The introduced ribbed mussel (Geukensia demissa) in Estero de Punta
Banda, Mexico: interactions with the native cord grass, Spartina foliosa
Mark E. Torchin1,*, Ryan F. Hechinger1, Todd C. Huspeni1, Kathleen L. Whitney1 &
Kevin D. La?erty1,2
1Marine Science Institute and Department of Ecology, Evolution and Marine Biology, University of
California, Santa Barbara, CA 93106, USA; 2US Geological Survey, Western Ecological Research
Center, Santa Barbara, CA 93106, USA; *Author for correspondence (Smithsonian Tropical Research
Institute, Apartado 0843-03092, Balboa, Ancon, Republic of Panama; e-mail: torchinm@si.edu;
fax: +1-507-212-8790)
Received 11 November 2003; accepted in revised form 19 April 2004
Key words: Baja California, biomass, bivalve, clapper rail, Estero de Punta Banda, Geukensia
demissa, introduced species, Mexico
Abstract
Introduced populations of Guekensia demissa occur on the west coast of North America. They have been
reported in San Francisco Bay, four southern California wetlands, and in Estero de Punta Banda (EPB),
Baja California Norte, Mexico. We randomly sampled benthic invertebrates in four habitat types within
EPB: marsh, channel, mud?at and pan. Geukensia demissa was the most abundant bivalve in the wet-
land at EPB. It was signi?cantly associated with the native cordgrass, Spartina foliosa, and occurred at
higher average densities in vegetated marsh sites (24/m2) and Spartina-dominated tidal channels (35/m2),
compared to mud?at (0/m2), and pan (0/m2) sites. We estimated that the total biomass of this invader
was over four times that of the next most abundant bivalve, Tagelus spp., in EPB. We examined G. dem-
issa for parasites and found that only a few native parasites colonized this introduced host at very low
prevalences and intensities. We performed bird surveys to determine the habitat overlap and potential
impact of this mussel on the EPB population of light-footed clapper rails (Rallus longirostrus levipes), an
endangered species in the United States. The high abundance of G. demissa in EPB, its presence in clap-
per rail habitat, and its known e?ects on salt marsh habitat in it?s native range, warrant further investi-
gations of the impact of this invader in EPB and elsewhere.
Introduction
The ribbed mussel, Geukensia demissa, is native
to the Atlantic Coast of North America where it
ranges from the Gulf of Saint Lawrence, Canada
to northeastern Florida, USA (Abbott 1974). It
was inadvertently introduced to San Francisco
Bay, California in the 1800s, presumably along
with eastern oysters (Crassostrea virginica) trans-
ported for aquaculture (Carlton 1979, 1992; Co-
hen and Carlton 1995). Populations were later
discovered in four embayments in Southern Cali-
fornia, including Alamitos, Anaheim, and New-
port Bays, and Bolsa Chica Lagoon (Carlton
1979, 1992; Cohen and Carlton 1995). Regional
introductions within California were likely due to
a combination of ballast water and shell?sh
movements (Carlton 1992). The ribbed mussel
has not been reported in the published literature
from locations south of Orange County, Califor-
nia, including relatively extensive and well-stud-
ied estuarine habitats in San Diego County
Biological Invasions (2005) 7: 607?614 
 Springer 2005
DOI 10.1007/s10530-004-5851-5
(Zedler et al. 1992; Levin et al. 1998). Also, in
recent surveys of benthic invertebrates in the Es-
tero de Punta Banda, Mexico (EPB), no G. dem-
issa were reported (Villareal-Chavez 1991;
Sinicrope Talley and Ibarra-Obando 2000),
albeit, in both studies, marsh habitat, where G.
demissa occurs in greatest densities, may not have
been surveyed. However, biologists from the Uni-
versidad Autonoma de Baja California (UABC)
are aware that G. demissa occurs at EPB. Mar-
cos Miranda-Aguilar identi?ed the species in the
early 1980s (Guillermo Viareal-Chavez, UABC,
personal communication). In addition, Fernando
Garcia-Pamanes (personal communication) and
colleagues collected a specimen ca. 1992.
We report on the distribution, abundance, and
biomass of the introduced ribbed mussel, G. dem-
issa, at EPB. We also examine the association
between the introduced mussel and native cord-
grass (Spartina foliosa), assess the degree of para-
sitism of the introduced mussel, and discuss
potential impacts of this invasion.
Materials and methods
The Estero de Punta Banda (EPB) is a tidal wet-
land located at 3145? N, 11637? W, on the Paci-
?c Coast of Baja California Norte, Mexico
(Figure 1). EPB is approximately 10 by 1 km in
dimension and possesses several habitat types:
tidal channels, subtidal lagoon, intertidal pans,
?ats, and salt marsh. This coastal area has a
Mediterranean climate with long, hot, dry sum-
mers and cool winters. Winter rainfall is unpre-
dictable with mean annual precipitation usually
around 200 mm and 500 mm in sporadic wet
years (Ibarra-Obando and Poumian-Tapia 1991).
There is no continual fresh water input, and
evaporation exceeds precipitation in summer and
fall (Pritchard et al. 1978).
We conducted a detailed sampling of the ben-
thic invertebrates of EPB in October and Novem-
ber 2002. Di?erent habitat types were digitized on
a georecti?ed 1 m resolution satellite image of
EPB taken on 13 April 2001 (Figure 1). Twenty-
three coordinates representing ?ve marsh, seven
channel, seven mud?at, and four pan sites were
randomly selected using ArcView 3.2. At each
site, we laid out six parallel transects spaced 2 m
apart. We placed adjacent 10 ? 50 cm quadrats
lengthwise along each transect, and recorded the
presence of epifaunal taxa in each quadrat. From
these, we randomly selected up to 20 quadrats
and collected all G. demissa in each, measuring
the length, height, and width of each mussel using
calipers (to 0.1 mm). We took a sub-sample of
mussels from each site for parasitological exami-
nation (total ? 78 mussels). We deposited vou-
cher G. demissa at the Santa Barbara Museum of
Natural History.
To examine the association between G. demissa
and the native cordgrass, S. foliosa, at each site,
Figure 1. Habitat map of Estero de Punta Banda, Baja Cali-
fornia, Mexico, produced from an April 1991 satellite image.
608
we also conducted a survey of the vegetation. We
used point-contact measurements every 10 cm
along three 10 m transects that extended across
our site. From this, we determined the percent
coverage of native cordgrass at each site. For a
more re?ned comparison, we also examined the
association between the introduced mussel and
the native cordgrass at one focal site. For this,
we divided the 10 ? 10 m2 site into a 100 1 m2
grid and determined the presence or absence of
mussels and cordgrass in each square. We then
randomly selected 60 of these squares, placed a
10 ? 10 cm2 quadrat at the lower right corner of
each, and counted individual mussels and cord-
grass stems within this quadrat.
We also examined the association between G.
demissa and S. foliosa density vs percent sand and
percentage of organic content in the sediment. At
each site, and at every other randomly selected
quadrat, we collected sediment samples using
6.5 cm long ? 2.5 cm diameter cores. We took 10
sediment samples per site, which were kept on ice
and then frozen for future analysis. In the labora-
tory, we determined percent sand and percent
organic content. Percent sand was determined by
sieving sediments through 63 lm mesh sieve, and
percent organic content was determined by com-
busting samples at 475 C for 12 h.
We estimated the total number and biomass of
G. demissa in EPB. The total area of each habitat
polygon was calculated using ArcView 3.2. We
then multiplied the density of mussels in each
habitat type (as obtained above) by the summed
area of each habitat type to generate an estimate
of total mussel abundance. To estimate biomass
of mussels in EPB, we ?rst determined the rela-
tionship between shell volume (length ? width ?
height) and shell and soft tissue mass, using lin-
ear regression analyses on two samples of mus-
sels. We used these relationships to convert the
size-frequency (volume) distribution of mussels to
a mass-frequency distribution for each habitat.
Then, using mean mass estimates for each habi-
tat, we estimated the biomass of mussels in the
entire wetland.
To determine whether mussels have the poten-
tial to impact clapper rails in EPB, we assessed
whether clapper rails occurred in the same habitat
as mussels (i.e., Spartina marsh). We performed
standardized bird surveys at each of our 23 ran-
domly selected sites. The bird survey area was a
100 m diameter circle around each site. We
conducted surveys between 25 June and 2 July
2003. Each site was surveyed four times between
dawn and dusk: twice during the lowest tidal
range (0?0.7 m) and twice at higher tides (0.75?
1.4 m). We used timed 15 min circular walking
transects at a 50 m radius from the center of the
site to detect birds in the vegetated habitat; no
taped calls or other means of eliciting territorial
response were used. We also mapped the location
of Spartina vegetation within the bird survey area.
For statistical analyses, we used non-paramet-
ric tests, as the data were not normally distrib-
uted (Shapiro?Wilks, P < 0.05 for density and
size data). We estimated 95% con?dence intervals
for habitat means of density and size by boot-
strapping the data with 10,000 iterations (Efron?s
percentile method (Manly 1997)). Bootstrapping
was preferable to parametric estimation since the
latter gives impossible lower con?dence limits
(i.e., less than zero). To examine the association
between G. demissa and S. foliosa, we used the
Spearman rank order correlation coe?cient (rs).
The Kruskal?Wallis test was used to compare G.
demissa density in the four habitats we sampled,
and to compare G. demissa density in sites with
and without S. foliosa. We used v2 to compare
the association between mussel presence and
cordgrass presence within blocks at the single
focal site.
Results
Mussel distribution, densities, size and association
with native cordgrass
Geukensia demissa was common at Estero de
Punta Banda and generally associated with native
cordgrass (Spartina foliosa). Sites where cord-
grass was present had an average density of
32 mussels/m2 compared to 1 mussel/m2 at sites
where cordgrass was not present in our transects
(v2 ? 7.6, P ? 0.006, df ? 1, n ? 23). There was
no association between G. demissa or S. foliosa
density and sediment grain size (rs ? )0.26, P >
0.05, n ? 23 and rs ? )0.36, P > 0.05, n ? 23,
respectively). There was however a positive asso-
ciation between both G. demissa and S. foliosa
609
density and organic composition of sediment
(rs ? 0.73, P < 0.0001, n ? 23 and rs ? 0.58,
P < 0.005, n ? 23, respectively).
Within our focal site, there was a signi?cant
association between mussel and cordgrass pres-
ence (v2 ? 71.9, P < 0.0001, df ? 1, n ? 100). At
our focal site, mussel density was positively cor-
related with cordgrass density (Figure 2,
rs ? 0.82, P = 0.0001, n ? 60). There was signi?-
cant variation in mussel density among habitats
(v2 ? 11.5, P ? 0.009, df ? 3, n ? 23). Vegetated
channel and marsh sites had high densities of
mussels and no mussels occurred at mud?at and
pan sites (Table 1). At four of the ?ve vegetated
marsh sites, cordgrass was the dominant ground-
cover. Interestingly, there were no mussels pres-
ent at the one marsh site where cordgrass was
absent, and where the vegetation was dominated
by Salicornia bigelovii. The mean length of mus-
sels was signi?cantly greater at channel sites
(63.3 mm, n ? 47) than at marsh sites (58.0 mm,
n ? 60) (Figure 3; v2 ? 4.3, P ? 0.037, df ? 1).
Mussel abundance and biomass in EPB
The total vegetated marsh area of EPB was 3.56
km2, with channels representing a negligible area
(0.15 km2) in comparison (Figure 1; Table 1).
Mean mussel density in marsh was 24/m2 and
there were an estimated 90.4 ? 106 total intro-
duced mussels at EPB (see Table 1). In compari-
son, the most common native bivalves (albeit in
di?erent habitats ? channels, pans, and mud?ats)
were the native jackknife clams, Tagelus californi-
anus, T. a?nis and T. subteres. Using random
coring and the same statistical method, the total
number of jackknife clams (pooling Tagelus spe-
cies) at EPB was 69.5 ? 106 (15.3 ? 106?150.4 ?
106) (R.F. Hechinger et al., in preparation).
For G. demissa the relationship between shell
mass and shell volume was: sm ? 0.00017 ?
(sv) )0.28 (R2 ? 0.98, P < 0.0001, n ? 38),
where sm is shell mass (g) and sv is shell volume
(mm3). The equation for the relationship between
wet soft tissue and shell volume was: tm ?
0.000086 ? (sv) + 1.4 (R2 ? 0.88, P < 0.0001,
n ? 49), where tm is tissue mass (g) and sv is
shell volume (mm3). The total estimated biomass
of G. demissa at EPB was 369.6 metric tons wet
soft-tissue weight and 837.6 metric tons total
weight (Table 1). For comparison, we estimated
the total biomass of the most abundant native
bivalves, Tagelus spp., to be 81.9 (28.8 ? 152.7)
metric tons wet soft-tissue weight and 183.3 (61.7
? 346.9) metric tons total weight.
Mussel parasites
Three species of parasites infected G. demissa at
EPB. Many mussels harbored an unidenti?ed cy-
clopoid copepod in the mantle cavity. Although
most mussels we examined had at least one cope-
pod, we cannot determine their prevalence in
EPB, as we suspect that transmission occurred in
aquaria used to hold our animals. Additionally,
2% and 8% of mussels were infected with the
native trematodes, Acanthoparyphium spinulosum
and Renicola sp., respectively. The mean number
of metacercarial cysts (?1 SE) per infected mussel
was 2 (?0) for A. spinulosum and 1.6 (?0.6) for
the Renicola sp. In comparison, we found eight
parasite species in the 83 native jackknife clams
we dissected. All jackknife clams were infected
with A. spinulosum and 60% were infected with
the Renicola sp. The mean number of metacercar-
ial cysts per infected jackknife clam was 355 (?81)
for A. spinulosum and 11 (?3) for Renicola sp.
Habitat overlap with clapper rails
We frequently detected light-footed clapper rails
in Spartina marsh during our general bird sur-
Figure 2. Scatter plot and trend line of native cordgrass,
Spartina foliosa, cover and introduced ribbed mussel, Geuken-
sia demissa, density within 10 ? 10 cm2 quadrats (rs ? 0.82,
P ? 0.0001, n ? 60) at a site in Estero de Punta Banda, Baja
California, Mexico.
610
veys at EPB. Spartina vegetation occurred within
the bird survey area at 17 of our 23 sites. We
observed territorial clapper rails in 15 of these 17
sites. Although general bird surveys undersample
clapper rails, we detected one to three clapper
rails at each of these sites, and every rail we
observed was in Spartina vegetation. Although
nest searching was not speci?cally performed, we
found active clapper rail nests at four of these
sampling sites; one site contained three incuba-
tion nests, and numerous ?edging platforms. In
total, we observed 27 clapper rails in our survey.
Discussion
The introduced Atlantic ribbed mussel, G. demissa,
is well established in Estero de Punta Banda, Baja
California, Mexico. The biomass of this invader
was 4.6 times that of the next most abundant
bivalve, Tagelus spp. The high abundance and
high biomass of G. demissa in EPB suggests it is
likely impacting this coastal wetland. Additionally,
our estimate of G. demissa abundance may be con-
servative considering that ribbed mussels generally
occur at highest densities on the lower edge of the
marsh (Nielsen and Franz 1995; Franz 1997),
areas which were not speci?cally targeted in our
study.Ta
bl
e
1.
D
en
si
ty
,
ab
un
da
nc
e,
an
d
bi
om
as
s
of
G
eu
ke
ns
ia
de
m
is
sa
in
E
st
er
o
de
P
un
ta
B
an
da
.
A
ve
ra
ge
A
ve
ra
ge
in
di
vi
du
al
m
as
s
(g
)
H
ab
it
at
ar
ea
(k
m
2 )
T
ot
al
nu
m
be
rs
(m
ill
io
ns
of
in
di
vi
du
al
s)
(9
5%
c.
i.)
T
ot
al
so
ft
ti
ss
ue
m
as
s
(m
et
ri
c
to
ns
)
(9
5%
c.
i.)
T
ot
al
bi
om
as
s
(m
et
ri
c
to
ns
)
(9
5%
c.
i.)
D
en
si
ty
(/
m
2 )
So
ft
ti
ss
ue
T
ot
al
(9
5%
c.
i.)
(9
5%
c.
i.)
(9
5%
c.
i.)
C
ha
nn
el
35
.4
(1
.2
?9
8.
3)
3.
9
(3
.4
?4
.3
)
10
.3
(8
.8
?1
1.
8)
0.
15
5.
3
(0
.3
?1
4.
4)
20
.7
(2
.2
?4
0.
5)
54
.9
(5
.8
?9
4.
9)
F
la
t
0
0
0
3.
15
0
0
0
M
ar
sh
23
.9
(1
1.
6?
33
.2
)
4.
1
(3
.6
?4
.7
)
9.
2
(7
.7
?1
0.
9)
3.
56
85
.1
(4
5.
7?
11
5.
0)
34
8.
9
(2
39
.2
?4
70
.3
)
78
2.
6
(6
29
.4
?8
97
.9
)
P
an
0
0
0
0.
01
0
0
0
Su
m
to
ta
l
E
P
B
6.
87
90
.4
(4
5.
9?
12
9.
3)
36
9.
6
(2
41
.4
?5
10
.8
)
83
7.
6
(5
32
.4
?1
18
2.
1)
Figure 3. Size frequency distribution of the introduced ribbed
mussel, Geukensia demissa in channel (open bars) and marsh
sites (solid bars) in Estero de Punta Banda, Baja California,
Mexico. Mean size in channel sites is signi?cantly greater than
in marsh sites which may be a result of di?erences in tidal ele-
vation between the two habitats (see Franz 2001).
611
Geukensia demissa exhibits su?ciently di?erent
habitat use compared to native bivalves and
probably does not compete directly with native
bivalve species for space in EPB. However, it is
more likely that G. demissa populations compete
with native bivalves and other suspension feeders
for food because G. demissa are capable of ?lter-
ing large volumes of water. In a small marsh
within its native range, the G. demissa population
?ltered a volume of water exceeding the total vol-
ume of water in the entire marsh each tidal cycle
(Jordan and Valiela 1982). Mussels ?ltered 1.8
times the particulate nitrogen exported from the
marsh and absorbed half of this, excreting the
other half, which resulted in biodeposition (Jor-
dan and Valiela 1982). Much of the resulting
mussel production entered the food web in the
form of gametes (Jordan and Valiela 1982). In its
native range, G. demissa has been shown to
ingest organic particles from a wide variety of
sources, including Spartina detritus (Kreeger and
Newell 2001). We note that a direct comparison
of EPB with east coast systems would require an
examination of di?erences in marsh tidal eleva-
tion between S. foliosa and S. alterni?ora. Recent
multiple stable isotope work in southern Califor-
nia indicates that S. foliosa supplies the bulk of
the organic matter that sustains wetland ?shes
via detrital-based food webs (Kwak and Zedler
1997). Geukensia demissa may therefore signi?-
cantly alter the EPB food web.
As is common for introduced species (Torchin
et al. 2003), compared to native bivalves, ribbed
mussels in EPB do not appear to serve as major
hosts for parasites. We found relatively few spe-
cies of parasites at low prevalences and low num-
bers per infected host. The potential exception to
this is the single species of copepod (for which
we lack conclusive data). This lack of parasites
could confer an advantage to G. demissa. How-
ever, we found no reports of metazoan parasites
infecting it on the east coast for a comparison.
Ford et al. (1995) indicate that G. demissa is not
known to become infected with the protozoan
oyster pathogen Haplosporidium nelsoni. Evaluat-
ing this hypothesis will require substantially more
information on parasites in both native and
introduced regions.
The most dramatic impact of G. demissa could
be its a?ect on vegetation. The association that
we found between the mussel and S. foliosa has
also been well documented in salt-marshes on the
Atlantic coast of North America, where G. dem-
issa attaches to the roots of S. alterni?ora by
byssal threads (Bertness 1980, 1984; Fell et al.
1982). Cordgrass production increases growth
and reproduction of the mussel, and the presence
of the mussel increases production of the cord-
grass, which results in increased marsh net pri-
mary production and stability (Bertness 1984). If
increased stability and deposition result in eleva-
tion increases su?cient for recruitment and
growth of cordgrass, this could expand marsh
habitat onto tidal ?ats and increase the represen-
tation of cordgrass in the wetland (Bertness
1984). Facilitation of cordgrass by G. demissa
could decrease habitat for other benthic mud-
dwelling invertebrates, and many of the birds
that tend to forage on tidal ?ats. However, an
increase in cordgrass habitat could favor some
species, such as clapper rails.
The most unusual impact of this introduced
mussel is that it has the potential to trap and
drown young birds foraging in marshes because
the posterior margins of the mussel?s shell typi-
cally protrude above the surface of the mud and
can clamp onto the beaks and feet of small birds.
This has been documented for endangered Cali-
fornia clapper rails (Rallus longirostris obsoletus),
which commonly feed on G. demissa in San
Francisco Bay (see de Groot 1927; Carlton 1979;
Takekawa 1993; Cohen and Carlton 1995). Here,
clamping by G. demissa causes toe loss in 75% of
adult clapper rails and may kill one or two
young rails of each brood (de Groot 1927). At
EPB, there is considerable habitat overlap
between G. demissa and the light-footed clapper
rails (Rallus longirostris levipes), creating the
potential for this to occur.
The widespread occurrence and high abun-
dance of G. demissa in Estero de Punta Banda,
suggests that the likely e?ects of this introduced
mussel are pronounced. The mussel?s ability to
facilitate cord grass growth and productivity, ?l-
ter the water column, and drive biodeposition
suggest it is altering the ecology and nutrient
cycling in EPB. These issues, as well as the
potential impact on clapper rail populations,
merit further exploration and evaluation. Should
the impact be deemed substantial, consideration
612
could be given to control. Fishing is one poten-
tial method for controlling introduced marine
pests (La?erty and Kuris 1996), and some have
suggested that G. demissa could be commercially
harvested from the Gulf coast of Mexico (Baque-
iro et al. 1993). The bene?t of reducing the
abundance of G. demissa in EPB via ?shing
would have to be weighed against the environ-
mental disturbance of ?shing for mussels in the
estuary, as well as the potential a growing market
demand could encourage the intentional spread
of G. demissa to other estuaries.
Acknowledgements
We thank C. Boch, M. Caun, E. Dunham, V.
McKenzie, and J. Shaw for assistance in the
?eld. L. Ladah and her lab helped with logistics
in the ?eld. We especially thank E. Dunham for
her work with GIS mapping. We thank P. Scott
(Santa Barbara Museum of Natural History) for
con?rming our identi?cation of Guekensia demis-
sa. We also thank J. Lorda for assistance with
the Spanish abstract. This manuscript has also
bene?ted from support received from the
National Science Foundation through the NIH/
NSF Ecology of Infectious Disease Program
(DEB-0224565), and a grant from the US Envi-
ronmental Protection Agency?s Science to
Achieve Results (STAR) Estuarine and Great
Lakes (EaGLe) program through funding to the
Paci?c Estuarine Ecosystem Indicator Research
(PEEIR) Consortium, US EPA Agreement #R-
882867601. However, it has not been subjected to
any EPA review and therefore does not necessar-
ily re?ect the views of the Agency, and no o?cial
endorsement should be inferred.
References
Abbott RT (1974) American Seashells. Van Nostrand Rein-
hold, New York
Baqueiro Cardenas E, C Castillo R, C Manuel Medina M
and M Huchin Maturel (1993) Potential pesquero del mej-
illon amarillo, Geukensia demissa, de las costas de los esta-
dos de Campeche y Yucatan, Mexico. Ciencia Pesquera 9:
63?72
Bertness MD (1980) Growth and mortality in the ribbed mus-
sel Geukensia demissa. The Veliger 23: 62?69
Bertness MD (1984) Ribbed mussels and Spartina alterni?ora
production in a New England salt marsh. Ecology 65:
1794?1807
Carlton JT (1979) History, Biogeography, and Ecology of
Introduced Marine and Estuarine Invertebrates of the Paci-
?c Coast of North America. University of California,
Berkeley, 904 pp
Carlton JT (1992) Introduced marine and estuarine mollusks
of North America: an end-of-the-20th-century perspective.
Journal of Shell?sh Research 11: 489?505
Cohen AN and Carlton JT (1995) Nonindigenous Species in a
United States Estuary: A Case History of the Ecological
and Economic E?ects of Biological Invasions in the San
Francisco Bay and Delta Region, US Fish and Wildlife
Service, Washington, DC
de Groot DS (1927) The California clapper rail: its nesting
habits, enemies and habitat. Condor 29: 259?270
Fell PE, Olmstead NC, Carlson E, Jacob W, Hitchcock D
and Silber G. (1982) Distribution and abundance of
macroinvertebrates on certain Connecticut tidal marshes,
with emphasis on dominant molluscs. Estuaries 5:
234?239
Ford SE, Ashton-Alcox KA and Kanaley SA (1995) In vitro
interactions between bivalve hemocytes and the oyster
pathogen Haplosporidium nelsoni (MSX). Journal of Para-
sitology 79: 255?265
Franz DR (1997) Resource allocation in the intertidal salt-
marsh mussel Geukensia demissa in relation to shore level.
Estuaries 20: 134?148
Franz DR (2001) Recruitment, survivorship and age struc-
ture of a New York ribbed mussel population (Geukensia
demissa) in relation to shore level ? a nine year study.
Estuaries 24: 319?327
Ibarra-Obando SE and Poumian-Tapia M (1991). The e?ect
of tidal exclusion on salt marsh vegetation in Baja Califor-
nia, Mexico. Wetlands Ecology and Management 1: 131-
148
Jordan TE and Valiela I (1982) A nitrogen budget of the
ribbed mussel, Geukensia demissa, and its signi?cance in
nitrogen ?ow in a New England salt marsh. Limnology
and Oceanography 27: 75?90
Kreeger DA and Newell RIE (2001) Seasonal utilization of
di?erent seston carbon sources by the ribbed mussel, Geu-
kensia demissa (Dillwyn) in a mid-Atlantic salt marsh.
Journal of Experimental Marine Biology and Ecology 260:
71?91
Kwak TJ and Zedler JB (1997) Food web analysis of south-
ern California coastal wetlands using multiple stable iso-
topes. Oecologia 110: 262?277
La?erty KD and Kuris AM (1996) Biological control of mar-
ine pests. Ecology 77: 1989?2000
Levin LA, Sinicrope Talley TS and Hewitt J (1998) Macro-
benthos of Spartina foliosa (Paci?c Cordgrass) salt marshes
in southern California: community structure and compari-
son to a Paci?c mud?at and a Spartina alterni?ora (Atlan-
tic Smooth Cordgrass) marsh. Estuaries 21: 129?144
Manly, BFJ (1997) Randomization, Bootstrap and Monte
Carlo Methods in biology, 2nd edn, Chapman and Hall,
London
613
Nielsen KJ and Franz DR (1995) The in?uence of adult con-
speci?cs and shore level or recruitment of the ribbed mus-
sel Geukensia demissa (Dillwyn). Journal of Experimental
Marine Biology and Ecology 188: 89?98
Pritchard DW, de La Paz Vela R, Cabrera Muro H, Farreras
Sanz S and Morales E. (1978) Hidrogra?a ?sica del Estero
de Punta Banda parte I: Analisis de datos. Ciencias Mari-
nas 5: 1?23
Sinicrope Talley TS and Ibarra-Obando SE (2000) Tidal ?at
macrofaunal communities and their associated environ-
ments in estuaries of the Southern California and Northern
Baja California, Mexico. Estuaries 23: 97?114
Takekawa J (1993) The California clapper rail: turning the
tide? Tideline (East Bay Regional Park District) 13: 1?3, 11
Torchin ME, La?erty KD, Dobson AP, McKenzie VJ and
Kuris AM (2003) Introduced species and their missing par-
asites. Nature 421: 628?630
Villareal-Chavez G (1991) Impacto de la construccion de un
dique arenoso en la ecologia bentonica de un estuario en el
paci?co norte de mexico. Ciencias Marinas 17: 105?119
Zedler JB, Nordby CS and Kus BE (1992) The Ecology of
Tijuana Estuary, California: A National Estuarine
Research Reserve. NOAA O?ce of Coastal Resource
Mangement, Washington, DC
614