SEARCH

SEARCH BY CITATION

Keywords:

  • branching;
  • compaction;
  • lateral root;
  • Musa acuminata (banana);
  • root diameter;
  • root system architecture

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • • 
    The relative importance of root system structure, plant carbon status and soil environment in the determination of lateral root diameter remains unclear, and was investigated in this study.
  • • 
    Banana (Musa acuminata) plants were grown at various moderate levels of soil compaction in two distinct experiments, in a field experiment (FE) and in a glasshouse experiment (GE). Radiant flux density was 5 times lower in GE. The distribution of root diameter was measured for several root branching orders.
  • • 
    Root diameters ranged between 0.09 and 0.52 mm for secondary roots and between 0.06 and 0.27 mm for tertiary roots. A relationship was found between the diameter of the parent bearing root and the median diameter of its laterals, which appears to be valid for a wide range of species. Mean lateral root diameter increased with distance to the base of the root and decreased with branching density [number of lateral roots per unit length of bearing root (cm−1)].
  • • 
    Typical symptoms of low light availability were observed in GE. In this case, lateral root diameter variability was reduced. Although primary root growth was affected by soil compaction, no effects on lateral root diameter were observed.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Root diameter is probably the root morphological characteristic that conveys the most information on root structural and functional properties. Apical diameter, which can be considered an indicator of root meristem size (Barlow & Pilet, 1984), has been related to axial growth (Wilcox, 1962; Hackett, 1973; Coutts, 1987; Cahn et al., 1989), growth duration (Cahn et al., 1989; Pagès, 1995; Eissenstat et al., 2000), root anatomy (Varney et al., 1991; Jordan et al., 1993), water transport properties (Varney & Canny, 1993; Vercambre et al., 2002), respiratory activity (Bidel et al., 2000), penetration ability (Misra et al., 1986) and gravitropism (Le Roux & Pagès, 1996). Root diameter is also sensitive to carbon nutrition (Thaler & Pagès, 1996a,b; Muller et al., 1998) and is affected by competition for carbon (Thaler & Pagès, 1996b, 1997).

The diameters of the lateral roots (secondary roots arising from nodal roots or taproots called primary roots, or tertiary roots arising from secondary roots) of a given species are highly variable (Yorke & Sagar, 1970; Lamond et al., 1983; Cahn et al., 1989; Varney et al., 1991; Jordan et al., 1993; Thaler & Pagès, 1996b) as a consequence of (i) the highly structured organization of root systems which results in characteristic root branching orders and (ii) the high degree of plasticity of root systems (Fitter, 2002; Waisel & Eshel, 2002). Root diameter variability for a given root order is increased in a heterogeneous soil environment (Jordan et al., 1993; Waisel & Eshel, 2002). One reason for this is that roots, depending on their position, do not encounter the same soil conditions. Another reason is that root morphological responses to unfavourable soil conditions are not independent of the environment experienced by the other roots of the system and their growth response to this environment. The results of several experimental studies have shown that compensatory growth caused by carbon redistribution occurs in unevenly compacted soils (Gersani & Sachs, 1992; Bingham & Stevenson, 1993; Thaler & Pagès, 1997; Mulholland et al., 1999; Montagu et al., 2001; Bingham & Bengough, 2003). Moreover, low carbon availability affects lateral roots more than primaries (Tester et al., 1986; Buttery & Stone, 1988; Rogers et al., 1992; Tatsumi et al., 1992; Bingham & Stevenson, 1993; Thaler & Pagès, 1996a; Muller et al., 1998). When primary roots are subjected to soil compaction or are truncated, lateral root diameters increase (Schuurman, 1965; Hackett, 1971; Crosset et al., 1975; Lamond et al., 1983; Thaler & Pagès, 1997), and this response is probably attributable to increased carbon availability for laterals (Thaler & Pagès, 1997). Overall, carbon partitioning into the entire root system is at least partly dependent on allocation from the shoots (Aguirrezabal et al., 1993; Farrar & Jones, 2000). Therefore, the distribution of lateral root diameters throughout the root system is the consequence of soil heterogeneity, root system structure, carbon availability and carbon partitioning amongst different roots. However, the relative contribution of each of these determinants remains largely unknown. On a given bearing root, lateral root diameters should vary within a determinate range, but this raises questions about how this range varies within the root system architecture, and how it is influenced by the environment. We therefore investigated, in two different experiments and at three levels of soil compaction, the variability of lateral root diameters within a banana (Musa acuminata) root system. The objectives were to quantify ranges of root diameters for the different root orders; to correlate variability in diameter to other architectural descriptors; and to evaluate the influence of moderate soil compaction on this variability.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Site and plant material

Two experiments, a field experiment (FE) and a glasshouse experiment (GE), were performed. The first experiment, FE, was carried out from December 1999 to March 2000 at the Centre Internationale de Recherche Agronomique pour le Développement (CIRAD) experimental farm in Neufchâteau, in Guadeloupe, French West Indies. The second experiment, GE, was also carried out in Guadeloupe, from June 2000 to August 2000, at the Institut National de la Recherche Agronomique (INRA) experimental domain in Petit Bourg. FE was carried out in a standard banana plantation with a plant density of 1890 plants ha−1. A complete description of the methods can be found elsewhere (Lecompte et al., 2003). GE was conducted in a glasshouse. In this experiment, 12 banana trees were planted in large square containers (2 m × 2 m × 0.5 m depth), separated into two compartments by a central polyvinylchloride (PVC) partition to obtain appropriate conditions for a split root system analysis. A semicircular hole, of 7.5 cm radius, was cut half-way across this central wall, at the top, where the banana corm was positioned. In both experiments the same clone of banana, Musa acuminata Colla cv. Grande Naine (AAA genome, Cavendish subgroup), was used.

Soil preparation and compaction

In both experiments, the soil used was an andosol (dystric andosol, World Reference Base classification). In FE, the field was subdivided into three subplots of equal size (20 m × 30 m). The soil was compacted by repeated passes of a tractor over the whole surface of the soil. The tractor was driven so that wheel tracks were contiguous across the subplots. Treatments were applied to the subplots as follows: one was left uncompacted (NC); one was compacted by two passes over the entire plot surface by a tractor weighing 3000 kg, with a tyre pressure of 70 kPa (medium compaction, MC); the last was compacted by five passes by a heavier tractor (5500 kg) with a 1200-kg axle load and a tyre pressure of 200 kPa (high compaction, HC). In both MC and HC treatments, tractor speed was 5 km h−1. Because of the compaction method used, it would have been impossible to generate, in a unique field, randomly distributed subplot replicates, and therefore only three large subplots corresponding to one treatment each were used. In the GE experiment, containers were filled with soil collected from the 0–40 cm layer of a field adjacent to the one used for the FE experiment. The soil was passed through a 2.5 cm × 2.5 cm sieve. The soil in the containers was compacted with a hydraulic press pushing a 1 m × 0.5 m metallic plate lying flat on the soil. A 25-cm-deep soil layer was added to each compartment (1 m × 2 m; i.e. half of the container), compacted using a four-step procedure, each step corresponding to a 1 m × 0.5 m (the metallic plate surface) soil strip. Once the whole layer was compacted, a second soil layer was added over the first layer, and compacted using the same method. Pressures were applied at three levels: 0 kg cm−2 (control, NC), 0.5 kg cm−2 (medium compaction, MC) and 1 kg cm−2 (high compaction, HC). The 12 containers allowed twofold replication of a combination of two levels of soil compaction (NC-NC, NC-MC, NC-HC, MC-HC, MC-MC and HC-HC).

In both experiments, soil cores (518 cm3) were collected at the positions of randomly selected primary root apices. In FE, cores were collected on three different sampling dates, and, on each date, for three different plants (see the Root sampling and measurements section). In GE, all containers were sampled. Between 16 and 26 soil cores were collected in each treatment in the two experiments. Soil was dried at 105°C for 72 h and mean bulk density was calculated for the different treatments in the two experiments. Air-filled porosity was also calculated, assuming a constant particle density of 2.41 g cm−3 for this soil, as mentioned in Dorel et al. (2000). In FE, no significant ‘plant’ or ‘date’ factor appeared from a hierarchical analysis of variance conducted to examine the relative variability of the different factors, and means were calculated by pooling data from the different subsamples, as for GE with measurements from the different containers. Bulk densities are given in Table 1. A two-way analysis of variance (ANOVA) indicated that the compaction treatments significantly affected soil bulk density (P < 0.0001) and that there were no differences between experiments (P = 0.14). Additional pairwise tests were undertaken (Table 1). The differences between compaction levels were highly significant in both experiments. The bulk density of NC soil in GE was lower than that in FE because the field soil in GE was collected, mixed and passed through a sieve before filling the containers. For the same reason, air-filled porosity was different in the NC treatments of GE (32%) and FE (24%). For the other treatment, air-filled porosities were similar between experiments (mean values of 17% and 11% for MC and HC, respectively), and higher than the threshold value of 10% under which it is generally considered that root growth can be affected. The field trial was rainfed, while the plants in the glasshouse experiment were irrigated. Soil water potential was monitored with tensiometers placed at 15, 40 and 60 cm depth in the FE experiment and at 10, 20 and 40 cm in the GE experiment, with two replicates per treatment (NC, MC and HC) and per depth. The mean soil water potential remained between −5 and −25 kPa in the FE experiment, except during a short 10-d period in which surface water potential fell to −50 kPa in the HC treatment. Apart from this period, soil compaction did not affect soil water status to a significant extent. In the GE experiment, minimum soil water potential was maintained above −30 kPa. It is assumed that these high water potentials did not restrict plant and root growth.

Table 1.  Means (± standard variations) of soil bulk density for the field (FE) and the glasshouse split root (GE) experiments, at three levels of soil compaction (not compacted, NC; moderate compaction, MC; high compaction, HC)
 Soil bulk density (g cm−3)
HCMCNC
  1. Different letters indicate significant differences for comparisons of means in repeated one-way analyses of variance (ANOVAs), with Duncan's test at the 1% significance level (n = 16–26 depending on group). The first letter corresponds to the comparison of means in a row, and the second letter to the comparison of means in a column.

FE0.82 ± 0.04 a,a0.77 ± 0.05 b,a0.67 ± 0.06 c,a
GE0.83 ± 0.05 a,a0.75 ± 0.07 b,a0.64 ± 0.04 c,b

Weather conditions and plant growth

Mean daily temperatures were fairly stable although slightly different in the two experiments: 22.6 ± 0.9°C (mean ± standard error) in the FE experiment and 27.0 ± 0.8°C in the GE experiment. Degree days were calculated using a base temperature of 14°C. Photosynthetic photon flux density (in the 400–700 nm waveband) was measured with a quantum sensor (Li-190; Li-Cor Inc., Lincoln, NE, USA). Mean cumulative daily radiation was 5 times higher in the FE experiment (29.2 ± 7.4 mol m−2 d−1) than in the GE experiment (5.8 ± 1.3 mol m−2 d−1). Plant leaf areas were recorded every 15 d in the FE experiment on five randomly selected plants in each treatment and every 5 d on all plants in GE.

Root sampling and measurements

Root sampling methods were identical in the two experiments. In FE, three series of excavations were made for each subplot at monthly intervals; for each subplot (NC, MC and HC) at each sampling time, three banana trees were sampled. A total of 27 plants were excavated (3 replicates × 3 treatments × 3 sampling dates), between 410 and 980 degree-days after planting. In GE, all 12 plants were excavated at random, one plant per day, between 621 and 817 degree-days. Leaf, pseudo-stem and corm were dried at 70°C for 72 h and then weighed. The total number of primary roots produced on each corm was counted. For each plant, 5 ± 1 (in FE) or 20 ± 3 (in GE) randomly selected primary nodal roots were excavated from the base to the apex by gently removing the surrounding soil. A sample of secondary and tertiary roots was also excavated. The position of each sampled lateral root was recorded as the distance between the point of insertion of the root and the point of insertion of the most distal root on the primary root. The distributions of root positions in the sample were very similar for all treatments in each experiment and between experiments, with 80% of the secondary roots located less than 30 cm away from the most distal root. Tertiary roots were subsampled from the secondary root sample, mainly within the first 10 cm from the most distal lateral. A total of 1220 (FE) and 1290 (GE) secondary roots were used for the analysis, and 764 and 162 tertiary roots were sampled in FE and GE, respectively. In each case, roots were evenly distributed between the compaction treatments. Once excavated, primary root segments bearing intact secondary and tertiary roots were immediately transferred to a diluted ethanol solution (1 : 5 volume:volume) and stored at 5°C. Necrotic primary roots whose apices were damaged were discarded, together with their laterals. Excavated root segments were subdivided into 2-cm-long root segments, and the positions of these segments were registered as the distance (cm) from the base of the root, i.e. for primary roots the distance from the point of insertion on the corm, and for secondary roots the distance from the point of insertion of the primary root. The length and apical diameter of each intact lateral of these segments were measured. The apical diameter was measured in the conical part of the apical region of the root, at one-third of the total distance between the root cap and the basal section of the cone, as described in Lecompte (2002). Lateral root apical diameters were measured with a monocular lens equipped with a reticule, with a precision of 0.01 mm (Cole Parmer, Vernon Hills, IL, USA; magnification ×60). Another lens (Leica, Bensheim, Germany; magnification ×5) was used for primary roots, whose diameters were determined with a precision of 0.05 mm. All laterals, whether sampled or not, were cut at their base, so that every lateral root branching point was visible. The root branching density (cm−1) was calculated for each 2-cm-long segment, as the number of branches per cm of bearing root. Primary root growth rate (cm d−1) was calculated as a linear function of the length of the apical unbranched zone, as detailed in Lecompte et al. (2001).

Statistical analysis

Apical diameter distributions for different root orders, experiments and compaction treatments were examined. The Kolmogorov–Smirnov test (see e.g. Hollander & Wolfe, 1973) was used for the comparisons of distributions of paired subgroups. This test relies on the calculation of the maximum deviations between the empirical distribution functions (EDFs) and on the probability of observing a larger value under the null hypothesis of equal EDFs. Because of the large amount of data collected, a 1% level of significance was used for group comparisons. Standard t-tests or other nonparametric tests, depending on data characteristics, were also used for comparing means. Because of the lack of subplot replications in FE, it was assumed in the analysis that no factors other than compaction and its consequences affected root growth in a different way in each treatment. All tests, regression analyses and analyses of variance were performed using SAS software procedures (SAS, Cary, NC, USA).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Shoot growth and primary root number

Expressed as degree-days, the increase in leaf area was steeper in GE than in FE (Fig. 1). This was accompanied by a higher specific leaf area in GE [198 ± 5 cm−2 g−1 leaf dry weight (d. wt)] than in FE (150 ± 14 cm−2 g−1 leaf d. wt). Significant differences were found between leaf areas in NC and the other treatments, from 400 degree-days to the end of FE (Fig. 1). There was a lower variability of shoot growth in GE, with no consistent significant effects of treatments. Leaf weight ratio (g leaf d. wt g−1 shoot d. wt) was higher in GE (0.54 vs 0.50 in FE; P = 0.03), but mean shoot biomass in GE was not significantly different from that in FE (calculated for banana trees excavated at the same times as those excavated in GE, between 620 and 820 degree-days; P = 0.27; data not shown). The number of primary roots was lower in GE than in FE (56.4 vs 64.5; P = 0.02).

image

Figure 1. Mean plant leaf areas as a function of degree-day sum (base temperature 14°C), for the field experiment (FE; solid line) and the glasshouse experiment (GE; dashed line), at various degrees of soil compaction: not compacted (NC; square), moderate compaction (MC; triangle) and high compaction (HC; circle). Combinations of treatments in the GE split containers were grouped as follows: NC-NC and NC-MC were considered equivalent to NC, MC-MC and NC-HC equivalent to MC and HC-HC and MC-HC equivalent to HC. Points are means of five (in FE) or four (in GE) replicates. Letters indicate significant differences between treatments in a given experiment; irrespective of their order in the plot, the upper letter always indicates NC, the middle letter MC and the lower letter HC.

Download figure to PowerPoint

Apical diameter and branching order

Apical diameter ranges for the different root orders in both experiments are shown in Fig. 2. Mean apical diameters were 1.46, 0.21 and 0.12 mm for primary, secondary and tertiary roots, respectively. None of the diameter distributions was normally distributed. Primary and lateral roots formed distinct, unmixed groups: the 1% percentile for primary root diameter distribution (0.6 mm) was higher than the 99% percentile for secondary and tertiary root diameter distributions (respectively, 0.52 and 0.27 mm). Secondary roots were significantly thicker than tertiary roots. However, the thinnest roots in each group order had comparable diameters: the thinnest secondary roots were 0.09 mm while the thinnest tertiary roots were 0.06 mm. Although the absolute diameter range was higher for primary roots, the coefficient of variation was nearly twice as high for secondary roots (40.1%) as for primary roots (22.5%), and 50% higher for tertiary roots (34.7%) than for primary roots. No significant difference was found between primary root diameter EDFs in the two experiments (P = 0.09; Kolmogorov–Smirnov test). Although not significant, two peaks in the diameter distribution at around 1.4 and 1.8 mm were perceptible in FE, but not in GE (illustrated later in Fig. 6a,b). Conversely, secondary root diameters were significantly lower in GE compared with FE (Fig. 2; P < 0.0001). Tertiary root diameters were lower still in GE. The range of variation of secondary root diameters was slightly higher in FE, with the same lower limit (0.09 mm) but a higher upper limit (0.86 mm in FE vs 0.74 in GE). The lack of thick roots in GE was clearer for tertiary roots, with diameters ranging from 0.06 to 0.32 mm in FE and 0.05 to 0.14 mm in GE. Overall variability in lateral root diameter was thus lower in GE.

image

Figure 2. Box plots of root apical diameters for primary (I), secondary (II) and tertiary (III) roots, in the field experiment (FE) and the glasshouse experiment (GE). Pooled data from all compaction treatments are shown. Bars join the 1st and 99th percentiles. The bottom and top edges of the box are located at the 25th and 75th percentiles. The central horizontal line is drawn at the median.

Download figure to PowerPoint

image

Figure 6. Distribution of root apical diameters at different levels of soil compaction [high compaction (HC; circle), moderate compaction (MC; triangle) and not compacted (NC; square)] for primary roots (a and b), secondary roots (c and d) and tertiary roots (e and f) in the field experiment (FE; a, c and e) and the glasshouse experiment (GE; b, d and f).

Download figure to PowerPoint

Variations in apical diameter within the architecture

There was a significant correlation between lateral and bearing root diameters, indicating a tendency for thick roots to bear larger laterals. This result is summarized in Fig. 3, where median lateral root diameters are plotted against bearing root diameters. Distinct regression lines were found for secondary roots (bearing root diameter approx. > 0.6 mm) and tertiary roots (bearing root diameter approx. < 0.6 mm). Testing for heterogeneity of slopes showed no difference between slopes in GE and FE, either for secondary (P = 0.42) or for tertiary roots (P = 0.4). Therefore, a model of covariance with median lateral root diameter as the response variable, bearing root diameter as the covariate and experiment as a group variable was undertaken for secondary and tertiary roots. Both models were highly significant with r2 = 0.39 and r2 = 0.56 for secondary and tertiary roots, respectively. A linear model accounting for the three factors studied here (Yijk = µ + αi + βj + xk + ɛijk, were Yijk is the median lateral root diameter of root k, µ is the constant, αi is the effect of root type i, βj is the effect of experiment j, and xk is the diameter of root k) explained 71% of median diameter variance.

image

Figure 3. Median lateral root apical diameter vs bearing root apical diameter (n = 148; each mean corresponds to on average 30 lateral roots on the same bearing root). Closed circles and solid lines: the field experiment (FE); open circles and dashed lines: the glasshouse experiment (GE). Lines represent predicted values from covariance models with bearing root diameter as the covariate and experiment as the group variable, for secondary (thick lines) and tertiary (fine lines) lateral roots.

Download figure to PowerPoint

Variability around the median lateral root diameter for a given bearing root was still large, and root diameter variability was higher for secondary than for tertiary roots [coefficient of variation (cv) of 28% for secondary roots vs 19% for tertiary roots; P < 0.0001]. Apical diameter variability along a bearing root was more pronounced in FE (cv = 31% for secondary roots and 20% for tertiary roots) than in GE (cv = 25% and 17%, respectively). Much of this variability could be detected on smaller root segments, because the mean coefficient of variation of lateral root diameter along short 2-cm-long root segments of bearing roots was still 19% for secondary roots and 17% for tertiary roots. Because there were large variations in the segment positions and the branching densities of the segments in the sample of cut primary roots, the influence of these two factors on secondary root apical diameter was tested. Mean branching density on primary roots was significantly lower (P < 0.0001) in GE (5.08 branches cm−1 parent root) than in FE (7.69 branches cm−1 parent root). The median apical diameter of the segments was positively correlated with their distance to the base of the root and negatively correlated with their branching density (Fig. 4). Similar tendencies were found in both experiments, so only results for pooled data are illustrated. An analysis of variance on these pooled data showed that both the factors distance and density, and their interaction, were significant (P < 0.0001 for the three factors). The increase with distance to the base of root diameter was less pronounced at high branching densities. Similarly, the decrease with branching density of root diameter was less pronounced at low distances. Note that these two variables are correlated, because branching density decreased exponentially with position along the root. At large distances from the base of the root, lateral root diameter was also more variable: the coefficient of variation of apical diameter increased almost linearly from 17% at the base of the root to 32% at more than 150 cm from the base of the root (data not shown).

image

Figure 4. Surface response of median secondary root apical diameter to branching density [number of lateral roots per centimetre of bearing root (cm−1)] and distance to the base of the primary root (cm). Branching density, distance, and median apical diameter were measured on 2-cm-long segments along primary roots. Pooled data from the field experiment (FE) and glasshouse experiment (GE) are shown.

Download figure to PowerPoint

Response of root diameter to soil compaction

Root elongation was reduced by 33% and 24% in FE and by 48% and 36% in GE in the HC and MC compaction treatments, respectively (Fig. 5). Subsampling effects (sampling date and plant) in FE were not significant. Roots in the compacted treatments of both experiments were tortuous and, when growing in cracks, flattened along the side in contact with the compacted soil, thus showing typical symptoms of impeded roots. Root growth in the uncompacted treatment was significantly less in FE than in GE, while the mean elongations achieved by impeded roots were comparable. Compaction did not affect primary root apical diameter distribution in FE (Fig. 6a). However, in GE, the EDF for the HC treatment was significantly different (P = 0.002; Kolmogorov–Smirnov test) from the EDF for NC, and slightly different, although not significantly so at the 1% level (P = 0.017), from the EDF for MC (Fig. 6b). The mean apical diameter was less in HC than in NC, irrespective of the associated treatment in the remaining half-container. A similar tendency was observed for elongation: root growth rate in each compaction treatment was not affected significantly by the compaction level encountered by the other half of the root system (data not shown).

image

Figure 5. Mean primary root growth rate (cm d−1) for the two experiments [the field experiment (FE; solid bars) and the glasshouse experiment (GE; cross-hatched bars)] at three levels of soil compaction: high compaction (HC; dark grey); moderate compaction (MC; light grey); not compacted (NC; black). Error bars represent the 99% confidence limit (n = 43–55 depending on group). Significant differences within a compaction treatment were assessed by a Wilcoxon rank sum test at the 1% level, and are indicated by an asterisk.

Download figure to PowerPoint

Lateral diameter distributions for secondary and tertiary roots in FE and GE are shown in Fig. 6c–f. As shown earlier, a larger proportion of thick laterals was found in FE. No significant differences in EDF were found between compaction treatments for secondary roots in FE (P = 0.12, 0.17 and 0.97 in paired comparisons between, respectively, HC and NC, MC and NC, and HC and MC, with all data pooled as subsampling factors showing no significant effects). Whereas in GE primary roots were thinner in the HC treatment than in the NC treatment, secondary root diameters were lower in HC than in NC (P = 0.007; data not shown). The absence of specific effects of soil compaction on secondary root EDFs was confirmed by an analysis of residuals in the model of covariance presented in Fig. 3. Once the variability accounting for the bearing root diameter was removed, no significant effect of soil treatment on lateral root diameter was found on residuals, either for secondary roots in GE or for tertiary roots in either experiment.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Banana root apical diameters ranged between 0.6 mm (percentile 1%) and 2.1 mm (percentile 99%) for primary roots, between 0.09 and 0.52 mm for secondary roots and between 0.06 and 0.27 for tertiary roots. Few data have been published on banana trees allowing comparisons with the diameters reported in this study. Apical diameters were measured at the conical end of the roots in order to obtain an estimate of meristem size. Conversely, diameters reported in the literature, even distal diameters, are usually measured at least several millimetres away from the root cap. However, our apical diameter values can be accurately converted into distal values using a power function regression, as proposed by Lecompte (2002). The values then obtained ranged between 2.2 and 4.3 mm for primary roots, between 0.3 and 1.5 mm for secondary roots and between 0.2 and 0.9 mm for tertiary roots. These values for primary roots are in agreement with those reported by Draye (2002) for three banana cultivars, but slightly lower than the ranges of 5–10 mm reported in earlier studies (Acquarone, 1930; Riopel & Steeves, 1964). A diameter decrease along successive root orders is inherent in the structure of root systems and has been used for the description of their fractal properties (Van Noordwijk et al., 1994). However, an original relationship between the diameter of bearing roots and the mean diameter of their laterals has been found here. Two distinct linear relationships were computed for secondary and tertiary roots; the slopes of the regression lines were not significantly different in the contrasting environments of the two experiments presented here. However, they differed in their intercept values. If data from both experiments are considered together, the order-dependent regression lines can be combined into a single power function grouping two successive root orders, still with a high coefficient of determination (R2 = 0.62). From an examination of the literature on root diameters, it appears that such a relationship would hold for a large range of species. Computing published data on the diameter of the three first root orders of Acer saccharum (Pregitzer et al., 1997), Fraxinus americana (Pregitzer et al., 1997), Viola pubescens (Pregitzer et al., 1997), Hydrophyllum canadense (Pregitzer et al., 1997), Hordeum vulgare (Hackett, 1968), Coix lacryma (Iijima & Kono, 1991), Oryza sativa (Iijima & Kono, 1991; Kimura & Yamazaki, 2001), Pisum sativum (Tsegaye & Mullins, 1994), Sorghum bicolour (Iijima & Kono, 1991), Zea mays (Donald et al., 1987; Cahn et al., 1989; Iijima & Kono, 1991; Jordan et al., 1993) and Hevea brasiliensis (Le Roux & Pagès, 1994), and adding the data on Musa accuminata presented here, a good regression between the mean diameter of a parent root and the mean diameter of laterals can be obtained, with an R2 of 0.85 (Fig. 7). The regression is still very good (R2 = 0.79), and estimated parameters are comparable, when not considering the isolated point corresponding to the thick primary roots of Hydrophyllum canadense. Thus, there could be a tendency for species with thick primary roots to bear relatively thin secondary roots, while the ratio between the mean diameters of two successive root orders tends to be lower as the root order increases.

image

Figure 7. Mean lateral root diameter vs bearing root diameter for secondary and tertiary laterals of 12 species. Data for Acer saccharum, Fraxinus americana, Viola pubescens, and Hydrophyllum canadense are from Pregitzer et al. (1997); those for Coix lacryma and Sorghum bicolour are from Iijima & Kono (1991); those for Oryza sativa are from Iijima & Kono (1991) and Kimura & Yamazaki (2001); those for Zea mays are from Donald et al. (1987), Cahn et al. (1989), Iijima & Kono (1991) and Jordan et al. (1993); those for Hordeum vulgare are from Hackett (1968); those for Pisum sativum are from Tsegaye & Mullins (1994); those for Hevea brasiliensis are from Le Roux & Pagès (1994), and those for Musa acuminata are from this study. In the case of the work of Cahn et al. (1989) on maize (Zea mays), mean diameter values by root order were not available and were estimated from histogram distributions. The solid line represents a power function fitted to the overall dataset (P < 0.0001, r2 = 0.85), and the dotted line the 1 : 1 line.

Download figure to PowerPoint

Although lateral root diameters are partly determined by the diameter of their bearing roots, considerable residual variability was observed. Large variations in lateral diameters have also been observed for pea (Pisum sativum; Yorke & Sagar, 1970), maize (Zea mays; Varney et al., 1991; Jordan et al., 1993), oak (Quercus robur; Pagès, 1995) and ruber tree (Hevea brasiliensis; Thaler & Pagès, 1996b). Interestingly, root diameters were less variable in the glasshouse experiment. In GE, radiant flux density was 5 times lower than in FE. Shoot biomasses were similar in the two experiments, but specific leaf area was lower in FE. There was also strong evidence that root biomass in GE was lower, as primary root number, primary root growth, branching density and lateral diameters were lower in GE. A parallel increase in leaf weight ratio and decrease in specific leaf area (Boardman, 1977; Dijkstra, 1989; Sinclair & Muchow, 1999), a decrease in shoot:root ratio (Aguirrezabal et al., 1993; Farrar & Jones, 2000), and a decrease in root branching density (Aguirrezabal & Tardieu, 1996; Freixes et al., 2002) and root diameter (Thaler & Pagès, 1996b; Muller et al., 1998) are typical plant responses to low light availability. As the other main factors that can affect root development (air and soil temperature, soil water content, and soil aeration) were similar in the two experiments, it can be hypothesized that the differences in root diameter observed between FE and GE are, at least partly, attributable to the lower radiation in GE, probably resulting in a lower assimilate availability for the root system. It is interesting to note that diameter decreases were more pronounced for roots of higher order. These roots could be penalized in at least two ways in the case of assimilate deficiency: directly because less resources are attributed to them, as shown in several studies (Tester et al., 1986; Buttery & Stone, 1988; Rogers et al., 1992; Tatsumi et al., 1992; Bingham & Stevenson, 1993; Thaler & Pagès, 1996a; Muller et al., 1998), and indirectly because, as shown here, a decrease in the diameter of the parent roots will result in a decrease in the diameter of their laterals. This case is illustrated in GE, where a decrease, compared with FE, in secondary root diameters was accompanied by a sharper decrease in tertiary root diameters. Higher root diameters and higher variability were also found in parts of the root system architecture where root branching density was low. This could also have been a result of lower competition for carbon when the number of lateral roots per unit length is decreasing. Aguirrezabal et al. (1994) found that secondary root growth rate decreased from the base to the apex of the sunflower taproot. Conversely, we found here a highly significant positive correlation between median secondary root apical diameter and distance from the base of the root. As higher apical diameters are generally associated with enhanced growth (Wilcox, 1962; Hackett, 1973; Coutts, 1987; Cahn et al., 1989; Pagès, 1995; Thaler & Pagès, 1996a), we could expect from this study an increase in lateral root elongation with distance to the base of the root. Unfortunately, as most lateral roots remained unbranched, root growth rate could not be estimated from a static indicator such as the length of the apical unbranched zone, as was done for primary roots. Nevertheless, a parameter such as the distance from the base of the root can, depending on the species and experimental conditions, be correlated with many other internal and environmental factors, so that variation with distance to the base of root diameters or growth rates could be largely circumstantial.

Soil compaction resulted in a maximal reduction of primary root elongation of 48%, accompanied by typical visual symptoms of mechanical impedance. The subplots corresponding to the different compaction treatments were not replicated in FE. Therefore, statistical conclusions cannot be drawn from these field data. However, it is very likely that these differences in root growth were caused by compaction per se; furthermore, the specific effect of compaction was ascertained in GE. No increase in primary root diameter was noted in impeded roots, perhaps because soil compaction was only moderate, as also noted by Tsegaye & Mullins (1994), and/or because this increase is generally a result of an increase in the number and diameter of mature cortical cell files (Wilson et al., 1977; Atwell, 1988; Croser et al., 1999), while in this study we measured root diameters very near the apex, in the meristematic region. Conversely, in GE, primary root apical diameters were smaller in the HC treatment than in the less compacted soils. Such differences were perceptible in each split root treatment of GE, while they did not appear in FE. Therefore, this decrease in primary root apical diameter might have been attributable to a cumulative effect of soil compaction and low light availability. Neither in FE nor in GE were lateral root diameters significantly affected by soil compaction. Many authors have reported enhanced lateral root growth in compacted soils when only primary roots are impeded (Schuurman, 1965; Hackett, 1971; Goss, 1977; Lamond et al., 1983; Donald et al., 1987; Misra & Gibbons, 1996; Thaler & Pagès, 1997). This compensatory response has recently been related to an increase in carbon availability for laterals (Thaler & Pagès, 1997). A similar absence of secondary root growth response has, however, also been reported by Tsegaye & Mullins (1994). Compensation seems therefore more easily illustrated with variables such as total root weight or length, accumulated for many roots, rather than with a morphological variable such as root length or root diameter averaged over a population, especially when considering a mature root system with several thousands of lateral roots. In our study, additional available carbon as a result of diminished primary root growth in MC and HC could have been ‘diluted’ into many insignificant morphological changes in the pool of secondary roots.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors wish to thank Thierry Bajazet, Philippe Artis and Daniel Taupe for their technical assistance in the construction and maintenance of the split root containers, Muriel Blin for her valuable help in collecting root samples in the glasshouse experiment, Magalie Wuillaume for her comments on the manuscript and Alan Scaife for correcting the English manuscript. This work received a financial contribution from the Centre International de Recherche Agronomique pour le Développement (CIRAD).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Acquarone P. 1930. The roots of Musa sapienta L. United Fruits Company Research Bulletin 26: 831868.
  • Aguirrezabal LAN, Deleens E, Tardieu F. 1994. Root elongation rate is accounted for by intercepted PPFD and source-sink relations in field and laboratory grown sunflower. Plant, Cell & Environment 17: 443450.
  • Aguirrezabal LAN, Pellerin S, Tardieu F. 1993. Carbon nutrition, root branching and elongation: can the present state of knowledge allow a predictive approach at a whole plant level? Environmental and Experimental Botany 33: 121130.
  • Aguirrezabal LAN, Tardieu F. 1996. An architectural analysis of the elongation of field grown sunflower root systems. Elements for modelling the effects of temperature and intercepted radiations. Journal of Experimental Botany 47: 411420.
  • Atwell BJ. 1988. Physiological responses of lupin roots to soil compaction. Plant and Soil 111: 277281.
  • Barlow PW, Pilet PE. 1984. The effect of abscisic acid on cell growth, cell division and DNA synthesis in the maize root meristem. Physiologia Plantarum 62: 125132.
  • Bidel LPR, Renault P, Pages L, Riviere LM. 2000. Mapping meristem respiration of Prunus persica (L.) Batsch seedlings: potential respiration of the meristems, O2 diffusional constraints and combined effects on root growth. Journal of Experimental Botany 51: 755768.
  • Bingham IJ, Bengough AG. 2003. Morphological plasticity of wheat and barley roots in response to spatial variation in soil strength. Plant and Soil 250: 273282.
  • Bingham IJ, Stevenson EA. 1993. Control of root growth: effects of carbohydrates on the extension, branching, and rate of respiration of different fractions of wheat roots. Physiologia Plantarum 88: 149158.
  • Boardman NK. 1977. Comparative photosynthesis of sun and shade plants. Annual Review of Plant Physiology 28: 355377.
  • Buttery BR, Stone JA. 1988. Some effects of nitrate and light intensity on root growth and development. Plant and Soil 106: 291294.
  • Cahn MD, Zobel RW, Bouldin DR. 1989. Relationship between root elongation rate and diameter and duration of growth of lateral roots of maize. Plant and Soil 119: 271279.
  • Coutts MP. 1987. Developmental processes in tree root systems. Canadian Journal of Forestry Research 17: 761767.
  • Croser C, Bengough AG, Pritchard J. 1999. The effect of mechanical impedance on root growth in pea (Pisum sativum). I. Rates of cell flux, mitosis, and strain during recovery. Physiologia Plantarum 107: 277286.
  • Crosset RN, Campbell DJ, Stewart HE. 1975. Compensatory growth in cereal root systems. Plant and Soil 42: 673683.
  • Dijkstra P. 1989. Cause and effect of differences in specific leaf area. In: LambersH, CambridgeML, KoningsH, PonsTL, eds. Causes and consequences of variation in growth rate and productivity of higher plants. The Hague, the Netherlands: SPB Academic Publishing, 125140.
  • Donald RG, Kay BD, Miller MH. 1987. The effect of soil aggregate size on early shoot and root growth of maize (Zea mays L.). Plant and Soil 103: 251259.
  • Dorel M, Roger-Estrade J, Manichon H, Delvaux B. 2000. Porosity and soil water properties of Caribbean volcanic ash soil. Soil Use and Management 17: 118.
  • Draye X. 2002. Consequences of root growth kinetics and vascular structure on the distribution of lateral roots. Plant, Cell & Environment 25: 14631474.
  • Eissenstat DM, Wells CE, Yanai RD, Whitbeck JL. 2000. Building roots in a changing environment: implications for root longevity. New Phytologist 147: 3342.
  • Farrar JF, Jones DL. 2000. The control of carbon acquisition by roots. New Phytologist 147: 4353.
  • Fitter AH. 2002. Characteristics and functions of root systems. In: WaiselY, EshelA, KafkafiU, eds. Plant roots: the hidden half. New York, NY, USA: Marcel Dekker, 1534.
  • Freixes S, Thibaud MC, Tardieu F, Muller B. 2002. Root elongation and branching is related to local hexose concentration in Arabidopsis thaliana seedlings. Plant, Cell & Environment 25: 13571366.
  • Gersani M, Sachs T. 1992. Development correlations between roots in heterogeneous environments. Plant, Cell & Environment 15: 463469.
  • Goss MJ. 1977. Effects of mechanical impedance on root growth in barley (Hordeum vulgare L.). I. Effects on the elongation and branching of seminal root axes. Journal of Experimental Botany 28: 96111.
  • Hackett C. 1968. A study of the root system of barley. I. Effects of nutrition on two varieties. New Phytologist 67: 287300.
  • Hackett C. 1971. Relations between the dimensions of the barley root system: effect of mutilating the root axes. Australian Journal of Biology Science 24: 10571064.
  • Hackett C. 1973. A growth analysis of the young sorghum root system. Australian Journal of Biology Science 26: 12111214.
  • Hollander M, Wolfe DA. 1973. Non parametric statistical methods. New York, NY, USA: Wiley.
  • Iijima M, Kono Y. 1991. Inter specific differences of the root system structures of four cereal species as affected by soil compaction. Japanese Journal of Crop Science 60: 130138.
  • Jordan MO, Harada J, Bruchou C, Yamazaki K. 1993. Maize nodal root ramification: absence of dormant primordia, root classification using histological parameters and consequences on sap conduction. Plant and Soil 153: 125143.
  • Kimura K, Yamazaki S. 2001. Root length and diameter measurement using NIH image: application of the line-intercept principle for diameter estimation. Plant and Soil 234: 3746.
  • Lamond M, Tavakol R, Riedacker A. 1983. Influence d’un blocage de l’extrémité du pivot d’un semis de chêne, sur la mophogénèse de son systeme racinaire. Annales des Sciences Forestières 40: 227250.
  • Le Roux Y, Pagès L. 1994. Développement et polymorphisme racinaire chez de jeunes semis d’hévéa (Hevea brasilensis). Canadian Journal of Botany 72: 924932.
  • Le Roux Y, Pagès L. 1996. Réaction géotropique des différents types de racines chez l’hévéa (Hevea brasiliensis). Canadian Journal of Botany 74: 19101918.
  • Lecompte F. 2002. Mise en place de l’architecture racinaire du bananier dans un andosol: effets des contraintes physiques liées au compactage. PhD thesis. Paris, France: Institut National Agronomique Paris-Grignon.
  • Lecompte F, Ozier-Lafontaine H, Pagès L. 2001. The relationship between static and dynamic variables in the description of root growth. Consequences for field interpretation of rooting variability. Plant and Soil 236: 1931.
  • Lecompte F, Ozier-Lafontaine H, Pagès L. 2003. An analysis of growth rates and direction of growth of primary roots of field-grown banana trees in an andisol at three levels of soil compaction. Agronomie 23: 209218.
  • Misra RK, Dexter AR, Alston AM. 1986. Maximum axial and radial growth pressures of plant roots. Plant and Soil 95: 315326.
  • Misra RK, Gibbons AK. 1996. Growth and morphology of eucalypt seedling-roots, in relation to soil strength arising from compaction. Plant and Soil 182: 111.
  • Montagu KD, Conroy JP, Atwell BJ. 2001. The position of localized soil compaction determines root and subsequent shoot growth responses. Journal of Experimental Botany 52: 21272133.
  • Mulholland BJ, Hussain A, Black CR, Taylor IB, Roberts JA. 1999. Does root-sourced ABA have a role in mediating growth and stomatal responses to soil compaction in tomato (Lycopersicon esculentum)? Physiologia Plantarum 107: 267276.
  • Muller B, Stosser M, Tardieu F. 1998. Spatial distributions of tissue expansion and cell division rates are related to irradiance and to sugar content in the growing zone of maize roots. Plant, Cell & Environment 21: 149458.
  • Pagès L. 1995. Growth patterns of the lateral roots of young oak (Quercus robur) tree seedlings. Relationship with apical diameter. New Phytologist 130: 503509.
  • Pregitzer KS, Kubiske ME, Yu CK, Hendrick RL. 1997. Relationships among root branch order, carbon, and nitrogen in four temperate species. Oecologia 111: 302308.
  • Riopel JL, Steeves TA. 1964. Studies on the root of Musa acuminata cv Gros Michel. I. The anatomy and development of main roots. Annals of Botany 28: 475495.
  • Rogers CE, Peterson CM, McCrimmon JN, Cure JD. 1992. Response of plant roots to elevated atmospheric CO2. Plant, Cell & Environment 15: 749752.
  • Schuurman JJ. 1965. Influence of soil density on root development and growth of oats. Plant and Soil 22: 352374.
  • Sinclair TR, Muchow RC. 1999. Radiation use efficiency. Advances in Agronomy 65: 215265.
  • Tatsumi J, Endo N, Kono Y. 1992. Root growth and partitioning of 13C labelled photosynthate in the seminal root of corn seedlings as affected by light intensity. Japanese Journal of Crop Science 61: 271278.
  • Tester M, Smith SE, Smith FA, Walker NA. 1986. Effects of photon irradiance on the growth of shoots and roots, on the rate of initiation of mycorrhizal infection and on the growth of infection units in Trifolium subterraneum L. New Phytologist 103: 375390.
  • Thaler P, Pagès L. 1996a. Periodicity in the development of the root system of young rubber trees (Hevea brasiliensis Müell. Arg.): relationship with shoot development. Plant, Cell & Environment 19: 5964.
  • Thaler P, Pagès L. 1996b. Root apical diameter and root elongation rate of rubber seedlings (Hevea brasiliensis) show parallel responses to photoassimilate availability. Physiologia Plantarum 97: 365371.
  • Thaler P, Pagès L. 1997. Competition within the root system of rubber seedlings (Hevea brasiliensis) studied by root pruning and blockage. Journal of Experimental Botany 48: 14511459.
  • Tsegaye T, Mullins CE. 1994. Effect of mechanical impedance on root growth and morphology of two varieties of pea (Pisum sativum L.). New Phytologist 126: 707713.
  • Van Noordwijk M, Spek LY, De Willigen P. 1994. Proximal root diameter as predictor of total root size for fractal branching models. I. Theory. Plant and Soil 164: 107117.
  • Varney GT, Canny MJ. 1993. Rates of water uptake into the mature root system of maize plants. New Phytologist 123: 775786.
  • Varney GT, Canny MJ, Wang XL, McCully ME. 1991. The branch roots of zea. I. First order branches, their number, sizes, and division into classes. Annals of Botany 67: 357364.
  • Vercambre G, Doussan C, Pagès L, Habib R, Pierret A. 2002. Influence of xylem development on axial hydraulic conductance within Prunus root systems. Trees: Structure and Function 16: 479487.
  • Waisel Y, Eshel A. 2002. Functional diversity of various constituents of a single root system. In: WaiselY, EshelA, KafkafiU, eds. Plant roots: the hidden half. New York, NY, USA: Marcel Dekker, 157174.
  • Wilcox HE. 1962. Growth studies of the root of incense cedar, Librocedrus decurrens II. Morphological features of the growth system and root behaviour. American Journal of Botany 49: 237245.
  • Wilson AJ, Robarts AW, Goss MJ. 1977. Effects of mechanical impedance on root growth in barley (Hordeum vulgare L.). II. Effects on cell development in seminal roots. Journal of Experimental Botany 28: 12161227.
  • Yorke JS, Sagar GR. 1970. Distribution of secondary root growth potential in the root system of Pisum sativum. Canadian Journal of Botany 48: 699704.