Journal of Experimental Botany, Vol. 51, No. 344, pp. 605-615,
March 2000
© 2000 Oxford University Press
Effects on the growth of carrots (Daucus carota L.), cabbage (Brassica oleracea var. capitata L.) and onion (Allium cepa L.) of restricting the ability of the plants to intercept resources
Horticulture Research International, Wellesbourne, Warwick CV35 9EF, UK
Received 5 May 1999; Accepted 25 October 1999
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The objective of this paper is to assess the size and penetration of edge effects in carrot, cabbage and onion field crops and the extent to which these edge effects are modified by the presence of aerial or soil competition between the crop rows. In all three crops, large weight differences developed between the plants in the edge rows and those in the central rows. There was no indication of plant weight fluctuating between large and small values with each successive row in from the edge, as suggested by others. In carrot and onion, edge effects were greatly reduced by the presence of either white reflective aerial partitions or soil partitions, indicating that these species competed for both light and soil resources in UK field conditions. In cabbage, the mere presence of clear aerial partitions between rows reduced edge effects and there was little effect of soil partitions. This indicates the predominance of shoot over root competition in this species. The differences between species are possibly related to the architectural flexibility of their shoots. These results suggest that, within crops, carrot and onion plants compete for light over a distance of about 20 cm in each direction and for below-ground resources over a distance of about 50 cm in each direction. For cabbage, interactions between plants appeared to be dominated by the requirement for sufficient space to deploy the shoots for efficient light interception.
Key words: Carrot, cabbage, onion, edge effects, competition, soil, light.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
As plants grow, they interact with and modify their surrounding environment and when plants are sufficiently close to one another they compete for resources causing a decrease in growth rate (Aspinall and Milthorpe, 1959
). Hence, competition is a major force behind the appearance and life history of plants, and the structure and dynamics of plant communities (Grace and Tilman, 1990).
The most direct method of investigating the relative importance of above- and below-ground resources in the competition between plants is by the use of shoot and/or root partitions. This approach has been used extensively (Donald, 1958; Aspinall, 1960; Schreiber, 1967; Snaydon, 1971; Rennie, 1974; Gamboa and Vandermeer, 1988), but all these studies used plants grown in pots. The use of pots has been criticized as it will always emphasize root competition (Harper, 1977). Partitions in the soil have been used in field studies to discriminate between soil competition and aerial competition. Nearly all these studies have investigated the competition between different species, using soil partitions. For example, metal plates were placed vertically between rows of a field intercrop to determine the importance of competition between roots systems of alfalfa and orchardgrass (Chamblee, 1958). Vertical polythene soil barriers have been similarly used for pearl millet and groundnut intercrops (Willey and Reddy, 1981). The relative importance of root and shoot competition between individual plants in ryegrass monocrops was determined by inserting metal tubes in the soil around seedlings to exclude root competition (Seager et al., 1992). These authors combined the root competition exclusion treatments with reduced shoot competition by trimming the leaves of the sward surrounding ryegrass seedlings.
Investigating edge effects is an alternative approach in determining the factors for which plants compete when grown in the field. An edge effect is the phenomenon in which a plant attains greater weight when growing adjacent to a gap (Salter et al., 1980). In effect, this plant has been partially released from competition. There have been a number of studies to quantify edge effects (Bleasdale, 1963; Thompson and Taylor, 1976; Austin and Blackwell, 1980; Salter et al., 1980; Hadjichristodoulou, 1983; Wright et al., 1986; Williams-Linera, 1990), but to our knowledge there have been no investigations as to their underlying cause.
The contribution of above- and below-ground competition to edge effects was investigated in field-grown plants by using aerial and soil partitions. The experiment was repeated for species of contrasting shoot and root morphologies: carrot, cabbage and onion, in separate years. Carrots and onions form below-ground storage organs and have flexible, long leaves. Plants were established at high densities to ensure that competition was intense and to give enough material to be harvested. The densities were within the range used commercially for onions and carrots, but approximately 10 times commercial densities for cabbage. The assumption was taken that competitive effects observed at high densities were representative of competition seen at lower densities at a later growth stage.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Three experiments were conducted, in separate years, in apparatus designed to investigate the importance of competition for above- and below-ground resources. Each apparatus formed a plot and contained no partitions, only aerial partitions, only soil partitions, or both. There were two types of aerial partition (clear or white polythene) and one type of soil partition (steel). Including the no partition control in each medium, six combinations of aerial and soil partition treatments were obtained. Figure 1a
shows the apparatus design, which was sunk into a trench to a depth of 40 cm (the soil partition height) with the partitions running east–west. Frames were levelled prior to back filling with shredded soil. The perforated steel base plate and plastic poles were installed in all plots (Fig. 1b), so that the drainage, and shading due to the poles, would be similar in all treatments. After construction and soil re-placement, in each experiment, the areas were allowed to settle for 2 weeks prior to planting.
|
Twelve plots, comprising of two replicates of each of the six treatments were randomly assigned within three blocks (A, B, C) in the experimental area (Fig. 1b). This allowed for two harvests to be taken per treatment per block if required. Two metre gaps were left between plots, and group C was sited 2 m east of group A (Fig. 1b). The experiment was surrounded by a windbreak. Each plot contained nine rows of plants, with a between-row spacing of 10 cm, arranged so as to isolate both adjacent rows and the outer rows from the surrounding environment when partitions were present.
The intention was that each row was to be recorded separately at harvest to assess the size and extent of edge effects within the different treatments. Therefore, each experiment was effectively arranged as a split-plot design, with partition treatments applied to main plots and ‘proximity to edge’ treatments being applied to rows (sub-plots) as there was systematic arrangement of sub-plot treatments within the main plots.
Aerial partitions were secured to two horizontal rods such that the top rod could be adjusted to raise the partition in order to stay level with the top of the growing crop (Fig. 1a). In this way, the top of the crop was in full sunlight and undesirable shield effects were minimized (Warren and Lill, 1975). The clear aerial partition treatment was imposed to prevent leaf overlap between rows of the crop, but allowed within-row leaf overlap and light to penetrate the crop. The white aerial partitions were imposed to prevent leaf overlap between rows and also to reduce the amount of direct, lateral light received by each row of crop, but would ensure each row received reflected and diffuse light. Comparison of these aerial treatments would determine whether the physical presence of barriers, and/or the incident light entering the crop, would affect crop growth edge effects.
The steel soil-partitions restricted the intermingling of roots of plants from different rows and prevented roots of edge row plants from extending into the soil between plots. An odd number of rows was chosen so that any inducement of alternating large and small plants in successive positions across the plot (as suggested by Vandermeer, 1986), would be re-enforced between the two edge rows across the plot. Nine rows were used to mimic the typical grouping that might produce edge effects, for example, in a bed system of crop production. In the analyses, the rows were numbered one to nine from south to north (Fig. 1a). Prior to each crop planting, nitrogen was applied at 80 kg ha-1, to each plot area.
Statistical methods
The aim of this study was to determine whether difference in mean weight occurred between outer and inner rows within plots, and how these differences were affected by the removal or reduction of root and/or shoot competition. The size of such an ‘edge effect’ can most easily be estimated by the ratio of mean weights in outer to inner rows, a value greater than unity indicating an advantage for the outer rows. Prior to analysis all dry weight data were subjected to a loge transformation to satisfy the ANOVA assumption of homogeneity of variance. An additional advantage of this transformation is that treatment differences on the transformed scale relate directly to the ratios described above.
To allow assessment of the size and extent of edge effects, the variation due to differences between rows was partitioned using a set of specific, orthogonal contrasts (Table 1). These contrasts allowed for the different replication of outer and inner rows by weighting the data according to the numbers of rows of each type. The nested set of contrasts was considered to be the most appropriate and efficient approach for the assessment of how far the edge effects penetrated the crop. An alternative approach using polynomial contrasts might have provided more information about the overall shape of the response to distance from the middle row, but would not facilitate the determination of the extent of penetration of any edge effect. In addition to the contrasts shown in Table 1, potential north–south differences were assessed using contrasts comparing each pair of outer rows (1 versus 9, 2 versus 8, 3 versus 7, and 4 versus 6). The significance of each single degree-of-freedom contrast, and hence the importance of each effect, was determined from the analysis of variance summary (Genstat 5 Committee, 1993).
|
The variation due to differences between the three aerial partition treatments was also partitioned using orthogonal contrasts, though a different set of contrasts was used in the analysis of the cabbage experiment from that used for the other two crops. For the carrot and onion crops it was anticipated that the restriction of light entering the crop would have the greater effect. The variation between aerial partition treatments was therefore partitioned using contrasts comparing the mean of the Nil and Clear aerial partition treatments with the White aerial partition treatment and comparing the Nil aerial partition treatment with the Clear partition treatment. However, for cabbage it was anticipated that the removal of physical competition between rows would have the greater effect. As a result, for this crop the variation was partitioned using contrasts comparing the Nil aerial partition treatment with the mean of the Clear and White aerial partition treatments, and comparing the Clear aerial partition treatment with the White aerial partition treatment. The modification of edge effects through the use of aerial and/or soil partitions was assessed through the interactions between the edge effect contrasts and the aerial partition contrasts described above and/or the soil partition factor. Again, the significance of each single degree-of-freedom contrast was determined from the analysis of variance summary.
Shoot:root dry weight ratios and percentage dry matter in the shoots of cabbages, storage roots of carrots and bulbs of onions were also subjected to analysis of variance. For onion, shoot:root dry weight ratio was subjected to a loge transformation prior to analysis.
Experiment 1
Carrot cv. Marathon pregerminated seed were sown into all plots of areas A, B, and C on 19 May 1994, 26 May 1994, and 9 June 1994, respectively. Fifty seeds were sown in each row, 2 cm apart, and subsequently thinned down or gapped-up to 30 plants per row.
Single plots of each treatment were harvested after 123 d and 166 d, (area A); 116 d and 159 d, (area B); and 102 d and 145 d, (area C). The outermost five plants at each end of each row were discarded in order to reduce/eliminate within-row edge effects. For each row, the number of plants remaining, and the distance they occupied per row, was recorded, prior to harvest. Plants were lifted and bulked by row, into prelabelled bags. Shoots and storage roots were separated, weighed, dried at 80 °C for 48 h, and then dry weights of plant parts per row were recorded.
Any plants that had been transplanted into rows to make up the stands were harvested separately and the number and weight of plant parts (fresh and dry) were recorded. In the analysis of data for this crop, covariate terms for day degrees (accumulated mean temperature per day above a base of 0 °C from sowing) and number of transplants were used. This was to account for the different sowing to harvest times for the three replicate areas and the number of transplants necessary to achieve 30 plants m-1 of row, respectively.
Experiment 2
Spring cabbage cv. Myatts Offenham Compacta seed was sown into modular trays, filled with Levington M2 compost (14 cm3 compost cell-1) and grown for 4 weeks in a glasshouse. Sufficient plants were transplanted into all plots on 3 May 1995 to allow for two harvests per treatment in each replicate area. Plant spacing within each row was 10 cm and no thinning was necessary for this crop, so the plants formed a square grid in each plot.
Immediately after transplanting, partitions were raised to approximately 10 cm to meet the top of the tallest plant per plot. Only one harvest was made for this experiment (110 d after transplanting) as the crop had started to senesce and decay by the time of the second scheduled harvest. The outermost plant at each end of each row was removed at harvest to reduce within-row edge effects. As the soil was very friable due to being sieved into all treatment plots, rows were lifted with a garden fork and the retrieved roots were taken as representative of the root system. Fresh and dry weights for leaf, stem and root were recorded for each bulked row, along with the number of harvested plants per row.
Experiment 3
Onion cv. Marco seed was sown into modular trays filled with Levington M2 compost (14 cm3 compost cell-1), and grown for 9 weeks in a glasshouse prior to transplanting into all plots of areas A and C on 19 April 1996. This allowed for two repeat plots per treatment in two replicate areas to be harvested at one time. Transplants were placed into the plots at 10 cm within-row spacing so the plants formed a square grid in each plot. The aerial partitions were then raised to the height of the tallest plant in each plot area (approximately 7 cm).
There was one scheduled harvest for this experiment, 103 d after transplanting, giving two plots per treatment per area. The procedure at harvest was the same as that followed in experiment 2. For each row, total fresh weight was recorded. In addition two plants per bag were removed and separated into leaf, pseudo-stem, bulb, and roots. Fresh and dry weights for these plant parts were then recorded.
Mean organ dry weight, 
d,p, was calculated for each row as-
 |
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Carrot
Mean shoot and mean storage root dry weights were greater in rows at or close to the crop edge for all treatments (Fig. 2
). Averaging across partition treatments, the ratio of mean weight in rows 1 and 9 to mean weight in rows 2–8 was significantly greater than unity for both shoots and storage roots (Tables 2, 3). Furthermore, the edge effects penetrated to benefit the growth of shoots and storage roots in rows 2 and 8, these rows producing significantly larger mean shoot dry weights and mean root dry weights than rows 3–7 (Tables 2, 3).
|
) No aerial partition present; (
) clear aerial partitions; (
) white aerial partitions; (
) no soil partitions; (
) soil partitions present. Left vertical bar is LSD
,0.05 for comparisons between treatments. Right vertical bar is LSD
|
|
The extent of the edge effects was affected by the presence of partitions between the rows (Fig. 2; Table 2). The presence of the white aerial partitions reduced the edge effect between rows 2 and 8 and their inner rows (3–7) compared with the nil and clear aerial partitions (Table 2). The presence of clear aerial partitions increased the edge effects for rows 1 and 9 and for rows 2 and 8 in comparison with the nil aerial partition treatment (Table 2; Fig. 2a). A similar pattern of responses was found for the effects of aerial partitions on edge effects for mean storage root dry weights. The steel soil-partitions significantly reduced the edge effects of rows 1 and 9 for mean shoot dry weight (Table 2) and for mean storage root dry weight (Table 3).
The presence of aerial and soil partitions had some interesting but statistically non-significant (P>0.05) effects on the mean shoot:root ratios and percentage dry matter. White aerial partitions increased overall shoot : root ratios in the carrot crop (Table 4) but both aerial partitions reduced percentage dry matter in the storage roots. The presence of soil partitions decreased shoot : root ratios in the carrot crop, but increased percentage dry matter in carrot storage roots (Table 4).
|
Onion
Mean shoot, bulb and root dry weights were greater in edge rows than in central rows. The magnitude of these edge effects was generally greater where partitions were absent (Fig. 3).
|
For shoot dry weight, the edge effect for rows 1 and 9 was smaller in the presence of white aerial partitions than for the nil and clear aerial partition treatments (Table 5a
). This edge effect was also significantly smaller for the clear aerial partition treatment than for the nil treatment (Table 5a). The presence of soil partitions reversed the edge effect for shoot dry weight, the ratio of mean shoot dry weight in rows 1 and 9 to that in rows 2–8 being less than unity (Table 5a). There was a complex interaction for ratio of mean shoot dry weights in rows 1 and 9 to rows 2–8 (F1,143 df=8.16; P=0.005). When soil partitions were present there was no effect of aerial partitions, whilst when soil partitions were absent, the presence of white aerial partitions eliminated the edge effect seen for the nil and clear aerial partition treatments (Table 5b).
|
|
The presence of clear or white aerial partitions generally reduced mean bulb weights, the reduction in weight being much greater when white partitions were present (Fig. 3b). Averaging over all partition treatments, the ratios of mean bulb weight in rows 1 and 9 to that in rows 2–8, and in rows 2 and 8 to rows 3–7 were significantly greater than unity (Table 6
). The sizes of these edge effects were not significantly altered by the different aerial partition treatments. When soil partitions were present, the sizes of the edge effect for both rows 1 and 9 and 2 and 8 were significantly reduced in comparison with the nil soil partition treatment (Fig. 3e; Table 6).
|
The presence of either the clear or white aerial partitions significantly reduced mean root weight overall (Fig. 3c), whilst the presence of steel partitions increased the mean root weight over the entire crop (Fig. 3f). Across all partition treatments the ratio of mean root weight in rows 1 and 9 to that in rows 2–8 was significantly greater than unity (P<0.001, data not shown). There were no interactions between the aerial and soil partition effects and the magnitude of this edge effect (P>0.05, data not shown).
The presence of either aerial partition increased shoot : root dry weight ratios but the presence of the steel soil-partition reduced these ratios (Table 7). The bulb percentage dry matter was greatest in the absence of aerial partitions and least in the presence white aerial partitions (Table 7). There was a similar significant reduction in onion shoot percentage dry matter in response to the presence of aerial partitions (P<0.001, data not shown). Soil partitions had no effect on the percentage dry matter in bulbs of onions (Table 7).
|
Cabbage
Mean shoot and mean root dry weights were greater in outer rows than central rows for all partition treatments, the edge effects penetrating to rows 2 and 8 for shoot dry weight (Tables 8, 9a). The size of these edge effects tended to be greater where partitions were absent (Fig. 4; Tables 8, 9a), the notable exception to this being for mean shoot dry weight in the presence of soil partitions (Fig. 4c). There were no apparent differences in the sizes of the edge effects for the white and clear aerial partitions.
|
|
|
A strong interaction was present for the edge effects in rows 1 and 9 (F1,94 df=8.67; P=0.004) and 2 and 8 (F1,94 df=10.24; P=0.002) for mean root dry weights (Table 9b
). In both cases the edge effect was much greater when no partitions were present than when either aerial partitions (clear or white) or steel soil partitions or both were present (Table 9b).
|
The presence of aerial partitions increased the shoot : root ratio (P=0.001), whilst the presence of the steel soil partition reduced this ratio (P=0.002) (Table 10
). The presence of aerial or soil partitions increased percentage dry matter in cabbage shoot material (Table 10), but this trend was not statistically significant (P>0.05).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The work presented here investigates whether aerial or soil factors affect edge effects by using soil and/or aerial partitions between rows of crop. This is similar to the work of other authors (Chamblee, 1958
; Corlett et al., 1992) who examined the mechanisms of competition between component species in intercrops using soil partitions. The authors are not aware of any study in monocrops of the causal mechanism of edge effects. The results presented here allow an estimate of the distance over which plants capture resources and hence give an estimate of the extent of the space over which individual plants interact with their neighbours, that is their ‘ecological neighbourhood’ (Mack and Harper, 1977). This is an important distinction between this study and those who used soil or aerial partitions in pot experiments.
It has been described how, for a single row of tomatoes, the second plant from an end grew poorly due to intense competition from the end plant (Vandermeer, 1986). In contrast, the third plant from the end benefited from the presence of a small competitor plant in this second position. It was reported that this phenomenon generated a wave of alternating large and small plants along the row of crop, but with decreasing amplitude. This experiment suggests that plant interactions are so local that there is no benefit of being close to the end of a row.
In this study, edge effects were present in all three species despite the use of species differing widely in shoot and root morphology. Findings presented here are strikingly different from those of Vandermeer. In particular, the smallest plants in Vandermeer's experiment were those one plant in from the end. In the experiments presented here, plants in the rows, one from the edge (rows 2 and 8), derived some benefit from being close to the edge. A similar effect has been reported in wheat and barley (Hadjichristodoulou, 1983). This suggests that interactions between plants extended further than their immediate neighbours.
To distinguish between light and other physical effects, two types of aerial partition were used in this study. If the effect of aerial partitions had been only on shelter effects, then the yield of crops in the clear aerial partition treatments would have exceeded that of the no aerial partition control (Marshall, 1967). There was no evidence for such an increase in growth. The lack of a shelter effect may be because all treatments had optimal shelter from the windbreak surrounding the entire experiment (Fig. 1b). For carrot shoots and storage roots and onion leaves, white polythene aerial partitions reduced the size of edge effects with a smaller reduction occurring in the presence of the clear polythene aerial partitions. A simple explanation is that competition for light was responsible, at least in part, for the edge effects and the small effect due to clear partitions was due to the small shading effect they produced. Penetration of edge effects to the penultimate edge rows, and its elimination by the presence of white aerial partitions, indicates that (i) these crops compete for light; and (ii) this competition operates over a distance greater than 10 cm. The penetration of edge effects into crops has been observed previously (Hadjichristodoulou, 1983). The studies presented here show that competition for light plays a role in the generation of edge effects and that light can be responsible for the penetration of the edge effect even in crops of uniform height. This has not been demonstrated in previous experiments.
For cabbage, the presence of either clear or white aerial partitions reduced shoot and root weights in edge rows. The presence of aerial partitions might have prevented optimal orientation of leaves for light interception in cabbage, which has large, rigid leaves. Hence, for cabbage, but not in the other two species, plants compete for space for optimal display of their leaves for light utilization. It has been similarly reported that plant yield is reduced when the aerial environment is fragmented by the presence of glass cylinders in the crop, especially in species that are architecturally inflexible (McConnaughay and Bazzaz, 1992b).
Edge effects were also reduced for onion leaves and bulbs, and for carrot in the presence of soil partitions, indicating that these species compete for below-ground resources under UK field conditions. Division of the soil medium without reducing volume causes reduction in root weight (McConnaughay and Bazzaz, 1992a). However, in cabbage and onion, the presence of soil partitions increased root weight, even in inner rows. Hence, these plants responded to the presence of soil partitions as though soil volume had been restricted (Fuleky and Nooman, 1991). If roots of individual plants in no soil-partition treatments spread laterally in the order of 50 cm in each direction even the plants of the innermost rows could ‘tap’ the resources of the gaps between plots. The imposition of soil partitions between rows would cause a disruption of the ability of root systems of plants in all rows to acquire resources. In the study presented here, this resulted in elevated percentage dry matter of plant parts (carrot roots and cabbage shoots) and root weight in response to the presence of soil partitions.
The partition treatments also influenced the partitioning of dry matter between shoot and fibrous roots. For cabbage and onion the responses were consistent with the concept of a functional equilibrium between these two organs or with Thornley's model based on carbon and nitrogen uptake and transport (Wilson, 1988). That is, white aerial partitions, which caused shading, increased shoot : root dry weight ratios, whereas soil partitions, which restricted root functioning, decreased these ratios. These results would also be explained by the soil partitions influencing endogenous hormone production by the root system (McDavid et al., 1973). Presented data do not allow the different hypotheses of control of shoot : root ratios (Wilson, 1988) to be distinguished.
The partitioning of dry matter between the shoots and storage organs of onion (the bulb) and of carrot (the root) was intriguing. Imposition of white aerial partitions or soil partitions reduced carbon gain by the plants; the white aerial partitions because they reduced light levels slightly; the soil partitions because they restricted the lateral spread of the root system. If partitioning of assimilate between shoot and storage organ was determined by carbon supply, then these two treatments would have similar effects upon shoot to storage organ weight ratio. In fact, white aerial partitions increased this ratio, whereas, soil partitions decreased this ratio. This suggests that the partitioning of assimilates is controlled from a mechanism sensitive to the soil environment, which is separate to the one sensitive to the radiation regime investigated previously (Hole and Dearman, 1993).
This study used partitions to determine whether above- and below-ground factors are responsible for edge effects. Such partitions have also been used to investigate separately the effects of plant density in the aerial and soil environments (Snaydon, 1979). The use of partitions to investigate competition has been criticized because aerial partitions can alter the microclimate (Warren and Lill, 1975). In this study, undesirable shield effects were minimized by increasing the height of the aerial partitions to match the height of the crop during the growing season. The use of pots and soil partitions gives an artificial restriction to the volume of soil available for rooting which overestimates the importance of root competition (Harper, 1977). In this study, the height of soil partitions above the base was 40 cm (Fig. 1b) so the volume of soil available for rooting was unlikely to be limiting. Furthermore, the base upon which the soil partitions were attached, was perforated to allow drainage and for roots to grow through. Potentially, roots had the capacity of growth through these perforations and therefore to re-establish edge effects. This seems implausible given the depth to which the bases were sunk, but the apparatus design is likely to underestimate the importance of soil factors as sources of edge effects.
In conclusion, the results of work presented here indicate that, in carrot and onion, competition between individual plants is for both above- and below-ground resources. The interactions between plants for above-ground resources are predominantly with their immediate neighbours, whereas interactions between plants for below-ground resources is probably with many neighbours and is unlikely to be confined only to the immediate ones.
For cabbage, their rigid shoot architecture confers a requirement for sufficient space to allow maximum light utilization.
|
Acknowledgments |
|---|
We thank MAFF for funding this work and Mr Steven Coggins, Miss Ann Hellemans, Dr Bindawa Awalu, and Dr Vincent Tenebe for their technical assistance. Part of this work was written for a Statistics project based at Coventry University, by Lindsey Peach and we thank Mrs C Wright for her invaluable help during that time.
|
Notes |
|---|
1 To whom correspondence should be addressed. Fax: +44 1789 470552. E-mail:laurence.benjamin{at}hri.ac.uk
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Aspinall D.1960. An analysis of competition between barley and white persicaria. II. Factors determining the course of competition. Annals of Applied Biology 48, 637–654.
Aspinall D, Milthorpe FL.1959. An analysis of competition between barley and white persicaria. I. The effects of growth. Annals of Applied Biology 47, 156–172.
Austin RB, Blackwell RD.1980. Edge and neighbour effects in cereal yield trials. Journal of Agricultural Science 94, 731–734.
Bleasdale JKA.1963. The bed-system of carrot growing. Ministry of Agriculture Fisheries and Food, Short term leaflet No. 27.
Chamblee DS.1958. Some above- and below-ground relationships of an alfalfa—orchardgrass mixture. Agronomy Journal 50, 434–437.
Corlett JE, Black CR, Ong CK, Monteith JL.1992. Above- and below-ground interactions in a Leucaena/millet alley cropping system. I. Experimental design, instrumentation and diurnal trends. Agricultural and Forest Meteorology 60, 53–72.
Donald CM.1958. The interaction of competition for light and nutrients. Australian Journal of Agricultural Research 9, 421–435.
Fuleky G, Nooman HJ.1991. Effects of soil volume on root growth and nutrient uptake. In: McMichael BL, Persson H, eds. Plant roots and their environment. Elsevier Science Publishers bv, 446–448.
Gamboa W, Vandermeer JH.1988. Root and shoot interactions between Phaseolus vulgaris L. and Cyperus rotundus. Tropical Ecology 29, 21–25.
Genstat 5 Committee.1993. Design and analysis of experiments. Genstat 5 release 3 Reference Manual. Oxford, UK: Clarendon Press, 461–538.
Grace JB, Tilman D. (eds.) 1990. Perspectives on plant competition. San Diego: Academic Press Inc.
Hadjichristodoulou A.1983. Edge effects on yield, yield components and other traits in mechanised durum wheat and barley trials. Journal of Agricultural Science 101, 383–387.
Harper JL.1977. Population biology of plants. New York: Academic Press.
Hole CC, Dearman J.1993. The effect of photon flux density on distribution of assimilate between shoot and storage root of carrot, red beet and radish. Scientia Horticulturae 55, 213–225.
Mack RN, Harper JL.1977. Interference in dune annuals:spatial pattern and neighbourhood effects. Journal of Ecology 65, 345–363.[Web of Science]
Marshall JK.1967. The effect of shelter on the productivity of grasslands and field crops. Field Crop Abstracts 20, 1–14.
McConnaughay KDM, Bazzaz FA.1992a. The occupation and fragmentation of space: consequences of neighbouring roots. Functional Ecology 6, 704–710.[Web of Science]
McConnaughay KDM, Bazzaz FA.1992b. The occupation and fragmentation of space: consequences of neighbouring shoots. Functional Ecology 6, 711–718.
McDavid CR, Sagar GR, Marshall C.1973. The effect of root pruning and 6-benzylaminopurine on the chlorophyll content, 14CO2 fixation and shoot/root ratio in Pisum sativum L. New Phytologist 72, 465–470.
Rennie JC.1974. Some effects of competition and density of plants on dry weight produced. Annals of Botany 38, 1003–1012.
Salter PJ, Currah IE, Fellows JR.1980. Further studies on the effects of plant density, spatial arrangement and time of harvest on yield and root size in carrots. Journal of Agricultural Science 94, 465–478.
Schreiber MM.1967. A technique for studying weed competition in forage legume establishments. Weeds 15, 1–4.
Seager NG, Kemp PD, Chu ACP.1992. Effect of root and shoot competition from established hill-country pasture on perennial ryegrass. New Zealand Journal of Agricultural Research 35, 359–363.
Snaydon RW.1971. An analysis of competition between plants of Trifolium repens L., populations collected from contrasting soils. Journal of Applied Ecology 8, 687–697.
Snaydon RW.1979. A new technique for studying plant interactions. Journal of Applied Ecology 16, 281–286.
Thompson R, Taylor H.1986. Plant competition and its implications for cultural methods in calabrese. Journal of Horticultural Sciences 51, 147–157.
Vandermeer J.1986. A computer-based technique for rapidly screening intercropping designs. Experimental Agriculture 22, 215–232.
Warren JF, Lill WJ.1975. The influence of aerial shields on plant growth environment in competition experiments. Annals of Botany 39, 1141–1142.
Willey RW, Reddy MS.1981. A field technique for separating above- and below-ground interactions in intercropping: an experiment with pearl millet/groundnut. Experimental Agriculture 17, 257–264.
Williams-Linera G.1990. Vegetation structure and environmental conditions of forest edges in Panama. Journal of Ecology 78, 356–373.
Wilson JB.1988. A review of evidence on the control of shoot : root ratio, in relation to models. Annals of Botany 61, 433–449.
Wright D, Ali SS, Hughes Ll G.1986. Edge effects in cereal yield trials. Journal of the National Institute of Agricultural Botany 17, 179–186.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||