Patterns in the Fate of Production in Plant Communities

vol. 154, no. 4 the american naturalist october 1999 Patterns in the Fate of Production in Plant Communities Just Cebrian* Boston University Mari...
Author: Mark Wright
1 downloads 0 Views 228KB Size
vol. 154, no. 4

the american naturalist

october 1999

Patterns in the Fate of Production in Plant Communities

Just Cebrian*

Boston University Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543 Submitted June 12, 1998; Accepted May 31, 1999

abstract: I examine, through an extensive compilation of published reports, the nature and variability of carbon flow (i.e., primary production, herbivory, detrital production, decomposition, export, and biomass and detrital storage) in a range of aquatic and terrestrial plant communities. Communities composed of more nutritional plants (i.e., higher nutrient concentrations) lose higher percentages of production to herbivores, channel lower percentages as detritus, experience faster decomposition rates, and, as a result, store smaller carbon pools. These results suggest plant palatability as a main limiting factor of consumer metabolical and feeding rates across communities. Hence, across communities, plant nutritional quality may be regarded as a descriptor of the importance of herbivore control on plant biomass (“top-down” control), the rapidity of nutrient and energy recycling, and the magnitude of carbon storage. These results contribute to an understanding of how much and why the trophic routes of carbon flow, and their ecological implications, vary across plant communities. They also offer a basis to predict the effects of widespread enhancement of plant nutritional quality due to largescale anthropogenic eutrophication on carbon balances in ecosystems. Keywords: plant community, primary production, herbivory, decomposition, carbon storage.

The amount of CO2 fixed by primary production in plant communities can follow diverse trophic routes (fig. 1). Herbivores remove a fraction of production, with the rest being accumulated as plant biomass and entering the degradable detrital compartment over the plant life span. This compartment may also receive plant detritus imported from other communities (Nixon 1980; Twilley et al. 1986). A fraction of plant detritus is exported off the community by means of physical agents, such as rivers in * Address for correspondence beginning January 1, 2000: Dauphin Island Sea Lab, 101 Bienville Boulevard, P.O. Box 369-370, Dauphin Island, Alabama 36528; e-mail: [email protected].

Am. Nat. 1999. Vol. 154, pp. 449–468. q 1999 by The University of Chicago. 0003-0147/1999/15404-0006$03.00. All rights reserved.

terrestrial communities or currents and waves in aquatic ones (Mann 1988; Pomeroy and Wiebe 1993). The rest of detritus undergoes decomposition within the community, with a small fraction of recalcitrant detritus escaping from further degradation and entering the refractory compartment (Schlesinger 1997). Temporal changes in the biomass and degradable detrital mass of plant communities are related to the trophic routes of production by the following mass-balance equations: (DB/t) = NPP 2 C 2 DP,

(1)

(DDB/t) = DP 1 I 2 D 2 E 2 RA,

(2)

where DB/t and DDB/t correspond to the increments in plant biomass and degradable detrital mass per unit time (g C m22 d21), respectively, and NPP, C, DP, I, D, E, and RA stand for net primary production, consumption by herbivores, detrital production, import, decomposition, export, and refractory accumulation, respectively (all variables in g C m22 d21). Under steady state assumptions, B and DB remain constant with time, and, thus, the sum of primary production and import equals the sum of herbivory, decomposition, export, and refractory accumulation. The assessment of the trophic routes of production in a plant community is important because these routes are indicative of the ecological role of the community. The extent of the fraction of production removed by herbivores reflects the importance of herbivores as controls of plant biomass (Petrusewicz and Grodzinski 1975; Cebrian and Duarte 1994). On the other hand, the absolute flux of production consumed by herbivores (in g plant C m22 d21) set limits to the abundance of herbivores supported by a plant community (McNaughton et al. 1989; Cyr and Pace 1993). The amount of detritus exported off a plant community points to the magnitude of allochthonous secondary production supported by the community because most exported detritus are consumed within the receiving communities (Mann 1988). In contrast, accumulation of refractory detritus within a community represents a net

450 The American Naturalist

Figure 1: Fate of primary production in plant communities

carbon sink since refractory carbon is preserved over longterm scales (Schlesinger 1997). A few studies have analyzed the extent and controls of the variability in specific trophic routes of production across diverse plant communities. Comparisons among a broad range of aquatic and terrestrial communities show that more productive communities support higher levels of absolute consumption by herbivores, and consequently, they maintain larger herbivore abundances (McNaughton et al. 1989; Cyr and Pace 1993). Absolute consumption increases linearly with primary production across communities (McNaughton et al. 1989; Cyr and Pace 1993). This tendency implies that the percentage of production removed by herbivores, which is indicative of their potential role as controls of plant biomass, is independent of the magnitude of primary production when aquatic and terrestrial communities are compared (Cebrian and Duarte 1994). The fraction of production removed is associated with the biomass turnover rate (i.e., fraction of biomass renewed per day [d21]) of the plant community, probably because faster turning-over communities have a greater palatability for herbivores (Cebrian and Duarte 1994; Duarte and Cebrian 1996). Plant detritus from communities with faster biomass turnover rates also seem to undergo faster decomposition rates probably because of their expected greater palatability for decomposers (En-

riquez et al. 1993; Cebrian and Duarte 1995). Plant communities having higher fractions of production removed by herbivores and faster decomposition rates store smaller biomass and detrital pools if differences in the intensity of heterotrophic consumption override differences in the magnitude of production (Cebrian et al. 1998). Marine plant communities typically exposed to substantial advection by waves and currents tend to export large amounts of detritus (Duarte and Cebrian 1996). Nevertheless, no attempt to compare all the trophic routes of production combined into thorough carbon balances across a broad range of aquatic and terrestrial plant communities has been made. Our knowledge about how primary production, consumption by herbivores, detrital production, decomposition, refractory accumulation, export, and import dovetail carbon balances across a continuum of diverse aquatic and terrestrial plant communities is meager. Likewise, although controls of the variability in some particular routes across communities have been identified, we lack knowledge of mechanisms responsible for differences in the entire carbon balance among communities. It has been suggested that plant communities with faster biomass turnover rates have higher fractions of production lost to herbivores and faster decomposition rates because they should have higher plant nutrient concentrations and,

Fate of Production in Plant Communities 451 thus, enhance the metabolism and feeding rates of heterotrophs (Cebrian and Duarte 1994, 1995; Duarte and Cebrian 1996). Nielsen et al. (1996) compare turnover rates and nutrient concentrations for a broad range of types of autotrophs (from microalgae to terrestrial macrophytes) under natural and experimental conditions and conclude a strong association between higher nutrient concentrations and faster turnover rates. Therefore, communities formed by plants with faster turnover rates should also have plants with higher nutrient concentrations. Yet, the association between plant nutrient concentrations and turnover rates across aquatic and terrestrial communities, and its effects on the levels of consumption by heterotrophs, remains to be tested. Moreover, greater levels of consumption by heterotrophs in communities with faster turnover rates could entail lower levels of accumulation of biomass and detrital mass within these communities, provided differences in consumption are greater than differences in production. Demonstrating these hypotheses would render the association between plant trophic quality and turnover rate as a main control of the carbon balance in plant communities: because plant communities with faster turnover rates are more palatable to heterotrophs, they would experience higher intensities of consumption (i.e., larger fractions of production removed and faster decomposition rates) and, as a result, they would accumulate less biomass and detrital mass. That would be so regardless of other numerous factors that could influence differences in consumption by herbivores (e.g., characteristics of the specific herbivore populations in the different plant communities compared) and in decomposition and carbon storage (e.g., temperature, humidity, sediment redox conditions) across the plant communities compared. In this article, I examine, through an extensive compilation of published reports, differences in the trophic routes of carbon flow (eqq. [1] and [2]) across a wide range of aquatic and terrestrial types of plant communities and analyze how these differences affect their ecological role. Then I test whether communities composed of plants with higher nutrient concentrations also have faster biomass turnover rates. I further test whether, as a result of the association between higher plant palatability and faster turnover rates, communities with faster turnover rates are subject to higher levels of consumption by herbivores and faster decomposition rates and, as a consequence, accumulate less plant biomass and detrital mass. I end by discussing the contribution of the patterns found to an understanding of how much and why carbon balances differ across plant communities.

Methods From over 200 published reports, I compiled an extensive data set of plant nutrient concentration, biomass, detrital mass, and magnitude and trophic fate (consumption by herbivores, detrital production, export, decomposition, and refractory accumulation) of production for the following types of plant communities: marine and freshwater phytoplanktonic communities, marine and freshwater benthic microalgal beds, marine macroalgal beds, freshwater macrophyte meadows, seagrass meadows, brackish and marine marshes, grasslands, mangroves, and shrublands and forests (see data set). The data set is available from the author upon request at http://www.mbl.edu/usr/ just/. Reports covering all the trophic routes of production for a given community were obviously uncommon, and, hence, the data set compiled combines reports covering different numbers of the variables considered. Moreover, the variables compiled only referred to the aboveground compartment in most of the reports for aquatic or terrestrial macrophyte communities, with only a few reports accounting for both the below- and aboveground compartments (see data set). Three initial criteria were applied when selecting papers: first, only studies of plant communities referring to natural conditions were considered (i.e., studies with manipulated conditions and artificial communities were discarded); second, studies focused on a single plant species were accepted only if that species accounted for most biomass and production within the community; and third, studies were considered if they covered at least 1 yr or the full growing season for seasonal producers (which usually spans from late winter to early fall). Plant nutrient concentrations were reported as nitrogen (N) and phosphorus (P) concentrations expressed as the percentage of plant tissue dry weight (%DW). Nitrogen and phosphorus are the two nutrients most often reported in the literature. Plant biomass corresponds to the stock of alive plant material (g C m22). The values compiled of plant detrital mass (g C m22) correspond to particulate detritus above the upper sediment layers for aquatic and terrestrial macrophyte communities, and to dead cells for microalgal communities. Hence, the detrital pool of plantdissolved carbon is disregarded because it was reported in very few articles. The values of detrital mass compiled for communities of macrophytes include both degradable and refractory detritus, although the total pools of degradable and refractory detritus were underestimated since only the upper sediment or soil layers were accounted for. When temporal changes in plant nutrient concentrations, biomass, and detrital mass through the study period were reported, yearly average values were calculated. Net primary production is defined as the net amount of atmos-

452 The American Naturalist pheric CO2 fixed by photosynthesis. It was expressed in g C m22 d21 by dividing the cumulative value for the study period by the number of days covered by the study. Plant turnover rate corresponds to the fraction of biomass renewed per day (d21), and it was estimated as the ratio of primary production to plant biomass when direct values were not provided in the report. Plant consumption by herbivores is defined as the ingestion of plant biomass by herbivores (g C m22 d21). Detrital production (g C m22 d21) corresponds to the sum of plant mortality by senescence, metabolical exudation of organic carbon, and plant material removed but nonconsumed by herbivores. Metabolical carbon exudation was not measured in many of the reports compiled for aquatic and terrestrial macrophytes, but the magnitude of plant carbon exudation in these communities is normally small in comparison with that of mortality by senescence (Grier and Logan 1977; Wetzel 1984). Hence, detrital production in communities of macrophytes should not be underestimated importantly. Plant mortality by senescence is estimated as “litterfall,” that is, senescent material shed from the plant, for communities of macrophytes, and as cell sedimentation and lysis rates for communities of microalgae. Direct measurements of lysis rates were provided in few reports, with most rates thus being estimated as the difference between primary production and the sum of any biomass increment, consumption by herbivores, sedimentation, metabolical exudation, and horizontal advection. Values of plant consumption and detrital production were expressed on a daily basis by dividing the cumulative values over the study period by the number of days covered. Plant decomposition (g C m22 d21) is the consumption of plant detritus by decomposers. Direct values of plant decomposition were not provided in most of the phytoplanktonic and benthic microalgal communities compiled. I estimated it from respiration measurements of the total micropelagic or microbenthic communities enclosed in incubation bottles or benthic chambers, respectively, as explained by Duarte and Cebrian (1996): plant decomposition within the bottles or chambers was estimated by subtracting the sum of phytoplankton or microphytobenthos respiration and grazing by microherbivores (i.e., those contained in the bottle or chamber) from the total community respiration value. Phytoplankton and microphytobenthos respiration was estimated from values of gross primary production for the same communities by assuming that respiration represents 35.4% 5 6.9% and 26.4% 5 6.4% of gross phytoplankton and microphytobenthos production, respectively. Grazing by microherbivores was supposed to be 50% 5 30% of the total grazing value (micro- and macroherbivores) obtained for the same community. These conversion values (mean 5 SD)

were obtained by Duarte and Cebrian (1996) through a compilation of published reports. Finally, decomposition of sedimenting dead phytoplanktonic cells was assumed to be 17% of the net primary production measured in the community (Martin et al. 1987) and added to the decomposition values obtained from the bottles. The value of 17% corresponds to a mean for a vast oceanic area, and the percentage of phytoplanktonic production decomposed during sedimentation can vary considerably among regions (Muller and Suess 1979; Suess 1980). Nevertheless, the use of mean values of autotrophic respiration, grazing by microherbivores and decomposition of sedimenting phytoplanktonic detritus should not entail a consistent bias on the derivation of decomposition for phytoplanktonic and microphytobenthic communities. This is shown by the comparison of the estimated carbon budgets among the communities considered (see below). Moreover, the error in the estimates of phytoplanktonic and microphytobenthic decomposition involved by the mean values used is unimportant in relation to the broad range of decomposition encompassed by the aquatic and terrestrial communities compared (see results on decomposition). When direct values of decomposition were not provided in communities of aquatic and terrestrial macrophytes, they were derived as D = (DP 2 E) # (1 2 e2kDt ),

(3)

where D, DP, and E denote cumulative decomposition, detrital production, and detrital export over the duration of the study (i.e., g C m22 [study duration]21), Dt corresponds to the duration of the study (d), and k is the decomposition rate (d21). Decomposition was expressed into a daily basis (g C m22 d21) by dividing the cumulative values by the duration of the study. Equation (3) assumes that the pool of degradable detrital mass in the community (fig. 1) does not change over the study period, and, hence, it was only applied in climax communities with steady pools of degradable detrital mass. When decomposition rates (k, d21) were not derived directly, they were estimated as the ratio of detrital production (g C m22 d21) to degradable detrital mass (g C m22) since the latter variable remained unchanged over the study period (Olson 1963). A few decomposition values in some types of communities of macrophytes were derived by using mean values of k for the specific types provided by Enriquez et al. (1993). The estimates of decomposition only cover the duration of the studies compiled (i.e., from one to a few years) and, hence, disregard long-term decomposition of refractory detritus. Export corresponds to the transportation of plant detritus off the community through rivers in terrestrial communities and waves, tides, and currents in aquatic com-

Fate of Production in Plant Communities 453 munities. Export off phytoplankton communities in enclosed lakes and open ocean was assumed to be 0 since phytoplankton detrital carbon, albeit horizontally advected, never traversed the community boundaries. Hence, export values for phytoplanktonic communities refer only to coastal communities. None of the reports compiled accounted for export of dissolved detrital material, and, hence, these estimates correspond only to export of particulate detrital material. Export values were expressed on a daily basis (g C m22 d21) by dividing the cumulative values over the study period by the number of days covered. Refractory accumulation corresponds to the amount of nonexported detrital production that is not decomposed during the duration of the study (i.e., from one to some few years). It thus represents the amount of refractory detritus accumulated during the duration of the study but disregards long-term accumulation once diagenetic losses have occurred (i.e., burial). Whenever direct values were not reported, refractory accumulation was calculated as the difference between net primary production and the sum of consumption by herbivores, export, and decomposition over the study period (Schlesinger 1997). Refractory accumulation was then expressed on a daily basis (g C m22 d21) by dividing the cumulative value by the number of days in the study. This procedure was valid only for climax communities with steady pools of plant biomass and degradable detrital mass. All the different assumptions, methods, and mean values employed for the estimation of the trophic routes of production bring out the questions whether these estimates are representative of the real carbon balances in the communities considered and, thus, whether they are comparable among different communities. To address these questions, I calculated the mean percentage of production represented by consumption by herbivores, decomposition, export, and refractory accumulation for each type of community from all the values in the data set. Biomass and detrital mass were steady and import negligible during the study interval in practically all the communities compiled. Therefore, if the mean percentages of production consumed, decomposed, exported, or accumulated sum up to about 100% for each community type, the values in the data set should be in general representative of real carbon balances in the communities considered. The sums obtained were (mean 5 SE): for phytoplankton, 112.2% 5 12.6%; for benthic microalgal beds, 102.4% 5 20.1%; for macroalgal beds, 98.5% 5 27.7%; for freshwater macrophyte meadows, 123.7% 5 23.8%; for seagrass meadows, 112.5% 5 22.6%; for marshes, 116.2% 5 18.7%; for grasslands, 126.7% 5 17.5%; for mangroves, 90.5% 5 22.2%; and for shrublands and forests, 95.6% 5 10.6%. None of these values is significantly dif-

ferent from 100% (t-test, P 1 .05). Hence, although the assumptions employed entail a certain bias in some of the estimates, the values in the data set generally represent valid carbon balances for the communities considered, and, therefore, the differences in the trophic fate of production found across these communities are reliable. Values reported in g plant DW were multiplied by 0.4 for conversion into g C (Wiebe 1988; Schlesinger 1997). Values for phytoplankton reported on a volumetric basis (i.e., g C m23 or g C m23 d21) were integrated through the depth limits of phytoplankton distribution in the water column and transformed into an areal basis (i.e., g C m22 or g C m22 d21). ANOVA was performed on differences in plant nutrient concentration, plant carbon pools (biomass and detrital mass), and routes of carbon flow (herbivore consumption, detrital production, decomposition, export, and refractory accumulation) across communities. Differences between two specific types of communities were analyzed with a HSD Tukey multiple comparison test (Zar 1984). Relationship between variables were analyzed with correlation and least-square regression techniques. Differences among correlation coefficients were tested with the Fischer Z statistic (Fischer 1921). The variables were log-transformed when necessary to comply with the assumptions of the techniques used. Results The NPP varied significantly among the communities considered (fig. 2A; ANOVA, P ! .05). Yet, when specific pairs of communities were compared, most pairs of communities showed similar ranges of NPP (Tukey HSD test, P 1 .05). Only benthic microalgal beds reached smaller values of production than most other community types (Tukey HSD test, P ! .05). In contrast, both turnover rate and N and P concentrations declined from microalgal communities to communities of aquatic macrophytes to communities of terrestrial macrophytes (fig. 2B– 2D; Tukey HSD test, P ! .05). In fact, faster turnover rates were closely associated with higher N and P concentrations across communities (fig. 3A, 3B; table 1), whereas primary production was independent of N and P concentrations and turnover rate (fig. 4A–4C; table 1). The nutrient concentrations reported for phytoplanktonic communities refer only to marine phytoplankton. Freshwater phytoplankton normally have lower concentrations (mean N range: 3.9%–5.2%; mean P range: 0.35%–0.62%; Hecky et al. 1993). The inclusion of these values would have certainly increased the range for the phytoplankton communities, but the differences found across communities would still hold (fig. 2C, 2D). Moreover, because the nature of the dependence of phytoplankton turnover rates on internal nutrient concentrations seems similar in marine and fresh-

454 The American Naturalist

Figure 2: Box plots showing the distribution of (A) net primary production, (B) plant turnover rate, (C) plant nitrogen concentration, (D) plant phosphorus concentration, (E) percentage of primary production consumed by herbivores, and (F) absolute consumption by herbivores across the types of communities considered. Boxes encompass 25% and 75% quartiles, and the central line represents the median. Bars encompass the range of values between the 25% quartile minus 1.5 times the difference between the quartiles 75% and 25% and the 75% quartile plus 1.5 times the difference between the quartiles 75% and 25%. Circles and asterisks represent values outside these limits.

Fate of Production in Plant Communities 455

Figure 3: The relationships (A) between plant turnover rate and nitrogen concentration and (B) between plant turnover rate and phosphorus concentration across the communities compiled. Open circle, phytoplankton; filled circle, benthic microalgae; open square, macroalgal beds; filled diamond, freshwater macrophyte meadows; filled square, seagrass meadows; filled triangle, marshes; open triangle, grasslands; open diamond, mangroves; and asterisk, forests. Lines represent the fitted equations.

water communities (Kilham and Hecky 1988), freshwater phytoplankton with lower nutrient concentrations should show proportionally slower turnover rates. Therefore, the associations between turnover and nutrients found across communities (fig. 3A, 3B) should not have been affected significantly had freshwater phytoplankton been considered. The percentage of production consumed by herbivores was higher in microalgal communities than in communities of macrophytes (fig. 2E; Tukey HSD test, P ! .05). The distribution of percentage of production consumed across communities resembled the distributions of nutrient concentrations and turnover rate. Indeed, the percentage of production consumed increased with faster turnover rates across communities (fig. 5A; table 1). In contrast, absolute consumption did not show a significant increase from communities of macrophytes to microalgal communities (fig. 2F; Tukey HSD test, P 1 .05), although it varied among communities (ANOVA, P ! .05). The distribution of absolute consumption across communities rather resembled that of primary production. In fact, absolute consumption was associated with primary production across communities, with more productive communities channeling larger fluxes of production toward herbivores (fig. 5B; table 1). Correspondingly, the percentage of production consumed and absolute consumption were very poorly related to primary production and turnover rate, respectively (fig. 5C, 5D). The percentage of production entering the detrital com-

partment was lower in microalgal communities than in communities of macrophytes (fig. 6A; Tukey HSD test, P ! .05). Absolute detrital production, however, was not smaller in microalgal communities (fig. 6B; Tukey HSD test, P 1 .05), but it displayed a distribution across communities similar to that of primary production. Accordingly, absolute detrital production and primary production were strongly related across communities, with more productive communities transferring proportionally larger fluxes of detrital production (fig. 7; table 1). Decomposition rates were faster in microalgal communities than in communities of macrophytes (fig. 6C; Tukey HSD test, P ! .05). Across communities, decomposition rates were associated with biomass turnover rates, with communities with faster turnover rates having also faster decomposition rates (fig. 8A; table 1). Conversely, absolute decomposition did not show an increase from macrophytic to microalgal communities (fig. 6D; Tukey HSD test, P 1 .05). Absolute decomposition was closely associated with primary production across communities, with more productive communities showing higher values of absolute decomposition (fig. 8B; table 1). Accordingly, decomposition rates and absolute decomposition were independent of primary production and turnover rates, respectively (fig. 8C, 8D; table 1). Refractory accumulation, both in absolute terms and as percentage of primary production, was smaller in microand macroalgal communities than in the rest of communities (fig. 6E, 6F; Tukey HSD test, P ! .05). Across

456 The American Naturalist Table 1: Relationships fitted across plant communities Dependent variable

Independent variable

PTR

N

PTR

P

PTR NPP NPP PPC

NPP N P PTR

PPC C

NPP NPP

C DP

PTR NPP

k

PTR

k D

NPP NPP

D RA

PTR PTR

RA B

NPP PPC

B

PTR

B

NPP

DM

k

DM

PTR

DM DM

NPP DP

Relationship

n

R2

F-ratio

P

log PTR = 22.6 (5.1) 1 1.9 (5.1) log N log PTR = 21.2 (5.1) 1 1.3 (5.1) log P ) ) ) log PPC = 1.9 (5.1) 1 .4 (5.04) log PTR ) log C = 2.9 (5.1) 1 .8 (5.06) log NPP ) log DP = 2.2 (5.03)11.1(5.03) log NPP log k = 2.8 (5.2)1.7 (5.1) log PTR ) log D = 2.4 (5.02) 1 1.1 (5.04) log NPP ) log RA = 2.7 (5.3) 2 3.4 (5.4) elog PTR ) log B = 3.3 (5.2) 2 1.6 (5.1) log PPC log B = 2.3 (5.1) 2 1.0 (5.03) log PTR log B = 2.1 (5.1) 1 1.1 (5.1) log NPP log DM = 2.1 (5.2) 2 .9 (5.1) log k log DM = .3 (5.2) 2 .7 (5.1) log PTR ) )

91

.78

321.2

!.00001

73

.63

124.2

!.00001

369 91 73 125

0 0 0 .50

.4 .4 1.2 122.6

.52 .55 .28 !.00001

135 135

.10 .53

9.0 153.7

!.00001

125 156

.17 .88

27.0 1161.9

!.00001 !.00001

59

.54

69.1

!.00001

59 215

0 .79

1.1 806.2

!.00001

126 98

.02 .52

115 125

.28 .49

45.1 118.3

!.00001 !.00001

369

.75

1130.2

!.00001

369

.27

138.7

!.00001

52

.7

119.8

!.00001

54

.56

68.5

!.00001

54 53

.12 .11

8.4 7.3

.005 .009

3.6 .79*

.003

.30

.06 !.00001

Note: PTR, plant turnover rate; N, plant nitrogen concentration; P, plant phosphorus concentration; NPP, net primary production; PPC, percentage of primary production consumed by herbivores; C, absolute consumption by herbivores; DP, plant detrital production; k, decomposition rate; D, absolute decomposition; RA, accumulation of refractory plant detritus; B, plant biomass; DM, plant detrital mass. Asterisk signifies x2 statistic.

communities, absolute refractory accumulation was better related to turnover rate than to primary production (figs. 9A, 9B; table 1; Z-test, H0: equality between correlation coefficients, P ! .05). Export, expressed both in absolute terms and as percentage of primary production, varied widely among communities (fig. 10A, 10B) and tended to be largest in macroalgal beds (Tukey HSD test, P ! .05). Plant biomass increased from microalgal communities

to communities of aquatic macrophytes to communities of terrestrial macrophytes (fig. 10C; Tukey HSD test, P ! .05). This distribution counters the distributions of turnover and percentage of production consumed across communities. Accordingly, communities with faster turnover rates and higher percentages of production consumed store smaller pools of biomass (fig. 11A, 11B; table 1). In contrast, biomass is only weakly related to primary pro-

Fate of Production in Plant Communities 457

Figure 4: The relationships (A) between net primary production and nitrogen concentration, (B) between net primary production and phosphorus concentration, and (C) between turnover rate and primary production across the communities compiled. Symbols as in figure 3.

duction across communities (table 1; Z-test, H0: equality between correlation coefficients, P ! .05). Similarly, plant detrital mass seems to increase from microalgal communities to communities of aquatic macrophytes to communities of terrestrial macrophytes (fig. 10D; ANOVA, P ! .05). Across communities, plant detrital mass is associated with turnover and decomposition rates, with a tendency for communities with faster turnover and decomposition rates to store smaller pools of detrital mass (fig. 12A, 12B; table 1). Conversely, detrital mass is very poorly related to primary production and detrital production (table 1). Estimates of detrital mass included only

leaf detritus in some of the forests and shrublands compiled (see data set). Therefore, I used the values of production, turnover rate, decomposition rate, and detrital production for the foliar compartments in those communities when deriving the relationships between detrital mass and primary production, biomass turnover rates, decomposition rates, and detrital production for all communities. Discussion These results allow a synthetic view of the fate of production in plant communities. With the exception of

458 The American Naturalist

Figure 5: The relationships (A) between percentage of net primary production consumed by herbivores and plant turnover rate, (B) between absolute consumption by herbivores and net primary production, (C) between percentage of net primary production consumed by herbivores and net primary production, and (D) between absolute consumption by herbivores and plant turnover rate across the communities compiled. Lines represent the equations fitted. Symbols as in figure 3.

microalgal communities, most production in all the communities considered is channeled as detrital production (fig. 6A). That generalizes the predominance of the detrital pathway in the trophic transfer of primary production to heterotrophs, in agreement with previous results in different communities (Schlesinger 1977; Thayer et al. 1984; Mann 1988; Cebrian and Duarte 1998). In general, most detrital production is decomposed within the community, with only modest fractions being exported even in communities typically exposed to substantial horizontal advection (fig. 10B). Macroalgal beds represent the only ex-

ception to this pattern, which, on the average, export about 50% of their production. This is because macroalgal beds often inhabit rocky shores exposed to intense wave scouring (Mann 1972; Marsden 1991). In general, only a small fraction of production is accumulated as refractory detritus within the community (fig. 6F), although this fraction may be sizable in grasslands, shrublands, and forests (Smith and Clymo 1984). In spite of these generalities, my results reveal important qualitative and quantitative differences in the fate of production among plant communities. These differences are

Fate of Production in Plant Communities 459

Figure 6: Box plots showing the distribution of (A) percentage of primary production channelled as detritus, (B) plant detrital production, (C) decomposition rates, (D) absolute decomposition, (E) accumulation of refractory detritus, and (F) percentage of primary production accumulated as refractory detritus across the types of communities considered. Boxes encompass the 25% and 75% quartiles, and the central line represents the median. Bars encompass the range of values between the 25% quartile minus 1.5 times the difference between the quartiles 75% and 25% and the 75% quartile plus 1.5 times the difference between the quartiles 75% and 25%. Circles and asterisks represent values outside these limits.

460 The American Naturalist

Figure 7: Relationship between plant detrital production and net primary production across the communities compiled. Line represent the equation fitted. Symbols as in figure 3.

related to differences either in the magnitude of production or in the quality of production for consumers (i.e., N and P concentrations in plant tissues). Across plant communities, the quality of plant tissues for consumers is closely associated with plant turnover rate, but it is independent of the magnitude of production (figs. 3 and 4; table 1). I identify a tendency for plant communities with a higher quality for consumers (i.e., faster plant nutrient concentrations and turnover rates) to lose a higher percentage of production to herbivores (fig. 5A). This tendency supports the hypothesis that the metabolism and feeding rates of herbivores are limited by the nutrient concentrations of their diets (Mattson 1980; Sterner and Hessen 1994; Hartley and Jones 1997; Sterner et al. 1997): plant tissues with higher nutrient concentrations enhance herbivore metabolism and feeding rates and, as a consequence, communities composed of plants with higher N and P concentrations have a higher percentage of production removed by herbivores. Greater percentages of production consumed in communities with more palatable plants should not be due to larger herbivore abundances. This is because across aquatic and terrestrial communities primary production is related to herbivore abundance (McNaughton et al. 1989; Cyr and Pace 1993), but independent of plant nutrient concentration (fig. 4A, 4B), and therefore plant nutrient concentration and herbivore abundance should be unrelated across communities. The tendency for communities composed of plants with higher nutrient concentrations to lose higher percentages

of production to herbivores implies that these communities channel smaller percentages of production as detritus (fig. 6A). Across communities, the importance of herbivory as a trophic route of primary production increases with plant palatability, as it has been previously hypothesized (Sterner et al. 1997). Moreover, this tendency supports the notion that across the range of communities compared, herbivores do not increase their feeding rates over plants with lower nutrient concentrations as a means of compensating for lower nutritional qualities. This mechanism has been hypothesized as implausible in an evolutionary sense because higher feeding rates on less nutritional, slower turning-over plants would eventually lead to plant depletion (Sterner and Hessen 1994). Yet, enhanced feeding rates as a herbivore response to lower nutritional quality have been found during transient periods within some particular plant communities (Williams et al. 1994; Lindroth et al. 1995; Hughes and Bazzaz 1997). The dependence of the intensity of herbivory (i.e., fraction of production consumed) on plant nutritional quality across communities overrides the effect of communityspecific herbivore characteristics. For instance, some of the communities compared host mainly migratory herbivores, whereas others have only resident herbivores. This migratory-resident duality may introduce considerable variability in the intensity of herbivory among communities (Nienhuis and Groenendijk 1986; Portig et al. 1994). In addition, some communities shelter mainly invertebrate herbivores (e.g., most submerged communities), whereas others host vertebrate homeotherm herbivores. These two types of herbivores have different metabolical needs, and these differences may cause substantial variability in the intensity of herbivory across communities (Crawley 1983; Begon et al. 1990). Moreover, the intensity of predation on herbivores also varies among communities, which may entail different cascade effects on the intensity of herbivory in different communities (Heck and Valentine 1995; Hartley and Jones 1997). Finally, differences in herbivore stoichiometry among communities may also be responsible for differences in herbivory. Although communities of more nutritional plants should have herbivores with higher nutrient contents, the range of nutrient concentrations in herbivore tissues may vary notably within a given community (Sterner and Hessen 1994). Because herbivores with higher nutrient concentrations also have higher nutritional demands, variations in the range of herbivore stoichiometry among communities could also generate differences in the intensity of herbivory. All these factors could perhaps explain the 11 order-of-magnitude variability observed in herbivory intensity among communities with similar turnover rates (fig. 5A). The tendency toward higher intensities of herbivory with higher plant palatability across communities suggests

Fate of Production in Plant Communities 461

Figure 8: The relationships (A) between decomposition rate and plant turnover rate, (B) between absolute decomposition and net primary production, (C) between decomposition rate and net primary production, and (D) between absolute decomposition and plant turnover rate across the communities compiled. Lines represent the equations fitted. Symbols as in figure 3.

that herbivores should exert a greater control of plant biomass in communities with more palatable plants. Accordingly, herbivores have been shown to limit plant biomass in communities with fast turnover rates (Lampert et al. 1986; McNaughton et al. 1988; Jing and Coley 1990), whereas they seem unimportant in communities with slow turnover rates (Bray 1964; Petrusewicz and Grodzinski 1975; Cebrian et al. 1996). My results suggest that, at least in part, communities with more palatable plants store smaller pools of biomass as a result of the higher intensities of herbivory supported: communities with faster turnover rates store smaller pools of biomass partially because they

reach relatively similar values of production (fig. 4C) but lose higher percentages to herbivores (fig. 5A). This hypothesis is supported by the tendencies toward lower biomasses with faster turnover rates and larger intensities of herbivory across communities (fig. 11A, 11B). Yet communities with faster turnover rates also have higher plant mortality rates (Odum 1971; Phillips and Gentry 1994), which can also contribute to the decline in plant biomass observed with faster turnover rates across communities. I used techniques of path analysis to differentiate the direct connection between faster turnover and decreasing biomass from the connection mediated

462 The American Naturalist

Figure 9: The relationships (A) between accumulation of refractory plant detritus and turnover rate and (B) between accumulation of refractory plant detritus and net primary production across the communities compiled. Lines represent the equations fitted. Symbols as in figure 3.

through increasing herbivory. These techniques allowed the categorization of variance in plant biomass across communities into the fraction directly explained by turnover, the fraction directly explained by herbivory, and the fraction explained by the interaction of both herbivory and turnover (Williams et al. 1990). The variance in plant biomass among communities explained by turnover is significantly reduced, albeit still significant (P ! .01), when the interaction with herbivory is eliminated (from R 2 = .69 to R 2 = .44). This suggests that the reduction in biomass with higher palatability (i.e., faster turnover) across communities is a partial effect of increasing herbivory. Furthermore, I used a randomization test (Pendleton et al. 1983) to test for spurious effects on the relationship between decreasing biomass and increasing turnover due to the presence of a biomass term in both variables (turnover rate was estimated as the quotient between production and biomass in many communities). The result demonstrates that the relationship obtained is different from that expected solely from the effect of a common term in both variables (H0: spurious slope = observed slope, P ! .001). Detritus from communities composed of more nutritional plants undergo faster decomposition (fig. 8A). This indicates that plant communities that turn over their biomass compartment faster also turn over their degradable detrital compartment faster. This implies that communities with faster biomass turnover rates have higher nutrient recycling rates, as expected to meet their faster rates of nutrient uptake (i.e., higher nutrient demand;

Chapin et al. 1986, 1987). This trend has been hypothesized recently (Sterner et al. 1997), and these results further support it. Accordingly, fast and efficient nutrient recycling generally plays a pivotal role in the maintenance of communities with fast biomass turnover rates and high nutrient demands (Crossland et al. 1991; Legendre and Rassoulzadegan 1995), whereas communities with slower biomass turnover rates normally exhibit slow recycling rates and low nutrient demands (Enriquez et al. 1993). The tendency toward faster decomposition rates with higher plant nutrient concentrations across communities is consistent with the idea that the metabolism and feeding rates of decomposers are limited by the nutrient concentrations in plant detritus (Enriquez et al. 1993). This limitation has been shown experimentally for a number of types of autotrophs (Swift et al. 1979; Goldman et al. 1987; Enriquez et al. 1993), and here it is suggested to exist across a broad range of plant communities. The tendency toward faster decomposition rates with higher plant nutrient concentrations overrides the effects on decomposition of different abiotic factors in the communities compared, such as temperature (Edwards 1975; Van Cleve et al. 1981), moisture (Wildung et al. 1977; Santos et al. 1984), and sediment redox conditions (Schlesinger 1977; Post et al. 1982). These factors could probably explain much of the residual variability in the tendency established. Moreover, these results indicate that communities with more palatable detritus store smaller detrital pools as a consequence of the faster decomposition rates supported (fig. 12A): communities with faster biomass turnover rates

Fate of Production in Plant Communities 463

Figure 10: Box plots showing the distribution of (A) export, (B) percentage of primary production exported, (C) plant biomass, and (D) detrital mass across the types of community considered. Boxes encompass the 25% and 75% quartiles, the central line represents the median, and bars encompass the range of values between the 25% quartile minus 1.5 times the difference between the quartiles 75% and 25% and the 75% quartile plus 1.5 times the difference between the quartiles 75% and 25%. Circles and asterisks represents values outside these limits.

store smaller pools of detrital mass (fig. 12B) because they reach relatively similar values of production and detrital production (fig. 4C) but have faster decomposition rates (fig. 8A). Another corollary of these tendencies is that communities composed of more palatable plants show lower levels of refractory accumulation (fig. 9A). Refractory accumulation corresponds to the amount of primary production nonconsumed by herbivores that is neither exported nor decomposed over the study period. Because both the percentage of production consumed and decomposition rates increase nonlinearly with turnover rate across communities (figs. 5A and 8A, respectively) while

being poorly related to primary production (figs. 5C and 8C), the decrease in refractory accumulation with faster turnover rates is exponential (fig. 9A). These results indicate that communities composed of more palatable plants are smaller carbon sinks because they experience higher intensities of consumption by heterotrophs, as it has been recently suggested (Cebrian et al. 1998). So far I have discussed the dependence of intensity of herbivory and decomposition rates on plant palatability across communities. Absolute consumption by herbivores, however, is very poorly related to plant palatability across communities (fig. 5D). Instead, differences in absolute

464 The American Naturalist

Figure 11: The relationships (A) between plant biomass and the percentage of net primary production consumed by herbivores and (B) between plant biomass and turnover rate across the communities compiled. Lines represent the equations fitted. Symbols as in figure 3.

consumption across communities are better related to differences in absolute production (fig. 5B), in agreement with results of other across-systems comparisons (McNaugton et al. 1989; Cyr and Pace 1993). The association between absolute consumption and production across communities is a mathematical consequence of the ranges of the variables compared: absolute consumption is the product between primary production and the percentage consumed by herbivores and, because primary production and the percentage consumed vary by four and three orders of magnitude across the communities compared, respectively (fig. 5C), differences in absolute consumption across communities are better correlated to primary production than to the percentage consumed. As a consequence, when a broad range of communities is compared, more productive communities tend to transfer greater fluxes of production to herbivores (fig. 5B). McNaugton et al. (1989) and Cyr and Pace (1993) also show that higher absolute consumptions in more productive communities are conducive to larger herbivore abundances. Therefore, differences in herbivore biomass across communities seem to be associated with differences in the magnitude of production rather than with the quality (i.e., nutrient concentration in plant tissues) of that production. These tendencies, if confirmed, would indicate that, across communities, increases in herbivore biomass due to larger absolute consumptions override any possible increases due to enhanced growth rates resulting from more palatable diets. Higher predation rates on herbivores feeding on more palatable plants could perhaps partially explain these

hypothesized trends. Similarly, absolute decomposition is closely related to production across communities (fig. 8B), and it is independent of plant palatability (fig. 8D). Again, the reason depends on the relative range of the variables compared across communities. At the time scales covered by the studies compiled (i.e., from one to some few years), the variability in production across communities is greater than the variability in the percentage decomposed, and because decomposition is the product of production and percentage decomposed, its variability across communities is dictated by variability in production. Differences in export, both as absolute magnitude and as percentage of production, across communities are independent of differences in plant palatability and are only weakly related to differences in production (least square regression of log export versus log production: R 2 = .23, P ! .05). Differences in export across communities may be related to differences in the intensity of horizontal advection through the community. This is supported by the fact that communities with higher export rates are typically exposed to substantial horizontal advection through waves, currents or tides, such macroalgal beds, seagrass meadows, and land-ocean fringing communities such as marshes and mangroves (fig. 10A, 10B). Communities with high export rates have the potential to support large levels of secondary production in adjacent communities (Nixon 1980; Twilley et al. 1986; Duarte and Cebrian 1996). Moreover, high export rates must imply substantial nutrient losses for the exporting communities since virtually all of the exported detritus are consumed and recycled elsewhere (Hedges and

Fate of Production in Plant Communities 465

Figure 12: The relationships (A) between plant detrital mass and decomposition rate and (B) between plant detrital mass and turnover rate across the communities compiled. Lines represent the equations fitted. Symbols as in figure 3.

Parker 1976; Smith and MacKenzie 1987). Hence, communities with high export rates must have high rates of nutrient import from adjacent communities to offset nutrient losses and maintain primary production. This argument suggests important implications for the interplay between “new” (i.e., fueled by allochthonous nutrients) and recycled production in plant communities: communities with higher percentages of production exported should have greater fractions of their production fueled by imported nutrients (i.e., higher factions of new production). In summary, plant nutritional quality for consumers and primary production are independent predictors of the nature and magnitude of the trophic routes of carbon flow in plant communities. Communities composed of more palatable plants lose higher percentages of production to herbivores, channel lower percentages as detritus, and experience faster decomposition rates. I present evidence that communities with more palatable plants store smaller carbon pools as a result of the higher intensities of herbivory and faster decomposition rates supported, but more experimental work is needed to confirm this hypothesis. These results may have important implications for broadscale carbon and nutrient balances because widespread anthropogenic eutrophication may enhance the nutritional quality of plant communities in large areas (Duarte 1995; Borum and Sand-Jensen 1996; Wedin and Tilman 1996). Absolute consumption and decomposition, however, are not associated with plant nutritional quality but with pri-

mary production across communities. The patterns in the nature and controls of the trophic fate of production reported here encompass contrasting communities ranging from unicellular autotrophs to structurally complex macrophytes. When the range of communities compared is reduced, factors other than plant nutritional quality and magnitude of production may be most important at explaining differences in the trophic fate of production. Hence, the dependence of the patterns reported on the range of communities encompassed needs investigation. Acknowledgments Research was supported by a postdoctoral fellowship from the Culture Ministry of the Spanish government. I thank M. Cole, M. Griffin, C. Jones, K. Kroeger, E. Stieve, E. Watson, and M. Williams for enlightening discussions on the ideas presented. M. Williams also provided valuable statistical help. I also thank R. Sterner and two anonymous referees for helpful comments on the manuscript. I am deeply indebted to the staff of the Marine Biological Laboratory/Woods Hole Oceanographic Institution Library for providing help with the literature research. Literature Cited Begon, M., J. L. Harper, and C. R. Townsend. 1990. Ecology. 2d ed. Omega, Oxford. Borum, J., and K. Sand-Jensen. 1996. Is total primary pro-

466 The American Naturalist duction in shallow coastal marine waters stimulated by nitrogen loading? Oikos 76:406–410. Bray, J. R. 1964. Primary consumption in three forest canopies. Ecology 45:165–167. Cebrian, J., and C. M. Duarte. 1994. The dependence of herbivory on growth rate in natural plant communities. Functional Ecology 8:518–525. ———. 1995. Plant growth-rate dependence of detrital carbon storage in ecosystems. Science (Washington, D.C.) 268:1606–1608. ———. 1998. Patterns in leaf herbivory on seagrasses. Aquatic Botany 60:67–82. Cebrian, J., C. M. Duarte, N. Marba, S. Enriquez, M. Gallegos, and B. Olesen. 1996. Herbivory on P. oceanica (L.) Delile: magnitude and variability in the Spanish Mediterranean. Marine Ecology Progress Series 130: 147–155. Cebrian, J., M. Williams, J. McClelland, and I. Valiela. 1998. The dependence of heterotrophic consumption and carbon accumulation on autotrophic content in ecosystems. Ecology Letters 1:165–170. Chapin, F. S., III, P. M. Vitousek, and K. Van Cleve. 1986. The nature of nutrient limitation in plant communties. American Naturalist 127:48–58. Chapin, F. S., III, A. J. Bloom, C. B. Field, and R. H. Waring. 1987. Plant responses to multiple environmental factors. BioScience 37:49–57. Crawley, M. J. 1983. Herbivory: the dynamics of animalplant interactions. Blackwell Scientific, Los Angeles. Crossland, C. J., B. G. Hatcher, and S. V. Smith. 1991. Role of coral reefs in global ocean production. Coral Reefs 10:55–64. Cyr, H., and M. L. Pace. 1993. Magnitude and patterns of herbivory in aquatic and terrestrial ecosystems. Nature (London) 361:148–150. Duarte, C. M. 1995. Submerged aquatic vegetation in relation to different nutrient regimes. Ophelia 41:87–112. Duarte, C. M., and J. Cebrian. 1996. The fate of marine autotrophic production. Limnology and Oceanography 41:1758–1766. Edwards, N. T. 1975. Effects of temperature and moisture on carbon dioxide evolution in a mixed deciduous forest floor. Soil Science Society of America Proceedings 39: 361–365. Enriquez, S., C. M. Duarte, and K. Sand-Jensen. 1993. Patterns in decomposition rates among photosynthetic organisms: the importance of detritus C : N : P content. Oecologia (Berlin) 94:457–471. Fischer, R. A. 1921. On the “probable error” of a coefficient of correlation deduced from a small sample. Metron 1: 3–32. Goldman, J. C., D. A. Caron, and M. R. Dennett. 1987. Regulation of gross growth efficiency and ammonium

regeneration in bacteria by substrate C : N ratio. Limnology Oceanography 32:1239–1252. Grier, C. C., and R. S. Logan. 1977. Old-growth Pseudotsuga menziesii communities of a western Oregon watershed: biomass distributions and production budgets. Ecological Monographs 47:373–400. Hartley, S. E., and C. E. Jones. 1997. Plant chemistry and herbivory, or, why the world is green. Pages 284–324 in M. J. Crawley, ed. Plant Ecology. 2d ed. Blackwell, Oxford. Heck, K. L., and J. F. Valentine. 1995. Sea urchin herbivory: evidence for long-lasting effects in subtropical seagrass meadows. Journal of Experimental Marine Biology and Ecology 189:205–217. Hecky, R. E., P. Campbell, and L. L. Hendzel. 1993. The stoichiometry of carbon, nitrogen and phosphorus in particulate matter. Limnology and Oceanography 38: 709–724. Hedges, J. I., and P. L. Parker. 1976. Land-derived organic matter in surface sediments from the Gulf of Mexico. Geochimica et Cosmochimica Acta 40:1019–1029. Hughes, L., and F. A. Bazzaz. 1997. Effect of elevated CO2 on interactions between the western flower thrips, Frankliniella occidentalis (Thysanoptera: Thripidae) and the common milkweed, Asclepias syriaca. Oecologia (Berlin) 109:286–290. Jing, S. W., and P. D. Coley. 1990. Diocey and herbivory: the effect of growth rate on plant defence in Acer negundo. Oikos 58:369–377. Kilham, P., and R. E. Hecky. 1988. Comparative ecology of marine and freshwater phytoplankton. Limnology and Oceanography 33:776–795. Lampert, W., W. Fleckner, R. Hakumat, and B. E. Taylor. 1986. Phytoplankton control by grazing zooplankton. A study on the spring clear-water phase. Limnology and Oceanography 31:478–490. Legendre, L., and F. Rassoulzadegan. 1995. Plankton and nutrient dynamics in marine waters. Ophelia 41: 153–170. Lindroth, R. L., G. E. Arteel, and K. K. Kinney. 1995. Responses of three saturniid species to paper birch grown under enriched CO2 atmospheres. Functional Ecology 9:306–311. Mann, K. H. 1972. Ecological energetics of the sea-weed zone in a marine bay on the Atlantic coast of Canada. II. Productivity of seaweeds. Marine Biology 14: 199–209. ———. 1988. Production and use of detritus in various freshwater, estuarine, and coastal marine ecosytems. Limnology and Oceanography 33:910–930. Marsden, I. 1991. Kelp-sandhopper interactions on a sandbeach in New Zealand. I. Drift composition and dis-

Fate of Production in Plant Communities 467 tribution. Journal of Experimental Marine Biology and Ecology 152:49–62. Martin, J. H., G. A. Knauer, D. M. Karl, and W. W. Broenkow. 1987. VERTEX: Carbon cycling in the northeast Pacific. Deep-Sea Research 34:267–285. Mattson, W. J. 1980. Herbivory in relation to plant nitrogen content. Annual Reviews of Ecology and Systematics 11:119–161. McNaughton, S. J., R. W. Ruess, and S.W. Seagle. 1988. Large mammals and process dynamics in African ecosystems. BioScience 38:794–800. McNaughton, S. J., M. Oesterheld, D. A. Frank, and K. J. Williams. 1989. Ecosystem-level patterns of primary productivity and herbivory in terrestrial habitats. Nature (London) 341:142–144. Muller, P. J. and E. Suess. 1979. Productivity, sedimentation rate, and sedimentary organic matter in the oceans. I. Organic carbon preservation. Deep-Sea Research 26A: 1347–1362. Nielsen, S. L., S. Enrı´quez, C. M. Duarte, and K. SandJensen. 1996. Scaling maximum growth rates among photosynthetic organisms. Functional Ecology 10: 1676–175. Nienhuis, P. H., and A. M. Groenendijk. 1986. Consumption of eel grass (Zostera marina) by birds and invertebrates: an annual budget. Marine Ecology Progress Series 29:29–35. Nixon, S. W. 1980. Between coastal marshes and coastal waters: a review of twenty years of speculation and research on the role of salt marshes in estuarine productivity and water chemistry. Pages 437–525 in P. Hamilton and K. B. MacDonald, eds. Estuarine and wetland processes. Plenum, New York. Odum, E. P. 1971. Fundamentals of ecology. Saunders, Philadelphia. Olson, J. S. 1963. Energy storage and the balance of producers and decomposers in ecological systems. Ecology 44:322–331. Pendleton, B. F., I. Newman, and R. S. Marshall. 1983. A Monte Carlo approach to correlational spuriousness and ratio variables. Journal of Statistics and Computation Simulation 18:93–124. Petrusewicz, K., and W. Grodzinski. 1975. The role of herbivore consumers in various ecosystems. Pages 64–70 in D. E. Reichle, J. F. Franklin, and D. W. Goodwall, eds. Productivity of world ecosystems. National Academy of Science, Warsaw. Phillips, O. L., and A. H. Gentry. 1994. Increasing turnover through time in tropical forests. Science (Washington, D.C.) 263:954–958. Pomeroy, L. R., and W. J. Wiebe. 1993. Energy sources for microbial food webs. Marine Microbial Food Webs 7: 101–108.

Portig, A. A., R. G. Mathers, W. I. Montgomery, and R. N. Govier. 1994. The distribution and utilisation of Zostera species in Strangford Lough, Northern Ireland. Aquatic Botany 47:317–328. Post, W. M., W. R. Emanuel, P. J. Zinke, and A. G. Stangenberger. 1982. Soil carbon pools and world life zones. Nature (London) 298:156–159. Santos, P. F., N. Z. Elkins, Y. Steinberger, and W. G. Whitford. 1984. A comparison of surface and buried Larrea tridentata leaf litter decomposition in North American hot deserts. Ecology 65:278–284. Schlesinger, W. H. 1977. Carbon balance in terrestrial detritus. Annual Review of Ecology and Systematics 8: 51–81. ———. 1997. Biogeochemistry: an analysis of global change. 2d ed. Academic Press, New York. Smith, L., and R. S. Clymo. 1984. An extraordinary peatforming community on the Falkland Islands. Nature (London) 309:617–620. Smith, S. V., and F. T. MacKenzie. 1987. The ocean as a net heterotrophic system: implications from the carbon biogeochemical cycle. Global Biogeochemical Cycles 1: 187–198. Sterner, R. W., and D. O. Hessen. 1994. Algal nutrient limitation and the nutrition of aquatic herbivores. Annual Review of Ecology and Systematics. 25:1–29. Sterner, R. W., J. J. Elser, E. J. Fee, S. J. Guildford, and T. H. Chrzanowski. 1997. The light : nutrient ratio in lakes: the balance of energy and materials affects ecosystem structure and process. American Naturalist. 150: 663–684. Suess, E. 1980. Particulate organic carbon flux in the oceans-surface productivity and oxygen utilization. Nature (London) 288:260–263. Swift, M. J., O. W. Heal, and J. M. Anderson. 1979. Decomposition in terrestrial ecosystems. Vol. 5. Studies in Ecology. Blackwell, Oxford. Thayer, G. W., K. A. Bjorndal, J. C. Ogden, S. L. Williams, and J. C. Zieman. 1984. Role of larger herbivores in seagrass communities. Estuaries 7:351–376. Twilley, R., A. Lugo, and C. Pattersen-Zuca. 1986. Litter production and turnover in basin mangrove forests in southwest Florida. Ecology 67:670–683. Van Cleve, K., R. Barney, and R. Schlentner. 1981. Evidence of temperature control of production and nutrient cycling in two interior Alaska black spruce ecosystems. Canadian Journal of Forest Research 11:258–273. Wedin, D. A., and D. Tilman. 1996. Influence of nitrogen loading and species composition on the carbon balance of grasslands. Science (Washington, D.C.) 274: 1720–1723. Wetzel, R. G. 1984. Detrital dissolved and particulate or-

468 The American Naturalist ganic carbon functions in aquatic ecosystems. Bulletin of Marine Science 35:503–509. Wiebe, P. H. 1988. Functional regression equations for zooplankton displacement volume, wet weight, dry weight, and carbon: a correction. Fisheries Bulletin 86: 833–835. Wildung, R. E., T. R. Garland, and R. L. Buschbom. 1977. The interdependent effects of soil temperature and water content on soil respiration rate and plant root decomposition in arid grassland soils. Soil Biology and Biochemistry 7:373–378.

Williams, R. S., D. E. Lincoln, and R. B. Thomas. 1994. Loblolly pine grown under elevated CO2 affects early instar pine sawfly performance. Oecologia (Berlin) 98: 64–71. Williams, W. A., M. B. Jones, and M. W. Demment. 1990. A concise table for path analysis statistics. Agronomy Journal 82:1022–1024. Zar, J. H. 1984. Biostatistical analysis. Prentice Hall, Upper Saddle River, N.J. Associate Editors: Robert W. Sterner Joel G. Kingsolver

Suggest Documents