August 4, 2007

Competitive Hierarchy for Four Common Old-Field Plant Species Depends on Resource Identity and Availability1

By Fortner, Allison M Weltzin, Jake F

FORTNER, A. M. (Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6351) AND J. F. WELTZIN (Department of Ecology and Evolutionary Biology, University of Tennessee, Knoxville, TN 37996-1610). Competitive hierarchy for four common old-field plant species depends on resource identity and availability. J. Torrey Bot. Soc. 134: 166-176. 2007.-Competition is an important factor structuring plant communities. Competitive hierarchies may facilitate prediction of the outcome of interactions among multi-species mixtures, but their utility depends upon invariance along environmental gradients. We examined interactions among four common old-field herbaceous dicots (Dactylis glomerata, Festuca elatior, Trifolium pratense, Plantago lanceolata) with pairwise monoculture and polyculture competition experiments at two levels each of light and water. We constructed competitive hierarchies for unique treatment combinations based on competitive effect and response scores. Biomass production and relative yield were affected by presence and identity of neighbor species and inter- and intraspecific competition were typically not distinguishable. Competitive effect and response scores were inversely correlated, but depended on treatment factor as well as levels within factors. Competitive rankings among these four common plant species varied along resource gradients, although the extreme ends of the hierarchy were relatively stable: under most conditions P. lanceolata was most competitive whereas F. elatior was least competitive. Results lay the groundwork for assessments of multispecies mixtures under field conditions along broader environmental gradients. Key words: competitive hierarchy, Dactylis glomerata, Festuca elatior, intraspecific competition, interspecific competition, old-field, Plantago lanceolata, Trifolium pratense.

Competition has been shown to be an important factor structuring plant communities (Weaver and Clements 1929, Harper 1977, Goldberg and Barton 1992, Keddy 2001). Resources important to the performance of individual plants (e.g., light, water, CO2, nutrients) are typically in limited supply (Tilman 1982); thus, intraspecific and interspecific interactions among neighboring plants can affect population- and community-level parameters of density, abundance, growth, and reproductive strategies (Connell 1983, Schoener 1983, Goldberg and Barton 1992, Greiling and Kichanan 2002, Suding et al. 2004). Although much of our current understanding of competition is derived from studies of pairwise interactions, there is growing evidence that competitive hierarchies, which consider multiple species simultaneously, may facilitate prediction of the outcome of interactions among multi-species mixtures (Keddy and Shipley 1989, Shipley and Keddy 1994, Keddy 2001).

Competitive hierarchies reflect a ranking of species-specific performance within communities from strongest to weakest competitor (Mitchley and Grubb 1986, Keddy and Shipley 1989, Goldberg and Landa 1991, Keddy 2001). Strong competitors have a great impact on performance of neighbor species, but are themselves little affected by the presence of those neighbors; in contrast, weak competitors have little effect on neighbors, but may respond strongly to the presence of those neighbors (Goldberg and Werner 1983, Miller and Werner 1987, Goldberg and Fleetwood 1987, Goldberg and Barton 1992, Liancourt et al. 2005a). Generally, competitive hierarchies are developed from empirical studies that investigate interactions among pairs of potentially interacting species simultaneously (e.g., Wilson and Keddy 1986, Mitchley and Grubb 1986, Barthram et al. 2002, Monzeglio and Stoll 2005). Competitive hierarchies have the potential to explain the organization of natural communities because positions of species in a hierarchy are often correlated to their relative abundance in the field (Shipley and Keddy 1994); however, few studies have specifically investigated this relationship (but see Canham et al. 2006).

Competitive hierarchies rest on the oft-demonstrated assumption that pairwise interactions are predominantly asymmetric, and that resultant competitive networks are primarily transitive (Keddy and Shipley 1989, Shipley 1993, Barthram et al. 2002, Sher et al. 2004, Fynn et al. 2005, Canham et al. 2006). Moreover, it has been argued that competitive hierarchies are invariant along environmental gradients, i.e., the relative ranking of competitive dominance is consistent at various levels of resources (Shipley et al. 1991, Shipley and Keddy 1994, Keddy et al. 2002, Liancourt et al. 2005b), although there has been considerable disagreement over the importance of relative competitive importance along resource gradients (e.g., Grime 1979, Grubb 1985, Tilman 1988, Silvertown and Dale 1991, Gurevitch et al. 1992, Keddy et al. 2002, Maestre et al. 2005).

Old-fields, or lands previously used for agriculture and now abandoned, are a common feature at the interface of managed and unmanaged systems throughout the world. North American old-fields are generally well-studied, as they represent early stages of secondary succession. Old-fields exhibit obvious and rapid changes in population and community structure and ecosystem function, including plant and animal species diversity, net primary production and standing biomass, and nutrient accumulation (e.g., Keever 1950, Beckwith 1954, Egler 1954, Odum 1960, Bazzaz 1968, Dayton 1975). Rates and patterns of succession in old-fields are driven by interactions between component taxa, which operate against a backdrop of intrinsic and extrinsic factors such as geological substrate, seed dispersal, microclimate and macroclimate, and perturbations (Barbour et al. 1999). Because of these features, as well as their tractability for experimental manipulation, species interactions within old-fields and among old-field species are well- documented (Goldberg and Fleetwood 1987, Goldberg 1987, Gurevitch et al. 1990, Gerry and Wilson 1995, Kosola and Gross 1999, Keddy et al. 2002, Schmitz 2003, Gibson et al. 2005, Schmitz et al. 2006).

The goal of this research was to investigate intra- and interspecific interactions, and the stability of competitive hierarchies along resource gradients, among four plant species common to old-fields in the eastern United States. We conducted a greenhouse study in which the four species were grown in pairwise monoculture and polyculture competition experiments along physical gradients of light and water. Our objectives were to (1) determine how the availability of light and water affect intraspecific and interspecific interactions among the four species in terms of growth and production, (2) determine the effect of initial plant size on competitive interactions in terms of relative yield, and (3) determine whether interactions among these species formed competitive hierarchies consistent across physical gradients of light and water. Our null predictions were that intensity of competition would be consistent within and among species, competition intensity would not vary across gradients of light and water, initial plant size would not affect competitive outcomes, and competitive hierarchies would be invariant along environmental gradients.

Materials and Methods. STUDY SPECIES. We selected four herbaceous plant species common to old-field plant communities of the eastern United States: two grasses, Dactylis glomerata L. (orchard grass) and Festuca elatior L. (fescue), and two herbaceous dicots, Plantago lanceolata L. (ribgrass) and Trifolium pratense L. (red clover). All four species (generic names are used hereafter for simplicity) were introduced to the United States, are typical dominants in early- successional abandoned agricultural fields or pastures, and are likely to interact within these communities (Keever 1950).

EXPERIMENTAL DESIGN. We conducted a competition experiment in a greenhouse in which the species described above were grown in pairwise monoculture and polyculture combinations along gradients of light and water. Plants were germinated in March 2003 in a greenhouse at the University of Tennessee, Knoxville. Planting treatments consisted of each species grown alone (i.e., one plant per pot; n = 80 plants) and in monoculture (i.e., two plants per pot; n = 160 plants) to assess the potential for intraspecific competition, and in four polyculture combinations to assess the potential for interspecific competition: Dactylis-Plantago, Dactylis- Trifolium, Festuca- Plantago, Festuca-Trifolium (i.e., two plants per pot; n = 160 plants). Polyculture treatments were selected to represent a subset of all possible combinations of plant species because of space limitations.

Prior to planting, we prepared standard polyethylene pots (15 cm diameter, 2.5 L) by drilling two holes in the side of each, at a depth of approximately 5 cm, for insertion of a probe to measure soil moisture; pots were filled with a 1:1 soil mixture composed of sterilized potting medium (Pro-mix) and silt-loam soil collected from an old-field near Oak Ridge, TN. Field soil was collected from a depth of 5-20 cm to minimize unwanted seeds and was sieved through 1 cm x 1 cm wire mesh prior to mixing with Pro-mix and filling pots. On 1 June 2003 we transplanted seedlings into the soil-filled pots at a spacing of about 5 cm. Seedlings were relatively small, and were selected to be relatively homogeneous in size. Upon transplanting, we recorded the number of leaves and the length of the longest shoot for all plants; mean +- SE length of longest shoots for Dactylis and Festuca (both of which had few leaves) were 15.7 cm +- 0.31 cm and 6.0 cm +- 0.10 cm, respectively, whereas mean +- SE number of leaves (of variable length) for Plantago and Trifolium were 3.9 +- 0.12 and 4.2 +- 0.10, respectively. Planted pots were placed on a greenhouse bench to acclimate to the greenhouse environment for one week, whereupon they were assigned at random to light and water treatments.

Half of the pots were assigned to a shade treatment (hereafter 'shaded'), whereas the other half of the pots remained unshaded (hereafter 'unshaded'). The shade treatment was created using a frame constructed of Polyvinylchloride pipe attached to the greenhouse bench and draped over the top and sides with 50% neutral density shade cloth.

Within each level of light, we assigned half of the pots to a 'wet' watering treatment and half of the pots to a 'dry' watering treatment. Pots in the 'wet' and 'dry' treatments were watered at least weekly, or as necessary, to maintain soil volumetric water content (VWC;%) at about 50% VWC (i.e., field capacity) and 20% VWC, respectively.

Plants were fertilized weekly during the regular watering event with Peters 20-20-20 water-soluble fertilizer mixed to 100 ppm within 0.1 L water. Plants were treated with Conserve (active ingredient Spinosad, 15.6 ml per liter water) for thrips (sprayed on leaves) and 1 [(6-chloro-3-pyridinyl)-methyl]-N-nitro-2- imidazolidinimine (Admire, 6 drops per 10 L water) for aphids weekly throughout the experiment, after the second week.

PLANT MONITORING. Every 14 d, we determined number of leaves (or leaf blades for grasses) and length of longest leaf on each plant. The experiment was terminated after 70 d because (1) the plants in most treatments were becoming pot-bound (i.e., with roots throughout the available soil volume) and (2) greenhouse bench space for shoots became limited. The duration of this experiment was comparable to, or longer than, other similar glasshouse competition experiments (e.g., Goldberg and Fleetwood 1987, Goldberg and Landa 1991). Aboveground biomass for each plant was determined after each plant was clipped at the soil surface and dried at 60 [degrees]C for 2 d. We did not collect root biomass because it was impractical to separate roots of each from one another and from the bulk soil.

ENVIRONMENTAL MONITORING. Photosynthetic photon flux density (PPFD; [mu]mol m^sup -2^s^sup -1^) in shaded and unshaded treatments was measured twice during the experiment using a line-integrating ceptometer (AccuPAR, Decagon Devices, Pullman, WA). Soil volumetric water content (VWC; %) in each pot was monitored at least weekly, 2- 3 d after a given watering event, using a handheld time-domain reflectometer (Hydrosense, Decagon Devices, Pullman, WA). Air temperature within the shaded and unshaded treatments was monitored with automated temperature dataloggers (HOBO Tidbit, Onset Computer Corp., Bourne, MA) for 14 consecutive days during the second quarter of the experiment; data were logged hourly.

DATA ANAYLSIS. We analyzed biomass of shoots and number of leaves at experiment termination with separate statistical models based on a completely randomized design (CRD) with replication and one nested treat- ment effect. Species, light and water were treated as fully crossed fixed effects, and neighbor was nested within species. Data were analyzed with general linear models (PROC GLM; SAS Institute 1999). Similarly, we analyzed soil volumetric water content within each pot for each of four representative sample dates. All data were tested for normality with the Shapiro-Wilk W-Statistic (Shapiro and WiIk 1965). We used Fisher's protected LSD (Fisher 1960) a posteriori mean separation tests to compare levels within factors for significant (P

Relative yield of biomass was calculated as

RY^sub ij^ = m^sub ij^/m^sub ii^

where m^sub ij^ = biomass of plant i when grown with plant j and m^sub ii^ = biomass of plant i when grown in monoculture (Keddy 2001, Weigelt and Jolliffe 2003). We calculated competitive effect and competitive response of species for (1) all treatment combinations combined, (2) light treatments pooled across water treatments, and (3) water treatments pooled across light treatments, by constructing matrices of relative yield values and summing across rows and columns to obtain target scores and neighbor scores (Miller and Werner 1987). High target scores indicate strong effect on neighbors, whereas low neighbor scores indicate little response to neighbors. Final competitive rank for each species within each set of treatment comparisons was determined by ranking neighbor (i.e., effect) and target (i.e., response) scores on a scale of 1 to 4, summing the neighbor and target ranks, and ranking the sums from 1 to 4, where 1 represents the strongest competitor and 4 represents the weakest competitor. We used Spearman rank correlation (r^sub s^; Zar 1996) to compare final competitive ranks between light treatments and between water treatments.

Size can be an important factor affecting competitive ability: larger plants may have a greater competitive effect simply because their size may enable them to intercept or acquire more resources than smaller plants (Connolly 1986, Wilson 1988). Therefore, we used regression analysis to investigate species-specific relationships between mass of individual plants at the initiation of the experiment, and their relative yield of biomass at experiment termination, for all interspecific neighbors pooled, and separately for each interspecific neighbor (e.g., Dactylis vs. Plantago and Trifolium, Dactylis vs. Plantago, and Dactylis vs. Trifolium). Initial mass and relative yield data were log-transformed prior to statistical analysis; however, non-transformed data are presented in the results to facilitate interpretation. Biomass of seedlings at experiment initiation was estimated using allometric relationships developed from a separate suite of seedlings established and destructively harvested for this purpose. Morphological characteristics (i.e., number of leaves and shoot length) were recorded for 10 seedlings of each species; seedlings were then harvested for determination of dry biomass. Regression analysis was used to determine the best morphological predictor of initial mass for each species; model r^sup 2^s were > 0.75. We were unable to develop a suitable relationship between morphological characteristics and mass for Festuca, so we did not include this species in our analysis of initial mass vs. relative yield of mass.

Results. PLANT BIOMASS. Production of biomass for all plant species by experiment termination was typically negatively affected by neighbors, although in most cases the identity of the neighbor was unimportant. For example, when Festuca was grown alone, it produced more biomass than when it was grown with another Festuca, or with Dactylis or Plantago (Fig. 1). Thus, inter- and intraspecific effects of competition on Festuca were indistinguishable. However, not all species responded to competitors in the same manner (species x neighbor interaction, P

Competitive relationships among species in terms of biomass production were not affected by the light or water treatments (neighbor x species x light x water, P = 0.67, F = 0.9, df = 24). However, biomass production in re- sponse to the watering treatment was species-specific (species x water interaction, P = 0.0005, F= 6.1, df = 3): Plantago produced 13% more biomass in wet pots than in dry pots, Trifolium produced 22% less biomass in wet pots than dry pots, and watering did not affect biomass of Dactylis or Festuca (Table 1). Biomass production in response to the light treatment also depended on species identity (species x light interaction, P = 0.04, F = 2.3, df = 3): Dactylis, Festuca, and Plantago each produced at least 16% more mass in sun than in shade, whereas biomass of Trifolium did not differ between sun and shade treatments (Table 1).

FIG. 1. Plant biomass (g plant^sup -1^; error bars represent 1 SE) (top) and number of leaves (mean with 1 SE) (bottom) at experiment termination for each species when grown with no neighbor (0) or with Dactylis (D), Plantago (P), Trifolium (T), or Festuca (F). Within species, means with different lower-case letters were different (P

Table 1. Plant biomass (g plant^sup -1^ +- SE) and number of leaves (mean +- SE) for each species at experiment termination in water treatments (wet, dry) and light treatments (shaded, unshaded). Within rows, means for water treatment and means for light treatment with different lower-case letters were different (P

NUMBER OF LEAVES. The response of our subject plants in terms of leaf production was similar to patterns observed in terms of biomass production. The number of leaf blades produced by Dactylis at experiment termination was negatively affected by the presence of Dactylis and especially Plantago, but not Trifolium (Fig. 1; species x neighbor interaction, P 0.09, F 0.05, ^sup 2^

Table 2. Competitive effect and response of four plant species with ranked neighbor (i.e., effect) and target (i.e., response) scores. 'Final rank' is rank order of sum of neighbor score and target score ranks.

COMPETITIVE EFFECT AND RESPONSE: A COMPETITIVE HIERARCHY. Assessments of pairwise interactions in terms of competitive effect and response, based on relative yield averaged across water and light treatments, produced the following competitive hierarchy: Plantago > Dactylis = Trifolium > Festuca (Table 2). Plantago had both the greatest effect on neighbors and the least response to neighbors, and was thus ranked as most competitive of the four species. In contrast, Festuca had the least effect on neighbors, and the greatest response to the presence of neighbors, so it was ranked as least competitive. Dactylis and Trifolium exhibited intermediate and inconsistent effect and response scores, and thus received equivalent overall ranks.

Table 3. Competitive effect and response of four plant species in dry and wet treatments, with ranked neighbor (i.e., effect) and target (i.e., response) scores. 'Final rank' is rank order of sum of neighbor score and target score ranks for each treatment.

COMPETITIVE HIERARCHIES ALONG ENVIRONMENTAL GRADIENTS. When hierarchies were constructed for separate treatments, competitive effect and response values varied by species, and final ranks were not concordant across light treatments or across water treatments (r^sub s^ 0.68, ? = 4). For example, in the dry treatment, Dactylis was the strongest competitor and Trifolium was the weakest competitor, whereas in the wet treatment, Plantago was the strongest competitor and Festuca was the weakest competitor (Table 3). Plantago ranked as the strongest competitor in both shaded and unshaded conditions, whereas Trifolium was the weakest competitor in unshaded conditions and Festuca was the weakest competitor in shaded conditions (Table 4).

ENVIRONMENTAL MONITORING. Throughout the experiment, soil VWC in the 'wet' pots was 2-4 times that in 'dry' pots, and within each treatment combination, VWC was relatively constant across sample dates (Fig. 2; P 1316, and df = 1 for each date). However, on all sample dates, light level also affected soil moisture, either as a main effect or in interaction with water treatment or species identity. For example, on 21 July, VWC in wet plots was greater in shaded than unshaded conditions, whereas VWC in dry plots was unaffected by the light treatment (light x water interaction, P

Table 4. Competitive effect and response of four plant species in unshaded and shaded treatments, with ranked neighbor (i.e., effect) and target (i.e., response) scores. 'Final rank' is rank order of sum of neighbor score and target score ranks.

PPFD, measured on two arbitrary dates during the course of the experiment, was reduced an average of 61% by the shade cloth. This reduction in PPFD was sufficient to affect production of leaves and biomass for all species except Trifolium (Table 1), although competitive relationships among species were unaffected (i.e., there were no light x species x neighbor interactions). Maximum daily air temperatures under the shadecloth averaged about 3[degrees]C lower than in the unshaded treatment, but the shadecloth had little effect on mean daily or minimum daily air temperatures (data not shown).

FIG. 2. Soil volumetric water content (VWC %; error bars represent 1 SE) for pots in all treatment combinations on four sampling dates throughout the experiment.

Discussion. Inter- and intraspecific competition among Dactylis, Trifolium, and Festuca were indistinguishable. This result is consistent with reviews of literature which found that intraspecific competition is not necessarily more intense than interspecific competition in terms of either competitive effect or response (Goldberg and Barton 1992, Gurevitch et al. 1992). However, for Plantago, intraspecific competition was more important than interspecific competition in terms of biomass production. Differential responses may be related to the relative competitive ability of these four taxa: Plantago was typically ranked as most competitive, whereas the other taxa had lower competitive rankings. Similarly, Canham et al. (2006) found that pairwise competitive interactions were strongly asymmetric, and that the intensity of pairwise competition depended strongly on the identity of neighbors.

Because of the lack of statistical interactions among species, neighbors, and light or water, we concluded that patterns and intensity of competition were consistent across our light and water treatments. These results support the theory that competition intensity is constant across productivity gradients (Tilman 1988), as well as results of a meta-analysis of primary producers in high- and low-productivity environments (Gurevitch et al. 1992). However, other reviews of competition intensity and environmental gradients are equivocal (Grime 1979, Fowler 1986, Goldberg and Barton 1992, Goldberg et al. 1999). Moreover, the selection of the estimator of plant performance, as well as the experimental approach employed, can affect the determined outcome of pairwise interactions and the effect of resource availability on apparent competitive abilities (Maestre et al. 2005, Carlyle and Fraser 2006).

For five of six pairwise interactions examined, initial size was unrelated to relative performance (cf. Wilson 1988, Gerry and Wilson 1995, Ramseier and Weiner 2006). This is important for admittedly short-term experiments such as this one, because the premature termination of an experiment may provide results influenced more by initial plant size than by competitive interactions (see Connolly 1997, Gibson et al. 1999, Freckleton and Watkinson 2001). In our experiment, positive relationships between initial plant size and biomass at experiment termination were observed only for the strongest competitor and largest species, Plantago. Moreover, the relationship between size and yield was greatest when this strong competitor was paired with the weakest competitor, Festuca.

However, the mixed outcome of relationships between initial mass and relative yield in this study is not inconsistent with the findings of other studies. For example, although several studies have shown that the ranking of competitive effect is correlated with the ranking of average plant size or total abundance of the neighbor species (e.g., Mitchley and Grubb 1986, Miller and Werner 1987, Goldberg and Fleetwood 1987, Gurevitch et al. 1990, Connolly and Wayne 1996, Keddy et al. 2002), others have found that initial size has little bearing on competitive outcome (e.g., Wilson 1988, Gurevitch et al. 1992, Gerry and Wilson 1995, Ramseier and Weiner 2006).

Although our experiment cannot clarify the exact mechanism for the observed relationship between size and yield (i.e., it could be driven by either the competitive strength of the subject species, or the relative weakness of its competitor, or both), our results do support the contention that relatively large differences in size between competitors results in a positive feedback that magnifies asymmetry of competition over time (Weiner 1986, Wilson 1988, Keddy and Shipley 1989, Goldberg 1990, Keddy et al. 2002).

In contrast with our null prediction that competitive hierarchies would be invariant along environmental gradients, we found that species competitive rankings varied between levels within a given treatment factor, and that this rank reorganization varied across treatment factors. That said, extreme ends of competitive abilities were relatively stable (i.e., under most conditions, Plantago was the most competitive whereas Festuca was least competitive). Results obtained from construction of competitive hierarchies are consistent with other studies which indicate that pairwise interactions are predominantly asymmetric, and that predictable competitive hierarchies exist among potentially interacting plants within a variety of plant communities (e.g., Mitchley and Grubb 1986, Goldberg and Fleetwood 1987, Miller and Werner 1987, Keddy and Shipley 1989, Gurevitch et al. 1990, Goldberg and Landa 1991, Panetta and Randall 1993, Keddy et al. 2002, Barthram et al. 2002, Sher et al. 2004, Fynn et al. 2005, Canham et al. 2006). Results from our analysis suggests that the outcome of competitive interactions, and the stability of competitive hierarchies along environmental gradients, depends on (1) the species in question, (2) the availability of resources important to plant performance, and (3) the identity of the resource in question (i.e., different resources will have different impacts on competitive outcomes). This conclusion is consistent with theoretical and empirical research, which indicates that species respond individualistically to changes in environmental conditions (Gleason 1926), and that competitive importance varies along resource gradients (e.g., Grubb 1985, Tilman 1988, Aerts et al. 1990, Silvertown and Dale 1991, Novoplansky and Goldberg 2001, Canham et al. 2006, Sammul et al. 2006).

Alternatively, a number of studies have demonstrated that competitive hierarchies are invariant along environmental gradients, i.e., the relative ranking of competitive dominance is consistent at various levels of resources (Shipley et al. 1991, Gurevitch et al. 1992, Shipley and Keddy 1994, Cahill and Casper 1999, Keddy et al. 2000, 2002, Liancourt et al. 2005b). Recent research further demonstrates that the quantitative approach used to construct the hierarchy, the level of replication, and the number of interacting species considered simultaneously may result in different hierarchies (Perkins et al. in press). For example, indirect species interactions could alter the competitive pressure on the focal species (e.g., Miller 1994), and interactions among species may be mediated by disturbance or other factors (e.g., Levine 2000, Center et al. 2005, Tripler et al. 2005), each of which could play an important role in the development of competitive hierarchies from empirical experiments. Finally, although we do not have the data with which to speculate about the potential for below-ground interactions, root competition can be stronger than shoot competition, and in more complex communities, may involve more neighbors than aboveground competition (Casper and Jackson 1997).

In conclusion, we determined that (1) intraspecific and interspecific interactions were relatively unaffected by the levels of light and water established for this experiment for most, but not all of the species investigated, (2) initial plant size had little effect on competitive interactions, except for one case where species with highly contrasting competitive abilities were paired, and (3) competitive hierarchies varied among and within environmental resource gradients, which suggests that competitive hierarchies may be insufficient to predict the outcome of species- specific interactions along environmental gradients (cf. Keddy et al. 2002, Canham et al. 2006). Nonetheless, results of this experiment, coupled with conflicting results and interpretations in the literature regarding the invariance of competitive hierarchies, illuminates the need for additional studies of the stability of competitive hierarchies along environmental gradients (see also Aarssen and Keogh 2002). Additional research should also consider the relevance of the observed hierarchies to multispecies mixtures, particularly in natural field conditions (e.g., Engel and Weltzin in press, Perkins et al. in press).

1 We thank P. Allen, E. Buckner, C. DeVan, C. Engel, E. Herndon, C. Iversen, K. Kennard, K. McFarland, J. Mynes, R. Norby and L. Souza for their assistance with this project. R. Norby and C. Engel provided comments that improved earlier drafts of the manuscript. Research was funded by the Undergraduate Summer Research Internship Program of Oak Ridge National Laboratory and the Office of the Provost at the University of Tennessee, and by the Office of Science (BER), U. S. Department of Energy Grant No. DE-FG02-02ER63366.

Literature Cited

AARSSON, L. W. AND T. KEOGH. 2002. Conundrums of competitive ability in plants: what to measure? Oikos 96: 531-542.

AERTS, R., F. BERENDSE, H. DE CALUWE, AND M. SCHMITZ. 1990. Competition in heathland along an experimental gradient of nutrient availability. Oikos 57: 310-318.

BARBOUR, M. G., J. H. BURK, W. D. PITTS, F. S. GILLIAM, AND M. W. SCHWARTZ. 1999. Terrestrial Plant Ecology, 3rd ed. Benjamin Cummings, Menlo Park, CA. 604 p.

BARTHRAM, G. T., D. A. ELSTON, C P. D. BIRCH, AND G. R. BOLTON. 2002. Defoliation and site differences influence vegetative spread in grassland. New Phytol. 155: 257-264.

BAZZAZ, F. A. 1968. Succession on abandoned fields in Shawnee Hills Southern Illinois. Ecology 49: 924-936.

BECKWITH, S. L. 1954. Ecological succession on abandoned farm lands and its relationship to wildlife management. Ecol. Monogr. 24: 349-1376.

CAHILL, J. F. JR, AND B. B. CASPER. 1999. Growth consequences of soil nutrient heterogeneity for two old-field herbs, Ambrosia artemisiifolia and Phytolacaa americana, grown individually and in combination. Ann. Bot. 83: 471-478.

CANHAM, C. D., M. J. PAPAIK, M. URIARTE, W. H. MCWILLIAMS, J. C JENKINS, AND M. J. TWERY. 2006. Neighborhood analyses of canopy tree competition along environmental gradients in New England forests. Ecol. Appl. 16: 540-554.

CARLYLE, C. N. AND L. H. FRASER. 2006. A test of three juvenile plant competitive response strategies. J. Veg. Sci. 17: 11-18.

CASPER, B. B. AND R. B. JACKSON. 1997. Plant competition belowground. Ann. Rev. Ecol. Syst. 28: 545-570.

CENTER, T. D., T. K. VAN, F. A. DRAY, S. J. FRANKS, M. T. REBELO, P. D. PRATT, AND M. B. RAYAMAJHI. 2005. Herbivory alters competitive interactions between two invasive aquatic plants. Biol. Control 33: 173-185.

CONNELL, J. H. 1983. On the prevalence and relative importance of interspecific competition: evidence from field experiments. Am. Nat. 122: 661-696.

CONNOLLY, J. 1986. On difficulties with replacement series methodology in mixture experiments. J. Appl. Ecol. 23: 125-137.

CONNOLLY, J. 1997. Substitutive experiments and the evidence for competitive hierarchies in plant communities. Oikos 80: 179-182.

CONNOLLY, J. AND P. WAYNE. 1996. Asymmetric competition between plant species. Oecologia 108: 311-320.

DAYTON, B. R. 1975. Early stages of vascular plant succession in a central New York old field. Am. Midl. Nat. 94: 62-71.

EGLER, F. E. 1954. Vegetational science concepts. I. Initial floristic composition, a factor in old-field vegetation development. Vegetatio 40: 412-417.

ENGEL, E. C AND J. F. WELTZIN. In press. Can community composition be predicted from pairwise species interactions? Plant Ecol.

FISHER, R. A. 1960. The design of experiments, 7th ed. Hafner, New York, NY. 248 p.

FOWLER, N. 1986. The role of competition in plant-communities in arid and semiarid regions. Annu. Rev. Ecol. Syst. 17: 89-110.

FRECKLETON, R. P. AND A. R. WATKINSON. 2001. Predicting competition coefficients for plant mixtures: reciprocity, transitivity, and correlations with life-history traits. Ecol. Lett. 4: 348-1357.

FYNN, R. W. S., C D. MORRIS, AND K. P. KIRKMAN. 2005. Plant strategies and trait trade-offs influence trends in competitive ability along gradients of soil fertility and disturbance. J. Ecol. 93: 384-394.

GERRY, A. K. AND S. D. WILSON. 1995. The influence of initial size on the competitive responses of six plant species. Ecology 76: 272-1279.

GIBSON, D. J., J. CONNOLLY, D. C. HARTNETT, AND J. D. WEIDENHAMER. 1999. Designs for greenhouse studies of interactions between plants. J. Ecol. 87: 1-16.

GIBSON, D. J., B. A. MIDDLETON, K. FOSTER, Y. A. K. HONU, E. W. HOYER, AND M. MATHIS. 2005. Species frequency dynamics in an old- field succession: Effects of disturbance, fertilization, and scale. J. Veg. Sei. 16: 415-422.

GLEASON, H. A. 1926. The individualistic concept of the plant association. Bull. Torr. Bot. Club 53: 1-20.

GOLDBERG, D. E. 1987. Neighborhood competition in and old-field plant community. Ecology 68: 1211-1223.

GOLDBERG, D. E. 1990. Components of resource competition in plant communities, p. 27-49. In J. B. Grace and D. Tilman [eds.], Perspectives on Plant Competition. Academic Press, San Diego, CA.

GOLDBERG, D. E. AND A. M. BARTON. 1992. Patterns and consequences of interspecific competition in natural communities: A review of field experiments with plants. Am. Nat. 139: 771-801.

GOLDBERG, D. E. AND L. FLEETWOOD. 1987. Competitive effect and response in four annual plants. J. Ecol. 75: 1131-1143.

GOLDBERG, D. E. AND K. LANDA. 1991. Competitive effect and response: hierarchies and correlated traits in the early stages of competition. J. Ecol. 79: 1013-1030.

GOLDBERG, D. E., T. RAJANIEMI, J. GUREVITCH, AND A. STEWART- OATEN. 1999. Empirical approaches to quantifying interaction intensity: competition and facilitation along productivity gradients. Ecology 80: 1118-1131.

GOLDBERG, D. E. AND P. A. WERNER. 1983. Equivalence of competitors in plant communities: a null hypothesis and a field experimental approach. Am. J. Bot. 70: 1098-1104.

GREILING, D. A. AND N. KICHANAN. 2002. Old-field seedling responses to insecticide, seed addition, and competition. Plant Ecol. 159: 175-183.

GRIME, J. P. 1979. Plant Strategies and Vegetation Processes. John Wiley & Sons, Chichester, England. 222 p.

GRUBB, P. J. 1985. Plant populations and vegetation in relation to habitat disturbance and competition: problems of generalization, p. 595-621. In J. White [ed.], The Population Structure of Vegetation. Dr. W. Junk Publishers, Dordrecht, The Netherlands.

GUREVITCH, J., P. WILSON, J. L. STONE, P. TEESE, AND R. J. STOUTENBURGH. 1990. Competition among old-field perennials at different levels of soil fertility and available space. J. Ecol. 78: 727-1744. GUREVITCH, J., L. L. MORROW, A. WALLACE, AND J. S. WALSH. 1992. A meta-analysis of competition in field experiments. Am. Nat. 140: 539-572.

HARPER, J. L. 1977. Population Biology of Plants. Academic Press, London, England. 892 p.

KEDDY, P. 2001. Competition. 2nd ed. Kluwer Academic Publishers, Dordrecht, The Netherlands. 552 p.

KEDDY, P., C. GAUDET, AND L. H. FRASER. 2000. Effects of low and high nutrients on the competitive hierarchy of 26 shoreline plants. J. Ecol. 88: 413-423.

KEDDY, P., K. NIELSON, E. WEIHER, AND R. LAWSON. 2002. Relative competitive performance of 63 species of terrestrial herbaceous plants. J. Veg. Sei. 13: 5-16.

KEDDY, P. AND B. SHIPLEY. 1989. Competitive hierarchies in herbaceous plant communities. Oikos 54: 234-241.

KEEVER, C. 1950. Causes of succession on old fields of the Piedmont, North Carolina. Ecol. Monogr. 20: 229-250.

KOSOLA, K. R. AND K. L. GROSS. 1999. Resource competition and suppression of plants colonizing early successional old fields. Oecologia 118: 69-75.

LEVINE, J. M. 2000. Complex interactions in a streamside plant community. Ecology 81: 3431-13444.

LIANCOURT, P., R. M. CALLOWAY, AND R. MICHALET. 2005a. Stress tolerance and competitive response ability determine the outcome of biotic interactions. Ecology 86: 1611-1618.

LIANCOURT, P., E. CORCKET, AND R. MICHALET. 2005b. Stress tolerance abilities and competitive responses in a watering and fertilization field experiment. J. Veg. Sci. 16: 713-722.

MAESTRE, F. T., F. VALLADARES, AND J. F. REYNOLDS. 2005. Is the change of plant-plant interactions with abiotic stress predictable? A meta-analysis of field results in arid environments. J. Ecol. 93: 748-757.

MILLER, T. E. 1994. Direct and indirect species interactions in an early old-field plant community. Am. Nat. 143: 1007-1025.

MILLER, T. E. AND P. A. WERNER. 1987. Competitive effects and responses between plant species in a first-year old-field community. Ecology 68: 1201-1210.

MITCHLEY, J. AND P. J. GRUBB. 1986. Control of relative abundance of perennials in chalk grassland in southern England. 1 . Constancy of rank order and results of pot-experiments and fieldexperiments on the role of interference. J. Ecol. 74: 1139-1166.

MONZEGLIO, U. AND P. STOLL. 2005. Spatial patterns and species performances in experimental plant communities. Oecologia 145: 619- 628.

NOVOPLANSKY, A. AND D. E. GOLDBERG. 2001. Effects of water pulsing on individual performance and competitive hierarchies in plants. J. Veg. Sei. 12: 199-208.

ODUM, E. P. 1960. Organic production and turnover in old field succession. Ecology 41: 34-49.

PANETTA, F. D. AND R. P. RANDALL. 1993. Emex australis and the competitive hierarchy of a grazed annual pasture. J. Appl. Ecol. 30: 3731379.

PERKINS, T. A., W. R. HOLMES, AND J. F. WELTZIN. In press. Consideration of multiple-species interactions in competitive hierarchies: methodological advancement and empirical test. J. Veg. Sei.

RAMSEIER, D. AND J. WEINER. 2006. Competitive effect is a linear function of neighbor biomass in experimental populations of Kochia scoparia. J. Ecol. 94: 305-309.

SAMMUL, M., L. OKSANEN, AND M. MAGI. 2006. Regional effects on competition-productivity relationship: a set of field experiments in two distant regions. Oikos 112: 138-148.

SAS INSTITUTE. 1999. SAS/STAT user's guide. Version 8.0. Cary, NC, USA: SAS Institute. 838

p. SCHMITZ, O. J. 2003. Top predator control of plant biodiversity and productivity in an old-field ecosystem. Ecol. Lett. 6: 156-163.

SCHMITZ, O. J., E. L. Kalies, and M. G. Booth. 2006. Alternative dynamic regimes and trophic control of plant succession. Ecosystems 9: 6591672.

SCHOENER, T. W. 1983. Field experiments on interspecific competition. Am. Nat. 122: 2401285.

SHAPIRO, S. S. AND M. B. WILK. 1965. An analysis of variance test for normality (complete samples). Biometrika 52: 591-611.

SHER, A. A., D. E. GOLDBERG, AND A. NOVOPLANSKY. 2004. The effect of mean and variance in resource supply on survival of annuals from Mediterranean and desert environments. Oecologia 141: 353-362.

SHIPLEY, B. 1993. A null model for competitive hierarchies in competition matrices. Ecology 74: 1693-1699.

SHIPLEY, B., P. A. KEDDY, C. GAUDET, AND D. R. J. MOORE. 1991. A model of species density in shoreline vegetation. Ecology 72: 1658- 1667.

SHIPLEY, B. AND P. A. KEDDY. 1994. Evaluating the evidence for competitive hierarchies in plant communities. Oikos 69: 340-345.

SILVERTOWN, J. AND P. DALE. 1991. Competitive hierarchies and the structure of herbaceous plant communities. Oikos 61: 441-444.

SMITH, T. M. AND M. A. HUSTON. 1989. A theory of the spatial and temporal dynamics of plant communities. Vegetatio 83: 49-69.

SUDING, K. N., J. R. LARSON, E. THORSOS, H. STELTZER, AND W. D. BOWMAN. 2004. Species effects on resource supply rates: do they influence competitive interactions? Plant Ecol. 175: 47-58.

TILMAN, D. 1982. Resource Competition and Community Structure. Princeton University Press, Princeton, NJ. 296 p.

TILMAN, D. 1988. Plant Strategies and the Structure and Dynamics of Plant Communities. Princeton University Press, Princeton, NJ. 360 p.

TRIPLER, C. E., C. D. CANHAM, R. S. INOUYE, AND J. L. SCHNURR. 2005. Competitive hierarchies of temperate tree species: interactions between resource availability and white-tailed deer. Ecoscience 12: 494-505.

WEAVER, J. E. AND F. E. CLEMENTS. 1929. Plant Ecology. McGraw- Hill, New York, NY. 601 p.

WEIGELT, A. AND P. JOLLIFFE. 2003. Indices of plant competition. J. Ecol. 91: 707-720.

WEINER, J. 1986. How competition for light and nutrients affects size variability in Ipomoea tricolor populations. Ecology 67: 1425- 1427.

WILSON, J. B. 1988. The effect of initial advantage on the course of plant competition. Oikos 51: 19-24.

WILSON, S. D. and P. A. Keddy. 1986. Species competitive ability and position along a natural stress/disturbance gradient. Ecology 67: 12361242.

ZAR, J. H. 1996. Biostatistical Analysis, 3rd ed. Prentice Hall, Upper Saddle River, NJ. 718 p.

Allison M. Fortner2 and Jake F. Weltzin3

Department of Ecology and Evolutionary Biology, University of Tennessee, 569 Dabney Hall,

1416 Circle Drive, Knoxville, TN 37996-1610

2 Present address: Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6351.

3 Author for correspondence: E-mail: [email protected]

Received for publication February 19, 2006, and in revised form January 23, 2007.

Copyright Torrey Botanical Society Apr-Jun 2007

(c) 2007 Journal of the Torrey Botanical Society. Provided by ProQuest Information and Learning. All rights Reserved.