
grassland. Ecol. Monogr. 4, 447 – 464.
270 (2002) 171 – 189
www.elsevier.com/locate/jembe
Patches of the ascidian Pyura stolonifera
(Heller, 1878): structure of habitat and associated
intertidal assemblages
S.M. Monteiro*, M.G. Chapman, A.J. Underwood
Centre for Research on Ecological Impacts of Coastal Cities, Marine Ecology Laboratories A11,
University of Sydney, Sydney NSW 2006, Australia
Received 5 July 2001; received in revised form 15 January 2002; accepted 17 January 2002
Abstract
Beds of the ascidian Pyura stolonifera on rocky shores around Sydney, Australia, provide an
important habitat for many organisms, such as algae, chitons, limpets and snails. Fishermen collect
Pyura for bait, which can change the structure of those beds. This may, in turn, cause changes in their
associated biota. To predict the effects of such disturbances, it is necessary to understand the
relationship between the structure of the habitat provided by beds of Pyura and their associated
assemblages. Beds of Pyura can provide two types of habitat: Clumped, with > 50% cover of closely
packed individuals or Sparse, with < 50% cover, where individuals are mostly isolated from contact
with others. The spaces amongst the Pyura can be bare rock or rock covered by numerous species of
algae and sessile animals. Nineteen algal species and 45 species of animals were identified in the field
in Clumped and Sparse habitats. Assemblages differed significantly between patches of Clumped and
Sparse habitat and much of the difference was attributable to relatively few species. In addition, a
number of species of algae and animals live on the surface of the Pyura themselves. Eighty-four taxa
were found at the scale of individual Pyura, many of which were relatively small and cryptic.
Organisms on Pyura differed between individual ascidians that were isolated from other Pyura (i.e. a
more common situation in Sparse habitat) and those that were surrounded by and in contact with other
Pyura (i.e. more common in Clumped habitat). These differences were mostly attributed to a few
families of small molluscs. It appears that changes in the structure of beds of Pyura have the potential
to make significant alterations to intertidal assemblages at the scale of individual Pyura and at the
scale of the habitats formed by beds of Pyura. D 2002 Elsevier Science B.V. All rights reserved.
Keywords: Patches; Pyura stolonifera; Habitat; Molluscs
*
Corresponding author. Tel.: +61-2-9351-4062; fax: +61-2-9351-6713.
E-mail address: smontess@bio.usyd.edu.au (S.M. Monteiro).
0022-0981/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 2 - 0 9 8 1 ( 0 2 ) 0 0 0 2 0 - 5
172 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
1. Introduction
Patch dynamics has long been a focus of ecological studies in different types of habitat
and geographical areas (Dayton, 1971; Paine and Levin, 1981; Connell and Keough, 1985;
Veblen, 1985; Ramage and Schiel, 1999), but no consistent definition of a ‘‘patch’’ can be
found in the literature. According to White and Pickett (1985), any definition of a patch
should always make reference to a defined spatial scale and structure and should relate to
the system being studied. In forestry, for instance, patches are usually defined as areas
dominated by different species of trees or shrubs (Veblen, 1985; Wu and Levin, 1994). On
the other hand, in intertidal and subtidal areas, patches are often described as areas of open
space that were created by some natural or man-made event (e.g. Connell and Keough,
1985; Sousa, 1985). This makes it impossible to have a general definition of patch that can
be applied to all habitats.
Nevertheless, patches have been divided into two general groups: Type I and Type II
patches. Type I patches are areas of cleared space in the middle of a matrix of a habitat, i.e.
they are surrounded totally or partially by the original habitat. For example, when clumps
of mussels are removed from beds of mussels by a disturbance, the patches created are of
Type I. Type II patches, on the other hand, are isolated from surrounding assemblages and
are the outcome of creation of new substratum. For example, Type II patches are created
by organisms such as barnacles, oysters or coralline algae, that can provide additional
surfaces for settlement of other organisms (Connell and Keough, 1985; Sousa, 1985).
Most landscapes are composed of Type I and/or Type II patches that differ in age, size,
structure, etc. (Veblen, 1985; Mauchamp et al., 1994). As a result, landscapes are often
˜
seen as mosaics of different phases (i.e. patches). For example, Montana (1992) described
the alternating densely vegetated patches and bare areas as a two-phase mosaic, character-
istic of scrublands in the Southern Chihuhuan Desert.
In rocky intertidal systems, primary substratum is often in short supply (Dayton, 1971;
Connell, 1972; Gaines and Roughgarden, 1985; Sousa, 1985) and many organisms live
among mussels and other organisms, which can provide additional substratum. Such
biologically generated habitat is important in enhancing biodiversity on rocky shores
(Lohse, 1993; Seed, 1996; Thompson et al., 1996).
The process of creation or maintenance of habitat, not involving direct trophic
interactions among species, was defined by Jones et al. (1994) as ecosystem engineering.
Ecosystem engineers are organisms that directly or indirectly control the availability of
resources (other than themselves) to other species, by causing changes in biotic or abiotic
materials (Jones et al., 1994). Habitats created by living organisms on intertidal shores are
structurally more complex than the surrounding rock surface. In addition, the amount of
available substratum can be greatly increased by their presence (Lohse, 1993). The
structure of habitat influences the composition of species associated with it and different
habitats can therefore have different associated assemblages (Ojeda and Dearborn, 1989;
Brown, 1991; Lintas and Seed, 1994; Seed, 1996; Thompson et al., 1996).
The structure of habitat can be defined in terms of heterogeneity (the relative abundance
per unit area or per unit volume of different structural components; McCoy and Bell,
1991) and complexity (the absolute abundance per unit area or per unit volume of
individual structural components; McCoy and Bell, 1991). Each of these elements can
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 173
strongly influence the small-scale distribution of species (Bourget et al., 1994; Lemire and
Bourget, 1996) and play an important role in structuring marine assemblages (Kohn and
Leviten, 1976; Bergeron and Bourget, 1986; Chapman and Underwood, 1994; Lemire and
Bourget, 1996). For example, for intertidal organisms, increased structural complexity can
alleviate environmental stresses during low tide (Thompson et al., 1996). Invertebrates are
protected from dislodgment by strong wave-action, insolation, etc., in mussel beds (Seed,
1996), and clumps of Fucus provide shelter for a variety of macro-invertebrates
(Thompson et al., 1996). In the early stages of development of a marine epibenthic
community in Canada, the small-scale distribution of species was strongly influenced by
the heterogeneity (panels with a smooth surface or with crevices of different depths) of the
substratum (Bourget et al., 1994).
The spatial and temporal patchiness of the habitat may influence the behaviour of
intertidal snails (Underwood and Chapman, 1989), competition (Dayton, 1971), the number
of species in assemblages (e.g. Kohn and Leviten, 1976) or structure of populations (e.g. the
snail Littorina unifasciata, Underwood and Chapman, 1992). The ways in which organisms
respond to patchiness can be complex and varied and, among other things, depend on
factors, such as the time of creation of the patch (Sousa, 1985), its area (Keough, 1984;
Sousa, 1984; Butler, 1991), its shape (Hamazaki, 1996) and the types of organisms present in
the patch and surroundings (Keough, 1984; Sousa, 1984; Eggleston et al., 1999). The almost
endless number of possible combinations of these elements at a certain point in time and
space results in the great patchiness and variability that characterize intertidal rocky shores
around the world. It is therefore of utmost importance to be able to describe natural
variability as a preliminary to trying to understand what factor(s) cause it.
Pyura stolonifera (Heller, 1878) (hereafter referred to as Pyura) is a large solitary
ascidian that occurs in dense beds on rocky intertidal reefs, from low intertidal habitats to a
depth of 10 to 12 m (Edgar, 1997). They are often more abundant in areas of strong wave-
action (Paine and Suchanek, 1983; Fielding et al., 1994). In Australia, they can be found
from Western Australia to Queensland and in Tasmania (Edgar, 1997). They are also found
along the coast of southern Africa (Fielding et al., 1994) and in Chile (Paine and
Suchanek, 1983; Edgar, 1997).
Like beds of mussels or kelps, beds of Pyura can greatly increase the range and amount
of habitat available for other organisms in low-shore areas where they dominate (Fielding
et al., 1994). The crevices and interstices among individuals in beds of Pyura provide a
sheltered environment for a wide variety of organisms. These range from macro-in-
vertebrates, such as chitons (in Australia, Onithochiton quercinus and Plaxiphora albida
tend to be more abundant on Pyura than on the surrounding rock), whelks and limpets
(Van Driel and Steyl, 1976; Fielding et al., 1994), to micro-invertebrates such as mic-
rogastropods, limpets and several species of algae (Van Driel and Steyl, 1976; Otway,
1989; Fielding et al., 1994). Living on and amongst Pyura may allow animals and plants
to get the benefits of a high energy environment, such as oxygenated water or a continuous
flow of food, without being subjected to such potentially destructive variables as strong
wave-action or currents (Fielding et al., 1994).
Across their distribution, Pyura are used as food and bait for fishing (Otway, 1989;
Fairweather, 1991; Chapman and Underwood, 1994). Otway (1989) described beds of
Pyura (areas of the shore where Pyura are dominant) as persistent and conspicuous
174 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
elements of low-shore assemblages in NSW, which provided much of the 3D complexity
at that level. Sixty-five percent of the loss of Pyura observed by Otway (1989) was due to
the animals being killed by fishermen. Preliminary observations by Fairweather (1991)
around Sydney suggested that collection by fishermen could potentially eliminate or, at
least, greatly reduce populations of Pyura. The loss of individuals from beds of Pyura may
lead to changes in the structure provided by those beds. Because the removal of Pyura
varies in time and space, patches with different types of structure may be present in
different areas at any time. At Cape Banks, NSW, Australia, the removal of Pyura by
fishermen or storms resulted in marked changes to the low-shore assemblage, possibly due
to the loss of the additional surface area provided by these animals (Otway, 1989) or to the
effects of the storm itself. Patches of bare space and/or patches of basal encrusting
coralline algae developed in spaces where Pyura were removed (Otway, 1989).
Unquantified observations suggested that beds of Pyura around Sydney were com-
prised of two different structural habitats:
Clumped: areas with large cover of Pyura ( > 50%) and with most Pyura in contact or
very close together to form large continuous patches of Pyura (Fig. 1a).
Sparse: areas with a small cover of Pyura ( < 50%) and with most Pyura not in contact
with others and therefore forming small patches of Pyura (Fig. 1b).
To test the model that these observations reflected natural patterns of structure of the
habitats, it was predicted that sampling random patches of Pyura on intertidal shores
around Sydney (Fig. 2) would show that all patches fall into one or the other of the above
categories (i.e. it was predicted that no areas with many isolated Pyura covering more than
50% and no areas with very few scattered patches of dense Pyura would be found).
Organisms associated with Pyura can vary in size from a few millimetres to several
centimetres. Individual Pyura can form clumps and be surrounded by other Pyura or be
found standing by themselves separated from other individuals. Animals and plants living
on the Pyura themselves may vary according to the proximity of surrounding Pyura.
Therefore, the prediction that different assemblages were associated with different types of
Fig. 1. (a) Clumped habitat; (b) Sparse habitat (circles represent individual Pyura and are placed in a way that
represents how Pyura individuals are organised in the field); (c) example of how the structure of habitats was
measured (— substratum not covered by Pyura; — substratum covered by Pyura).
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 175
Fig. 2. Map showing all locations and sites sampled.
habitat was tested at two spatial scales: the scale of patches of Pyura and that of individual
Pyura, because when examining the diversity associated with any habitat or assemblage, it
is important to use a range of resolutions (e.g. Thompson et al., 1996).
2. Materials and methods
Field work was done at four shores (Cape Banks, Long Bay, Little Congwong and
Congwong; Fig. 2) near Sydney, Australia. The experiments relating to the associated
assemblages were only done in the Cape Banks Scientific Marine Research Area in Botany
Bay, NSW, Australia (Underwood et al., 1983; Fig. 2).
176 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
2.1. Patches of Pyura
Low intertidal areas of approximately 3 m  3 m, where Pyura were numerous were
randomly chosen from those available at each site and defined a priori to be either
Clumped or Sparse habitat, according to the qualitative descriptions given above. In each
of these areas, four replicate 1 m  1 m quadrats were randomly placed. Using a 100-cm
tape measure, the substratum covered by Pyura and the substratum not covered by Pyura
were measured along four transects set 20 cm from each other (i.e. 20, 40, 60 and 80 cm
from the top of the frame of the quadrat; Fig. 1c). When either category extended beyond
the frame of the quadrat, the length outside the quadrat was also measured, up to a
maximal distance of 1 m away from the frame (distances greater than 1 m were considered
to be outside the patch being measured).
These measurements were used to estimate three variables to characterize the structure
of each patch of habitat: (i) percentage cover of Pyura; (ii) size of patches of substratum
covered by Pyura and (iii) size of patches of substratum without Pyura. Percentage cover
was calculated based on the total length of Pyura measured in each quadrat and the total
possible area (i.e. 400 cm).
To sample the associated assemblages at the scale of patch (i.e. 1 m  1 m quadrat), the
cover and diversity of the different species of animals and plants were measured by
recording the percentage cover of algae and sessile animals and numbers of mobile
animals in four, randomly placed 30 cm  30 cm sub-quadrats. Percentage covers were
recorded under 100 points in each sub-quadrat. Any sessile organism present, but not
recorded under a point of intersection, was given a cover of 0.5%. Any organism not
identified in the field was taken to the laboratory for identification. Organisms were
generally identified to species.
2.2. Individual Pyura
In order to sample the assemblages at the scale of individual Pyura, five Isolated
(standing by themselves separated from other individuals) and five Clumped Pyura (part
of a clump and surrounded by other Pyura) were collected from each of two sites (CB and
HF; Fig. 2). The individuals collected were randomly selected from those available at each
site. Clumped individuals were collected from Sparse or Clumped habitats, whenever
possible and from different clumps, so that the widest possible range of individuals was
sampled. Isolated individuals were only collected from Sparse habitat (there were no
Isolated individuals in Clumped habitat). Each individual was removed from the rock by
wedging a knife between the bottom of its test and the substratum and using it as a lever.
To avoid the possible loss of any associated animals during collection, a plastic bag was
put around each individual prior to removal. The samples were then taken to the laboratory
and preserved in 7% formalin.
The covers of algae and sessile animals on the tests of each Pyura were estimated
using a 3 cm  3 cm quadrat with 10 random points (five replicate quadrats were
sampled per Pyura; the average area of a typical Pyura is 180 cm2). Mobile animals
were removed by placing the Pyura in a 500-Am sieve and washing it under running
water. When no more animals were found in the sieve, the Pyura was examined under
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 177
a microscope. Any animals still present were removed with forceps. All animals were
then identified and counted. Algae and sessile animals were identified to species, but
mobile animals were only identified to family. This taxonomic resolution allowed
spatial patterns of these taxa to be identified and saved considerable time during
sorting.
3. Results
3.1. Two habitats
Clumped and Sparse habitats were compared, using analysis of variance, in terms of
percentage cover of Pyura and sizes of patches covered by Pyura or covered by other
Table 1
Summary of analyses comparing (a) the average percentage cover of Pyura at three sites, at Cape Banks;
variances were homogenous (Cochran’s test, p > 0.05); (b) the mean size of patch of substratum not covered by
Pyura at three sites, at Cape Banks. Variances were heterogeneous, so data were transformed to Ln(x + 1); (c) the
mean size of patch of Pyura at three sites, at Cape Banks. Variances were heterogeneous, so data were
transformed to Ln(x + 1)
Source of variation df Mean squares F-ratio P
(a)
S 2 45.55 1.68 ns
H 1 2375.06 13.39 ns
Q(S Â H) 18 27.05 1.43 ns
SÂH 2 177.43 6.56 *
Residual 72 18.86
SNK tests (‘‘ = ’’ not significant at P = 0.05); Site 1: C p S; Site 2: C p S; Site 3: C p S
(b)
S 2 2.92 2.74 ns
H 1 25.35 10.33 ns
Q(S Â H) 18 1.07 2.50 **
SÂH 2 2.45 2.30 ns
Residual 72 0.43
(c)
S 2 2.19 2.81 ns
H 1 1.85 0.08 ns
Q(S Â H) 18 0.78 1.55 ns
SÂH 2 22.24 28.51 ***
Residual 72 0.50
SNK tests (‘‘ = ’’ not significant at P = 0.05); Site 1: C = S; Site 2: C p S; Site 3: C p S
S = Site (three levels; random), H = Habitat, i.e. Clumped versus Sparse (two levels; fixed), Q(S Â H) = Quadrats
(four levels in each habitat at each site; random). n = 4 replicate transects.
ns, p > 0.05; *p < 0.05; **p > 0.01; ***p > 0.001.
178 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
substratum. In cases where Cochran’s tests were significant, data were transformed
(Tables 1 and 2). The hypothesis that patches of Pyura would fall into two structural
groups was tested at three sites in Cape Banks (HF, CB and P, see Fig. 2). The
generality of the findings was then tested at the three other locations around Cape Banks
(Fig. 2).
Even though the difference varied among sites (S Â H interactions in Tables 1a and 2a),
the cover of Pyura was greater in Clumped than in Sparse habitat at all sites sampled at
Cape Banks and at all other locations. Across all locations, the cover of Pyura was always
greater in Clumped habitat (Fig. 3a,e). The size of patch of Pyura in Clumped habitat
varied more from site to site and from location to location. At two out of three sites at Cape
Banks, patches of Pyura were larger in Clumped habitat (Fig. 3b,e), but this was not
significant (Table 1b). At all other locations sampled, there were significant differences
between habitats (Table 2b). The size of patches of substratum other than Pyura was
consistently greater in Sparse habitat (Fig. 3c,f ). At Cape Banks, there was a site  habitat
interaction. Patches without Pyura were smaller in Clumped than in Sparse habitats at two
Table 2
Summary of the analyses comparing (a) the average percentage cover of Pyura at three locations; variances were
homogenous (Cochran’s test, p > 0.05); (b) the mean size of patch of substratum not covered by Pyura at three
locations. Variances were heterogeneous, but data could not be transformed; (c) the mean size of patch of Pyura at
three locations. Variances were heterogeneous, so data were transformed to Mx
Source of variation df Mean squares F-ratio P
(a)
L 2 622.17 1.49 ns
H 1 22 330.89 53.5 ***
Q(L Â H) 12 417.42 1.81 ns
LÂH 2 1764.39 4.23 *
Residual 54 230.13
SNK tests (‘‘ = ’’ not significant at P = 0.05); Loc 1: C = S; Loc 2: C p S; Loc 3: C p S
(b)
L 2 6.05 1.72 ns
H 1 78.12 22.17 **
Q(L Â H) 12 3.52 2.42 *
LÂH 2 10.39 2.95 ns
Residual 54 1.45
(c)
L 2 1217.06 2.10 ns
H 1 8975.95 15.52 **
Q(L Â H) 12 578.44 2.41 *
LÂH 2 1463.03 2.53 ns
Residual 54 240.12
L = Locations (three levels; random), H = Habitat, i.e. Clumped versus Sparse (two levels; fixed),
Q(L Â H) = Quadrats (three levels; random). n = 4 replicate transects.
ns, p > 0.05; *p < 0.05; **p > 0.01; ***p > 0.001.
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 179
Fig. 3. Mean ( + SE): (a,d) percentage cover of Pyura; (b,e) size of patches of substratum covered by Pyura; (c,f )
size of patches of substratum not covered by Pyura, (n = Clumped, 5 = Sparse). HF, CB and P are three sites at
Cape Banks (a – c); LB (Long Bay), LC (Little Congwong) and C (Congwong) are three other shores.
out of three sites (SNK tests and Table 1c). At all remaining locations, the patches of
substratum not covered by Pyura were smaller. Using just one of these variables to classify
habitats would not be useful, but combining all three variables makes the distinction
between Clumped and Sparse habitats easier. The cover of Pyura and the sizes of patches
of Pyura were greater in Clumped habitat, while the sizes of patches of other substratum
were greater in Sparse habitat.
180 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
Fig. 4. nMDS plot of assemblages associated with Pyura stolonifera at the scale of patch, from the three sites at
Cape Banks combined (circles, Clumped habitat; triangles, Sparse habitat). There were four quadrats in three
patches of each type of habitat (different shading of symbols represents the three sites).
3.2. Assemblages in the two habitats
Assemblages associated with the two habitats were different at all sites at Cape Banks,
but this difference was only significant in ANOSIM tests (on ranked Bray – Curtis
dissimilarities, Clarke, 1993; P < 0.06) at two of the sites (HF: ANOSIM R = 0.08,
P < 0.06; CB: R = 0.27, P < 0.0005; P: R = 0.53, P < 0.0005). To test the generality of these
Table 3
The contribution towards Bray – Curtis measures of dissimilarity for the most important species in each site,
between assemblages in Clumped versus Sparse habitats
Dissimilarity
Clumped versus Sparse habitat
Site 1 Site 2 Site 3 All sites
Bare % (Rank) 2 (13) 0.3 (10) 7 (5) 3 (6)
Corallina officinalis % (Rank) 21 (1=) 26 (1) 42 (1) 29 (1)
Non-geniculate Corallina % (Rank) 6 (3=) 20 (2) 4 (7) 11 ( = 3)
Sargassum sp. % (Rank) 4 (5) 2 (7=) 1 (9) 2 (7=)
Hildenbrandia rubra % (Rank) 21 (1=) 7 (4) 9 (3) 13 (2)
Ralfsia verrucosa % (Rank) 13 (2) 16 (3) 8 (4) 11( = 3)
Laurencia sp. % (Rank) 1 (14=) 4 (5=) 0.1 (27) 2 (7=)
Tesseropora rosea % (Rank) 1 (14=) 2 (7=) 12 (2) 5 (5)
Montfortula rugosa % (Rank) 6 (3=) 4 (5=) 6 (6) 6 (4)
Small black siphonarians % (Rank) 4 (4) 3 (6) 0 (36) 2 (7=)
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 181
patterns across Cape Banks, the data from all three sites were combined. Assemblages
associated with Clumped and Sparse habitats were significantly different (R = 0.19,
P < 0.0005; Fig. 4). Cover of Pyura was not included in these analyses because Pyura
was considered the provider of habitat. Percentage covers of algae and sessile animals were
corrected after the removal of the cover of Pyura. Where significant patterns were detected,
SimDiss (an equivalent procedure to SIMPER, Primer package, Clarke and Warwick, 1994)
was used to determine the contribution of different species to those patterns.
Fig. 5. Mean ( + SE): (a) number of mobile species; (b) mean number of sessile species; (c) mean total number of
species associated with habitats of Pyura, at the three sites sampled at Cape Banks (n = Clumped, 5 = Sparse).
182 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
Nineteen species of algae and 45 species of animals were found in the habitats provided
by Pyura, across the three sites. Molluscs were the dominant group. Others included
sponges, echinoderms, anemones and crustaceans. Of the 64 taxa, only a small group was
responsible for the majority of the dissimilarities between assemblages (Table 3). Across
all sites, Corallina contributed most to the differences. Others included the encrusting
algae Hildenbrandia and Ralfsia. All these taxa are generally associated with the space
among Pyura and tended to be more common in Sparse habitat. The fissurellid limpet
Montfortula rugosa was the only mobile animal that contributed more than 5% to the
Fig. 6. Mean ( + SE): (a) percent cover of Corallina officinalis; (b) number of Cellana tramoserica; (c) number of
Thais orbita; (d) number of Cabestana spengleri found in habitats of Pyura, at the three sites sampled at Cape
Banks (n = Clumped, 5 = Sparse).
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 183
differences between assemblages. This species was present in both habitats and found on
Pyura and on the surrounding area.
Some species varied from one habitat to the other, but the numbers of taxa in each
habitat appeared to be similar. In light of this, the hypotheses that the number of mobile,
sessile and the total number of taxa were similar between the habitat were tested. The
numbers of mobile and sessile species and the total number of species were similar in each
habitat (Fig. 5a– c). Some species, such as C. officinalis and the limpet Cellana tramoserica
were usually associated with the spaces amongst Pyura and were more abundant in Sparse
habitat (Fig. 6a,b). The whelks Thais orbita and Cabestana spengleri were generally
found attached to Pyura. Thais was found in both habitats and had no specific pattern
of abundance (Fig. 6c). Cabestana, on the other hand, were mostly found in Clumped
habitat (Fig. 6d).
3.3. Assemblages on individual Pyura
Eighty-four taxa were found living on the tests of individual Pyura. Again, molluscs
were the dominant group, although several species of algae, polychaetes and isopods were
also present. The differences (Fig. 7) between the types of Pyura (for HF and CB,
respectively, ANOSIM R = 0.60, P < 0.0008; R = 0.59, P < 0.008) were mainly due to
limpets of the families Siphonariidae and Acmaeidae, bivalves of the family Erycinidae
and micro-gastropods of the family Eatoniellidae. Analysis of variance was initially used
to test the hypothesis that abundances of these taxa differed between Isolated and Clumped
Fig. 7. nMDS plot of assemblages on individual Pyura. Five individuals were collected from Clumped or Sparse
habitat at each of the two sites (C = Clumped Pyura; I = isolated Pyura) to close square.
184
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
Table 4
Analyses of numbers and frequencies of occurrence of types of gastropods on Pyura
Taxon Site Individual Pyura Clumped Pyura v2 (a)
(1 df ) P v2 (b)
(1 df ) P Fi Fc
No. occupied No. individuals No. occupied No. individuals
Siphonaridae 1 4 99 1 13 1.8 ns 66.0 ***
2 5 144 0 0 5.0 144.0 ***
1+2 0.5 1.00
Acmaeidae 1 4 77 2 5 0.7 ns 63.2 ***
2 5 88 3 5 0.5 ns 74.1 *
1+2 0.66 0.5
Erycinidae 1 5 557 5 127 0.0 ns 270.3 ***
2 5 173 5 142 0.0 ns 3.0 ns
1+2 0.67 0.5
Eatoniellidae 1 5 280 5 102 0.0 ns 82.9 ***
2 4 54 5 22 0.1 ns 13.5 ***
1+2 0.59 0.67
For each family of gastropods, at each site, v2 (a) tested the null hypothesis that the frequencies of Isolated and Clumped Pyura occupied by that family were equal; v2 (b)
tested the null hypothesis that the total number of that family on 10 Isolated Pyura equals the total number on 10 Clumped Pyura. Fi is the probability from Fisher’s exact
test of the null hypothesis that the proportion of Isolated Pyura occupied by that family in site 1 equals the proportion occupied in site 2; Fc is the same test for Clumped
individuals.
*p < 0.05; ***p < 0.001.
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 185
Pyura, but the heterogeneity of the data did not allow conclusive results and the
transformation of data did not solve this problem. To overcome this difficulty, three
different hypotheses were tested: (i) the proportion of Isolated Pyura occupied by taxon
i p proportion of Clumped Pyura occupied by taxon i; (ii) the proportion of taxon i on
Isolated Pyura p proportion of taxon i on Clumped Pyura; (iii) the proportion of Clumped
or Isolated Pyura with taxon i at site 1 = proportion of Clumped or Isolated Pyura with
taxon i at site 2. In the cases, where the expected values (under the null hypothesis of no
difference) were smaller than 5, Fisher’s exact tests were used. The tests were done for the
taxa (chosen on the basis of SIMDISS results) that accounted for most of the differences
between the types of Pyura.
For all four taxa tested, with the exception of Siphonariidae at CB, the proportions of
Isolated and Clumped Pyura occupied by taxon were similar (Table 4). At each site,
however, more individuals were found on Isolated than on Clumped Pyura. The only
exception to this trend was Erycinidae at CB. The proportions of Isolated and Clumped
Pyura occupied by each taxon were similar at the two sites.
4. Discussion
Pyura are a conspicuous and abundant component of rocky shores across its area of
distribution in Australia (Kott, 1985; Otway, 1989; Edgar, 1997). Their presence provides
added surface area for the settlement of other organisms, and by ameliorating environ-
mental conditions they may provide different microhabitats for those organisms. Living on
and amongst P. stolonifera may allow animals and plants to get the benefits of a high
energy environment, such as oxygenated water or continuous flow of food, without being
subjected to potentially destructive variables as strong wave-action or currents (Fielding et
al., 1994). Pyura may therefore be considered an example of an intertidal ecosystem
engineer (Jones et al., 1994).
Patches of other substratum (generally occupied by different species of foliose and/or
encrusting algae) in habitats of Pyura fall under the category of Type I patches — areas of
habitat surrounded by a matrix of some other habitat (Connell and Keough, 1985), in this
case Pyura. Patches of Pyura can, on the other hand, be considered Type II patches
because they are a different substratum and provide an additional area for the settlement of
other organisms (Connell and Keough, 1985).
As originally predicted, habitats in beds of Pyura separated into two different structural
types based on a combination of percentage cover and relative sizes of patches of Pyura
and patches of other substratum. Samples were randomly chosen, at all sites and locations
so that the widest possible range of patches was sampled. This also assured that if there
were any overlapping between the two types of habitat, it would be identified in the
analyses. There was only a handful of cases where a patch could have possibly been
considered intermediate but in every such case statistical analyses showed them to be
either Clumped or Sparse and not intermediate.
Clumped habitats were characterized by a large cover of Pyura and by larger patches of
Pyura in relation to those of other substratum. Sparse habitats, on the other hand, were
characterized by a small cover of Pyura, with patches of other substratum being on
186 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
average larger than those of Pyura. These patterns were consistent across all locations
sampled. The fact that no intermediate areas were found and the generality of the patterns
observed support the original idea that areas of the shore where Pyura is found can be
classified either as Clumped or Sparse habitat. The differences between Clumped and
Sparse habitats are presumably the result of combinations of differences in dispersal, rates
of settlement or post-settlement mortality, including differences due to disturbances.
Assemblages can vary at different spatial scales and the type of patterns observed at one
scale may or may not be present at a different one emphasizing the need to sample at
different spatial scales in order to reach a better assessment of the diversity in a given area
(Thompson et al., 1996). In the present study, the scales chosen were patches of Pyura and
individual Pyura.
Various studies have shown that different intertidal habitats are associated with different
assemblages (e.g. Hawkins, 1981; Lohse, 1993; Thompson et al., 1996). This was also the
case for the habitats of Pyura at Cape Banks. Assemblages were different at all three sites.
When the data from all sites were combined, the assemblages were significantly different.
Species normally associated with bare spaces, such as the algae Corallina and Hilden-
brandia, were responsible for most of the differences between the assemblages in the two
habitats. The algae Corallina and the limpet Cellana were more abundant in Sparse habitat,
which can possibly be explained by the greater availability of substratum not covered by
Pyura. Corallina may settle and grow better on rock (at all sites Corallina was mostly
found on the space between Pyura rather than on the Pyura themselves). C. tramoserica
are grazers and the algae they tend to feed on are more common on the substratum that
surrounds the Pyura. Species like the whelks T. orbita and C. spengleri were generally
found attached to Pyura and were more frequent in Clumped habitat.
As was the case at the scale of patch, the assemblages were different at the scale of
individual Pyura, supporting the idea that two different habitats were present in the beds of
Pyura. Differences at this scale were mostly due to differences in the abundances of
organisms rather than the number or types of taxa. What varied between Isolated and
Clumped Pyura were the numbers of animals present and the proportions of Pyura
occupied by those organisms rather than the types of animals. In this case, contrary to what
happened at the scale of patch, where differences were mainly due to algae, the taxa
responsible for most of the differences were mobile molluscs (three families of gastropods
and one family of bivalves).
The results of this study suggest that the variables used, percentage cover of Pyura, the
size of the patches of substratum covered and not covered by Pyura, and whether Pyura
are isolated or in clumps were appropriate for identifying patterns in the structure of
habitat and assemblages associated with the beds of Pyura around Sydney.
Pyura is commonly killed and used as bait by recreational fishermen (Underwood and
Kennelly, 1990) in Australia. Fishermen usually cut off the top of the test and use the
insides as bait. The portion still left attached to the rock usually disappears after about two
weeks. This type of collection can be selective; larger or more readily accessible
individuals are preferred. The numbers of Pyura can be seriously reduced by this type
of collection (Fairweather, 1991) and the structure of the patches of Pyura can also be
affected. A reduction in the number and cover of Pyura may lead to changes in the
structure of the habitat available for other organisms. Removal of Pyura at Cape Banks, by
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 187
fishermen or storms, has been shown to result in marked changes to the assemblages in
low-shore areas (Otway, pers. comm.).
Changes in the structure of a habitat may affect the associated assemblages in different
ways, so, in order to be able to predict the possible effects of such changes, it is important
to know the underlying natural patterns in those assemblages. Possible responses of the
assemblages to changes in the structure of the habitat can include decreases or increases in
abundances of some or all species and/or increases or decreases in the number of species.
Some species can disappear and new ones can replace them or be added to an assemblage.
In the case of habitats provided by Pyura, there are differences in the associated
assemblages, depending on whether the ascidians are clumped or sparsely scattered on
the shore. To determine the extent to which changing the arrangement of the Pyura
themselves may lead to changes in associated assemblages requires manipulative experi-
ments. In order to be able to plan and design such experiments, the natural patterns of
occurrence of the assemblages associated with Pyura must be known. Describing those
patterns quantitatively is the important first step in that understanding.
Acknowledgements
This study was supported by funds from the Australian Research Council through the
Centre for Research on Ecological Impacts of Coastal Cities. S.M. Monteiro was sup-
ߘ ˆ
ported by a scholarship from the Programa Praxis XXI of the Fundacao para a Ciencia e
Tecnologia, Lisbon, Portugal. We thank Dr. P. Archambault and two anonymous referees
for comments on earlier versions of this manuscript and numerous colleagues in the Centre
who helped with fieldwork. [RW]
References
Bergeron, P., Bourget, E., 1986. Shore topography and spatial partitioning of crevice refuges by sessile epibenthos
in an ice disturbed environment. Mar. Ecol. Prog. Ser. 1 – 2 (28), 129 – 145.
Bourget, E., DeGuise, J., Daigle, G., 1994. Scales of substratum heterogeneity, structural complexity, and the
early establishment of a marine epibenthic community. J. Exp. Mar. Biol. Ecol. 181 (1), 31 – 51.
Brown, V.K., 1991. In: Bell, S.S., McCoy, E.D., Mushinsky, H.R. (Eds.), The Effects of Changes in Habitat
Structure During Succession in Terrestrial Communities. Chapman & Hall, New York, USA, pp. 141 –
168.
Butler, A.J., 1991. Effect of patch size on communities of sessile invertebrates in Gulf St. Vincent, South
Australia. J. Exp. Mar. Biol. Ecol. 153, 255 – 280.
Chapman, M.G., Underwood, A.J., 1994. Dispersal of the intertidal snail, Nodilitorina pyramidalis, in response to
the topographic complexity of the substratum. J. Exp. Mar. Biol. Ecol. 179 (2), 145 – 169.
Clarke, K.R., 1993. Non-parametric multivariate analysis of changes in community structure. Austr. J. Ecol. 18,
117 – 143.
Clarke, K.R., Warwick, R.M., 1994. Change in Marine Communities: An Approach to Statistical Analysis and
Interpretation. Plymouth Marine Laboratory, Plymouth, 144 pp.
Connell, J.H., 1972. Community interactions on marine rocky intertidal shores. Annu. Rev. Ecol. Syst. 3, 169 –
192.
Connell, J.H., Keough, M.J., 1985. In: White, P.S., Pickett, S.T.A. (Eds.), Disturbance and Patch Dynamics of
Subtidal Marine Animals on Hard Substratum. Academic Press, New York, USA, pp. 125 – 151.
188 S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189
Dayton, P.K., 1971. Competition, disturbance and community organisation: the provision and subsequent uti-
lisation of space in a rocky intertidal community. Ecol. Monogr. 41 (4), 351 – 389.
Edgar, G.J., 1997. Australian marine life — the plants and animals of temperate waters. Australia, Reed Books,
544 pp.
Eggleston, D.B., Elis, W.E., Etherington, L.L., Dahlgren, C.P., Posey, M.H., 1999. Organism responses to habitat
fragmentation and diversity: habitat colonization by estuarine macrofauna. J. Exp. Mar. Biol. Ecol. 236 (1),
107 – 132.
Fairweather, P.G., 1991. A conceptual framework for ecological studies of coastal resources: an example of a
tunicate collected for bait on Australian seashores. Ocean Shoreline Manage. 15, 125 – 142.
Fielding, P.J., Weerts, K.A., Forbes, A.T., 1994. Macroinvertebrate communities associated with intertidal and
subtidal beds of Pyura stolonifera (Heller) (Tunicata: Ascideacea) on the Natal coast. S. Afr. J. Zool. 29 (1),
46 – 53.
Gaines, S., Roughgarden, J., 1985. Larval settlement rate: a leading determinant of structure in an ecological
community of the marine intertidal zone. Proc. Natl. Acad. Sci. U. S. A. 82, 3707 – 3711.
Hamazaki, T., 1996. Effects of patch shape on the number of organisms. Landscape Ecol. 11 (5), 299 – 306.
Hawkins, S.J., 1981. The influence of Patella grazing on fucoid-barnacle mosaic on moderately exposed rocky
shores. Proceedings of the 15th European Marine Biology Symposium: Lower Organisms and Their Impor-
tance in the Food Web. Kiel. Meeresforsch., Sonderh. vol. 5, pp. 537 – 543.
Jones, C.G., Lawton, J.H., Shachak, M., 1994. Organisms as ecosystem engineers. Oikos 69 (3), 373 – 386.
Keough, M.J., 1984. Effects of patch size on the abundance of sessile marine invertebrates. Ecology 65 (2), 423 –
437.
Kohn, A.J., Leviten, P.J., 1976. Effect of habitat complexity on population density and species richness in a
tropical intertidal predatory gastropod assemblage. Oecologia 25 (3), 199 – 210.
Kott, P., 1985. The Australian ascidiacea: Part I. Phlebobranchia and stolidobranchia. Mem. Queensl. Mus. 23,
328 – 331.
Lemire, M., Bourget, E., 1996. Substratum heterogeneity and complexity influence microhabitat selection of
Balanus sp. and Tubularia crocea larvae. Mar. Ecol. Progr. Ser. 135, 77 – 87.
Lintas, C., Seed, R., 1994. Spatial variation in the fauna associated with Mytilus edulis on a wave exposed rocky
shore. J. Molluscan, 165 – 174.
Lohse, D.P., 1993. The importance of secondary substratum in a rocky intertidal community. J. Exp. Mar. Biol.
Ecol. 166, 1 – 17.
Mauchamp, A., Rambal, S., Lepart, J., 1994. Simulating the dynamics of a vegetation mosaic — a spatialized
functional model. Ecol. Modell. 1 – 3, 107 – 130.
McCoy, E.D., Bell, S.S., 1991. Habitat structure: the evolution and diversification of a complex topic. In: Bell,
S.S., McCoy, E.D., Mushinsky, H.R. (Eds.), Habitat Structure: The Physical Arrangement of Objects in
Space. Chapman & Hall, London, pp. 3 – 27.
˜
Montana, C., 1992. The colonization of bare areas in two phase mosaics of an arid ecosystem. J. Ecol. 80, 315 – 327.
Ojeda, F.P., Dearborn, J.H., 1989. Community structure of macroinvertebrates inhabiting the rocky subtidal zone
in the Gulf of Maine: seasonal and bathymetric distribution. Mar. Ecol. Progr. Ser. 57 (3), 147 – 161.
Otway, N.M., 1989. The effects of grazing by chitons on mid and low shore intertidal communities. PhD Thesis,
University of Sydney.
Paine, R.T., Levin, S.A., 1981. Intertidal landscapes: disturbance and the dynamics of pattern. Ecological Mono-
graphs 51 (2), 145 – 178.
Paine, R.T., Suchanek, T.H., 1983. Convergence of ecological processes between independently evolved com-
petitive dominants: a tunicate – mussel comparison. Evolution 37 (4), 821 – 831.
Ramage, D.L., Schiel, D.R., 1999. Patch dynamics and response to disturbance of the seagrass Zostera nova-
zealandica on intertidal platforms in southern New Zealand. Mar. Ecol. Prog. Ser. 189, 275 – 288.
Seed, R., 1996. Patterns of biodiversity in the macro invertebrate fauna associated with mussel patches on rocky
shores. J. Mar. Biol. Assoc. U. K. 76, 203 – 210.
Sousa, W.P., 1984. Intertidal mosaics: patch size, propagule availability and spatially variable patterns of suc-
cession. Ecolology 65 (6), 1918 – 1935.
Sousa, W.P., 1985. In: White, P.S., Pickett, S.T.A. (Eds.), Disturbance and Patch Dynamics on Rocky Intertidal
Shores. Academic Press, New York, USA, pp. 101 – 124.
S.M. Monteiro et al. / J. Exp. Mar. Biol. Ecol. 270 (2002) 171–189 189
Thompson, R.C., Wilson, B.J., Tobin, M.L., Hill, A.S., Hawkins, S.J., 1996. Biologically generated habitat
provision and diversity of rocky shore organisms at a hierarchy of spatial scales. J. Exp. Mar. Biol. Ecol.
202, 73 – 84.
Underwood, A.J., Chapman, M.G., 1989. Experimental analyses of the influences of topography of the sub-
stratum on movements and density of an intertidal snail, Littorina unifasciata. J. Exp. Mar. Biol. Ecol. 134,
175 – 196.
Underwood, A.J., Chapman, M.G., 1992. Experiments on topographic influences on density and dispersion of
Littorina unifasciata in New South Wales. In: Grahame, J., Mill, P.J., Reid, D.G. (Eds.), Proceedings of the
Third International Symposium on Littorinid Biology. The Malacological Society of London, London, pp.
181 – 195.
Underwood, A.J., Kennelly, S.J., 1990. Pilot studies for design of surveys of human disturbance of intertidal
habitats in New South Wales. Aust. J. Freshwater Res. 41, 165 – 173.
Underwood, A.J., Denley, E.J., Moran, M.J., 1983. Experimental analyses of the structure and dynamics of mid
shore intertidal communities in New South Wales. Oecologia 56, 202 – 219.
Van Driel, C.D., Steyl, C.D., 1976. Proceedings of the first interdisciplinary conference on marine and freshwater
research in Southern Africa. Port Elizabeth, 5 – 10 July.
Veblen, T.T., 1985. Stand dynamics in Chilean Nothofagus forests. In: White, P.S., Pickett, S.T.A. (Eds.), The
Ecology of Natural Disturbance and Patch Dynamics. Academic Press, New York, USA, pp. 35 – 51.
White, P.S., Pickett, S.T.A., 1985. In: White, P.S., Pickett, S.T.A. (Eds.), Natural Disturbance and Patch Dynam-
ics: An Introduction. Academic Press, New York, USA, pp. 3 – 13.
Wu, J.G., Levin, S.A., 1994. A spatial patch dynamics modelling approach to pattern and process in an annual
grassland. Ecol. Monogr. 4, 447 – 464.