Full text of DOI:10.1017/S0025315401004842

J. Mar. Biol. Ass. U.K. 92001), 81, 917^930
Printed in the United Kingdom
Spatial patterns of diverse macrofaunal assemblages
in coralline turf and their associations with
environmental variables
B.P. Kelaher*, M.G. Chapman and A.J. Underwood
Centre for Research on Ecological Impacts of Coastal Cities, Marine Ecology Laboratories A11,
University of Sydney, NSW, Australia, 2006. *Present address: Department of Ecology and Evolution,
Life Sciences Building, State University of NewYork, Stony Brook, NY 11974-5245, USA.
E-mail: bkelaher@life.bio.sunysb.edu
Mats of articulated coralline algal turf are common on many rocky intertidal shores. The dense fronds
provide a habitat for extremely diverse and abundant macrofaunal assemblages. Despite a large contribu-
tion to faunal biodiversity of rocky shores, little has been published about these assemblages. This study
describes patterns of distribution and abundance of macrofauna in coralline turf on rocky shores around
Sydney. In addition, the potential of environmental variables 9sediment, epiphytes, length and density of
coralline fronds) for determining these patterns was also investigated. Relatively consistent di¡erences were
found between macrofauna in low- and mid-shore areas at all times of sampling and on all shores.
Although there was some variation among shores, there was generally signi¢cant variation in macrofauna
between sites separated by tens of metres. Generally, a relatively small number of taxa were responsible for
the great majority of dissimilarity between assemblages. Apart for the small bivalve Lasaea australis,
however, these taxa varied between heights on the shore, among times of sampling and among shores.
These data illustrate the important contribution that coralline turf has for biodiversity of faunal assem-
blages on rocky shores around Sydney. They also provide a basis for investigating biological processes and
physical factors responsible for structuring patterns of biodiversity of macrofaunal assemblage in coralline
turf.
INTRODUCTION
have been determined experimentally 9Denley
&
Underwood, 1979; Underwood et al., 1983; Fairweather,
1985; Chapman & Underwood, 1994). The majority of
this research has focused on large and conspicuous
animals and algae. Rocky shores, however, have many
complex mat-like habitats with extremely diverse assem-
blages of small and cryptic invertebrates 9Hicks, 1971;
Myers & Southgate, 1980; Seed, 1996). In contrast to the
more conspicuous organisms on rocky shores, published
studies on these cryptic macrofaunal assemblages are rela-
tively scarce 9Underwood & Petraitis, 1993; Menge, 1995).
Common forms of mat-like habitats on rocky shores
include mussel beds 9Tsuchiya & Nishihira, 1985; Peake
& Quinn, 1993; Seed, 1996), ascidians 9Fielding et al.,
1994; Otway, 1994), algal turf 9Hicks, 1980; Myers &
Southgate, 1980; Dean & Connell, 1987a,b) and lichens
9Seed & O'Connor, 1980; Healy, 1996).
Environmental variables relating to the habitat are
generally considered to have a strong in£uence on macro-
faunal assemblages in mat-like habitats 9Hicks, 1980; Dean
& Connell, 1987a; Johnson & Scheibling, 1987; Gibbons,
1988; Gee & Warwick, 1994). For example, the amount of
sediment within a mat-like habitat often increases the
diversity and abundance of associated macrofauna
9Wigham, 1975; Hicks, 1980; Gibbons, 1988, 1991; Whor¡
et al., 1995). It has been postulated that this is because
sediment creates a more heterogeneous habitat for infaunal
and epifaunal species 9Hicks, 1985; Gibbons, 1991) and
The quantitative description of patterns of distribution
and abundance is the basis of understanding processes
that structure assemblages of organisms 9Andrew &
Mapstone, 1987; Underwood et al., 2000). There has been
a long history of description of patterns of distribution and
abundance of organisms on rocky intertidal shores 9see
Southward, 1958; Connell, 1972, for reviews). In the past,
much of this research focused on variation in assemblages
of organisms along obvious environmental gradients, such
as among di¡erent heights on the shore 9e.g. Connell, 1961;
Dayton, 1971; Paine, 1974) or among shores with di¡erent
exposure to waves 9e.g. Crisp & Southward, 1958; Lewis,
1964) and often small-scale variation was overlooked 9but
see Connell, 1972). More recently, it has been shown that a
considerable amount of variability in abundances of organ-
isms on rocky shores occurs at scales of tens of metres or less
9Lohse, 1993; Chapman, 1994; Underwood & Chapman,
1996). Consequently, a greater emphasis is being placed
on describing and explaining variation in the diversity
and abundance of organisms at a range of di¡erent
spatial scales 9Morrisey et al., 1992; Thompson et al.,
1996; Underwood & Chapman, 1996, 1998).
The spatial scales of variability of many organisms on
rocky shores of south-eastern Australia have been well-
described 9Underwood & Chapman, 1996). In many
cases, the major processes responsible for these patterns
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

918 B.P. Kelaher et al. Special patterns of macrofauna in coralline turf
retains moisture when the habitat is emersed 9Gibbons,
1988). Similarly, the amount of epiphytic algae attached
to a mat-like habitat increases the diversity of the asso-
ciated faunal assemblage by providing food and shelter
9Johnson & Scheibling, 1987; Du¡y, 1990; Martin-Smith,
1993). In addition, the amount and complexity of physical
structure of mat-like habitats may increase the diversity of
associated organisms by providing greater available
surface area 9Connor & McCoy, 1979; McGuinness &
Underwood, 1986), providing a better refuge from pre-
dation 9Russ, 1980; Coull & Wells, 1983; Dean &
Connell, 1987b), reducing environmental harshness 9e.g.
increase relative humidity; Nixon et al., 1971) or reducing
the impact of disturbances 9e.g. reduce e¡ects of wave
action; Dommasnes, 1968; Whor¡ et al., 1995).
Articulated coralline algae often form mat-like habitats
and are a major component of algal assemblages on many
rocky shores 9Stewart, 1982; Grahame & Hanna, 1989;
Dye, 1993; Benedetti-Cecchi & Cinelli, 1994). They are
extremely abundant in low-shore areas and relatively
abundant in mid-shore areas of wave-exposed and
sheltered rocky shores in New South Wales, Australia
9Underwood, 1981; Underwood & Chapman, 1998). The
dense fronds provide a habitat for extremely diverse macro-
faunal assemblages 9Hicks, 1971; Akioka et al., 1999) and
may contain in excess of 200,000 individuals m72 9Brown
& Taylor, 1999). Given the abundance of coralline algal
turf and diversity of the associated macrofaunal assem-
blage, it is clear that turfs generally support a large
component of faunal biodiversity of many rocky shores.
Despite this, there has only been one published study
investigating macrofaunal assemblages associated with
coralline turf in Australia 9Davenport et al., 1999, South
Australia) and only a handful world-wide 9Dommasnes,
1968, 1969; Hicks, 1971; Ballesteros, 1988; Grahame &
Hanna, 1989; Hull, 1997; Lopez & Stolz, 1997; Akioka et
al., 1999; Brown & Taylor, 1999). Of these studies, some
are limited to only one day of sampling 9e.g. Davenport et
al., 1999) or a single site 9e.g. Hull, 1997; Davenport et al.,
1999) or use very broad taxonomic groups 9e.g. Grahame
& Hanna, 1989; Brown & Taylor, 1999). Detailed quanti-
tative information on the spatial patterns at a range of
scales of diverse assemblages associated with coralline turf
is therefore conspicuously lacking.
The present study describes tests of the hypotheses that
macrofaunal assemblages in coralline turf: 9i) di¡er
between low- and mid-shore areas; and 9ii) di¡er at
scales of tens of metres at each height on the shore. This
study also investigated the potential for environmental
variables to explain the patterns by testing the hypothesis
that the amount of sediment and epiphytes associated with
the turf and the structural complexity of the coralline turf
will show similar patterns of spatial variation as does the
macrofaunal assemblage. In addition, it was hypothesized
that each of these environmental variables will be posit-
ively associated with the diversity of macrofauna. Finally,
it has been shown that supposed general patterns of distri-
bution of organisms can vary among times of sampling
and shores 9Foster, 1990; Underwood & Chapman, 1998).
In this study, the generality of spatial patterns of macro-
faunal assemblages in coralline turf was evaluated by re-
testing hypotheses about spatial variability three times at
one shore near Sydney, Australia and then at three other
shores in the Sydney region.
MATERIALS AND METHODS
Study sites and characterization of coralline turf
This study was mostly done at Cape Banks Scienti¢c
Marine Research Area 9hereafter called Cape Banks),
located on the northern headland of the entrance to
Figure 1. Shores sampled around Sydney, New South Wales, Australia.
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

Special patterns of macrofauna in coralline turf B.P. Kelaher et al. 919
Botany Bay 9348000S 1518150E; Figure 1). On this shore,
given area. The heights of mid-shore sites varied between
0.58 and 0.82 m above ILWS 9Indian Low Water Springs
at Fort Denison, Sydney). Low-shore sites were de¢ned as
the lowest patches of coralline turf that could be safely
sampled on a calm day with a low tide less than 0.30 m
above ILWS. Heights of low-shore sites varied between
0.21 and 0.42 m above ILWS. All sampling was done on
days with low swell, calm seas and a low tide less than
0.3 m above ILWS because it was not possible to sample
low-shore areas with the same care as mid-shore areas on
days with smaller tides or rough conditions.
The macrofauna in natural coralline turf was sampled
using a sharpened metal corer, which had an internal
diameter of 8 cm 9ꢀ50 cm2). This size of core was the
most appropriate compromise between the precision of
estimates of diversity and abundance and time required to
sort each core 9Kelaher, 2000). In each site, four
randomly-placed replicate cores were collected. For each
replicate, the corer was pushed into the coralline turf and
the algae and sediment inside the corer were scraped o¡
at the level of the rock, placed in a plastic bag and taken to
the laboratory. Because other species of algae were some-
times found amongst the coralline turf, only areas with
greater than 95% primary cover of Corallina o¤cinalis
were sampled. All replicates were preserved in a 7%
formalin solution.
In the laboratory, each replicate core was washed thor-
oughly in a 500-mm sieve. The fauna remaining in the
sieve was identi¢ed and counted using a binocular micro-
scope 9Â16 magni¢cation). Sessile animals that were
permanently attached to the fronds or substrata were
commonly found in the turf 9e.g. sponges, bryozoans,
serpulid and spirorbid polychaetes, barnacles, etc.). These
organisms were not included in this study because the
methods used were not appropriate to quantify these
animals accurately. In total, 71,776 individual animals
were counted and identi¢ed into 147 di¡erent taxa
9Table 1). Taxonomic resolution of macrofauna varied
exposure to waves, slope and aspect of platforms vary
from place to place 9Underwood et al., 1983; Fairweather
& Underwood, 1991). All the sites used in this shore had
medium to heavy exposure to waves. Three other rock
platforms were also sampled. From north to south, these
were Green Point 9338330S 1518180E), Harbord 9338470S
1518180E) and Coledale 9348110S 1508570E; Figure 1).
Harbord and Coledale are fully exposed to waves and
Green Point has an increasing gradient of wave exposure
from south to north 9Underwood, 1981). On this shore,
sites were selected in areas exposed to wave-action.
All shores had large patches of coralline turf in low- and
mid-shore areas. This turf was almost entirely Corallina
o¤cinalis Linnaeus. Nevertheless, at di¡erent places and
di¡erent times, Haliptilon roseum 9Lamarck) Garbary and
Johansen, Jania spp. Lamouroux and Amphiroa spp.
Lamouroux, were found in the tur¢ng matrix. Often,
non-coralline algae occupied small amounts of primary
space amongst the turf 9e.g. Sargassam spp. C. Agardh,
Hormosira banksii 9Turner) Decaisne). The coralline turf
was associated with a diverse epiphytic assemblage.
The most obvious and abundant epiphytic species were
Ulva lactuca Linnaeus, Endarachne binghamiae J. Agardh,
Colpomenia sinuosa 9Mertens ex Roth) Derbes and Solier
and Enteromorpha spp. Linnaeus. Many other species of
¢lamentous and foliose algae were found in small abund-
ances, but were extremely variable in space and time.
Sampling methods
At Cape Banks, patches of turf were sampled in May
1997, September 1997 and January 1998. At each time of
sampling, two sites 92Â4 m patches of coralline turf ) not
previously sampled were haphazardly selected in low-
and mid-shore areas. At each height, sites were approxi-
mately 50 70 m apart. The top of mid-shore areas was
^
de¢ned by the uppermost patches of coralline turf in any
Table 1. Summary of the 147 macrofaunal taxa found in coralline turf during this study.
Phylum
Class
No. of taxa and resolution
Cnidaria
Anthozoa
Turbellaria
^
1 Taxon
1 Taxon
1 Taxon
1 Taxon
1 Taxon
19 Families
1 Taxon
3 Species
1 Taxon
1 Taxon
1 Taxon
1 Taxon
29 Families*
6 Species
65 Species
11 Species
2 Species
1 Taxon
1 Species
Platyhelminthes
Nematoda
Nemertea
^
Annelida
Oligochaeta
Polychaeta
^
Sipuncula
Arthropoda
Insecta
Pycnogonida
Arachnida
Ostracoda
Copepoda
Malacostraca
Polyplacophora
Gastropoda
Bivalvia
Mollusca
Echinodermata
Asteroidea
Ophiuroidea
Echinoidea
*, 17 of the 29 families of malacostraceans were in the order Amphipoda.
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

920 B.P. Kelaher et al. Special patterns of macrofauna in coralline turf
among taxa because many of the species have not been
described in Australia and many of the animals were juve-
niles and could not be reliably identi¢ed to species using
taxonomic keys 9when keys were available).
To test the hypothesis that spatial patterns of macro-
fauna in coralline turf at Cape Banks are general to rocky
shores in the Sydney Region, the four shores described
above 9including Cape Banks) were sampled in summer,
on 27 and 28 January 1998. Each shore was sampled
using similar methods to those described for Cape Banks,
except ¢ve cores were taken from each site to increase the
power and resolution of analyses.
To test hypotheses about environmental variables, the
dry weight of sediment and epiphytes and the average
length and density of fronds were measured for each core
sampled at each of the four shores in January 1998. For
each core, the dry weight of epiphytes was measured by
carefully removing epiphytes from the coralline fronds.
The epiphytes were dried for 48 h at 608C, cooled in a
desiccator for four hours and weighed. Occasionally, non-
coralline algal species occupied primary space. It was not
possible to separate these from epiphytic algae in the
laboratory. Therefore, all non-coralline algae in each core
were de¢ned as epiphytes. For each core, the sediment
particles greater than 63 mm were collected, dried in an
oven for 48 h at 808C, cooled in a desiccator for four
hours and weighed. Because the sediment was relatively
coarse, the component of sediment less than 63 mm only
had a small contribution to the total weight 9less than
2%) and was therefore not quanti¢ed 9see also Gibbons,
1988).
The average length and density of fronds were used to
quantify characteristics of the physical structure of coral-
line turf. The average length of fronds for each core was
determined from measurements of length of four
randomly-selected fronds. Each frond measured had an
intact holdfast and was not broken. For each core, the
density of coralline turf 9gcm73) was determined by
dividing the dry weight of coralline fronds per cm2 by the
average length. To measure the dry weight of coralline
fronds, the epiphytes and animals on each frond were
carefully removed. The coralline fronds were then dried
and weighed using similar methods to those used for
epiphytes.
Analyses of data
Speci¢c hypotheses were tested for entire macrofaunal
assemblages and the number of taxa of polychaetes,
amphipods and gastropods. Gastropods, polychaetes and
amphipods were analysed separately because they were
the most diverse and dominant faunal groups 9Table 1).
Non-parametric multivariate analyses of variance
9NP-MANOVA) were used to test hypotheses about
Table 2. Results of NP-MANOVA analyses for macrofaunal assemblages sampled ꢀA) at Cape Banks and ꢀB) at the four shores
sampled in January 1998. Analyses were done ꢀi) comparing heights at each time or shore, or ꢀii) comparing times or shores at each
height. `he' is the ¢xed comparison between mid- and low-shore areas; `ti' is the comparison among randomly-chosen times; `sh' is the
comparison among randomly-chosen shores; `si' is the comparison of sites nested in height, times or shores.
A. Cape Banks.
Analyses
Post-hoc tests
mid-shore
9i) Analyses at each time
si9he)
df
he
df
low-shore
May 1997 9T1)
September 1997 9T2)
January 1998 9T3)
***
***
***
2, 12 df
2, 12 df
2, 12 df
**
**
*
1, 2 df
1, 2 df
1, 2 df
S1=S2
S1=S2
S1=S2
S1ˆS2
S1=S2
S1=S2
9ii) Analyses at each height
si9ti)
df
ti
df
times
mid-shore
low-shore
***
***
2, 12 df
2, 12 df
Ns
**
2, 3 df
2, 3 df
T1ˆT2ˆT3
T1ˆT2=T3
B. Four shores.
Analyses
Post-hoc tests
mid-shore
9i) Analyses at each shore
si9he)
df
He
df
low-shore
Green Point 9GP)
Harbord 9HA)
Cape Banks 9CB)
Coledale 9CD)
***
ns
***
***
2, 16 df
2, 16 df
2, 16 df
2, 16 df
**
***
*
1, 2 df
1, 2 df
1, 2 df
1, 2 df
S1=S2
^
S1=S2
S1=S2
S1=S2
^
S1=S2
S1=S2
***
9ii) Analyses at each height
si9sh)
df
Sh
df
shores
mid-shore
low-shore
***
***
4, 32 df
4, 32 df
***
***
3, 4 df
3, 4 df
GPˆHAˆCB=CD
GPˆHAˆCB=CD
*, P50.05; **, P50.01; ***, P50.001; ns, not signi¢cant. Post-hoc comparisons are presented for sites, times and locations.
For post-hoc comparisons: 1, P50.05; ˆ, not signi¢cant.
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

Special patterns of macrofauna in coralline turf B.P. Kelaher et al. 921
1992; Anderson Legendre, 1999). Because NP-
multivariate di¡erences among macrofaunal assemblages
9Anderson, 2001). The NP-MANOVAs were followed by a
posteriori pair-wise comparisons on appropriate terms in
the model found to be signi¢cant at P40.05. For these
tests, only probability values are presented because the
multivariate F-statistics and T-statistics were generated by
permutations. For some comparisons, tests were done by
permuting the residuals of the full model to obtain a
signi¢cance level with P40.05 9described by Ter Braak,
&
MANOVA can only analyse one or two factors at one
time, temporal and spatial di¡erences in macrofaunal
assemblages at Cape Banks were tested in two ways. First,
two-way nested analyses were done to test for di¡erences
between `heights' 9two levels, orthogonal and ¢xed) and
`sites' 9two levels, nested within heights) at each time.
Second, two-way nested analyses were done to test for
di¡erences among `times' 9three levels, orthogonal and
Figure 2. Two-dimensional nMDS ordinations comparing macrofaunal assemblages from mid- 9un¢lled) and low-shore 9¢lled)
areas. Data are averaged among replicates within sites. 9A) Assemblages sampled at Cape Banks in May 1997 9circles), September
1997 9triangles) and January 1998 9squares); 9B) assemblages sampled at Green Point 9circles), Harbord 9triangles), Cape Banks
9squares) and Coledale 9pentagons) in January 1998.
Table 3. Percentage contributions of individual taxa to dissimilarity between sites ꢀi) in mid-shore, ꢀii) in low-shore areas and ꢀiii)
mean of comparisons among every replicate in low-shore sites with each replicate in mid-shore sites. Dissimilarities are from the ꢀa) three
times of sampling at Cape Banks and ꢀb) at the four shores sampled in January 1998. The ten taxa selected contributed more than 60%
of the total dissimilarity for all comparisons and each taxon made a relatively large contribution to the total dissimilarity compared to
other taxa found in coralline turf.
Taxa
A. Cape Banks
9ii) low-shore
T1 T2 T3
B. Four shores
9i) mid-shore
9iii) mid vs low
9i) mid-shore
9ii) low-shore
9iii) mid vs low
T1 T2 T3
T1 T2 T3
GP HA CB CD GP HA CB CD GP HA CB CD
Nematodes
0
2
2
1
1
2
0
3
5
2
0
3
1
33
0
0
3
6
15
0
1
0
1
0
6
24
2
9
1
3
1
2
0
4
1
4
11
0
1
5
23
5
10
4
1
2
4
1
4
5
0
0
3
8
1
5
2
5
0
1
0
3
2
4
6
2
1
9
12
5
2
4
4
2
0
4
1
33
0
0
3
5
0
5
0
0
0
0
3
1
6
0
0
0
5
6
1
3
0
0
1
3
1
4
10
0
1
5
4
9
0
0
1
0
7
3
6
1
12
0
8
0
1
2
0
5
0
2
8
2
5
5
2
1
10
10
4
2
2
0
4
0
0
1
1
1
Syllid polychaetes
Orbinid polychaetes
Limonia marina
Janirid isopods
Tanaids
Melitid amphipods
Isherosid amphipods
Eatoneilla atropurpureua
Lasaea australis
16 10
0
7
0
0
0
0
14
0
6
1
0
0
0
6
10 14
7
0
0
3
1
0
0
0
2
7
12 22
36 32
1
12 11
5
10
5
10 13
6
22 48 27 68
6
6
6
1
1
6
41 53 19
32 26
52 54 27
31 53 20 54
27
Total dissimilarity
66 74 69
64 72 64
72 79 74
70 80 70 70
67 64 63 57
70 81 71 79
T1, May 1997; T2, September 1997; T3, January 1998; GP, Green Point; HA, Harbord; CB, Cape Banks; CD, Coledale.
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

922 B.P. Kelaher et al. Special patterns of macrofauna in coralline turf
random) and `sites' 9two levels, nested within times) for
each height separately. A similar approach was used to
test the generality of the spatial patterns at Cape
Banks, except in these analyses the factor `shores' 9four
levels, orthogonal and random) replaced the factor
`time'.
Non-metric multidimensional scaling 9nMDS, Field
et al., 1982; Clarke, 1993) was used to produce two-
dimensional ordination plots to show relationships among
macrofaunal assemblages. Similarity percentage analysis
9SIMPER) 9Clarke, 1993) was used to determine the taxa
contributing most to the dissimilarity between groups. All
BIO-ENV 9Clarke & Ainsworth, 1993) was used to
investigate the relationships between macrofaunal assem-
blages and environmental variables. For these analyses, a
weighted Spearman's rank coe¤cient was used 9recom-
mended by Clarke & Ainsworth, 1993). For univariate
comparisons, patterns of association were tested using
Pearson's r correlation coe¤cient 9Winer et al., 1991). To
decrease the probability of excessive Type I errors,
multiple comparisons were only considered signi¢cant if
P50.01. Because of large di¡erences between macrofaunal
assemblages at di¡erent heights on the shore 9see Results),
analyses investigating relationships between environ-
mental variables and macrofauna were done separately in
low- and in mid-shore areas.
multivariate analyses were done using the Bray Curtis
^
similarity coe¤cient 9Bray & Curtis, 1957; Clarke &
Green, 1988; Clarke, 1993) and the data were untrans-
formed. Analysis of variance was used to test hypotheses
about di¡erences in the number of taxa of the main
faunal groups. These analyses were preceded by
Cochran's test for homogeneity of variances 9Winer et
al., 1991; Underwood, 1997). Where variances showed
signi¢cant heterogeneity, the data were transformed
using 9x‡1)0.5 or ln9x‡1) transformations where appro-
priate 9Underwood, 1997). For univariate comparisons,
all three factors were included in analyses.
RESULTS
Patterns of macrofaunal assemblages at Cape Banks
At each time of sampling at Cape Banks, there were
signi¢cant di¡erences between assemblages from low- and
from mid-shore areas 9Table 2A; Figure 2A). There were
also signi¢cant di¡erences between sites at each height,
Table 4. Analyses of number of taxa of the three main taxonomic groups from ꢀA) the three times of sampling at Cape Banks and ꢀB)
the four shores sampled in January 1998. `ti' is the comparison among three times of sampling at Cape Banks. `sh' is the comparison
among the four randomly-chosen shores; `he' is the ¢xed comparison between mid- and low-shore areas; `si' is the comparison of sites
nested in `ti' and `he' or `sh' and `he'.
A. Cape Banks.
9Nˆ4)
Species of Gastropods
none
Families of Polychaetes
none
Families of Amphipods
none
Transform
df
MS
F
MS
F
MS
F
ti
he
2
1
2
6
36
103.14
161.33
77.77
11.16
8.32
9.24
2.02
7.14*
12.33
54.18
4.00
16.56
2.10
0.74 ns
13.55 ns
0.24 ns
38.58
72.52
36.58
10.64
1.27
3.62 ns
1.98 ns
3.44 ns
8.38***
^
^
tiÂhe
site9tiÂhe)
residual
SNK
1.34 ns
7.87***
T1: NS T2: NS T3: M5L
M: T1ˆT2ˆT3 L: T1ˆT25T3
B. Four shores.
9Nˆ5)
Species of Gastropods
none
Families of Polychaetes
Families of Amphipods
Transform
ln9x + 1)
ln9x + 1)
df
MS
F
MS
F
MS
F
sh
he
3
1
3
8
64
83.01
180.00
84.10
14.27
4.60
5.82 ^
2.14 ^
5.89*
3.10**
8.51
10.51
13.84
19.33
3.86
0.44 ns
0.76 ns
0.72 ns
8.29***
30.13
396.05
8.31
7.67
0.44
3.93 ns
47.62**
1.08 ns
shÂhe
site9shÂhe)
residual
SNK
17.30***
GP: NS HA: NS CB: M5L CD: M5L
M: GPˆHAˆCBˆCD L: GPˆHAˆCD5CB
M5L
*, P50.05; **, P50.01; ***, P50.001; ns, not signi¢cant; ^, no test possible because of signi¢cant interaction. For Student Newman
^
^
Keuls 9SNK) comparisons: T1, May 1997; T2, September 1997; T3, January 1998; GP, Green Point; HA, Harbord; CB, Cape Banks;
CD, Coledale; M, mid-shore areas; L, low-shore areas; 5, P40.05; ˆ, not signi¢cant.
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

Special patterns of macrofauna in coralline turf B.P. Kelaher et al. 923
but these di¡erences varied from time to time. There were
large contribution to the dissimilarity between sites in
low-shore areas in September 1997 and melitid amphipods
9mostly Elasmopus sp.) were major contributors in January
1998. At the ¢rst two times of sampling, L. australis
contributed more than 50% of the di¡erence between
assemblages in low- and mid-shore levels. In contrast, a
range of species contributed to the dissimilarity at the
third time of sampling 9L. australis contributing only 27%;
see Table 3).
no signi¢cant temporal di¡erences among assemblages in
mid-shore areas, but in low-shore areas, assemblages in
January 1998 were signi¢cantly di¡erent from those at
the other two times of sampling 9Table 2A; Figure 2A).
The magnitude of temporal change between the ¢rst two
times of sampling was similar between assemblages in low-
or mid-shore, but di¡ered greatly at the third time of
sampling 9Figure 2A).
The small and abundant bivalve, Lasaea australis
9Lamarck), made large contributions to the dissimilarity
between sites in low- and mid-shore areas and between
mid- and low-shore areas at all times of sampling at Cape
Banks, except low-shore areas in January 1998 9Table 3A).
Janirid isopods 9mostly Ianiropsis sp.) made large contribu-
tions to the dissimilarities among sites in mid-shore areas
in May and September 1997, but not in January 1998. In
contrast, tanaids made relatively small contributions to the
dissimilarity among mid-shore sites at the ¢rst two times
of sampling, compared with their contribution in January
1998. Orbinid polychaetes and the small gastropod,
Eatoniella atropurpurea 9Fauenfeld), made consistently large
contributions to the dissimilarity between sites in low-
shore areas at all times of sampling. Nematodes made a
There were no signi¢cant di¡erences in the number of
families of polychaetes or amphipods between low- and
mid-shore areas 9Table 4A; Figure 3). There were signi¢-
cantly more species of gastropods in low- than in mid-
shore areas in January 1998, but not at other times of
sampling 9Table 4A; Figure 3). There were also signi¢-
cantly more species of gastropods in January 1998 than at
other times of sampling. Although not signi¢cantly
di¡erent, at each time of sampling, there were more
families of polychaetes and amphipods and more species of
gastropods in low- than in mid-shore areas 9Figure 3).
There was also signi¢cant variability in the number of
families of amphipods and polychaetes between sites,
but not in the number of species of gastropods
9Table 4A).
Figure 3. Mean 9SE) number of taxa of the three main faunal groups for 9A) the three times of sampling at Cape Banks 9T1,
May 1997; T2, September, 1997 and T3, J anuary, 1998); and 9B) the four shores 9GP, Green Point; HA, Harbord; CB, Coledale
and CD, Coledale) sampled in January 1998. Data are from mid- 9un¢lled) and low-shore areas 9¢lled) and are averaged
across sites.
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

924 B.P. Kelaher et al. Special patterns of macrofauna in coralline turf
Generality of patterns of macrofaunal assemblages
areas 9Table 4B; Figure 3). There were also signi¢cantly
more species of gastropods in low- than in mid-shore
areas at Cape Banks and Coledale, but not at the other
shores 9Table 4B; Figure 3). There was signi¢cant vari-
ability in the number of taxa of amphipods, polychaetes
and gastropods among sites within shores 9Table 4B).
At each shore, there were signi¢cant di¡erences
between assemblages from low- and mid-shore areas
9Table 2B; Figure 2B) and the magnitude of these di¡er-
ences were relatively similar 9Figure 2B). Nevertheless, the
assemblages in low- and in mid-shore areas at Coledale
were signi¢cantly di¡erent from those at other shores
9Table 2B). There were also signi¢cant di¡erences
between sites in low- and mid-shore areas at all shores,
except Harbord 9Table 2B).
Syllid polychaetes, the insect larvae, Limonia marina
Skuse, the small gastropod, E. atropurpurea, and the small
bivalve, Lasaea australis, made relatively large contribu-
tions to the dissimilarity between sites in mid-shore areas
9Table 3B). Syllid polychaetes, melitid amphipods,
ischyrocerid amphipods and the small bivalve, Lasaea
australis, made a relatively large contribution to the
dissimilarity between sites in low-shore areas. Lasaea
australis made an extremely large contribution to dissimi-
larity between low- and mid-shore areas at all locations
sampled. Melitid and ischyrocerid amphipods were
relatively large contributors between low- and mid-shore
areas at all locations, except Coledale. Syllid polychaetes
and the insect larvae, Limonia marina, made relatively large
contributions to dissimilarity at all locations.
Environmental variables and their relationship with
macrofaunal assemblages
Except for epiphytes, all environmental variables
di¡ered signi¢cantly between sites in low- and in mid-
shore areas 9Table 5). In addition, the length of fronds
and amount of epiphytes were signi¢cantly greater in
low- than in mid-shore areas at all locations 9Table 5;
Figure 4). There were no overall signi¢cant di¡erences in
the amount of sediments or density of fronds between low-
and mid-shore areas 9Table 5; Figure 4). All environ-
mental variables varied signi¢cantly among shores,
although the length of fronds only varied among shores in
low-shore areas 9Table 5; Figure 4).
Of all the environmental variables, sediment showed the
strongest relationship with the macrofaunal assemblage in
either low- or mid-shore areas 9Table 6). In mid-shore
areas, the strength of the relationship between sediment
and macrofauna became greater as architectural
characteristics of turf 9length and density of fronds) were
also incorporated. In low-shore area, the relationship
between sediment and macrofauna became weaker as
extra environment variables were included in analyses.
There were no signi¢cant di¡erences in the number of
families of polychaetes between low- and mid-shore areas
9Table 4B; Figure 3). There were, however, signi¢cantly
more families of amphipods in low- than in mid-shore
Table 5. Analyses of environmental variables at the four shores sampled in January 1998. `sh' is the random comparison among the
four shores; `he' is the ¢xed comparison between mid- and low-shore areas; `si' is the random comparison of sites nested in `sh' and `he'.
Sediment
none
Epiphytes
sq9x + 1)
Transform
df
MS
F
MS
F
Sh
He
3
1
3
8
64
16.49
0.03
1.18
2.04
0.12
8.09***
0.03 ns
0.58 ns
0.17
0.61
0.02
0.03
0.02
5.39*
26.14*
0.75 ns
1.37 ns
shÂhe
si9shÂhe)
Residual
SNK height
Location
16.29***
M5L
GPˆHAˆCBˆCD
GPˆHA5CBˆCD
Length
none
MS
Density
none
Transform
df
F
MS
F
Sh
He
3
1
3
8
64
0.32
108.44
1.88
0.16
0.08
1.97
57.42
11.67**
2.03*
59.45
13.89
30.68
11.74
2.48
5.07*
^
^
0.45 ns
2.67 ns
4.73***
shÂhe
si9shÂhe)
Residual
SNK height
GP: M5L HA: M5L CB: M5L CD: M5L
M: GPˆHAˆCBˆCD L: GPˆCDˆCB5HA
CD5GPˆHAˆCD
*, P50.05; **, P50.01; ***, P50.001; ns, not signi¢cant. For Student Newman Keuls 9SNK) tests: GP, Green Point; HA, Harbord;
^
^
CB, Cape Banks; CD, Coledale; L, low-shore areas; M, mid-shore areas; 5, P40.05; ˆ, not signi¢cant.
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

Special patterns of macrofauna in coralline turf B.P. Kelaher et al. 925
Figure 4. Mean 9SE) measures of environmental variables in each core from the four shores 9GP, Green Point; HA, Harbord; CB,
Coledale; and CD, Coledale) sampled in January 1998. Data are from mid- 9un¢lled) and low-shore areas 9¢lled) and are averaged
across sites.
Table 6. Correlation coe¤cients ꢀNˆ40) for comparisons between the macrofaunal assemblage and combinations of environmental
variables from BIO-ENV analyses. Values indicate the strength of relationships relative to other comparisons at each height on the shore.
Bold values indicate the strongest relationships for comparisons involving either one, two or three variables.
Mid-shore sites
Low-shore sites
sediment ꢀse)
0.19
0.15
0.06
0.42
0.10
0.02
0.02
epiphytes ꢀep)
length ꢀle)
packedness ꢀdensity) ꢀpa)
70.12
se‡ep
se‡le
se‡pa
ep‡le
ep‡pa
le‡pa
0.16
0.21
0.17
0.09
0.06
0.18
0.26
0.32
0.32
0.06
0.08
0.13
se‡ep‡le
0.18
0.18
0.22
0.13
0.20
0.27
0.26
0.34
0.11
0.28
se‡ep‡pa
se‡le‡pa
ep‡le‡pa
se‡ep‡le‡pa
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

926 B.P. Kelaher et al. Special patterns of macrofauna in coralline turf
Table 7. Summary of correlations between environmental variables and the number of taxa in the three main faunal groups.
Correlations are divided into ꢀi) mid- and ꢀii) low-shore areas.
Sediment
Epiphtyes
Length
Density
A. Mid-shore areas.
r
r
r
r
Families of Amphipods
Families of Polychaetes
Species of Gastropods
70.15
0.51**
0.04
0.57**
70.54**
0.02
0.53**
70.26
70.04
70.25
0.28
0.09
B. Low-shore areas.
Families of Amphipods
Families of Polychaetes
Species of Gastropods
0.38
0.43**
0.65**
0.31
0.13
70.05
70.33
70.14
70.46**
70.20
70.14
70.54**
r, Pearson's correlation coe¤cient; **, P50.01; Nˆ40.
In mid-shore areas, there was a signi¢cant positive corre-
lation between the number of families of polychaetes and
the amount of sediment, between the number of families of
amphipods and the amount of epiphytes and length of
fronds 9Table 7A) and there was a signi¢cant negative
correlation between the number of families of polychaetes
and the amount of epiphytes 9Table 7A). In low-shore
areas, there was a signi¢cant positive correlation between
the number of families of polychaetes and number of
species of gastropods and the amount of sediment 9Table
7A). In addition, there was a signi¢cant negative correla-
tion between number of species of gastropods and the
length and density of fronds in low-shore areas 9Table 7A).
been tested, it has been found that patterns may vary
from shore to shore 9Foster, 1990; Underwood &
Chapman, 1998). In this study, assemblages at Coledale
were signi¢cantly di¡erent from those at other shores.
Despite this, however, the di¡erences between low- and
mid-shore areas appeared to be relatively consistent
among shores.
There was a trend for greater numbers of taxa of gastro-
pods, amphipods and polychaetes in low- than in mid-
shore areas at each time of sampling at Cape Banks,
although these di¡erences were not always signi¢cant. In
contrast to the multivariate results, however, when the
patterns of richness of gastropods and polychaetes were
tested, there was no generality among shores. For
example, there were more families of polychaetes in mid-
than in low-shore areas at Coledale and more species of
gastropods in mid- than low-shore areas at Green Point.
This lack of generality indicates that no common process
is responsible for structuring patterns of richness of gastro-
pods and polychaetes between low- and mid-shore areas
over the region studied 9120 km) and that local processes
are extremely important for structuring some aspects of
the macrofaunal assemblage. In contrast to gastropods
and polychaetes, there were more families of amphipods
in coralline turf in low- than in mid-shore areas at each
shore sampled. Similar results have been reported in
other studies 9Tararam et al., 1986; Underwood &
Verstegen, 1987; Lintas & Seed, 1994), indicating the
possibility of general biological processes or environmental
factors being responsible for the vertical distribution of
amphipods on rocky shores.
At every time of sampling at Cape Banks and at the
other shores, except Harbord, there were signi¢cant
di¡erences among macrofaunal assemblages and the rich-
ness of polychaetes, amphipods and gastropods in sites
separated by tens of metres. In previous studies investi-
gating macrofaunal assemblages in mat-like habitats on
rocky shores, replication at the scales of tens of metres has
rarely been considered 9e.g. Dommasnes, 1968; Lintas &
Seed, 1994). This is probably because sorting and
identifying macrofauna from these habitats is extremely
time-consuming 9Hicks, 1977) and replication is usually
minimized. Nevertheless, this study showed that macro-
fauna in mat-like habitats can vary at the scales of tens of
metres, as do other intertidal assemblages on rocky shores
DISCUSSION
Prior to this study, little was known about the bio-
diversity of macrofauna in coralline turf on rocky shores
around Sydney. In this study, 147 taxa of mobile macro-
fauna were identi¢ed and densities of macrofauna often
exceeded 250,000 m72 9Table 1; Kelaher, 2000). Given
the great abundance of coralline turf on rocky shores of
New South Wales 9Underwood, 1981; Underwood &
Chapman, 1998), it is clear that this assemblage makes an
extremely large contribution to biodiversity of these habi-
tats. Even considering di¡erences in methods of sampling,
the macrofauna in coralline turf on the shores investigated
in this study appears to be relatively rich compared with
shores in other countries 9Japan: 91 taxa 9Akioka et al.,
1999); Chile: 36 taxa 9Lopez & Stolz, 1997); Norway: 72
taxa 9Dommasnes, 1969); New Zealand: 106 using similar
taxonomic resolution of the phyla in Table 1 9Hicks, 1971);
Ireland: 92 taxa using the same taxonomic resolution as
here, B.P.K., unpublished data).
At each time of sampling at Cape Banks and at the other
shores, there were di¡erences between assemblages in low-
and mid-shore areas. These results were not unexpected
because this pattern is generally found for other assem-
blages of organisms on rocky intertidal shores 9Lewis,
1964; Dayton, 1971; Menge, 1976; Lubchenco et al., 1984;
Underwood & Chapman, 1998). Despite a large body of
literature documenting the patterns of vertical distribution
of algae and animals, the generality of these patterns has
rarely been tested. In the few cases where generality has
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

Special patterns of macrofauna in coralline turf B.P. Kelaher et al. 927
9Ca¡ey, 1985; Archambault & Bourget, 1996; Underwood
not vary between di¡erent heights on the shore, these vari-
ables probably explain little of the variation between
assemblages in low- and mid-shore areas.
& Chapman, 1998). These results with those of other
studies emphasize the need for within-shore or within-
height replication before di¡erent shores or di¡erent
heights on the shore can be compared 9see Hurlbert,
1984; Underwood, 1994).
Like the macrofaunal assemblages in coralline turf, all
environmental variables, except epiphytes, varied at scales
of tens of metres. Moreover, all environmental variables
were signi¢cantly correlated with some aspect of the
assemblage in either low- or mid-shore areas. Never-
theless, except for the positive association between rich-
ness of polychaetes and the amount of sediment, the
relationships between macrofauna and the environmental
variables were much more complicated than the simple
positive relationships found in other studies 9e.g. Dean &
Connell, 1987a; Gibbons, 1988; Gee & Warwick, 1994).
For example, the length of coralline fronds was negatively
associated with the number of families of amphipods in
low-shore areas, but positively associated in mid-shore
areas. In addition, the directions of correlations were
strongly dependent on the type of organisms being investi-
gated, e.g. the amount of epiphytes in mid-shore areas was
positively correlated with the richness of amphipods and
negatively correlated with the richness of polychaetes.
Although the results suggest that variation in environ-
mental variables may explain some of the spatial vari-
ability in the richness of macrofauna in coralline turf,
they also show that direction of the e¡ects depends on the
height on the shore and the taxa being investigated.
Of the four environmental variables measured, sedi-
ment had the strongest and most consistent associations
with the macrofauna. It has been postulated that sediment
creates a more heterogeneous habitat for infaunal and
epifaunal species and thereby increases opportunities for
di¡erent types of macrofauna 9Hicks, 1980, 1985;
Gibbons, 1991). This cannot explain the lack of positive
associations between the amount of sediment and the rich-
ness of amphipods and gastropods in mid-shore areas.
Gibbons 91988) argued that sediment enhances the diver-
sity of macrofauna in mat-like habitats in intertidal areas
because it retains moisture when the habitat is exposed
during low tide. It is possible that the positive associations
between sediment and the richness of amphipods and
gastropods existed in low- but not mid-shore areas,
because the water-retaining properties of sediment in
low-shore areas allowed species more commonly found in
sub-tidal areas 9e.g. dexaminid amphipods or the small
opisthobranch, Runcina australis) to remain in low-shore
intertidal areas.
Much of the variation in assemblages between low- and
mid-shore areas and between sites in each height on the
shore can be attributed to relatively few taxa. Of these
taxa, the small bivalve, Lasaea australis, made an extremely
large contribution to dissimilarity among assemblages at
nearly all times of sampling and shores. This is not un-
expected because L. australis often reached densities of
more than 100,000 m72, was spatially variable and was
more abundant in mid- than in low-shore areas 9Kelaher,
2000). Other studies have shown a similar dominance of
Lasaea spp. in macrofaunal assemblages in other mat-like
habitats on rocky shores 9Ong Che & Morton, 1992; Peake
& Quinn, 1993). Other than L. australis, the taxa most
contributing to spatial variation tended to vary from
shore to shore and from time to time. For example,
Isherosid amphipods were relatively large contributor to
dissimilarity between sites in low-shore areas and
between low- and mid-shore areas at Green Point and
Harbord, but not at Cape Banks or Coledale. Never-
theless, this taxon rarely contributed to di¡erences
between sites in mid-shore areas at any time of sampling.
In contrast, tanaids made an extremely large contribution
to dissimilarity among sites in mid-shore areas in January
1998 at Cape Banks, but not at other shores or at other
times. At any one time of sampling or shore, there was
generally a relatively small group of taxa responsible for
most of the dissimilarity between assemblages. Apart
from L. australis, it is di¤cult to identify speci¢c taxa that
were consistently responsible for large amount of dissimi-
larity among assemblages at all times of sampling and
shores.
Environmental variables, such as the amount of sedi-
ment, epiphytes or the amount or type of physical struc-
ture of the habitat, are generally considered important for
structuring macrofaunal assemblages in algal turf 9Hicks,
1980; Dean & Connell, 1987a; Johnson & Scheibling, 1987;
Gibbons, 1988; Gee & Warwick, 1994). In this study, it is
obviously not possible to determine the e¡ects of environ-
mental variables on the spatial variability of macrofaunal
assemblages in coralline turf because the data are descrip-
tive. Nevertheless, the patterns of spatial variability of
environmental variables and correlations between
environmental variables and macrofauna may provide
indications of their relative importance and guide future
research. For example, the average length of fronds and
the amount of epiphytes were greater in low- and in mid-
shore areas. Coralline turf with long fronds may have a
greater surface area available for colonization and reduce
the e¡ects of waves more e¡ectively than turf with short
fronds 9Dommasnes, 1968; Whor¡ et al., 1995). In addi-
tion, epiphytes provide both food and shelter for asso-
ciated assemblages of organisms 9Johnson & Scheibling,
1987; Du¡y, 1990; Schneider & Mann, 1991a,b; Martin-
Smith, 1993). Therefore, each of these variables may
contribute to di¡erences between the macrofaunal assem-
blages in low- and mid-shore areas. Nevertheless, because
the amount of sediment and density of coralline fronds did
In this study, the length and density of fronds provided
measures of the amount and complexity of physical struc-
ture of the habitat. It has usually been shown that as the
amount and complexity of structure increases, the richness
of the associated assemblage of organisms also increases
9MacArthur & MacArthur, 1961; Kohn & Leviten, 1976;
Heck & Wetstone, 1977; Stoner, 1980; Dean & Connell,
1987a; Beck, 2000). Nevertheless, apart from the positive
association between richness of amphipods and the length
of fronds, the majority of associations between length and
density of fronds were negative. The results found here
therefore directly contrast with the positive relationships
shown by other similar studies 9e.g. Dean & Connell,
1987a; Gibbons, 1988). It is di¤cult to explain these
results, except that they imply that there are other
biological processes or environmental variables that are
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

928 B.P. Kelaher et al. Special patterns of macrofauna in coralline turf
Ca¡ey, H.M., 1985. Spatial and temporal variation in settlement
and recruitment of intertidal barnacles. Ecological Monographs,
strongly interacting with the physical characteristics of the
coralline turf 9see Stoner, 1982; Gibbons, 1988). These
results do provide, however, initial evidence that general-
izations about positive relationships between the physical
structure of habitat and the diversity of the associated
assemblage of organisms may not be valid in all situations.
Careful manipulative experiments that separate the e¡ects
of the physical structure of coralline turf on associated
macrofauna from other factors are needed before more
concrete conclusions can be reached.
55, 313 332.
^
Chapman, M.G., 1994. Small- and broad-scale patterns of
distribution of the upper-shore littorinid Nodilittorina
pyramidalis in New South Wales. AustralianJournal of Ecology, 19,
83 95.
^
Chapman, M.G. & Underwood, A.J., 1994. Dispersal of the
intertidal snail, Nodilittorina pyramidalis, in response to topo-
graphic complexity of the substratum. Journal of Experimental
Marine Biology and Ecology, 179, 145 169.
^
Clarke, K.R., 1993. Non-parametric analyses of changes in
It is clear from this study that coralline algal turf make a
large contribution to patterns of variation of invertebrates
on rocky intertidal shores around Sydney. Despite some
temporal variation and some di¡erences among shores,
much of the variability of these assemblages was at small
spatial scales 9between low and mid-shore areas and
between sites separated by tens of metres). Therefore,
understanding the processes or environmental variables
that create and maintain the structure of these assemblages
will be best achieved by initially testing hypotheses about
processes that a¡ect small-scale spatial variability, rather
than focusing on larger-scale variation among shores.
community structure. AustralianJournal of Ecology, 18, 117 143.
^
Clarke, K.R. & Ainsworth, M., 1993. A method of linking multi-
variate community structure to environmental variables.
Marine Ecology Progress Series, 92, 205 219.
^
Clarke, K.R. & Green, R.H., 1988. Statistical design and
analysis for a `biological e¡ects' study. Marine Ecology Progress
Series, 46, 213 226.
^
Connell, J.H., 1961. The in£uence of interspeci¢c competition
and other factors on the distribution of the barnacle
Chthamalus stellatus. Ecology, 42, 710 723.
^
Connell, J.H., 1972. Community interactions on marine rocky
intertidal shores. Annual Review of Ecology and Systematics, 3,
169 192.
^
Connor, E.F. & McCoy, E.D., 1979. The statistics and biology of
the species area relationship. American Naturalist, 113, 791 833.
Coull, B.C. & Wells, J.B., 1983. Refuges from ¢sh predation:
experiments with phytal meiofauna from the New Zealand
^
^
This study was supported by an Australian Post-Graduate
Award 9to B.P.K.) and by funds from the Australian Research
Council through the Centre for Research on Ecological Impacts
of Coastal Cities. Three anonymous referees made helpful
comments on an earlier draft of the paper.
rocky intertidal. Ecology, 64, 1590 1609.
^
Crisp, D.J. & Southward, A.J., 1958. The distribution of inter-
tidal organisms along the coasts of the English Channel.
Journal of the Marine Biological Association of the United Kingdom,
37, 157 208.
^
REFERENCES
Davenport, J., Butler, A. & Cheshire, A., 1999. Epifaunal
composition and fractal dimensions of marine plants in rela-
tion to emersion. Journal of the Marine Biological Association of
Akioka, H., Baba, M., Masaki, T. & Johansen, W., 1999. Rocky
shore turfs dominated by Corallina 9Corallinales, Rhodophtya)
in northern Japan. Phycological Research, 47, 199 206.
Anderson, M.J., 2001. A new method for non-parametric multi-
the United Kingdom, 79, 351 355.
^
^
Dayton, P.K., 1971. Competition, disturbance, and community
organization: the provision and subsequent utilization of
space in a rocky intertidal community. Ecological Monographs,
variate analysis of variance. Austral Ecology, 26, 32 46.
^
Anderson, M.J. & Legendre, P., 1999. An empirical comparison
of permutation methods for tests of partial regression co-
e¤cients in a linear model. Journal of Statistical Computation and
41, 352 389.
^
Dean, R.L. & Connell, J.H., 1987a. Marine invertebrates in
algal succession. II. Tests of hypotheses to explain diversity
with succession. Journal of Experimental Marine Biology and
Simulation, 62, 271 303.
^
Andrew, N.L. & Mapstone, B.D., 1987. Sampling and the
description of spatial pattern in marine ecology. Oceanography
Ecology, 109, 217 247.
^
and Marine Biology. Annual Review, 25, 39 90.
Dean, R.L. & Connell, J.H., 1987b. Marine invertebrates in
algal succession. III. Mechanisms linking habitat complexity
with diversity. Journal of Experimental Marine Biology and
^
Archambault, P. & Bourget, E., 1996. Scales of coastal hetero-
geneity and benthic intertidal species richness, diversity and
abundance. Marine Ecology Progress Series, 136, 111 121.
Ecology, 107, 249 273.
^
^
Ballesteros, E., 1988. Composicion y estructura de la comunidad
infralitoral de Corallina elongata Ellis and Solander, 1786 de la
Costa Brava 9Mediterraneo occidental). Investigaciones Pesqueras.
Denley, E.J. & Underwood, A.J., 1979. Experiments on factors
in£uencing settlement, survival and growth of two species of
barnacles in New South Wales. Journal of Experimental Marine
Barcelona, 52, 135 151.
Biology and Ecology, 36, 269 293.
Dommasnes, A., 1968. Variations in the meiofauna of Corallina
o¤cinalis L. with wave exposure. Sarsia, 34, 117 124.
Dommasnes, A., 1969. On the fauna of Corallina o¤cinalis L. in
^
^
Beck, M.W., 2000. Separating the elements of habitat structure:
independent e¡ects of habitat complexity and structural
components on rocky intertidal gastropods. Journal of
^
Experimental Marine Biology and Ecology, 249, 29 49.
Benedetti-Cecchi, L. & Cinelli, F.,1994. Recovery of patches in an
assemblage of geniculate coralline algae: variability at di¡erent
western Norway. Sarsia, 38, 71 86.
Du¡y, J.E., 1990. Amphipods on seaweeds: partners of pests?
^
^
Oecologia, 83, 267 276.
^
successional stages. Marine Ecology Progress Series, 110, 9 18.
Bray, J.R. & Curtis, J.T., 1957. An ordination of the upland forest
communities of southern Wisconsin. Ecological Monographs, 27,
Dye, A.H., 1993. Recolonization of intertidal macroalgae in rela-
tion to gap size and molluscan herbivory on a rocky shore on
the east coast of southern Africa. Marine Ecology Progress Series,
^
325 349.
95, 263 271.
^
^
Brown, P.J. & Taylor, R.B., 1999. E¡ects of trampling by humans
on animals inhabiting coralline algal turf in the rocky inter-
tidal. Journal of Experimental Marine Biology and Ecology, 235,
Fairweather, P.G., 1985. Di¡erential predation on alternative
prey and the survival of rocky intertidal organisms in New
South Wales. Journal of Experimental Marine Biology and Ecology,
45 53.
^
109, 135 156.
^
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

Special patterns of macrofauna in coralline turf B.P. Kelaher et al. 929
Fairweather, P.G. & Underwood, A.J., 1991. Experimental
Lohse, D.P., 1993. The importance of secondary substratum in a
removals of a rocky intertidal predator: variations within two
rocky intertidal community. Journal of Experimental Marine
habitats in the e¡ects on prey. Journal of Experimental Marine
Biology and Ecology, 166, 1 17.
^
Biology and Ecology, 154, 29 75.
Lopez, C.A. & Stolz, W.B., 1997. Description of the fauna asso-
ciated to Corallina o¤cinalis L. in the intertidal of the rocky
shore of `Palo Colorado' 9Los Vilos, IV-Region, Chile). Revista
^
Field, J.G., Clarke, K.R. & Warwick, R.M., 1982. A practical
strategy for analysing multispecies distribution patterns.
Marine Ecology Progress Series, 8, 37 52.
de Biologia Marina y Oceanogra¢a, 32, 17 35.
Lubchenco, J. et al., 1984. Structure, persistence, and role of
^
^
Fielding, P.J., Weerts, K.A.
&
Forbes, A.T., 1994.
Macroinvertebrate community associated with intertidal
and subtidal beds of Pyura stolonifera 9Heller) 9Tunicata:
Ascidiacea) on the Natal coast. South AfricanJournal of Zoology,
consumers in
a tropical rocky intertidal community
9Taboguilla Island, Bay of Panama). Journal of Experimental
Marine Biology and Ecology, 78, 23 74.
MacArthur, R.H. & MacArthur, J.W., 1961. On bird species
^
29, 46 53.
^
Foster, M.S., 1990. Organization of macroalgal assemblages in
the northeast Paci¢c: the assumption of homogeneity and the
diversity. Ecology, 42, 594 598.
Martin-Smith, K.M., 1993. Abundance of mobile epifauna: the
^
illusion of generality. Hydrobiologia, 192, 21 33.
Gee, J.M. & Warwick, R.M., 1994. Metazoan community
structure in relation to the fractal dimensions of marine
role of habitat complexity and predation by ¢shes. Journal of
^
Experimental Marine Biology and Ecology, 174, 261 275.
^
McGuinness, K.A. & Underwood, A.J., 1986. Habitat structure
macroalgae. Marine Ecology Progress Series, 103, 141 150.
and the nature of communities on intertidal boulders. Journal
^
Gibbons, M.J., 1988. The impact of sediment accumulations,
relative habitat complexity and elevation on rocky shore meio-
fauna. Journal of Experimental Marine Biology and Ecology, 122,
of Experimental Marine Biology and Ecology, 104, 97 123.
^
Menge, B.A., 1976. Organization of the New England rocky
intertidal community: role of predation, competition and
environmental heterogeneity. Ecological Monographs, 46,
225 241.
^
Gibbons, M.J., 1991. Rocky shore meiofauna: a brief over-
335 393.
^
view. Transactions of the Royal Society of South Africa, 47,
Menge, B.A., 1995. Indirect e¡ects in marine rocky intertidal
interaction webs: patterns and importance. Ecological
595 603.
^
Grahame, J. & Hanna, F.S., 1989. Factors a¡ecting the distribu-
Monographs, 65, 21 74.
^
tion of the epiphytic fauna of Corallina o¤cinalis 9L.) on an
Morrisey, D.J., Howitt, L., Underwood, A.J. & Stark, J.S., 1992.
Spatial variation in soft-sediment benthos. Marine Ecology
exposed rocky shore. Ophelia, 30, 113 129.
^
Healy, B., 1996. The distribution of Oligochaeta on an exposed
Progress Series, 81, 197 204.
^
rocky shore in southeast Ireland. Hydrobiologia, 334, 51 62.
Heck, K.L. & Wetstone, G.S., 1977. Habitat complexity and
invertebrate species richness and abundance in tropical
Myers, A.A. & Southgate, T., 1980. Arti¢cial substrates as a
means of monitoring rocky shore cryptofauna. Journal of the
^
Marine Biological Association of the United Kingdom, 60, 963 975.
^
seagrass meadows. Journal of Biogeography, 4, 135 142.
Hicks, G.R.F., 1971. Check list and ecological notes on the fauna
Nixon, S.W., Oviatt, C.A., Rodgers, C. & Taylor, R.K., 1971.
^
Mass and metabolism of a mussel bed. Oecologia, 8, 21 30.
^
associated with some littoral corallinacean algae. Bulletin of
Ong Che, R.G. & Morton, B., 1992. Structure and seasonal
variations in the abundance of the macroinvertebrate commu-
nity associated with Septifer virgatus 9Bivalvia: Mytillidae) at
Natural Sciences, 2, 47 58.
^
Hicks, G.R.F., 1977. Species association and seasonal population
densities of marine phytal harpacticoid copepods from Cook
Strait. New Zealand Journal of Marine and Freshwater Research, 11,
Cape d'Aguilar, Hong Kong. Asian Marine Biology, 9, 217 233.
^
Otway, N.M., 1994. Population ecology of the low-shore chitons
621 643.
Onithochiton quercinus and Plaxiphora albida. Marine Biology, 121,
^
Hicks, G.R.F., 1980. Structure of phytal harpacticoid copepod
assemblages and the in£uence of habitat complexity and
turbidity. Journal of Experimental Marine Biology and Ecology, 44,
105 116.
^
Paine, R.T., 1974. Intertidal community structure: experimental
studies on the relationship between a dominant competitor and
its principal predator. Oecologia, 15, 93 120.
157 192.
^
^
Hicks, G.R.F., 1985. Meiofauna associated with rocky shores
Peake, A.J. & Quinn, G.P., 1993. Temporal variation in species
^
algae. In The ecology of rocky coasts 9ed. P.G. Moore and R.
area curves for invertebrates in clumps of an intertidal mussel.
Seed), pp. 36 56. London: Hodder and Stoughton.
Ecography, 16, 269 277.
^
^
Hull, S.L., 1997. Seasonal changes in diversity and abundance of
ostracods on four species of intertidal algae with di¡ering
structural complexity. Marine Ecology Progress Series, 161,
Russ, G.R., 1980. E¡ects of predation by ¢shes, competition, and
the structural complexity of the substratum on the establish-
ment of a marine epifaunal community. Journal of Experimental
71 82.
^
Marine Biology and Ecology, 42, 55 69.
^
Hurlbert, S.H., 1984. Pseudoreplication and the design of ecolo-
Schneider, F.I. & Mann, K.H., 1991a. Species speci¢c relation-
ships of invertebrates to vegetation in a seagrass bed. I.
Correlational studies. Journal of Experimental Marine Biology and
gical ¢eld experiments. Ecological Monographs, 54, 187 211.
^
Johnson, S.C. & Scheibling, R.E., 1987. Structure and dynamics
of epifaunal assemblages on intertidal macroalgae Ascophyllum
nodosum and Fucus vesiculosus in Nova Scotia, Canada. Marine
Ecology, 145, 101 117.
^
Schneider, F.I. & Mann, K.H., 1991b. Species speci¢c relation-
ships of invertebrates to vegetation in a seagrass bed. II.
Experiments on the importance of macrophyte shape epiphyte
cover and predation. Journal of Experimental Marine Biology and
Ecology Progress Series, 37, 209 227.
^
Kelaher, B., 2000. Biodiversity of macrofaunal assemblages in
coralline algal turf. PhD thesis, University of Sydney, Sydney,
Australia.
Ecology, 145, 119 139.
^
Kohn, A.J. & Leviten, P.J., 1976. E¡ect of habitat complexity on
population density and species richness in tropical intertidal
Seed, R., 1996. Patterns of biodiversity in the macro-invertebrate
fauna associated with mussel patches on rocky shores. Journal of
predatory gastropod assemblages. Oecologia, 25, 199 210.
Lewis, J.R., 1964. The ecology of rocky shores. London: English
University Press.
the Marine BiologicalAssociationofthe United Kingdom,76, 203 210.
Seed, R. & O'Connor, R.J., 1980. Shell shape and seasonal
changes in population structure in Lasaea rubra 9Bivalvia:
^
^
Lintas, C. & Seed, R., 1994. Spatial variation in the fauna asso-
ciated with Mytilus edulis on a wave-exposed rocky shore.
Erycinidae). Journal of Molluscan Studies, 46, 66 73.
Southward, A.J., 1958. The zonation of plants and animals on
^
Journal of Molluscan Studies, 60, 165 174.
rocky sea shores. Biological Reviews, 33, 137 177.
^
^
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)

930 B.P. Kelaher et al. Special patterns of macrofauna in coralline turf
Stewart, J.G., 1982. Anchor species and epiphytes in intertidal
Underwood, A.J. & Chapman, M.G., 1996. Scales of spatial
patterns of distribution of intertidal invertebrates. Oecologia,
algal turf. Paci¢c Science, 36, 45 59.
Stoner, A.W., 1980. The role of seagrass biomass in the organiza-
^
107, 212 224.
^
tion of benthic macrofaunal assemblages. Bulletin of Marine
Underwood, A.J. & Chapman, M.G., 1998. Spatial analyses of
intertidal assemblages on sheltered rocky shores. Australian
Science, 30, 537 551.
^
Stoner, A.W., 1982. The in£uence of benthic macrophytes on the
Journal of Ecology, 23, 138 157.
^
foraging behaviour of pin¢sh, Lagodon rhomboides 9Linnaeus).
Underwood, A.J., Chapman, M.G. & Connell, S.D., 2000.
Observation in ecology: you can't make progress on processes
without understanding the patterns. Journal of Experimental
Journal of Experimental Marine Biology and Ecology, 8, 271 284.
^
Tararam, A., Wakabara, Y. & Leite, F.P., 1986. Vertical distribu-
tion of amphipods living on algae of a Brazilian intertidal
Marine Biology and Ecology, 250, 97 115.
^
rocky shore. Crustaceana, 51, 183 187.
Ter Braak, C.J., 1992. Permutation versus bootstrap signi¢cance
tests in multiple regression and ANOVA. In Boot strapping and
Underwood, A.J., Denley, E.J. & Moran, M.J., 1983.
Experimental analyses of the structure and dynamics of mid-
shore rocky intertidal communities in New South Wales.
^
related techniques 9ed. K.H. Jockel et al.), pp. 79 86. Berlin:
Oecologia, 56, 202 219.
^
^
Springer Ve rl ag.
Underwood, A.J. & Petraitis, P.S., 1993. Structure of intertidal
assemblages in di¡erent locations: how can local processes be
compared? In Species diversity in ecological communities: historical
and geographical perspectives 9ed. R.E. Ricklefs and D. Schluter),
^
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. Journal of Experimental Marine Biology and Ecology,
pp. 38 51. Chicago: University of Chicago Press.
^
202, 73 84.
Underwood, A.J. & Verstegen, P.H., 1987. Experiments on
the association between the intertidal amphipod Hyale
media Dana and the limpet Cellana tramoserica 9Sowerby).
Journal of Experimental Marine Biology and Ecology, 119,
^
Tsuchiya, M. & Nishihira, M., 1985. Islands of Mytilus edulis as a
habitat for small intertidal animals: e¡ect of island size on
community structure. Marine Ecology Progress Series, 25, 71 81.
^
Underwood, A.J., 1981. Structure of a rocky intertidal commu-
nity in New South Wales: patterns of vertical distribution and
seasonal changes. Journal of Experimental Marine Biology and
83 98.
^
Whor¡, J.S., Whor¡, L.L. & Sweet, M.H., 1995. Spatial varia-
tion in an algal turf community with respect to substratum
slope and wave height. Journal of the Marine Biological
Ecology, 51, 57 85.
^
Underwood, A.J., 1994. Things environmental scientists 9and
statisticians) need to know to receive 9and give) better statis-
tical advice. In Statistics in ecology and environmental monitoring
Association of the United Kingdom, 75, 429 444.
Wigham, G.D., 1975. The biology and ecology of Rissoa parva 9Da
Costa) [Gastropoda: Prosobranchia]. Journal of the Marine
^
9ed. D. Fletcher and B.F.J. Manly), pp. 33 61. Dunedin:
Biological Association of the United Kingdom, 55, 45 67.
^
^
University of Otago Press.
Winer, B.J., Brown, D.R. & Michels, K.M., 1991. Statistical
principles in experimental design. NewYork: McGraw-Hill.
Underwood, A.J., 1997. Experiments in ecology: their logical design and
interpretation using analysis of variance. Cambridge: Cambridge
University Press.
Submitted 5 February 2001. Accepted 17 October 2001.
Journal of the Marine Biological Association of the United Kingdom ꢀ2001)