BULLETIN OF MARINE SCIENCE, 71(3): 1343–1351, 2002
THE ECOLOGICAL IMPORTANCE OF AN INVERTEBRATE
CHEMOAUTOTROPHIC SYMBIOSIS TO PHANEROGAM
SEAGRASS BEDS
M. A. Johnson, C. Fernandez and G. Pergent
ABSTRACT
The symbiotic chemoautotrophic bivalve Loripes lacteus was found to inhabit
Cymodocea nodosa seagrass beds in a lagoon in Upper Corsica. Clams were observed at
a mean density of 775 ind m−2. Mean clam wet weight for the site was 0.099 mg and the
gill, organ in which are found the sulfur-oxidizing endosymbiotic bacteria, accounted for
32.5% of total body weight. Total wet tissue weight due to these animals in this sediment
was therefore in the order of 77 g m−2. The percentage of carbon is 11.2% of wet weight.
A rough estimate of net clam production within the seagrass bed yields a value of 1.73 g
C m−2 yr−1. The autotrophic potential of Loripes lacteus was calculated to be in the order
of 47.2 g C m−2 yr−1, which represents roughly 16% of the seagrass bed’s primary production. The role of these symbioses, in terms of carbon flux, within phanerogam seagrass
beds is discussed.
Chemoautotrophic symbioses are nutritionally based associations between sulfur-oxidizing chemoautotrophic bacteria and marine invertebrate or protist hosts. Marine molluscs that contain symbiotic sulfur-oxidizing chemolithotrophic bacteria have been identified from a variety of sulfide-rich habitats, such as sewage outfall areas (Felbeck et al.,
1981; Johnson et al., 1994), pulp mill effluent sites (Reid, 1980), anoxic basins (Felbeck
et al., 1981), mangrove swamps (Schweimanns and Felbeck, 1985; Frenkiel et al., 1996),
seagrass beds (Fisher and Hand, 1984), and hydrothermal vents (Cavanaugh, 1983). A
common feature of these habitats is the simultaneous availability of hydrogen sulfide and
molecular oxygen. This dissolved sulfide (HS− and H2S) provides the energy source used
to drive the reductive bacterially-mediated reactions of carbon fixation (Childress and
Mickel, 1982; Felbeck, 1983; Cavanaugh, 1983; Fisher, 1990).
This sulfide can originate either from geothermal sources, such as at hydrothermal
vents, or from decaying organic matter as, for example, in seagrass meadows. A large
portion of the plant material produced in these latter habitats is eventually deposited on
the bottom as detritus, leading to high organic content and high oxygen demand. As a
result, the abundance of organic material in seagrass sediments is often greater than the
oxygen available for its degradation, leading to an elevated activity of sulfate-reducing
bacteria and the production of high levels of hydrogen sulfide (Jorgensen and Fenchel
1974). Indeed, Solemya velum (Solemyidae) has been reported from eelgrass beds that
‘smelled strongly’ of hydrogen sulfide (Cavanaugh, 1983), Codakia orbicularis (Lucinidae)
occurs in beds of the seagrass Thalassia testudinum in zones of ‘high sulphide concentration’ (Berg and Alatolo, 1984) and decaying seagrass leaves leading to the production of
sulfide are characteristic of the habitat of Lucina floridana (Lucinidae) (Fisher and Hand,
1984).
Seagrass beds are now recognized as one of the most productive of marine communities. Primary production values ranging from 500 to over 3000 g C m−2 yr−1 have been
proposed (Fenchel, 1977; Zieman and Wetzel, 1980; Stevenson, 1988). Cymodocea nodosa
(Ucria) Ascherson is widely distributed throughout the Mediterranean (Terrados and Ros,
1343
1344
BULLETIN OF MARINE SCIENCE, VOL. 71, NO. 3, 2002
1992) and can form vast underwater meadows in sheltered bays and coastal lagoons
(Ballesteros et al., 1989). Marine mollusks containing sulfur-oxidizing chemoautotrophic
bacteria can be quite abundant in such seagrass beds, with densities that range from 4 to
1500 ind m−2 (Dando et al., 1986a; Monnat, 1970) and can form up to 50% of the infaunal
biomass (Buchanan, 1963). The symbiotic bacteria are believed to provide the host with
part of its nutritional requirements by chemosynthetically fixing carbon dioxide via aerobic oxidation of sulphide (Cary et al., 1989; Dando and Spiro, 1993; Johnson et al., 1994).
In light of their population densities, these symbiotic associations may play a non-negligible role in the cycling of carbon in seagrass ecosystems, a contribution which has received very little attention to date (Fisher and Hand, 1984).
The present work is a study of a lucinid bivalve, Loripes lacteus, inhabiting C. nodosa
seagrass beds in a lagoon in upper Corsica (France). An attempt was made to determine
the prevalence of these symbiotic associations and to estimate their possible role in the
flux of carbon within seagrass ecosystems.
MATERIALS AND METHODS
Sampling was performed within the Urbino lagoon from January to October 2000. Clams were
sampled quantitatively by SCUBA diving. A sediment corer of approximately 162 cm2 was used for
a sediment depth of 0–20 cm. This is the maximum depth at which lucinids have been found within
the sediment (Dando et al., 1986b). The sediment samples were sieved on a 1 mm mesh to retrieve
the lucinid clams. The clams were dissected, and the clam soft tissue was removed, blotted to
remove excess water and weighed. The gills were then dissected, blotted and weighed. Bivalve
densities were calculated on a m2 basis as was clam biomass.
Clam organic content was determined by weighing 15 clams (wet weight), dessicating these
samples at 70°C until constant weight (dry weight) and by the subsequent combustion in a muffle
furnace at 500°C for 48 h (% carbon).
Means, when calculated, are expressed as the mean + standard deviation. These results were
used to approximate net annual clam production based on the dry weight to carbon conversion
factor established here and assuming 5 yrs to reach adult size (Dando et al., 1986b). This net clam
production is thus the total carbon production of the population calculated on a yearly basis and
which does not consider production losses due to respiration, reproduction, predation or mortality.
The bacterial member of these chemoautotrophic symbioses uses sulfide to drive the reductive
reactions of carbon fixation. This is performed via the Calvin Benson cycle using the CO2-fixing
enzyme ribulose 1,5-bisphosphate carboxylase (RuBCase). Another means of evaluating the production of this association therefore involves calculating the autotrophic potential using RuBCase
activity. Activity values for this enzyme within the host gill tissue have been determined for a
number of lucinid species, with activities that range from roughly 10 to 800 units g−1 wet weight
(Dando et al., 1986b; Fisher and Childress, 1986). Here, a RuBCase value of 300 units g−1 wet
weight was adopted as this value represents an average value found in lucinid chemoautotrophic
symbioses. RuBCase activity units are in nmoles CO2 fixed min–1.
RESULTS
Clam densities ranged from 242 to 2666 ind m−2. Mean clam densities were in the order
of 775 ± 364 ind m−2 (Fig. 1). Total clam individual weight ranged from 0.0017 to 0.4874
g wet weight. Mean clam wet weight for the site was thus 0.099 ± 0.086 g ind−1 (Fig. 2).
Mean water content of the clams was determined to be in the order of 52%. The gill
accounted for 32.5 ± 6.3% of total body weight.
JOHNSON ET AL.: IMPORTANCE OF AN INVERTEBRATE CHEMOAUTOTROPHIC SYMBIOSIS
1345
Figure 1. Mean clam density (±S.D.) throughout the sampling period.
Based on a mean wet weight of 0.099 g, the total wet tissue weight due to L. lacteus in
this sediment was 77 g m−2. The wet weight to carbon ratio was calculated to be 11.2 ±
1.5%. Thus, converting wet weight to carbon using this ratio and assuming 5 yrs to reach
adult size (Dando et al., 1986b) gave a fixation rate of 1.73 g C m−2 yr−1.
Using a RuBCase value of 300 units g−1 fresh weight, the autotrophic potential of L.
lacteus was estimated to be in the order of 47.2 g C m−2 yr−1.
DISCUSSION
The density of lucinid populations within seagrass beds appears to vary considerably
depending on both the locality and seagrass species. Indeed, densities recorded in the
literature can be seen to range from 3.8 to 1500 ind m−2 (Table 1). The densities recorded
in the present study are thus among the highest values to have been recorded. The great
variability in density between the sampling months is to be expected and can be attributed to such factors as recruitment, predation and natural mortalities. For calculation
purposes, however, a mean of these monthly values was calculated.
Mean biomass per individual also varied throughout the sampling period with lower
than average biomass during periods of recruitment (January, May and June). Here again,
a mean value was generated for calculation purposes.
The net clam production for L. lacteus was calculated to be in the order of 1.73 g C m−2 yr−1.
A value of 2.24 g C m−2 yr−1 was obtained for Lucina floridana (calculated from data in
Fisher and Hand, 1984, a gill percentage of 30% was assumed (mean of several studies)). For Lucinoma borealis and Myrtea spinifera, values of 0.02 g C m−2 yr−1 (calculated from data in Dando et al., 1986a) and 0.0026 g C m−2 yr−1 (based on data from
Dando et al., 1986b) were calculated (Table 2). These values are a rough estimate of the
net clam production within the seagrass bed. As can be observed in Table 2, this net clam
1346
BULLETIN OF MARINE SCIENCE, VOL. 71, NO. 3, 2002
Figure 2. Mean wet weight of clams (±S.D.) throughout the sampling period.
production will depend on the density of the clams and the weight of the species examined. The value found here is comparable to that of Fisher and Hand (1984) for L.
floridana. This last species was observed at lower densities than that of the present study
but is a particularly large bivalve. The very low values for L. borealis and M. spinifera
(Dando et al., 1986a) can be attributed to both the very low densities at which they are
found and the relatively small size of the animals.
Another means of evaluating the production of this association is by making use of the
RuBCase activity within the gill tissue. Activity values for this enzyme within the host
Table 1. Lucinid species found in seagrass beds and their densities, data from the present study
and values obtained from the literature. ND (no data) = bivalves observed but no density values
given.
Bivalve species
Seagrass
Wallucina assimilis
Posidonia australis
Amphibolis antarctica
Mixed seagrass bed
Thalassia
Thalassia
Codakia orbicularis
C. costata and
Ctena orbiculata
(together)
Linga pensylvanica
Solemya velum
Lucina floridana
L. floridana
Lucinoma borealis
L. borealis
Lucinella divaricata
Loripes lacteus
Thalassia
Zostera
Thalassia
Ruppia
Zostera
Zostera
Zostera
Cymodocea
Density
(ind m−2)
718 ± 357
565 ± 259
1,048 ± 267
ND
26
Rare
ND
84 ± 12
83 ± 11
74 ± 34
3.8
120−1500
200−300
775 ± 364
Reference
Barnes and Hickman, 1999
Berg and Alatolo, 1984
Aurelia, 1969
Schweimans and Felbeck, 1985
Cavanaugh, 1983
Fisher and Hand, 1984
Fisher and Hand, 1984
Dando et al., 1986a
Monnat, 1970
Monnat, 1970
Present study
JOHNSON ET AL.: IMPORTANCE OF AN INVERTEBRATE CHEMOAUTOTROPHIC SYMBIOSIS
1347
Table 2. Net clam production determined for Loripes lacteus as well as for several other lucinids
present in seagrass meadows (values calculated based on literature data).
Species
Loripes lacteus
(present study)
Lucina floridana (Fisher
and Hand, 1984)
Myrtea spinifera
(Dando et al., 1985, 1986)
Lucinoma borealis
(Dando et al., 1986a)
Density
(ind m−2)
775
Mean individual
weight (g)
0.099
Biomass
(g m−2)
77
Net production
(g C m−2 yr−1)
1.73
80
1.4
112
2.24
20
0.065
0.13
0.0026
4
0.25
1.0
0.02
Based on a 11.2% wet weight to carbon conversion factor.
Assuming 5 yrs to reach adult size.
gill tissue have been determined for a number of lucinid species, with activities that range
from roughly 10 to 800 nmoles CO2 fixed min−1 g−1 fresh weight (Dando et al., 1986a;
Spiro et al., 1986; Herry et al., 1989; Johnson et al., 1994; Dando et al., 1986b; Dando et
al., 1985; Spiro et al., 1986; Fisher and Childress, 1986; Schweimanns and Felbeck, 1985).
These values have been used to calculate the autotrophic potential of such a symbiotic
relationship (Table 2) (Fisher and Hand, 1984). In this last study, the authors used RuBCase
values determined in vitro to estimate that L. floridana could potentially contribute 336 ±
96 g C m−2 yr−1 to gross carbon fixation in seagrass beds. Similar calculations were performed here for L. borealis, producing a value of 0.35 g C m−2 yr−1 (based on data from
Dando et al., 1986a, calculated assuming a RuBCase activity of 200 units g−1 fresh weight
(average value from several studies)). A value of 0.074 g C m−2 yr−1 was similarly calculated for M. spinifera (based on data from Dando et al., 1986b). For L. lacteus of the
present study, a RuBCase value of 300 units g−1 wet weight was adopted as this value
represents an average value found in lucinid symbioses. It was thus calculated that the
autotrophic potential of L. lacteus in a C. nodosa seagrass bed was in the order of 47 g C
m−2 yr−1. Of course, this fixation value is only an estimate, since in vitro enzyme activity
values are at best only reflective of in situ metabolic potential (Fisher and Hand, 1984).
Also, they do not take into consideration losses due to respiration, maintenance and reproduction.
Table 3. Autotrophic potential determined for Loripes lacteus as well as for several other lucinids
present in seagrass meadows (values calculated based on literature data).
Species
Loripes lacteus
(present study)
Lucina floridana
(Fisher & Hand, 1984)
Myrtea spinifera
(Dando et al, 1985,
1986b)
Lucinoma borealis
(Dando et al, 1986a)
RuBCase
Weight of gill Bivalve density
Autotrophic
(nmoles min−1 g−1)
(g)
(bivalves m−2) potential (g C m−2 yr−1)
300.*
0.032
775
47
1,510
40
200
0.42
80
0.0147
20
0.074
4
0.35
0.07
* mean value from lucinid data available from the literature.
320
1348
BULLETIN OF MARINE SCIENCE, VOL. 71, NO. 3, 2002
Figure 3. Schematic representation of carbon fluxes in a seagrass ecosystem modified to incorporate
the chemoautotrophic-bearing bivalve populations.
Autotrophic potential is 20 times greater than net clam production in L. borealis, 30
times greater in M. spinifera, 150 times greater in L. floridana and 27 times greater for L.
lacteus of the present study. The one but last value is considerably higher than the others
due to the inordinately high RuBCase value found for this organism by Fisher and Hand
(1984). These differences are most probably due to the carbon losses mentioned above as
well as to losses in the carbon routing mechanisms between symbiont and host.
C. nodosa (Ucria) Ascherson is widely distributed throughout the Mediterranean (Terrados
and Ros, 1992) and can form vast underwater meadows in sheltered bays and coastal lagoons
(Ballesteros et al., 1989). Published values for this species indicate that production values
(leaf + rhizome + root) range from approximately 300 to 950 g dw m−2 yr−1 (Terrados and
Ros., 1992; Cébrian et al., 1997; Peduzzi and Vukovic, 1990; Pérez and Romero, 1994). For
comparative purposes, a median value of 610 g dw m−2 yr−1 was adopted here. This corresponds to roughly 290 g C m−2 yr−1 (based on mean carbon content values in C. nodosa
(Peduzzi and Vukovic, 1990)). This would indicate that the ‘primary production potential’
from chemoautotrophic symbiosis represents roughly 16% of the seagrass bed’s primary
production. This symbiotic production, of course, is not truly primary, unlike that at the
hydrothermal vents. Indeed, it is a secondary process (Fenchel and Riedl, 1970) since the
reduced sulphur compounds that fuel the system are produced by anaerobic heterotrophic
bacteria that use organic matter originally fixed by the marine plants. This organic production based on oxidation of sulphur compounds is thus ultimately dependent on photosynthetic sources (Spiro et al., 1986).
Chemoautotrophic symbioses are thus prevalent in many phanerogam seagrass beds
where they represent, in terms of their net annual production and autotrophic potential,
ecologically important components of seagrass ecosystems. These findings provide us
with new insight concerning the flux of carbon in seagrass beds. Indeed, it has been
traditionally perceived that seagrass primary production buried in the sediment as refrac-
JOHNSON ET AL.: IMPORTANCE OF AN INVERTEBRATE CHEMOAUTOTROPHIC SYMBIOSIS
1349
tory detritus involves a net loss for heterotrophic use and represents the seagrasses’ capacity to act as a sink for organic carbon (Hemminga et al., 1991; Romero et al., 1994).
Thus, all organic matter that escapes both the Microbial Loop and Below-ground Food
Web will become incorporated into the seagrass bed’s carbon sink. The presence of
chemoautotrophic symbioses generates a slightly modified view of this carbon flux (Fig.
3). Indeed, the mineralization of deposited organic matter will lead to an elevated activity
of sulfate-reducing bacteria and the production of high levels of hydrogen sulfide
(Jorgensen, 1977). A portion of this hydrogen sulfide will be used by the bacterial member of the chemoautotrophic symbiosis to drive the reductive reactions of carbon fixation.
This symbiotically fixed carbon will contribute to providing the host with the organic
matter necessary for its survival (maintenance, reproduction, growth). Indeed, studies
involving δC13 values in the closely related species L. lucinalis reveal that, on average,
63% of the host’s carbon input is of chemoautotrophic origin (Johnson et al., 1994). In
addition, the presence of, at times, large numbers of empty pierced lucinid shells reveals
that there are predation pressures exerted on the symbiotrophic bivalve population (unpubl.
results). This represents a flux of originally photosynthetically derived carbon through
the Microbial Loop back to the above-ground food web via the symbiotrophic bivalve
population. The release of bivalve gametes into the water column can be considered as an
additional means of cycling this carbon back to the pelagic compartment. These chemoautotrophic associations thus contribute towards reducing the carbon sink in seagrass communities and should be taken into consideration when addressing the problem of carbon
and sulphur cycling in reducing coastal habitats.
LITERATURE CITED
Aurelia, M. 1969. The habitats of some subtidal pelecypods in Herrington Sound, Bermuda. Pages
39–52 in R. N. Ginsburg and S. M. Stevens, eds. Seminar on organism-sediment interelationship.
Berm. Stn. Spec. Publ. 6.
Ballesteros, S., J. D. Ros, J. Flos and R. Margalef. 1989. Els ecosistemes bentonics. Pages 119–176
in R. Folch, ed. Historia naturals dels països catalans. Sistemes naturals. Fundacio Enciclopedia
Catalana, Barcelona.
Barnes, P. A. G. and C. S. Hickman. 1999. Lucinid bivalves and marine angiosperms: a search for
causal relationships. Pages 215–238 in D. I. Walker and F. E. Wells, eds. The seagrass flora and
fauna of Rottnest Island, Western Australia. Western Australian Museum, Perth.
Berg, C. J. and P. Alatolo. 1984. Potential of chemosynthesis in molluscan mariculture. Aquaculture 39: 165–179.
Buchanan, J. B. 1963. The bottom fauna communities and their sediment relationships off the coast
of Northumberland. Oikos 14: 154–175.
Cary, S. C., R. D. Vetter and H. Felbeck. 1989. Habitat characterization and nutritional strategies of
the endosymbiont-bearing bivalve Lucinoma aequizonata. Mar. Ecol. Progr. Ser. 55: 31–45.
Cavanaugh, C. M. 1983. Symbiotic chemoautotrophic bacteria in marine invertebrates from sulphide-rich habitats. Nature 302 (5903): 58–61.
Cebrian J., C. M. Duarte, Marba Nuria and S. Enriquez. 1997. Magnitude and fate of the production of four co-occurring Western Mediterranean seagrass species. Mar. Ecol. Prog. Ser. 155:
29–44.
Childress, J. J. and T. J. Mickel. 1982. Oxygen and sulfide consumption rates of the vent clam
Calyptogena pacifica. Mar Biol. Lett. 3: 73–79.
1350
BULLETIN OF MARINE SCIENCE, VOL. 71, NO. 3, 2002
Dando, P. R., S. A. Ridgway and B. Spiro. 1994. Sulphide ‘mining’ by lucinid bivalve molluscs:
demonstrated by stable sulphur isotope measurements and experimental models. Mar. Ecol.
Progr. Ser. 107: 169–175.
__________, A. J. Southward and E. C. Southward. 1986a. Chemoautotrophic symbionts in the
gills of the bivalve mollusc Lucinoma borealis and the sediment chemistry of its habitat. Proc.
R. Soc. Lond. 227: 227–247.
__________, _____________, E. C. Southward and R. L. Barrett. 1986b. Possible energy sources
for chemoautotrophic prokaryotes symbiotic with invertebrates from a Norwegian Fjord. Ophelia
26: 135–150.
__________, _____________, ______________, N. B. Terwilliger and R. C. Terwilliger. 1985.
Sulphur-oxidizing bacteria and haemoglobin in gills of the bivalve mollusc Myrtea spinifera.
Mar. Ecol. Prog. Ser. 23: 85–98.
__________ and B. Spiro. 1993. Varying nutritional dependence of the thyasirid bivalves Thyasira
sarsi and Thyasira equalis on chemoautotrophic symbiotic bacteria, demonstrated by isotope
ratios of tissue carbon and shell carbonate. Mar. Ecol. Prog. Ser. 92: 151–158.
Felbeck, H. 1983. Sulfide oxidation and carbon fixation by the gutless clam Solemya reidi: an
animal-bacteria symbiosis. Science 152: 3–11.
_________, J. J. Childress and G. N. Somero. 1981. Calvin-Benson cycle and sulphide oxydation
enzymes in animals from sulphide-rich habitats. Nature 293: 291–293.
Fenchel, T. 1977. Aspects of the decomposition of seagrasses. Pages 123–145 in C. P. McRoy and
C. Helfferich, eds. Seagrass ecosystems: A scientific perspective, Marcel Dekker, Inc., New
York.
_________ and R. J. Riedl. 1970. The sulfide system: a new biotic community underneath the
oxidized layer of marine sand bottoms. Mar. Biol. 7: 255–268.
Fisher, C. R. 1990. Chemoautotrophic and methanotrophic symbiosis in marine invertebrates. Rev.
Aquat. Sci. 2: 399–436.
__________ and J. J. Childress. 1986. Translocation of fixed carbon from symbiotic bacteria to
host tissues in the gutless bivalve Solemya reidi. Mar. Biol. 93: 59-–68.
Fisher, M. R. and S. C. Hand. 1984. Chemoautotrophic symbionts in the bivalve Lucina floridana
from seagrass beds. Biol. Bull. 167: 445–459.
Frenkiel, L., O. Gros and M. Mouëza. 1996. Gill structure in Lucina pectinata (Bivalvia: Lucinidae)
with reference to hemoglobin in bivalves with symbiotic sulphur-oxidizing bacteria. Mar. Biol.
125: 511–524.
Hemminga, M. A., P. G. Harrison and F. van Lent. 1991. The balance of nutrient losses and gains in
seagrass meadows. Mar. Ecol. Prog. Ser. 71: 85–96.
Herry, A., M. Diouris and M. Le Pennec. 1989. Chemoautotrophic symbionts and translocation of
fixed carbon from bacteria to host tissues in the littoral bivalve Loripes lucinalis (Lucinidae).
Mar. Biol. 101: 305–312.
Johnson, M. A., M. Diouris and M. Le Pennec. 1994. Endosymbiotic bacterial contribution in the
carbon nutrition of Loripes lucinalis (Mollusca: Bivalvia). Symbiosis 17: 1–13.
Jorgensen, B. B. 1977. The sulfur cycle of a coastal marine sediment (Limfjorden, Denmark).
Limnol. Oceanogr. 22(5): 814–832.
Monnat, J. Y. 1970. Introduction à l’étude de la reproduction chez Lucinoma borealis (Linné),
Bivalvia, Lucinacea. Thèse 3e cycle, Fac. Sc. Brest.
Peduzzi, P. and A. Vukovic. 1990. Primary production of Cymodocea nodosa in the Gulf of Trieste
(Northern Adriatic Sea): a comparison of methods. Mar. Ecol. Prog. Ser. 64: 197–207.
Pérez, M. and J. Romero. 1994. Growth dynamics, production and nutrient status of the segrass
Cymodocea nodosa in a Mediterranean semi-estuarine environment. PSZN I. Mar. Ecol. 15:
51–64.
Reid, R. 1980. Aspects of the biology of a gutless species of Solemya (Bivalvia: Protobranchia).
Can. J. Zool. 58: 386–393.
JOHNSON ET AL.: IMPORTANCE OF AN INVERTEBRATE CHEMOAUTOTROPHIC SYMBIOSIS
1351
Romero, J., M. Pérez, M. A. Mateo and E. Sala. 1994. The belowground organs of the Mediterranean seagrass Posidonia oceanica as a biogeochemical sink. Aquat. Bot. 47: 13–19.
Schweimanns, M. and H. Felbeck. 1985. Significance of the occurrence of chemoautotrophic bacterial endosymbionts in lucinid clams from Bermuda. Mar. Ecol. Prog. Ser. 24: 113–120.
Spiro, B., P. B. Greenwood, A. J. Southward and P. R. Dando. 1986. 13C/12C ratios in marine invertebrates from reducing sediments: confirmation of nutritional importance of chemoautotrophic
endosymbiotic bacteria. Mar. Ecol. Prog. Ser. 28: 233–240.
Stevenson, J. C. 1988. Comparative ecology of submersed grassbeds in freshwater, estuarine, and
marine environments. Limnol. Oceanogr. 33: 867–893.
Terrados, J. and J. D. Ros. 1992. Growth and primary production of Cymodocea nodosa (Ucria)
Ascherson in a Mediterranean coastal lagoon: the Mar Menor (SE Spain). Aquat. Bot. 43: 63–74.
Zieman, J. C. and R. G. Wetzel. 1980. Productivity in seagrasses: methods and rates. Pages 115–
131 in R. C. Phillips and C. P. McRoy, eds. Handbook of seagrass biology: An ecosystem
perspective, Garland STPM Press, New York.
ADDRESS: Faculty of Sciences, University of Corsica, BP 52, 20250 Corte, France.