Present address: Division of Biology, 125 Ackert Hall, Kansas State University, Manhattan, KS 66506, USA.
Interspecific and spatial differences in nitrogen uptake in monocultures and two-species mixtures in north European grasslands
Article first published online: 1 AUG 2002
Volume 16, Issue 4, pages 454–461, August 2002
How to Cite
Jumpponen, A., Högberg, P., Huss-Danell, K. and Mulder, C. P. H. (2002), Interspecific and spatial differences in nitrogen uptake in monocultures and two-species mixtures in north European grasslands. Functional Ecology, 16: 454–461. doi: 10.1046/j.1365-2435.2002.00642.x
- Issue published online: 1 AUG 2002
- Article first published online: 1 AUG 2002
- Received 24 January 2002;accepted 25 January 2002
- nutrient uptake;
- root length;
- stable isotope
1. To study the potential for complementarity in nitrogen acquisition from different soil depths, we injected an isotope tracer (15NH4Cl) at 5 and 20 cm depths in plant communities containing Achillea millefolium L. and Festuca ovina L. or Phleum pratense L. and Trifolium pratense L. in monocultures and two-species mixtures.
2. In monoculture, Festuca and Phleum took up tracer at 5 and 20 cm depths. In contrast, Achillea and Trifolium monocultures acquired the tracer mainly from 5 cm depth. In two-species mixtures, all four species took up tracer at 5 cm depth.
3. Achillea N acquisition from 20 cm depth increased in mixture with Festuca in comparison to that in monoculture; Festuca N acquisition from 20 cm depth decreased, although not significantly. Trifolium N acquisition remained unchanged when grown in mixture with Phleum. Phleum behaved like Festuca: its N acquisition from 20 cm depth in mixture was reduced in comparison to monoculture.
4. Our data on Festuca and Achillea support spatial partitioning in resource acquisition. This was not evident in Phleum and Trifolium mixture, potentially because Trifolium relied on N2 fixation as N source.
5. These results demonstrate spatial variation among plant species and plant communities in their N acquisition in the field.
Explanatory mechanisms for the coexistence of plants in diverse communities have been sought in niche partitioning. Spatial distribution of roots and temporal separation of nutrient uptake have been proposed as possible below-ground mechanisms of such partitioning in plant communities (Berendse 1981; Berendse 1982; McKane, Grigal & Russelle 1990; Parrish & Bazzaz 1976; Pechácováet al. 1999; Veresoglou & Fitter 1984; Yeaton, Travis & Gilinsky 1977). The spatial distribution of roots may be affected by a variety of biotic and abiotic factors. Root growth and proliferation are frequently responsive to resource availability (Arredondo & Johnson 1999; Caldwell, Manwaring & Jackson 1991b; Crick & Grime 1987; Fitter 1994; Robinson 1994; Tibbett 2000). Similarly, environmental stresses such as drought can modify rooting depth (Mamolos, Elisseou & Veresoglou 1995; Reader et al. 1992). Finally, interference by adjacent roots, via either competition and resource depletion (Caldwell, Manwaring & Durham 1991a; Caldwell, Manwaring & Durham 1996; D’Antonio & Mahall 1991; Mahall & Callaway 1991) or root-originating allelopathic or inhibitory compounds (Mahall & Callaway 1991; Mahall & Callaway 1992) have also been proposed to be important mechanisms in governing root distribution in soil matrix. Taken together, community composition and resource availability may influence the distribution of roots in plant communities and subsequent nutrient acquisition.
A model presented by Berendse (1979) predicts that a partitioning of vertical rooting depth would result in greater resource acquisition and increase in relative yield totals (according to de Wit & van den Bergh 1965) in the community. Indeed, when deep-rooting Plantago lanceolata was combined in a mixture with a shallow-rooting Antoxanthium odoratum, P. lanceolata– potentially the inferior competitor in the shallower soil profiles – was forced to utilize deeper soil layers (Berendse 1982). Berendse's model does not directly address spatial or temporal heterogeneity of soil resources. However, it suggests that species may differ in their abilities to exploit soil resources, whether or not they are heterogeneously distributed in time or space. Similarly to Berendse's predictions (Berendse 1979) and experimental work (Berendse 1982), D’Antonio & Mahall (1991) recorded greater rooting depths of the native Haplopappus spp. when an adjacent, shallow-rooted invader (Carpobrotus edulis) was present in Californian coastal shrublands. Other studies have also shown that species differ in their rooting depths, but have not explicitly addressed whether rooting depths were altered by community composition, or whether or not these differences in rooting depths result in different resource exploitation capabilities (Fitter 1986; Mamolos et al. 1995; McKane et al. 1990).
Recently, the connection between community composition or species richness and ecosystem function has received considerable attention. Although a larger proportion of the positive effect of the species richness on productivity may be attributable to a probability of including certain species in the communities – the ‘sampling effect’ (Aarssen 1997; Hector 1998; Tilman 1999) – complementary resource use may have an additional effect (Hector 1998; Hector et al. 1999; Spehn et al. 2000; Tilman 1999). We aimed to test if it is possible to detect differential capabilities for nutrient acquisition among species growing in monocultures and two-species mixtures. We examined if naturally occurring species in northern Sweden would acquire nitrogen at different depths when grown in monoculture than when grown in mixture. Differing abilities in N acquisition from different soil depths suggest potential among species to utilize resources in different soil depths, whether they are heterogeneously or homogeneously distributed.
Materials and methods
Site description and experimental design
The field site was located at the Swedish University of Agricultural Sciences in Umeå, Sweden (63°45′ N, 20°17′ E, 12 m a.s.l.). The soil type was a fine silty sand with little clay (4·1% clay, 57·9% silt, 38·0% fine sand), pH 6·0. In the 3 years prior to the experiment, the field had been used for barley, potato and barley cultivation. It had been fertilized every year, the last being 1995, with an application of 400 kg ha−1 N-P-K (11-5-18). During the summer of 1995, 7·5 g ha−1 herbicide (Expresspreparat, DuPont Agro, Malmö, Sweden) was applied to reduce weed growth. The last barley crop was harvested in the fall of 1995, and the site was ploughed without removal of straw. In spring 1996 the site was repeatedly harrowed. No fertilizer was added from 1996 onwards. As a result of this agricultural history, the field was considered adequately homogeneous to allow comparisons among the different, adjacent, human-made communities established for this and other experiments.
Six plant communities, four monocultures and two mixtures, containing perennial, native vascular plants (Festuca ovina L. and Achillea millefolium L. or Phleum pratense L. and Trifolium pratense L.) were established in June, 2 years before conducting the tracer injection experiment. We selected these two pairs of species because within each pair, individuals reach approximately the same above-ground height, but the species have different rooting patterns with respect to root length. Furthermore, species within each pair share similar habitat requirements: Festuca and Achillea occur typically on dry sites, whereas Phleum and Trifolium are commonly cultivated as forage species and occur naturally on roadsides and meadows. The two herbs form a dominant tap root, whereas the two grasses have fibrous root systems.
The six plant communities were sown on six separate rectangles, each measuring 5·0 × 2·2 m. All communities received a total density of 2000 seeds m−2, i.e. monocultures received 2000 seeds m−2 of one single species, while two-species mixtures received 1000 seeds m−2 of each of the two species.
Within each plant community, at the peak of the growing season in early August 1998, two random soil samples 20 cm deep were collected with a 4 cm diameter corer. The samples were divided into two depths: 0–10 and 10–20 cm. Each sample was cleaned free of soil on a screen under running tap water, spread evenly on a transparent screen, and digitized using a flatbed scanner. We were unable to separate roots in the mixtures. Therefore the total root lengths for both monocultures and mixtures are presented in the results. The root lengths were estimated from the digitized images with the DELTA-T SCAN image analysis system (DELTA-T Devices Ltd, Cambridge, UK).
Injection of15N and foliar sample processing
Within each of the six communities with different species composition, we selected 10 representative plots measuring 10 × 10 cm. To minimize isotope dilution effects, individual plants of any given species were visually judged to be equal in size among the 10 plots. These 10 plots were then randomly assigned to two different N tracer treatments (see below), five for each injection depth. This design totalled five replicates of two different injection depths within six plant communities; 60 plots in total (Table 1). Individuals of all four species were flowering at the time of injection in monocultures and mixtures, indicating close similarities in phenology.
|Monoculture||Achillea millefolium L.|
|Monoculture||Festuca ovina L.|
|Mixture||Achillea millefolium L. and Festuca ovina L.|
|Monoculture||Phleum pratense L.|
|Monoculture||Trifolium pratense L.|
|Mixture||Phleum pratense L and Trifolium pratense L.|
All plant communities were injected with 15N tracer within a 2 h period in mid-July 1998. No precipitation occurred during the injection experiment. A well for 15N injection, 5 or 20 cm deep, was created by pushing an open-ended plastic cylinder (≈7 mm diameter) into the ground in the four mid-points of the lines defining the 10 × 10 cm plot. A total of 12·5 ml of 2·133 mm15NH4Cl (98 at %15N) solution was injected into each of the four wells in each plot. The total N injected per plot equalled ≈0·16 g m−2, a minute quantity compared to the total soil N pool (total N 1·3 mg g−1), but substantial with regard to (0·06 mg soil water and 0·33 mg soil water in plots with Phleum and Phleum plus Trifolium, respectively) and (0·4 µg soil water and 0·8 µg soil water in plots with Phleum and Phleum plus Trifolium, respectively; Cecilia Palmborg, unpublished lysimeter data, August 1998).
Approximately 10 mg (dry weight) of the topmost, green leaves were collected before injection. The individual plants were always adequately large, or the plots contained more than one individual, to allow repeated harvests from the same layer of the canopy within a plot. The sampling was repeated 2, 6 and 24 h after injection. All foliar samples were dried at 60 °C for 24 h and homogenized with a ball mill. The samples were analysed for 15N abundance using a Europa Scientific (Crewe, UK) isotope ratio mass spectrometer interfaced with a Europa Scientific ANCA-NT preparation module (Ohlsson & Wallmark 1999). Results for 15N abundance before injection are expressed in the standard notation (δ15N) in parts per thousand () relative to the international standard (atmospheric N2 at 0·3663 at %, Junk & Svec 1958; Mariotti 1983):
- δ15N = [(Rsample/Rstandard) − 1] × 1000
where R = mole ratio of 15N/14N.
Results after injection are expressed as increase in concentration of the heavy isotope in dried foliar tissue (nmol of 15N mg−1 dry foliar tissue) resulting from the injection, i.e. the difference between the 15N concentration before injection and at 2, 6 and 24 h after injection. We assumed that the applications of N would not significantly affect the growth of plants, and that creating an N-enriched patch would not affect the spatial distribution of roots during the short 24 h period. Furthermore, we mainly estimated whether or not the uptake occurred, rather than making quantitative comparisons among the target species.
After the final foliar sample was collected, 24 h after injection, a soil core 2·5 cm in diameter was removed from a depth of 30 cm in the centre of each plot. Two samples representing the 0–10 and 15–25 cm horizons were separated from each soil core. The soil samples were dried at 80 °C for 48 h and homogenized using a rod mill. The samples were analysed for 15N abundance, as described above.
Although the differences were frequently non-significant (Figs 1 and 2, least squares means pairwise comparisons; SAS 1989), greater root lengths generally occurred at 0–10 than 10–20 cm depth (Tables 2 and 3, anova; SAS 1989). Achillea monoculture had few roots deeper in the soil, significantly fewer than Festuca monoculture. Phleum had significantly greater root length at 0–10 than at 10–20 cm, while Trifolium had the same root length at both soil depths. These data agree with our earlier observations and thus supported our choice of species for this study.
Tracer injection at the two different soil depths was successful (soil δ15N before experiment 4·4 ± 1·0; after injection 11·6 ± 4·7; mean ± 1 SD). Although there were no significant differences in 15N isotope composition among the soil samples (anova; SAS 1989), the tracer mainly remained in the soil layer to which it was injected (data not shown). The fact that is less mobile than is an advantage in this type of experiment, and may have contributed to minimal 15N leakage between the two soil horizons studied.
The δ15N data for leaves collected before injection were left untransformed and analysed by anova (SAS 1989). Festuca (δ15N 4·52 ± 0·60) and Achillea (δ15N 5·41 ± 0·47) monoculture δ15N did not differ before injection (Table 4). However, plants in mixed communities were more 15N-enriched than in monocultures (Festucaδ15N 6·17 0·95, Achilleaδ15N 6·80 2·34). In contrast, monocultures of Phleum (δ15N 7·58 ± 1·17) and Trifolium (δ15N 0·31 ± 0·16) clearly differed at the time of injection (Table 5): Phleum was more enriched than Trifolium in 15N. When in mixture, Phleum (δ15N 5·09 ± 0·71) was less enriched in 15N than in monoculture, while Trifolium in mixture (−0·32 ± 0·58) did not differ from Trifolium in monoculture.
The foliar 15N excess data were analysed separately by anova (SAS 1989) for different sampling times after tracer injection, because of increasing variance with time since injection. The data were log10-transformed because of large differences in variance among the different treatments within a given sampling time. All species took up the injected 15N, and this was usually evident from sampling of the topmost leaves within 2 h after injection. The 15N excess increased with time since injection. For brevity and clarity, the results are presented separately for each species and each point in time (Tables 6–9), but focus on the final sampling time, 24 h after injection.
|Time (h)||Community||Injection (cm)||15N excess (nmol mg−1)||anova|
|2||Monoculture||5||22·6 ± 24·6a|
|Monoculture||20||2·51 ± 4·16a||Mixture||1||0·05||0·829|
|Mixture||5||1·00 ± 0·30a||Depth||1||2·26||0·157|
|Mixture||20||0·41 ± 2·29a||Interaction||1||1||0·336|
|6||Monoculture||5||32·7 ± 24·4a|
|Monoculture||20||0·99 ± 0·83a||Mixture||1||0·46||0·507|
|Mixture||5||5·74 ± 7·87a||Depth||1||0||0·959|
|Mixture||20||0·51 ± 0·39a||Interaction||1||2·29||0·153|
|24||Monoculture||5||185 ± 256c|
|Monoculture||20||3·97 ± 3·73a||Mixture||1||3·92||0·069|
|Mixture||5||86·0 ± 30·9bc||Depth||1||27·0||0·001|
|Mixture||20||26·7 ± 18·9b||Interaction||1||5·55||0·035|
|Time (h)||Community||Injection (cm)||15N excess (nmol mg−1)||anova|
|2||Monoculture||5||4·68 ± 2·58a|
|Monoculture||20||1·75 ± 1·23a||Mixture||1||0·28||0·604|
|Mixture||5||6·01 ± 5·89a||Depth||1||2·43||0·141|
|Mixture||20||1·48 ± 0·45a||Interaction||1||0·23||0·642|
|6||Monoculture||5||20·2 ± 25·9a|
|Monoculture||20||1·92 ± 0·48a||Mixture||1||1·93||0·186|
|Mixture||5||8·98 ± 5·98a||Depth||1||5·51||0·034|
|Mixture||20||0·72 ± 0·77a||Interaction||1||1·78||0·204|
|24||Monoculture||5||87·2 ± 83·0a|
|Monoculture||20||88·3 ± 85·2a||Mixture||1||0·37||0·554|
|Mixture||5||73·5 ± 25·8a||Depth||1||0·91||0·356|
|Mixture||20||21·6 ± 23·7a||Interaction||1||2·94||0·109|
|Time (h)||Community||Injection (cm)||15N excess (nmol mg−1)||anova|
|2||Monoculture||5||12·3 ± 11·7a|
|Monoculture||20||1·07 ± 0·52a||Mixture||1||0·84||0·373|
|Mixture||5||2·82 ± 0·82a||Depth||1||2·81||0·113|
|Mixture||20||1·83 ± 1·50a||Interaction||1||0·02||0·892|
|6||Monoculture||5||162 ± 261b|
|Monoculture||20||21·1 ± 13·4b||Mixture||1||4·21||0·059|
|Mixture||5||45·8 ± 23·0b||Depth||1||15·8||0·001|
|Mixture||20||5·20 ± 2·20a||Interaction||1||1·22||0·286|
|24||Monoculture||5||336 ± 275b|
|Monoculture||20||82·8 ± 23·5b||Mixture||1||3·43||0·091|
|Mixture||5||86·7 ± 59·7b||Depth||1||3·30||0·097|
|Mixture||20||5·89 ± 3·98a||Interaction||1||0·28||0·604|
|Time (h)||Community||Injection (cm)||15N excess (nmol mg−1)||anova|
|2||Monoculture||5||1·71 ± 1·00a|
|Monoculture||20||−0·24 ± 1·18a||Mixture||1||4·23||0·058|
|Mixture||5||1·96 ± 0·98a||Depth||1||0·83||0·378|
|Mixture||20||1·15 ± 0·42a||Interaction||1||1·10||0·311|
|6||Monoculture||5||29·8 ± 24·4a|
|Monoculture||20||0·82 ± 0·29a||Mixture||1||0||0·981|
|Mixture||5||15·0 ± 10·1a||Depth||1||0·21||0·651|
|Mixture||20||0·67 ± 0·55a||Interaction||1||0||0·975|
|24||Monoculture||5||29·0 ± 13·6b|
|Monoculture||20||1·42 ± 1·42a||Mixture||1||0·82||0·379|
|Mixture||5||26·9 ± 12·6b||Depth||1||53·3||0·001|
|Mixture||20||1·56 ± 0·55 a||Interaction||1||0·96||0·343|
Achillea when in monoculture took up tracer exclusively from 5 cm depth (Table 6). In contrast, Festuca when in monoculture acquired tracer equally from both 5 and 20 cm depths (Table 7). When in mixture, both Festuca and Achillea took up tracer at 5 cm depth. However, Festuca N acquisition from 20 cm depth in mixture was reduced in comparison to that in monoculture, although not significantly. Achillea N acquisition in mixture, in contrast, was greater from 20 cm depth in comparison to that in monoculture. Achillea N acquisition from 20 cm depth in monoculture barely increased over the experimental period.
Phleum had a significant uptake of tracer both from 5 and 20 cm depth when in monoculture (Table 8). Although the differences in the acquired quantities were large, on average four times greater at 5 than at 20 cm, they were not significant. Trifolium acquired soil 15N nearly exclusively from 5 cm depth (Table 9). When in mixture, both Phleum and Trifolium acquired tracer from 5 cm depth. Phleum N acquisition from 20 cm depth in mixture remained small throughout the experiment, and was very small in comparison to monoculture. Trifolium acquired soil N nearly exclusively from 5 cm depth even when in mixture: its N acquisition remained unchanged when grown in mixture with Phleum.
The goals of the present study were to observe N uptake from two different soil horizons among four native grassland species when a pulse of was injected directly into soil. We created uniquely 15N-enriched patches which were placed either at 5 or 20 cm soil depth. The added solution had an N concentration five orders of magnitude greater than the soil solution. Although isotopic exchange with endogenous N pools may confound the interpretation of injection studies similar to ours, the patches in our study were so concentrated in N that minor differences in endogenous available N between soil depths or between monocultures or species mixtures should not affect our interpretations. Additionally, our emphasis lies on the depth of soil N acquisition in one plant species when it occurred with another species, as compared to when it grew alone. Finally, to minimize the potential isotope dilution effects, we specifically selected plant individuals of comparable size.
All species acquired soil N from the upper horizon, whereas the deeper horizon was utilized to a lesser extent. However, our data suggest that the species differed in their ability to utilize N from deeper soil, and the N acquisition was different between individuals grown in a monoculture and in a mixture.
After 24 h, Achillea had acquired only minimal quantities of N tracer from the 20 cm depth in monoculture. When grown with Festuca, Achillea N acquisition from that horizon increased. In contrast, Festuca N acquisition from that horizon decreased, although not significantly. These data support niche differentiation among plant species in mixtures: N acquisition depends on whether or not neighbours are present. It remains unclear from our study whether or not the observed differences in N uptake represent changes in rooting depth or root activity. Earlier theory (Berendse 1979), as well as greenhouse and field experiments (Berendse 1981; Berendse 1982; D’Antonio & Mahall 1991) suggest that rooting depths can be altered depending on the presence of neighbours.
Changes in the spatial distribution of roots have frequently been reported in response to uneven or patchy distribution of nutrients (Arredondo & Johnson 1999; Caldwell et al. 1991b; Hodge et al. 2000; Jackson & Caldwell 1991; Robinson 1994; Tibbett 2000; van Vuuren, Robinson & Griffiths 1996) or other abiotic factors such as drought (Reader et al. 1992; Rhizopoulou & Davies 1991). We collected foliar tissues within 24 h after tracer injection. Although exceptionally fast proliferation and nutrient uptake in response to addition of liquid fertilizer has been reported (Jackson & Caldwell 1989; Jackson, Manwaring & Caldwell 1990), plant response to fertile patches in soil usually requires more time – days or weeks rather than hours (Tibbett 2000; van Vuuren et al. 1996). Plant proliferation responses to nutrient availability also require relatively large nutrient patches (Hodge et al. 2000; Jackson & Caldwell 1991; van Vuuren et al. 1996). Therefore it is unlikely that the plants responded to increased supply of N or other abiotic changes resulting from tracer injection. Resource competition or inhibitory mechanisms in mixed communities may control the spatial distribution of roots (Caldwell et al. 1991a; Caldwell et al. 1996; Mahall & Callaway 1991; Mahall & Callaway 1992). Such mechanisms could result in the observed changes in N acquisition when the community complexity increases.
Our data suggest that Achillea and Festuca in monocultures and mixtures differ in the depth of N acquisition. As indicated by our root length data, both species occupied both soil depths. However, when in monoculture Achillea had a greater root length in the upper soil horizon. Plants frequently concentrate roots in the uppermost soil layers (Cook & Ratcliff 1984; Fitter 1986; Mamolos et al. 1995). Nevertheless, species differ in their ability to obtain nutrients from different soil layers (Berendse 1982; Fitter 1986; Mamolos et al. 1995; McKane et al. 1990; Veresoglou & Fitter 1984).
Several factors may explain our observations. First, we do not know to what extent the plants took up 15N as ammonium or nitrate. We have seen previously (Näsholm, Huss-Danell & Högberg 2000) that Phleum and Trifolium can take up 15N from enriched nitrate at this field site. Although we added 15N as ammonium, it is possible that some of it was nitrified during the 24 h period. Nitrification rates at the two soil depths are not known. Second, plant species may differ in their ability to compete for ammonium as well as for nitrate. Third, different plant species may also differ in their rhizosphere flora (Grayston & Campbell 1996; Westover, Kennedy & Kelley 1997) and in the extent to which nitrate can be immobilized in the rhizosphere. Nitrogen acquisition by plants involves more than root length and spatial root distribution.
Fitter (1986) questioned if spatial separation of roots or their activity could be observed in more diverse alluvial grassland communities. Based on the results from natural, species-rich grassland communities, it was concluded that temporal segregation of resource acquisition was essential, whereas there was little support for differentiation in the depth of root activity (Fitter 1986; Veresoglou & Fitter 1984). Our data on Festuca and Achillea support spatial partitioning in N acquisition. It is unclear whether N acquisition would differ within the growing season; the current study provides only a snapshot in time. Our data from the Trifolium and Phleum mixture, however, provide little support for vertical resource partitioning: Trifolium did not acquire N from the deeper horizon in monoculture or in mixture with Phleum. It is possible that Trifolium depends more heavily on its symbiotic N2 fixation in monoculture and in mixed communities, and that its use of soil N is confined to the uppermost soil layer. In support of this, we also observed differences among species in their isotopic composition prior to tracer injection. The initial δ15N data indicated two different patterns: (i) Festuca and Achillea were more enriched in 15N when grown together than they were in monoculture; (ii) when in mixture with Trifolium, Phleum was less enriched in 15N than in monoculture. Furthermore, Phleum had nearly double the foliar N concentration [26·4 vs 14·3 mg g−1 (DW)] in the mixed community with Trifolium when compared to the monoculture. This indicates that N2 fixation associated with Trifolium probably contributed to the N economy of the mixed plant communities.
The results from our study utilizing two sets of monocultures and mixtures suggest that spatial partitioning in N uptake may be important in some plant communities, but not in others. The Achillea and Festuca mixture seemed to support vertical separation in N acquisition, whereas the Phleum and Trifolium mixture showed no such separation, but Trifolium probably relied on its symbiotic N2 fixation, thus avoiding competition for available soil N. In conclusion, complementary resource use may help to maintain species diversity and minimize competition for resources in some plant communities, but plays a lesser role in others.
This project is part of the ‘Biodiversity and Ecosystem Function in Terrestrial Herbaceous Plant Communities’ (BIODEPTH) project funded by the European Commission (Framework IV, Environment and Climate Program, ENV-CT95-0008). Additional funding was generously provided by the Kempe Foundation (Sweden) and the National Science Foundation EPSCoR Grant no. 9874732, with matching support from the State of Kansas. Elisabeth Bärlund, Sanna Forsberg and Sebastian Marklund assisted in various stages of experiment maintenance and sample processing. Håkan Wallmark performed all isotope composition analyses. The authors are grateful for permission to use Dr Cecilia Palmborg's unpublished soil nutrient data.
- 1997) High productivity in grassland ecosystems: effected by species diversity or productive species? Oikos 80, 183–184. (
- 1999) Root architecture and biomass allocation of three range grasses in response to nonuniform supply of nutrients and shoot defoliation. New Phytologist 143, 373–385. & (
- 1979) Competition between plant populations with different rooting depths. I. Theoretical considerations. Oecologia 43, 19–26. (
- 1981) Competition between plant populations with different rooting depths. II. Pot experiments. Oecologia 48, 334–341. (
- 1982) Competition between plant populations with different rooting depths. III. Field experiments. Oecologia 53, 50–55. (
- 1991a) The microscale distribution of neighboring plant roots in fertile soil microsites. Functional Ecology 5, 765–772. , & (
- 1991b) Exploitation of phosphate from fertile soil by three Great Basin perennials when in competition. Functional Ecology 5, 757–764. , & (
- 1996) Species interactions at the level of fine roots in the field: influence of soil nutrient heterogeneity and plant size. Oecologia 106, 440–447. , & (
- 1984) A study of the effects of root and shoot competition on the growth of green panic (Panicum maximum var. trichoglume) seedlings in an existing grassland using root exclusion tubes. Journal of Applied Ecology 21, 971–982. & (
- 1987) Morphological plasticity and mineral nutrient capture in two herbaceous species of contrasted ecology. New Phytologist 107, 453–460. & (
- 1991) Root profiles and competition between the invasive, exotic perennial, Carpobrotus edulis, and two native shrub species in California coastal scrub. American Journal of Botany 78, 885–894. & (
- 1986) Spatial and temporal patterns of root activity in a species-rich alluvial grassland. Oecologia 69, 594–599. (
- 1994) Architecture and biomass allocation as components of plastic response of root systems to soil heterogeneity. Exploitation of Environmental Heterogeneity by Plants (eds M. M.Caldwell & R. W.Pearcy), pp. 305–323. Academic Press, San Diego, CA. (
- 1996) Functional biodiversity of microbial communities in the rhizopheres of hybrid larch (Larix eurolepis) and Sitka spruce (Picea sitchensis). Tree Physiology 16, 1031–1038. & (
- 1998) The effect of diversity on productivity – detecting the role of species complementarity. Oikos 82, 597–599. (
- 1999) Plant diversity and productivity experiments in European grasslands. Science 286, 1123–1127. , , et al. (
- 2000) Competition between roots and soil micro-organisms for nutrients from nitrogen rich patches of varying complexity. Journal of Ecology 88, 150–164. , , , & (
- 1989) The timing and degree of root proliferation in fertile-soil microsites for three cold-desert perennials. Oecologia 81, 149–153. & (
- 1991) Kinetic responses of Pseudoroegneria roots to localized soil enrichment. Plant and Soil 138, 231–238. & (
- 1990) Rapid physiological adjustment of roots to localized soil enrichment. Nature 344, 58–59. , & (
- 1958) The absolute abundance of the nitrogen isotopes in the atmosphere and compressed gas from various sources. Geochimica et Cosmochimica Acta 14, 234–243. & (
- 1991) Root communication among desert shrubs. Proceedings of the National Academy of Sciences, USA 88, 874–876. & (
- 1992) Root communication mechanisms and intracommunity distribution of two mojave desert shrubs. Ecology 73, 2145–2151. & (
- 1995) Depth of root activity of coexisting grassland species in relation to N and P additions, measured using nonradioactive tracers. Journal of Ecology 83, 643–652. , & (
- 1983) Atmospheric nitrogen is a reliable standard for natural delta 15N abundance measurements. Nature 303, 685–687. (
- 1990) Spatiotemporal differences in 15N uptake and the organization of an old-field plant community. Ecology 71, 1126–1132. , & (
- 2000) Uptake of organic nitrogen in the field by four agriculturally important plant species. Ecology 8, 1155–1161. , & (
- 1999) Novel calibration with correction for drift and nonlinear response for continuous flow isotope ratio mass spectrometry applied to the determination of δ15N, total nitrogen, δ13C and total carbon in biological material. Analyst 124, 571–577. & (
- 1976) Underground niche separation in successional plants. Ecology 57, 1281–1288. & (
- 1999) Species-specific spatial pattern of below-ground plant parts in a montane grassland community. Journal of Ecology 87, 569–582. , , & (
- 1992) A comparative study in seedling rooting depth in drying soil. Journal of Ecology 81, 543–550. , , & (
- 1991) Influence of soil drying on root development, water relations and leaf growth of Ceratonia siliqua L. Oecologia 88, 41–47. & (
- 1994) The responses of plants to non-uniform supplies of nutrients. New Phytologist 127, 635–674. (
- SAS (1989) SAS/STAT User's Guide, Vol. 1, Version 6, 4th edn. SAS Institute Inc., Cary, NC.
- 2000) Above-ground resource use increases with plant species richness in experimental grassland ecosystems. Functional Ecology 14, 326–337. , , , & (
- 2000) Roots, foraging and the exploitation of soil nutrients: the role of mycorrhizal symbiosis. Functional Ecology 14, 397–399. (
- 1999) The ecological consequences of changes in biodiversity: a search for general principles. Ecology 80, 1455–1474. (
- 1984) Spatial and temporal patterns of growth and nutrient uptake of five co-existing gresses. Journal of Ecology 72, 259–272. & (
- 1996) Nutrient inflow and root profileration during the exploitation of a temporally and spatially discrete source of nitrogen in the soil. Plant and Soil 178, 185–192. , & (
- 1997) Patterns of rhizosphere microbial community structure associated with co-occurring plant species. Journal of Ecology 85, 863–873. , & (
- 1965) Competition between herbage plants. Netherlands Journal of Agricultural Sciences 13, 212–221. & (
- 1977) Competition and spacing in plant communities: the Arizona upland association. Journal of Ecology 65, 587–595. , & (