ELSEVIER Journal of Experimental Marine Biology and Ecology

184 (1994) l-20

JOURNAL OF EXPERIMENTAL MARINE BIOLOGY AND ECOLOGY

Nitrogenase activity in marine sediments from a temperate saltmarsh lagoon: modulation by complex polysaccharides,

ammonium and oxygen

B. J. Tibbles a,*, M. I. Lucas a, V. E. Coyneb, S.T. Newton a

d Marine Biology Research Institute. Department of Zoology, University of Cape Town, Private Bag Rondebosch, 7700 South Africa

b Department of Microbiology, University of Cape Town, Private Bag Rondebosch, 7700 South Africa

Received 2 October 1993; revision received 20 May 1994; accepted 7 June 1994

Abstract

Nitrogenase activity (acetylene reduction) was examined in marine sediments associated with seagrass beds and thalassinidean prawn burrows, in a temperate saltmarsh lagoon. Nitrogenase activity in unamended sediments from different microhabitats ranged from undetectable values to ~0.75 nmol C,H, (g dry sediment)-‘.hour-‘. The bulk of organic material available to heterotrophic bacteria in saltmarsh lagoons is probably derived from primary production.

Amendment of sediments with structural plant polysaccharides showed that xylan and alginate stimulated nitrogenase activity on average by 5 to 18-fold relative to unamended sediments, whereas cellulose and carrageenan were less effective. Amendment of sediments with storage plant polysaccharides produced the greatest stimulation of nitrogenase activity. Addition of laminarin and glycogen (amylopectin) significantly (p <O.OS) stimulated nitrogenase activity by 19 to 92-fold. In contrast to these polysaccharides of plant origin, chitin (a polymer of prawn exoskeletons) did not significantly enhance nitrogenase activity in these sediments. Aerobic conditions stimulated nitrogenase activity by 2 to 20-fold in surface sediments, but not in sub- surface sediments, indicating that aerobic or microaerophilic respiration was a significant source of energy for nitrogenase activity in surface sediments. Oxygen-stimulation of nitrogenase activity was less marked in aerobic sediments around the rim of prawn burrow openings where subsurface sediments had been displaced to the surface by bioturbation. Ammonium did not appear to play a major role in the regulation of nitrogenase activity in these sediments as in situ concentrations were z 20-fold lower than the concentration (50 PM) which resulted in a slight (O-15%) reduc- tion in nitrogenase activity.

Keywords: Bioturbation; Heterotrophic nitrogen fixation; Organics; Thalassinids; Saltmarsh; Seagrass

* Corresponding author.

0022-0981/94/$7.00 0 1994 Elsevier Science B.V. All rights reserved SSDl 0022-0981(94)00104-9

B.J. Tibbles et al. /J. Exp. Mar. Bid. Ed. 184 (1994) I-20

1. Introduction

Saltmarsh and seagrass environments are characterized by high rates of primary

production, ranking among the most productive ecosystems known (Valiela et al., 1976; McRoy & McMillan, 1977; Schubauer & Hopkinson, 1984). Besides supporting rich biological diversities of their own (Ogden, 1980), seagrass communities can also con- tribute significant quantities of nutrients to the food chains of adjacent ecosystems (Valiela & Teal, 1979; Wolff, 1980; Whiting et al., 1987).

High rates of primary production in saltmarsh and seagrass ecosystems may nev- ertheless be limited by low levels of inorganic nitrogen, which are a characteristic of many marine environments (Patriquin, 1972; Short, 1983; Moriarty et al., 1985). Al- though much of the primary production in these ecosystems may be sustained by in- ternal recycling of reduced nitrogen, other sources of nitrogen are essential to replen- ish that lost from the system by tidal exchange (Valiela & Teal, 1979; Whiting et al., 1987) and denitrification (Seitzinger et al., 1984). In the absence of sufficient new ni- trogen (NO,-N) being introduced by tidal exchange, nitrogen fixation can be a major input of nitrogen to saltmarsh and seagrass systems (Patriquin & Knowles, 1972; Hanson, 1977a; Capone et al., 1979; Capone, 1982; O’Donohue et al., 1991). While such a source of nitrogen provides an attractive solution to the problem, the precise mechanisms which control nitrogen fixation in situ remain elusive. The importance of the control of nitrogen fixation in marine environments has thus been emphasized

(Capone, 1988; Paerl, 1990). Factors that have been shown to modulate nitrogenase activity in coastal ecosystems

include the availability of organic substrates, inorganic nitrogen, and oxygen (Capone, 1988; Paerl, 1990). Several workers have reported on the stimulation of nitrogenase activity in natural samples by mono- and disaccharides (Capone & Budin, 1982; O’Neil

& Capone, 1989). However, there have been no reports of similar studies which have examined the potential of polysaccharides to support nitrogen fixation in saltmarsh ecosystems, even though most of the carbon available for bacterial use in saltmarsh ecosystems is derived primarily from such complex polysaccharide-based plant mate- rials. Bacterial use of complex carbohydrates is quite different from their use of simple ones (Gottschalk, 1986) with possible implications for energy-demanding processes such as nitrogen fixation.

A further ecological implication of diazotrophic growth on detrital materials in salt- marsh ecosystems concerns the nutritional value of these materials to consumers. Saltmarsh macrophytes have a high fibre content, and low protein/nitrogen content (Atkinson & Smith, 1983), and as such are of limited direct nutritional value to many invertebrate consumers. Consequently, it is more usual in these systems for energy and materials from primary production to flow through detritus food webs rather than through grazer food webs (Ogden, 1980). Microbial colonization of detrital particles has been shown to increase the nitrogen content of the food resource, and thereby improve its nutritional value to consumers (Newell, 1965; Pomeroy, 1980). Microbial growth on detritus is facilitated in part by the ability of some microbes to utilize complex polysac- charides (e.g. cellulose) and other refractory carbon compounds as energy and carbon sources, while the nitrogen demands of microbial protein synthesis are supplemented

B.J. Tibbles et al. 1 J. Exp. Mar. Bid. Ecol. 184 (1994) l-20 3

with inorganic nitrogen obtained from the environment. A further source of nitrogen

for growth on these substrates could be derived from bacterial dinitrogen fixation.

Nitrogenase activity may be influenced by the community structure of lagoonal

ecosystems, since these ecosystems comprise a “mosaic” of microhabitats. For in- stance, several workers have measured nitrogenase activity associated with beds of different macrophyte species (Hanson, 1977a,b; Dicker & Smith, 1980; Gandy & Yoch, 1988) and with seagrass roots (Capone & Budin, 1982; Capone, 1983; Smith & Hayasaka, 1982a). However, few comparisons have been made of nitrogenase activ- ity associated with “nutrient rich” seagrass beds and “nutrient depleted” open flats

(Capone, 1982; O’Donohue et al., 1991). Moreover, there have been no reports on the effects of benthic invertebrates on nitrogen fixation in saltmarsh sediments, although the importance of benthic animals in other biogeochemical cycling processes has long

been recognised (Koike & Mukai, 1983). The benthic macrofauna of many estuaries in southern Africa is dominated by the

thalassinidean prawns, Callianassa kraussi and Upogebia africana, which reach densi- ties of 350 and 400 individuals.m-’ sediment, respectively (Hanekom, 1980). Thalassinidean prawns exert considerable influence on ecosystem processes through

their activities while burrowing and feeding (Bird, 1982; Branch & Pringle, 1987). For instance, the burrowing activities of thalassinids result in profound bioturbation, which

has been shown to contribute to sediment transportation (Roberts et al., 1981) increase in oxygenation (Dye, 1978), alteration of the depth-distribution of microalgae (Branch & Pringle, 1987), and has been correlated with enhanced numbers of bacteria (Yingst & Rhoads, 1980; Branch & Pringle, 1987). Bioturbation by thalassinids may also modulate cover by seagrass beds (Suchanek, 1983). Thus, the activities of thalassinids could directly, and potentially indirectly through effects on nitrogen fixation, affect

primary production of diatoms and macrophytes, which provide the food source of these invertebrates as well as many other consumers.

In the present study, acetylene reduction assays were used to measure nitrogenase activity (Hardy et al., 1968) in sediment and water samples from a saltmarsh lagoon on the west coast of South Africa. Measurements of nitrogenase activity were compared on a macro-scale (two sites at opposite ends of the lagoon) and on a micro-scale

(different microhabitats related to seagrass beds and prawn burrows at each site). The modulation of nitrogenase activity by the availability of oxygen, ammonium, and dif- ferent complex carbohydrates, in relation to different sites, was examined.

2. Materials and methods

2. I. Study site

Langebaan Lagoon is a partially enclosed marine system situated between 33 ’ 00’ to 30” 13’ S and 17”57’ to 18” 08’ E on the south-west coast of South Africa (Fig. 1). This estuary is unusual in that it is entirely marine in origin, with no riverine input, and receives very little fresh water from precipitation or runoff. Consequently, salinity values are relatively constant and approximate those of seawater, or may be-

B.J. Tibbies et al. 1 J. Exp. Mar. Bioi. Ecol. 184 (1994) 1-20

_~._...

Key:

Intertidal flats

n Saltmarsh

gzg Zostem beds

1

0 L

Fig. 1. Location of Langebaan Lagoon, showing sampling sites Oesterwal and Geelbek, the extent of inter- tidal sand- and mudflats, and the distribution of macrophyte beds.

come hypersaline in the summer months. Two dissimilar sites in the lagoon were se-

lected for study: Oesterwal, near the lagoon mouth is characterized by sandflats and is burrowed by the sandprawn ~a~~i~~as~~ krausi, while Geelbek, in the southern reaches has anaerobic mudBats and burrows are mostly due to the mudprawn U. afiicana. Both sites have a saitmarsh component and beds of Zostera cupensis

(Fig. 1). Five sedimentary microhabitats were chosen for study at each site: (1) surface sediment to a depth of 5 mm, away from the seagrass beds and between prawn bur- row openings; (2) sediment to a depth of 5 mm from within the Zostera beds; (3) sediment from the rims of prawn burrow openings; (4) sediment from burrow linings; and (5) sediment from a depth of 10 cm below the surface (extruded with a PVC corer and subsampled using a 10 ml syringe with the luer end removed).

B.J. Tibbles et al. /J. Exp. Mar. Bid. Ecol. 184 (1994) l-20 5

2.2, Sampling procedure

Samples were always taken on the morning of spring tides, and the temperatures of the samples at time of collection were recorded. There was at least a 10 “C difference in temperature between seasons for each site. Morning temperatures of surface sedi- ment ranged from Il. 1 “C in winter to 24.5 “C in summer, while sediment temperatures rose above 28 “C on summer afternoons. Sediment samples were taken in triplicate from randomly selected sites within each microhabitat, transported to the laboratory

in an insulated container, and processed within 2 h of collection.

2.3. Acetylene reduction assays

Sediment samples (25 g wet weight) were weighed into glass bottles and amended as below. The bottles were then sealed, and the headspace replaced with 85% argon

and 15% acetylene (Fedgas, SA) unless otherwise stated. Flasks were incubated in the dark at 20 “C. Gas samples (100 ~1) were withdrawn with a gas-tight syringe at pre- determined time intervals for determination of ethylene content in a Hewlett Packard

588OA gas chromatograph. Gases were separated on a Porapak N column (3 m x 3 mm) at 80 “C, with nitrogen as the carrier gas at a flow rate of 20 ml.min -I. The in- jector and detector temperatures were 200 “C and 250 “C, respectively, and the flow rates of hydrogen and air were 30 ml min -’ and 400 ml.min -I, respectively. Peak areas were integrated on a Hewlett Packard 5880A Series integrator and expressed as nmol C,H, by comparison with peak areas of known C,H, concentration. Preliminary ex- periments examined the effect of pC,H, on acetylene reduction. Acetylene reduction

activity plateaued between pC,H, of 0.10 and 0.20 atm, and therefore all subsequent assays were conducted with a pC,H* of 0.15 atm as above. No ethylene production was detected in the absence of added acetylene, nor was ethylene production detected in autoclaved controls.

2.4. Assa_v amendments

The effects of NH,+, organic substrates, and 0, on acetylene reduction were exam- ined. To determine the effects of O,, sediment samples were incubated in 100 ml glass bottles with 50 ml 0.2 pm filter-sterilised seawater, either aerobically (i.e. 85 7; air, 15 Tf;

C,H,) or anaerobically, as above. Bottles were incubated in the dark on a shaker to ensure gaseous equilibration. The effects of NH,+ or different polysaccharides on acetylene reduction were also examined. Sediments were amended in 25 ml wide-neck glass bottles by addition of these substrates in 2 ml 0.2 pm filter-sterilised seawater. Ammonium chloride was added at concentrations of 0, 50, 100, 500, and 1000 PM. Polysaccharides, including Avicel microcrystalline cellulose (FMC Corp.), glycogen (Merck), Na-alginate (BDH), oat spelt xylan, carrageenan, laminarin, starch, and crab shell chitin (all Sigma), were added at 0.9 mg g-l wet sediment, which was the mass of substrate approximately equivalent to mono- and disaccharide amendments used by others (Capone & Budin, 1982; Gandy & Yoch, 1988). Chitin substrate was prepared according to the method of Reichenbach & Dworkin (198 1).

6 B.J. Tibbles et al. /J. Exp. Mar. Bid. Ed. 184 (1994) l-20

2.5. Determination of ammonium concentrations, organic carbon content, and statistical

analysis of data

Interstitial waters were extracted from different sediments by centrifugation within 5 minutes of collection, and assayed immediately for ammonium content using the phenol-hypochlorite method of Koroleff (1983). Total organic content of replicate (n

= 3) sediment samples was determined by weight loss of dried samples when com- busted at 450 “C for 6 h. Statistical analysis of data was carried out by analysis of variance (ANOVA) on untransformed data sets where Bartlett’s test for homogeneity of variance indicated homoscedasticity. ANOVA of transformed (log,,) data sets was carried out when heteroscedasticity was encountered in untransformed data. Student’s t-test was used for statistical comparison of paired means.

3. Results

Representative time courses of acetylene reduction in different sediments from Langebaan Lagoon are presented in Fig. 2. Rates were usually linear for 12-24 h, and hence comparisons of controls and treatments (see below) considered data within

z 24-h incubation.

3.1. Ammonium regulation

The effect of ammonium on nitrogenase activity in sediments from Langebaan La- goon was examined during summer 1992. Nitrogenase activity in different surface sediments from Langebaan Lagoon responded differently to additions of NH,’ (Fig. 3). Nitrogenase activity in the Geelbek muds showed little response to O-l mM NH,+ In contrast, additions of NH,+ to sandy sediments (Oesterwal) caused a concentration-dependent inhibition of nitrogenase activity, with 90% inhibition of ac- tivity when 1 mM NH2 was added (Fig. 3). To assess the relevance of these effects, in situ NH,+ concentrations were determined for interstitial and burrow waters (Table 1). In situ concentrations of NH,’ during summer were z 20-fold lower than the lowest addition of NH,+ (50 PM), which had no affect on nitrogenase activity in

muddy sediments, and caused only a 15% drop in rates in sandy sediments (Fig. 3). Ammonium concentrations in interstitial and prawn-burrow waters in Langebaan

Lagoon were lower than those reported for other saltmarsh and seagrass ecosystems (Table 1). During summer, ammonium concentrations in interstitial water from muddy sediments at Geelbek (3.51 k 10.49 PM) were higher than those in the sands at Oester- wal (1.5 1 +_ 3.73 PM). During autumn, NH,+ concentrations at Oesterwal (11.6 + 10.1 PM) were slightly higher than at Geelbek (7.4 & 8.6 PM). Interstitial ammonium con- centrations (means for both sites) were higher in autumn (9.5 t 9.6 ,uM) than in sum- mer (2.5 + 7.9 PM). This seasonal comparison corresponds with a lower demand for nitrogen by primary producers in autumn, when plants are senescent, than in the summer growth-season. Ammonium concentrations in water from C. kraussi burrows at Oesterwal, were higher than those in U. africana burrow waters at Geelbek during

B.J. Tibbles et al. /J. Exp. Mar. Bid Ed. 184 (1994) l-20

0 0 20 40 60 80 100 1

U Surface A Zosters bed + Burrow opening -+- Burrow lining -b- 1 Ocm depth

!O

Time (hours)

Fig. 2. Time course reduction of acetylene by different sediments from Langebaan Lagoon.

both summer and autumn (Table 1). Previous studies have shown that water from C. juponica burrows had a higher ammonium content than water from U. major burrows (Table l), and that burrows of both organisms exhibited ammonium production com- parable to rates of the adjacent surface sediment (Koike & Mukai, 1983).

3.2. Organic carbon regdation

Rates of nitrogenase activity in unamended (control) sediments from Oesterwal and Geelbek are shown in Fig. 4 (microhabitats at each site) and Table 2 (means k SD for each site). Comparison of nitrogenase activity between the two sites (macroscale) showed that rates in unamended samples were greater on average (by 2.4-fold, not significant, p > 0.05) in the fine, muddy sediments from Geelbek than in coarse, sandy sediments from Oesterwal (Table 2). The organic contents of sediments from Geelbek were also greater than those from Oesterwal (Table 1). Spatial comparison on a mi- croscale (different microhabitats at each site) showed that unamended nitrogenase

B.J. Tibbles et al. 1 J. Exp. Mar. Biol. Ecol. 184 (1994) l-20

No addition (Control)

;, *; et & *; I ,

100

NITROGENASE ACTIVITY (percent of control)

0

Fig. 3. Effect of ammonium chloride on nitrogenase activity in surface sediments. Experiments were carried out on 17.2.92. Values are expressed as percentage of the activity of unamended controls (0.005 ?r 0.004 nmol C,H, (g dry sed.))‘.h-’ for Oesterwal; 0.24 + 0.08 nmol CzH4 (g dry sed.))‘.h-’ for Geelbek), which were made to equal 100%. Values are means of triplicate samples assayed after 20 h.

activity was highest in surface and Zostera sediments, and decreased with sediment depth (Fig. 4). Organic contents varied between sediment microhabitats; Zostera and burrow lining sediments supported larger organic contents than surface, burrow- opening, or subsurface sediments (Table 1). Zostera bed sediment from Geelbek had the highest organic content (Table 1) and also supported the highest rates of nitroge-

nase activity (Fig. 4). Since nitrogen fixation is an energy-demanding process, the availability of suitable

organic substrates may also control rates of heterotrophic nitrogen fixation in situ. Since most of the organic material available for bacterial use in saltmarsh ecosystems should

be derived from primary production, the effects of complex polysaccharides on nitro- genase activity were examined. Four structural and three storage polysaccharides of plant origin were chosen for comparison. Moreover, because of the high prawn den- sities in this lagoon, the effect of chitin on nitrogenase activity was also tested.

Benthic nitrogen fixation in saltmarsh and seagrass environments can be stimulated by additions of easily-utilizable carbon sources such as glucose (Dicker & Smith, 1980; Capone, 1982; Capone & Budin, 1982). Similar effects were measured in Langebaan Lagoon where addition of glucose to sediments stimulated nitrogenase activity by > lo-fold relative to unamended controls (Fig. 4). The glucose-effect is included here because it shows that heterotrophic nitrogen fixation in Langebaan Lagoon may be limited by the availability of suitable organic carbon sources.

Amendment of sediments with defined, complex polysaccharides stimulated nitro-

Tab

le

1

Am

mon

ium

co

ncen

trat

ions

an

d or

gani

c m

atte

r co

nten

t of

sed

imen

t in

ters

titia

l w

ater

an

d th

alas

sini

d bu

rrow

w

ater

Mic

roha

bita

t L

ocat

ion

Dep

th

NH

: co

ncen

trat

ion”

O

rgan

ic

cont

ent’

R

efer

ence

(cm

) (P

M)

(mg

orgg

’ D

W)

Exp

osed

se

dim

ent

Zos

teru

be

d se

dim

ent

Bur

row

op

enin

g se

dim

ent

Tha

lass

inid

bu

rrow

w

ater

’

Subs

urfa

ce

sedi

men

t

Oes

terw

al-s

umm

er

0.5

Oes

terw

al-a

utum

n 0.

5

Gee

lbek

-sum

mer

0.

5

Gee

lbek

-aut

umn

0.5

Oes

terw

al

(Z.

cqen

sis)

-sum

mer

0.

5

Oes

terw

al

(Z.

Cap

ens&

)-au

tum

n 0.

5

Gee

lbek

(Z

. ca

pens

is)-

sum

mer

0.

5

Gee

lbek

(Z

. cu

pens

is)-

autu

mn

0.5

Que

ensl

and,

A

ustr

alia

(Z

. ca

pric

orn

i)

2

Ala

ska,

U

SA

(Z.

mar

ina)

5

New

Y

ork,

U

SA

(Z. ma

rina

) o-

2

Oes

terw

al

(C.

krau

ssi)

0.

5

Gee

lbek

(U

. u

fica

nu

) 0.

5

Oes

terw

al

(C.

krau

ssi)

-sum

mer

1

Oes

terw

al

(C.

krau

ssi)

-aut

umn

1

Gee

lbek

(U

. u/

?icu

na)-

sum

mer

1

Gee

lbek

(U

. af

rica

na)-

autu

mn

1

Japa

n,

limor

ia

(C.

jupo

mic

u)

n.d.

Japa

n,

limor

ia

(U.

maj

or)

n.d.

Oes

terw

al-s

umm

er

10

Oes

terw

al-a

utum

n 10

Gee

lbek

-sum

mer

10

Gee

lbek

-aut

umn

10

0.59

*

0.76

6.

34

k 0.

87

4.50

+

_ 1.0

0 nd

.

1.25

k3.1

2 14

.4 t

2.

16

6.40

+

4.80

nd

.

2.68

+

2.32

25.7

t

10.6

7.18

k

15.8

12.5

+ 1

6.0

34

50

20-3

3

8.86

&

1.21

nd.

29.1

f

7.02

nd.

n.d.

nd.

n.d.

n.d.

n.d.

5.36

+ 0

.84

11.9

~2.2

0

2.38

k

6.95

5.40

+ 2

.00

1.09

k2.7

4

3.7O

k3.1

0

11.9

-59.

2

8.5-

19.9

n.d.

n.d.

n.d.

n.d.

nd.

n.d.

0.41

5

0.81

10.9

+ 3

.10

4.51

~ 13

.2

7.00

&

1.4

0

6.49

+

1.42

n.d.

12.7

+_ 1

.43

n.d.

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Mor

iart

y et

al.

(198

5)

Shor

t (1

983)

Hor

riga

n an

d C

apon

e (1

985)

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Koi

ke

and

Muk

ai

(198

3)

Koi

ke

and

Muk

ai

(198

3)

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

Thi

s st

udy

ta

i

a V

alue

s fo

r L

ange

baan

re

pres

ent

mea

ns

+ SD

of

repl

icat

es

(n =

12)

, de

term

ined

du

ring

D

ecem

ber,

19

92

(sum

mer

) an

d M

ay,

1994

(a

utum

n).

h V

alue

s fo

r L

ange

baan

re

pres

ent

annu

al

mea

ns

+ S

D

of t

ripl

icat

e se

ason

al

mea

sure

men

ts.

’ B

urro

w

wat

er

is a

ssum

ed

to

appr

oxim

ate

amm

oniu

m

leve

ls

in

borr

ow-a

ssoc

iate

d se

dim

ents

.

n.d.

=

not

dete

rmin

ed.

10 B.J. Tibbh et al. 1 J. Exp. Mar. Biol. Ecol. 184 (1994) I-20

CONTROL 0.8

0.6

q Zostera bed sediment

n Burrow opening sediment

q Burrow lining sedimenl

0 Subsurface sediment

GLUCOSE T T

0

OESTERWAL GEELBEK

Fig. 4. Nitrogenase activity in sediments without amendment (control) or amended with glucose. Note that scales of y-axes are not equal. Sediments were collected from five microhabitats at Oesterwal (12.3.93) and Geelbek (24.3.93). Values represent means + SD (n = 2) of rates between z 12-24-h incubation. Overall means (sites and microhabitats) for each treatment are: Control, 0.19 + 0.25; Glucose, 4.32 + 2.87 nmol C,H, (g dry sed.))‘*h-‘.

genase activity differently according to the substrate and sediment type (Figs. 5 amd 6). Of the four structural plant-polysaccharides tested, xylan and alginate were more effective than cellulose and carrageenan at stimulating nitrogenase activity in most sediment types (Fig. 5, Table 2). Activities (means for sites) stimulated by xylan were 5 to 15-fold greater (not significant, p >0.05) than those of unamended controls, whereas activities stimulated by alginate were 5 to 18-fold greater (significant, p < 0.05) than those of controls (Table 2). Amendment of sediments with carrageenan caused sporadic stimulation of nitrogenase activity among microhabitats and sites (Fig. 5), but the mean rates stimulated by carrageenan were not significantly (p > 0.05) greater than those of the controls (Table 2). The cellulose polymer, microcrystalline cellulose, stim- ulated nitrogenase activity only after extended lag phases ( > 60 h), and only in certain types of sediments (results not shown).

B.J. Tibbles et al. /J. Exp. Mar. Biol. Ecol. 184 (1994) I-20 II

Table 2

Effects of different polysaccarides and glucose on nitrogenase activity in sediments from Langebaan lagoon

Treatment Site Nitrogenase activity” Ratio”

None (control)

Glucose

Cellulose

Xylan

Alginatc

Carrageenan

Glycogen

Laminarin

Starch

Chitin

Oesterwal

Geelbek

Oesterwal

Geelbek

Oesterwal

Geelbek

Oesterwal

Geelbek

Oesterwal

Geelbek

Oesterwal

Geelbek

Oesterwal

Geelbek

Oesterwal

Geelbek

Oesterwal

Geelbek

Oesterwal

Geelbek

0.11 kO.17

0.26 f 0.3 1 4.17k2.93

4.4723.15

0.05 f 0.07

0.18 + 0.27

l.69+_ 1.80

1.40 f 1.49

1.98 k 1.67

1.40 +_ 0.95

1.03 f 2.15

0.36 +_ 0.35

10.07 f 7.19

5.27 k 1.53

6.59 + 3.93

4.96 f 2.24

0.07 f 0.05

0.35 k 0.40

0.16kO.19

0.38 + 0.44

1

38*

17*

0

I 15

5

18’ 5*

9

92*

20*

60*

19*

a Mean k SD of all microhabitats at each site. Units: nmol C,H,(g- ’ dry sed.).h- ‘. Rates of acetylene re-

duction were determined between z 12-24-h incubation.

b Activity of treatment/activity of control.

* Denotes treatment activity significantly (ANOVA, ~~0.05) different to that of the unamended control.

Of all the polysaccharides examined, the storage polysaccharides, glycogen and laminarin, were the most effective at stimulating nitrogenase activity (Fig. 6, Table 2). Glycogen significantly (p < 0.05) stimulated nitrogenase activity (means for

sites) by 20 to 92-fold, while laminarin significantly (p < 0.05) stimulated rates by 19 to 60-fold, relative to the control values (Table 2). Starch required incubations of > 24

h for stimulation of nitrogenase activity (results not shown). In general, rates of nitro- genase activity stimulated by plant polysaccharides (Figs. 5 and 6) reflected the pat- tern among microhabitats of the unamended samples (Fig. 4); i.e. rates decreased with sediment depth, with Zosteru bed sediment generally supporting the greatest activity. In contrast to polysaccharides of plant origin, additions of chitin were not effective at stimulating nitrogenase activity relative to the control (Table 2).

These data for the effects of plant polysaccharides on nitrogenase activity (Fig. 5 & 6) are supported by results of previous experiments. Similar experiments carried out during November, 1991 (Oesterwal), December, 1991 (Oesterwal), January, 1992 (Geelbek), and May, 1992 (Oesterwal and Geelbek) also concluded that glycogen, alginate, and xylan were effective substrates for supporting nitrogenase activity in these sediments (results not shown). Laminarin was not tested in those experiments.

12 B.J. Tibbtes et ul. /J. Exp. Mur. Bid. Ed. 184 (1994) I-20

3.3. Oxygen regulation

The effect of oxygen on nitrogenase activity in Langebaan sediments is shown in

Table 3. Nitrogenase activity was stimulated in the presence of air in surface sediments. This effect was most dramatic at Geelbek where nitrogenase activity in Zosteru bed sediment was x20-fold greater in the presence of air than in its absence. Oxygen (air) did not stimulate nitrogenase activity in subsurface sediments. Furthermore, it was noted that O,-stimulation of nitrogenase activity in burrow opening sediment was

less marked than in adjacent surface sediment. Sediment from the rims of prawn burrow openings includes subsurface sediments deposited onto the surface due to bioturbation.

,-

,-

I-

I-

ALGINATE

OESTERWAL GEELBEK

CARRAGEENAN

OESTERWAL GEELBEK

Fig. 5. Effects of additions of structural ptant polysaccharides on nitrogenase activity in sediments from Oesterwal and Geelbek. Sediments were collected from five microhabitats at Oesterwal(I2.3.93) and Gecl- bek (24.3.93). Values represent means&SD (n = 2) of rates between z 12-24 h incubation. Overall means (sites and mjcrohabitats) for each treatment are: Cellulose, 0.11 + 0.20; Xylan, 1.55 & 1.57; Alginate. 1.69 t 1.32; Carrageenan, 0.70& 1.50 nmol C,H, (g dry sed.)“.h-‘.

B.J. Tibbles et al. 1 J. Exp. Mm. Bid. Ed. 184 (1994) l-20 13

_”

T GLYCOGEN 20-

15-

10 -

s-

0-V

Surface sediment Zostera bed sediment Bunow opening sediment Burrow lining sediment Subsurface sediment

i

LAMINARIN 20

T

CL”

20 -

15 -

10 -

5-

STARCH

Q1 OESTERWAL GEELBEK

Fig. 6. Effects of additions of storage plant pofysaccharides on nitrogenase activity in sediments from Oesterwal and Geelbek. Sediments were collected from five microhabitats at Oesterwal (12.3.93) and Geel- bek (24.3.93). Values represent means t_ SD (n = 2) of rates between = 12- and 24-h incubation. Overall means (sites and microhabitats) for each treatment are: Starch, 0.21 _+ 0.31; Glycogen, 7.65 _+ 5.52; Lami- narjn, 5.78 f 3.14 nmol C,H, (g dry sed.)-‘,h-‘.

14

Table 3

B.J. Tibbles et al. / J. Exp. Mar. Bid. Ed. 184 (1994) l-20

Effect of 0: on nitrogenase activity in different sediments

Sediment nmol C2H, (g dry sed.)- ‘.h lil

Oesterwal

AnO, 0,

Geelbek

AnO, 0,

Surface 0.03 f 0.00 0.08 _+ 0.06 0.07 _+ 0.01* 0.73 i 0.06

Zostera bed 0.07 * 0.03 0.29 + 0.20 0.59 f 0.13* 11.29 f 1.62

Burrow opening 0.02 f 0.01 0.03 + 0.03 0.04 + 0.02* 0.33 * 0.11

Burrow lining 0.01 * 0.01 0.02 f 0.02 0.03 + 0.01 0.02 * 0.01

10 cm depth 0.01 + 0.01 0.0 1 f 0.00 0.02 + 0.01 0.01 f 0.01

” Values represent means + SD of replicates (~2 = 3), measured after 15h.

* Represents significant (p< 0.05) difference between treatments as determined by Student’s t-tes(

4. Discussion

The concentrations of porewater ammonium from sediments (range 0 to 53 PM) were lower in Langebaan Lagoon than those reported for other seagrass ecosystems (Short, 1983; Horrigan & Capone, 1985; Moriarty et al., 1985). Large populations of burrowing invertebrates are present at Oesterwal and Geelbek. These animals pump a significant volume of water through their burrows while feeding, causing subsurface sediments to be well ventilated (Dye, 1978). Benthic infauna have been shown to increase the flux of nutrients from sediments (Aller, 1978; McCaffrey et al., 1980; Blackburn & Hen- ricksen, 1983; Pomroy et al., 1983), which could account for the unusually low am- monium concentrations in porewater from Oesterwal and Geelbek. Concentrations of porewater ammonium were highest in Zostera bed sediment and subsurface sediments, where prawn densities, and thus ventilation rates, are lowest.

The regulation of nitrogenase activity by ammonium is well-known (Postgate, 1982). Several workers have reported inverse correlations between in situ ammonium con- centrations and nitrogenase activity in saltmarsh sediments (Carpenter et al., 1978; Patriquin & Keddy, 1978; Teal et al., 1979). However, other reports have not concluded that it has a major and consistent role in this regard (Hanson, 1977a,b; Dicker & Smith, 1980). Additions of ammonium (O-l mM) to sediments from Langebaan Lagoon had little effect on nitrogenase activity in muddy sediments, but caused a concentration- dependent inhibition of nitrogenase activity in sandy sediments. Ammonium-control of nitrogen fixation in these sediments may not be significant, however, since in situ NH4+ concentrations were z20-fold lower than additions (50 PM) which inhibited nitrogenase activity in sandy sediments by only 15%, and had no effect on rates in muddy sediments. These findings, especially with respect to the Geelbek muds, con- trast with the reports of other workers who have noted lower threshold concentrations for the inhibition of nitrogenase activity of 100-200 PM NH,’ in other saltmarsh ecosystems (Carpenter et al., 1978; Teal et al., 1979).

Few studies have addressed the modulation of nitrogenase activity in marine envi- ronments by the availability of polysaccharides. Paerl et al. (1987) showed that the

B.J. Tibbles et al. 1 J. Exp. Mar. Biol. Ecol. 184 (1994) I-20 15

amendment of seawater from the coast of North Carolina with combinations of mono-

or disaccharides and homogenized 2. marina or Spartina alteml$ora leaves, enhanced

rates of nitrogen fixation. Whereas the sugars may have provided an energy source for bacterial metabolism and nitrogen fixation, Paerl et al. (1987) attributed the effect of detrital particles to their function as a surface for O,-depleted microzone formation, and thus as suitable sites for O,-sensitive nitrogen fixation, DetritaI material may also be utilized by bacteria as a substrate for growth.

The results of the present study show different responses of nitrogenase activity to

amendments with defined mono- and polysaccharides. These responses varied accord- ing to both the substrate and microhabitat type. Additions of xylan to lagoonal sedi- ments stimulated nitrogenase activity in all microhabitats at both sites; rates stimulated by xylan were w 7-fold greater than those of unamended controls. Alginate was another structural polysaccharide which was effective at stimulating nitrogenase activity, with

the overall magnitude of stimulation similar to that produced by xylan (Z &fold). Al- ginate is a structural polysaccharide in the Phaeophyta (brown algae) which, although largely absent from the study sites in Langebaan Lagoon, are more abundant in the adjacent Saldanha Bay (Fig. 1). Acylated alginic acid is also produced in an extracel- lular mucilage in certain bacteria (Gorin & Spencer, 1966), which have been isolated from saltmarsh sediments (Dicker & Smith, 1981).

Cellulose and hemicellulose (xylan) constitute the major structural polysaccharides

of saltmarsh and seagrass macrophytes, and some algae. Hemicellulose (xylan) ac- counts for 15-30 y0 of plant material and may shield the cellulose polymer from attack by cellulase enzymes (Dekker & Lindner, 1979; Orpin, 1988). In contrast to the effects of xylan and alginate, additions of microcrystalline cellulose did not stimulate nitroge- nase activity in incubations of ~30 h. The effective degradation of microcrystalline cellulose is a complex process usually requiring a synergistic attack by three classes of cellulase enzymes. Such enzyme systems are probably not ubiquitous among natural assemblages (Btguin, 1990). Carrageenan did support nitrogenase activity in some sediment types, but the pattern of its effect was not marked nor consistent between sites.

Of all the polysaccharides tested, glycogen and laminarin were the most effective at stimulating nitrogenase activity. Laminarin is the characteristic glucan of Phaeophyta (e.g. Laminaria spp.). Instead of starch, some algae include amylopectins or glycogens, which are closely related in chemical structure. Starch only stimulated nitrogenase

activity after extended incubations. Starch is a mixture of arnylose and amylopectin. and is the storage polysaccharide of many macrophytes. Stimulation of nitrogenase activity by the storage polysaccharides glycogen and laminarin was significantly (p

< 0.05) greater than that produced by structural polysaccharides (xylan, alginate, cel- lulose, and carrageenan). Furthermore, of those tested, polysaccharides common to many algae (alginate, glycogen, and laminarin) were more effective at stimulating ni- trogenase activity than those dominant in saltmarsh and seagrass macrophytes (cellu- lose, xylan, and starch), even in sediments from beds of 2. marina. The dominant macroalga at Oesterwal and Geelbek is Gracillaria verrucosa, although this is not a significant producer relative to the production of saltmarsh and seagrass macrophytes.

Chitin is the major structural polymer in crustacea, and since saltmarsh lagoons often support large populations of these animals, ecdysis of their exoskeletons may represent

16 B.J. Tibbles et al. / J. Exp. Mar. Biol. Ecol. 184 (1994) l-20

a major input of carbon into these environments. In the context of these studies chitin

therefore provided an interesting comparative substrate to polysaccharides of plant origin. Both cellulose and chitin are /3(1-4)-linked polysaccharides, but whereas cellu- lose consists of monomers of glucose, chitin consists of monomers of N-acetyl-D- glucosamine. The anaerobic degradation of chitin in saltmarsh sediments has been previously addressed with respect to two other biogeochemical cycles, namely sulfate reduction and methanogenesis (Boyer, 1986). In those studies, no sulfate-reducing or methanogenic isolates were capable of chitin utilization. However, in mixed cultures of chitin degraders and sulfate reducers or methanogens, additions of chitin stimulated

sulfide or methane production, respectively (Boyer, 1986). However, the present study showed that chitin was a poor stimulator of nitrogenase activity, indicating that this substrate is not an important regulator of nitrogen fixation in the habitats studied at

Langebaan Lagoon. Despite the extremely oxygen-labile nature of nitrogenase, nitrogen-fixing bacteria

have diverse physiologies, ranging from strict anaerobes to strict aerobes. Aerobic diazotrophs (e.g. Azotobacter spp.) possess physiological mechanisms which shield their

nitrogenase from contact with oxygen. The trade-off between the energy-requirement and oxygen-sensitivity of nitrogenase is emphasized by many reports that have de- scribed different responses of in situ nitrogenase activity to the presence of oxygen. Capone and Budin (1982) found that nitrogenase activity associated with roots and rhizomes of Z. marina was greatest under microaerophilic conditions. Smith & Ha- yasaka (1982a,b) detected O,-enhanced rates of nitrogenase activity in rhizoshere sediments of seagrass ecosystems, while Dicker & Smith (1980) found large popula- tions of aerobic diazotrophs (Azotobacter spp.) were present in sediments from a Dela- ware saltmarsh. Oxygen also significantly (p < 0.05) stimulated nitrogenase activity in surface sediments from Geelbek, Langebaan Lagoon. This oxygen-effect was more pronounced in sediments from beds of Z. caper&, but was also recorded in exposed, surface sediments located away from macrophyte cover. These data indicate that populations of aerobic diazotrophs were active in these sediments, and prompted comparison with the effect of oxygen on rates in burrow lining (a subsurface sediment/ water interface). Oxygen did not stimulate nitrogenase activity in burrow lining. It is not clear why evidence for aerobic diazotrophy was found in surface sediments, but not in burrow lining, since no significant interaction between light and oxygen on nitroge-

nase activity was noted in surface sediments (Tibbles, unpubl. data). Bioturbation of sediments by benthic infauna, such as thalassinids, may increase the

aeration of subsurface sediments (Dye, 1978). Indeed, Capone (1988) has commented “the effect of O,, along with those factors controlling 0, penetration into sediments (temperature, rate of consumption, organic load, bioturbation, etc.) deserve further investigation.” It was noteworthy that O,-stimulation of nitrogenase activity in burrow opening sediment was not as marked as in adjacent surface sediment or Zostera bed sediment (Table 3). Turnover of sediments by prawns (bioturbation) may limit the extent of O,-stimulation of nitrogenase activity in surface sediments, if microaerophilic or aerobic diazotrophs are displaced from the surface around the burrow opening by the effects of bioturbation. At depth, particularly in muddy sediments, bacterial popu- lations may be predominantly anaerobic. Under such conditions O,-inhibition of ni-

B.J. Tibbles et al. /J. Exp. Mar. Bid. Ecol. 184 (1994) I-20 17

trogenase activity might be expected. However, in subsurface sediments from Lange-

baan Lagoon, oxygen had no significant (p > 0.05) effect on nitrogenase activity.

Spatial patterns of nitrogenase activity were evident in Langebaan Lagoon sedi- ments. Nitrogenase activity was greater in the fine, muddy sediments of Geelbek than in the coarser, sandy sediments of Oesterwal. This pattern reflects the organic carbon content of these sites (Table l), and correlates with a previous report that the mudflats of Geelbek support greater bacterial densities than the sandflats of Oesterwal (Tibbles et al., 1992). O’Neil & Capone (1989) noted a similar relationship between particle size and nitrogenase activity for tropical carbonate marine sediments. Patterns of nitroge- nase activity between microhabitats at each site were similar; rates were highest in surface sediments and decreased with depth. Highest rates of nitrogenase activity were

measured in Zostera bed sediment from Geelbek. Seasonal die-back of macrophyte beds in temperate saltmarsh lagoons, such as

Langebaan, results in an increased flux of detrital material for bacterial use (Mazure

& Branch, 1979). This material, typically low in N, becomes N-enriched by bacterial colonization. The N-requirements of bacterial growth on plant materials can be sub- sidized by inorganic nutrients taken from the environment or, as this study shows, by bacterial dinitrogen fixation. It appears that certain polysaccharide constituents of this detrital material are more available for heterotrophic, diazotrophic growth than others.

These findings may assist our understanding of the dynamics of energy flow and the carbon and nitrogen cycles of these ecosystems. Aerobic or microaerophilic respiration also appears to be a significant energy source for nitrogenase activity in these ecosys- tems, despite the O,-sensitivity of nitrogenase. Bioturbation has a negative effect on 02-stimulation of nitrogenase activity when surface sediments in these microhabitats are displaced by subsurface sediments. It is tempting to speculate on the consequences of a feedback loop between bioturbation, nitrogen fixation, and primary production of

diatoms and saltmarsh/seagrass macrophytes, in studies of the quantity and nutritional value of food-sources available to benthic invertebrate detritivores, which rely prima- rily on plant material as an energy source.

Acknowledgements

We would like to thank Dr. D.G. Capone for criticisms of an earlier draft of this manuscript, and Mr. B.P. Emanuel for assistance with fieldwork. This work was funded by an award from the University of Cape Town Research Committee to M.I. Lucas and B.J. Tibbles.

References

Aller, R.C., 1978. Experimental studies of changes produced by deposit feeders on porewater, sediment, and

overlaying water chemistry. Am. J. Sci., Vol. 278, pp. 1185-1234.

Atkinson, M.J. & S.V. Smith, 1983. C:N:P ratios of benthic marine plants. Limnol. Oceanogr., Vol. 28,

pp. 568-574.

Bkguin, P., 1990. Molecular biology of cellulose degradation. Annu. Rev. Microbial., Vol. 44, pp. 219-248.

18 R.J. Tibbtes et ul. 1 J. Exp. Mar. Riot. Ecol. 184 f‘f994J l-20

Bird, E.W., 1982. Population dynamics of the Thalassinidean shrimps and their community eflects through sediment modification. Ph.D. dissertation, University of Maryland, College Park, USA.

Blackburn, T.H. & K. Henriksen, 1983. Nitrogen cycling in different types of sediments from Danish wa-

ters. Limnol. Uceunogr., Voi. 28, pp. 471-493.

Boyer, J.N., 1986. End products of anaerobic chitin degradation by saltmarsh bacteria as substrates

for dissimilatory sulfate reduction and methanogenesis. Appt. Environ. Microbial., Vol. 52, pp. 1415-

1418.

Branch, G.M. & A. Pringle, 1987. The impact of the sand prawn Calliunassa kraussi stebbing on sediment

turnover and on bacteria, meiofauna, and benthic microfauna. J. Exp. Mar. &of. Ecof., Vol. 107, pp. 2 1% 235.

Capone, D.G., 1982. Nitrogen fixation (acetylene reduction) by rhizosphere sediments of the eelgrass Zosteru

marina. Mar. Ecol. Prog. Ser., Vol. 10, pp. 67-75.

Capone, D.G., 1983. N, fixation in seagrass communities. Mar. Technol. Sot. J., Vol. 17, pp. 32-37.

Capone, D.G., 1988. Benthic nitrogen fixation. In, Nitrogen cycling in the marine enuironmenr, edited by T.H.

Blackburn & J. Sorensen, John Wiley and Sons Ltd., New York, pp. 85-123.

Capone, D.G. & J.M. Budin, 1982. Nitrogen fixation associated with rinsed roots and rhizomes of the ee-

lgrass i&era marina. Plum Physiol., Vol. 70, pp. 1601-1604.

Capone, D.G., P.A. Penhale, R.S. Oremland & B.F. Taylor, 1979. Relationship between productivity and

N, (C,H,) fixation in a Thalassia testudinum community. Limnoi. Oceanogr., Vol. 24, pp. 117- 125.

Carpenter, E.J., C.D. van Raalte, I. Vahela, 1978. Nitrogen fixation by algae in a Massachusetts salt marsh.

Limnol. Oceanogr., Vol. 23, pp.318-327.

Dekker, R.F.H. & W.A. Lindner, 1979. Bioutilization of lignocellulosic waste materials: a review. S. Afr. J. Sci., Vol. 15, pp. 65-71.

Dicker, H.J. & D.W. Smith, 1980. Enumeration and relative importance of acetylene-reducing (nitrogen-

fixing) bacteria in a Delaware salt marsh. Appl. Environ. Microbioi., Vol. 39, pp. 1019-1025.

Dicker, H.J. & D.W. Smith, 1981. Effects of salinity on acetylene reduction (nitrogen fixation) and respi-

ration in a marine Azotobacter. Appl. Environ. Microbiot., Vol. 42, pp. 740-744.

Dye, A.H., 1978. An ecophysiological study ofthe meiofauna ofthe Swartkops estuary. I. The sampling sites:

physical and chemical features. Zool. A,fr., Vol. 13, pp. l-18.

Gandy, E.L. & D.C. Yoch, 1988. Relationship between nitrogen-fixing sulfate reducers and fermenters in

salt marsh sediments and roots of Spartina a~terni~~~ru. Appl. Environ. M~~robio~., Vol. 54, pp. 2031-

2036.

Gorin, P.A.J. & J.F.T. Spencer, 1966. Exocellular alginic acid from Azotohacter vine/andii. Con. J. Chem.. Vol. 44, pp. 993-998.

Gottschalk, G., 1986. Bacterial metabolism. Springer-Verlag, New York, second edition, 359 pp.

Hanekom, N.M., 1980. n study of two thalassinid prawns in the jron-Spartina regions of‘ the Swrtkops estu- ary. Ph.D. thesis, University of Port Elizabeth, South Africa.

Hanson. R.B., 1977a. Comparison ofnitrogen fixation activity in tall and short Spartinu ultermj?ora salt marsh

soils. Appl. Environ. Microbial. Vol. 33, pp. 596-602.

Hanson, R.B., 1977b. Nitrogen fixation (acetylene reduction) in a saltmarsh amended with sewage sludge

and organic carbon and nitrogen compounds. Appl. Environ. Microbial., Vol. 33, pp. 846-852.

Hardy, R.W.F., R.D. Holsten, E.K. Jackson & R.C. Burns, 1968. The acetylene-ethylene assay for N?-

fixation: laboratory and field evaluation. PIunr Ph_v.riol., Vol. 43, pp. 1185-1207.

Horrigan, S.G. & D.G. Capone, 1985. Rates of nitrification and nitrate reduction in nearshore marine

sediments at near ambient substrate concentrations. Mar. Chem., Vol. 16, pp. 317-327.

Koike, 1. & H. Mukai, 1983. Oxygen and inorganic contents and fluxes in burrows of the shrimps Callianassu japonica and Upogebia major. Mar. Ecol. Prog. Ser., Vol. 12, pp. 185-190.

Koroleff, F., 1983. Determination of nutrients. In, ~eth~~~~ c?frea~uteranalr:Fis. edited by K. Grasshoifet al.,

Verlag Chemie, pp. 125-187.

Mazure, H.G.F. & G.M. Branch, 1979. A preliminary analysis of bacterial numbers and biomass in Lange-

baan Lagoon. Trans. R. Sot. S. A/?.. Vol. 44, pp. 43-54.

McCaffrcy, R.J., A.C. Myers, E. Davey, G. Morrison, M. Bender, N. Luedtke, D. Cullen, P. Froelich, &

G. Klinkhammer, 1980. The relation between pore water chemistry and benthic fluxes of nutrients and

manganese in Narragansett Bay, Rhode Island. Lintnot. Oceanogr., Vol. 25. pp. 31-44.

B.J. Tibbles et (11. / J. E-up. Mur. Biol. Ecol. 184 (1994) I-20 19

McRoy, C.P. & C. McMillan, 1977. Production ecology and physiology of seagrasses. In, Seagrass eco.yJs-

tents, edited by C.P. McRoy & C. Helfferich, Marcel Dekker Inc., New York, pp. 53-87.

Moriarty, D.J.W., P.I. Boon, J.A. Hanson, W.G. Hunt, I.R. Poiner, P.C. Pollard, G.W. Skyring & D.C.

White, 1985. Microbial biomass and productivity in seagrass beds. Geomicrobiol. J., Vol. 4, pp. 21-51.

Newell. R., 1965. The role of detritus in the nutrition of two marine deposit feeders, the prosobranch Hydrobiu

ulvrte and the bivalve Macoma bulthica. Proc. 2001. SIX., Vol. 144, pp. 25-45.

O’Donohue, M.J., D.J.W. Moriarty & I.C. Mac Rae, 1991. Nitrogen fixation in sediments and the rhizo-

sphere of the seagrass Zostera cqrieorni. Microb. Ecol., Vol. 22, pp. 53-64.

Q’Neil, J.M. & D.G. Capone, 1989. Nitrogenase activity in tropical carbonate marine sediments. Mrrr. Ecol.

Prog. Ser., Vol. 56, pp. 145-156.

Ogden. J.C., 1980. Fauna1 relationships in Caribbean seagrass beds. In, Handbook of seagrass biology; wt

rco.y~tem pevwctive, edited by R.C. Phillips & C.P. McRoy, Garland STPM Press. New York, pp. I73-

198.

Orpin, C.G., 1988. Genetic approaches to the improvement of lignocellulosic degradation in the rumen. In.

Biochemistry rrrzd genetics of cellulose degrudution. edited by J-P. Aubert, P. Beguin & J. Millet, Academic

Press, London. pp. 171-179.

Pacrl, H.W., 1990. Physiological ecology and regulation of N, fixation in natural waters. Adv. Microh. Ecol..

Vol. 11, pp. 305-344.

Paerl. H.W., K.M. Cracker & L.E. Prufert, 1987. Limitation of Nz fixation in coastal marine waters: rela-

tive importance of molybdenum, iron, phosphorus, and organic matter availability. Limnol. Uceano~r. Vol.

32, pp. 525-536.

Patriquin, D.G., 1972. The origin of nitrogen and phosphorus for growth of the marine angiosperm Thalrxsirr

testudinum. Mar. Biol., Vol. 15, pp. 35-46.

Patriquin, D. & R. Knowles, 1972. Nitrogen fixation in the rhizospherc of marine angiosperms. Mur. Biol..

Vol. 16. pp. 49-58.

Pomcroy. L.R., 1980. Detritus and its role as a food resource. In, Fundementals of uquatic ecosystems, ed-

itcd by R.S.K. Barnes & K.H. Mann, Blackwell Scientific Publications, Oxford, pp. 84-102.

Pomroy. A.J.. I.R. Joint & K.R. Clarke, 1983. Benthic nutrient flux in a shallow coastal environment.

Oecologia, Vol. 60, pp. 306-312.

Postgate. J.R., 1982. The Fundementuls q/ nitrogerz,fixation. Cambridge University Press, London.

Reichenbach, H. & M. Dworkin, 1981. The order Myxobacteriules. In, Theprokaryotes, edited by M.P. Starr.

H. Stolp, H.G. Truper, A. Balows, H.C. Schlegel, Springer-Verlag, Berlin, Vol. 1, pp. 360-376.

Roberts, H.H.. T.H. Suchanek & W.J. Wiseman, 1981. Lagoon sediment transport: the significant effect of

Culliun~.wa bioturbation. Proc. 4th Irrt. Corul Red Symposium. Manila, pp. 459-465.

Schubauer, J.P. & C.S. Hopkinson, 1984. Above- and belowground emergent macrophyte production and

turnover in a coastal marsh ecosystem, Georgia. Limnol. Uceanogr., Vol. 29, pp. 1052-1065.

Seitzinger, S.P., S.W. Nixon & M.E.Q. Pilson, 1984. Denitrification and nitrous oxide production in a coastal

marine ecosystem. Limno/. Oceunogr., Vol. 29, pp. 73-83.

Short. F.T., 1983. The response of interstitial ammonium in eelgrass (Zostera marina L.) beds to environ-

mental pertubations. J. Exp. Mar. Biol. Ecol., Vol. 68, pp. 195-208.

Smith, G.W. & S.S. Hayasaka, 1982a. Nitrogenase activity associated with Hulodule wightii roots. Appl.

Emiron. Microbial., Vol. 43, pp. 1244- 1248.

Smith, G.W. & S.S. Hayasaka, 1982b. Nitrogenase activity associated with Zosteru marina from a North

Carolina estuary. Can. J. Microbial.. Vol. 28, pp. 448-451.

Sichanek. T.H., 1983. Control of seagrass communities and sediment distribution by CaNiarzassa (Crusta-

tea, Thalassinidea) bioturbation. J. Mar. Res., Vol. 41, pp. 281-298.

Teal. J.M.. 1. Valiela & D. Berlo, 1979. Nitrogen fixation by rhizosphere and free-living bacteria in saltmarsh

sediments. Limttol. Oceunogr., Vol. 24, pp. 126-132.

Tibbles, B.J., C.L. Davis, J.M. Harris & M.I. Lucas, 1992. Estimates of bacterial productivity in marine

sediments and water from a temperate saltmarsh lagoon. Microb. Ecol. Vol. 23, pp. 195-209.

Valiela, I. & J.M. Teal, 1979. The nitrogen budget of a salt marsh ecosystem. Nature, Vol. 280, pp. 652-

656.

Valicla, I., J.M. Teal & N.Y. Persson, 1976. Production and dynamics of experimental] enriched salt marsh

vegetation: belowground biomass. Limnol. Oceatzogr., Vol. 21, pp. 245-252.

‘0 E.J. Tihbies et ai. / J. Exp. Mar. Biol. Ecol. I84 (19941 I-20

Valiela, l., J.M. Teal, S. Volksmann, D. Shafer & E.J. Carpenter, 1978. Nutrient and particulate thrxes in

a salt marsh ecosystem: tidal exchanges and inputs by precipitation and groundwater. Limnol. Oeeanogr., Vol. 23, pp. 798-8 12.

Whiting, G.J., H.N. McKellar, B. Kjerfve & J.D. Spurrier, 1987. Nitrogen exchange between a southeast-

ern USA salt marsh ecosystem and the coastal ocean. Mar. Biol., Vol. 95, pp. 173-182.

Wolff, T., 1980. Animals associated with seagrass in the deep-sea. In, Handbook of seagrass biology: on ecns,wem persective, edited by R.C. Phillips & C.P. McRoy, Garland STPM Press, New York, pp. 199-

224.

Yingst. J.Y. & D.C. Rhoads, 1980. The role of bioturbation in the enhancement of bacteria1 growth rates

in marine sediments, In, MC&e henthic d~no~jc.~, edited by K.R. Tenore & B.C. Could, University of South

Carolina Press, Columbia, USA, pp. 407-421.