Agronomy Journal Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (11)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Liedgens, M.
Right arrow Articles by Richner, W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Liedgens, M.
Right arrow Articles by Richner, W.
Agricola
Right arrow Articles by Liedgens, M.
Right arrow Articles by Richner, W.
Related Collections
Right arrow Spatial Distribution
Right arrow Maize
Right arrow Plant Analysis
Right arrow Root Development
Agronomy Journal 93:1097-1104 (2001)
© 2001 American Society of Agronomy

ROOT DEVELOPMENT

Minirhizotron Observations of the Spatial Distribution of the Maize Root System

Markus Liedgens*,a and Walter Richnerb

a ETH Zürich, Inst. of Plant Sci., FEL, Eschikon 33, CH-8315 Lindau, Switzerland
b ETH Zürich, Inst. of Plant Sci., LFW A4, ETH Zentrum, Universitätstasse 2, CH-8092 Zürich, Switzerland

* Corresponding author (markus.liedgens{at}ipw.agrl.ethz.ch)

Received for publication November 29, 1999.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The vertical and horizontal distribution of maize (Zea mays L.) roots was studied using minirhizotrons in drainage lysimeters for 3 yr. Ten minirhizotrons (60-mm o.d.) were placed horizontally at depths of 5 to 100 cm, perpendicular to the maize row. Root density (roots cm-2) on minirhizotron images (2.43 cm2) was observed at leaf developmental stages 3, 6, 9, and 12 and at pollen shed. Root density increased to a maximum at 25-cm depth and decreased at greater depths. This pattern was observed in all years and at all developmental stages except for early in the season. The density of roots decreased with increasing distance from the plant row. Soil depth influenced root density more than the distance from the plant row, and its pattern was more complex. Root density was influenced by an interaction between both factors. Significant interactions of the spatial components of root density with maize developmental stage, but not with years, were identified although years strongly influenced maize leaf area. These results suggest that there is a basic pattern of maize root distribution in the soil, which is modified, but not fundamentally changed, by the ability of the roots to adapt to varying environmental conditions. Our results also indicated that the maize crop can explore soil resources only to a limited extent at early developmental stages, in deep soil layers, and at increasing distances from the plant row.

Abbreviations: DP, deep percolation • ETP, potential evapotranspiration • LA, leaf area • LDS, leaf developmental stage • PS, pollen shed • SWS, soil water storage


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE SPATIAL DISTRIBUTION of roots in the soil determines the ability of a crop to tap soil nutrients and water necessary to sustain plant growth. Although information on root distribution is fundamental for understanding a wide range of vegetation and especially crop processes, such as water and nutrient uptake and root competition, the methodological difficulties associated with data sampling are responsible for the lack of knowledge of root distribution.

Variation of maize root system characteristics in the soil profile is the aspect of spatial distribution most frequently described in the literature (Foth, 1962; Sharp and Davies, 1985). Four general patterns of root distribution with depth can be depicted: (i) a steady decrease with depth, which is the most frequently described pattern; (ii) a steep decrease from top soil layers to deeper soil layers; (iii) an increase in the top soil layers down to the depth of maximum root growth, followed by a decrease in soil layers below; and (iv) an irregular distribution with depth. These patterns have been shown to be influenced by maize plant development (Foth, 1962; Nakamoto et al., 1992) and soil environmental conditions (Box et al., 1989; Fiskell et al., 1968). The distribution of roots at various soil depths has occasionally been described by empirical mathematical functions (Dwyer et al., 1996; Phene et al., 1991), but the various forms of root distribution in the profile prevented the establishment of any definitive model for the maize crop irrespective of the soil environment.

Reports on the root distribution at varying distances from the plant row are more scarce. Roots are concentrated near the plant row (Logsdon and Allmaras, 1991; Voorhees, 1989) as a consequence of the large distances between the rows and the root branching characteristics: About 70 axile roots (Hoppe et al., 1986) form a large number of short lateral roots and only a small number of long and branching laterals (Pagès and Pellerin, 1994). Reported patterns of horizontal root distribution can be classified as follows: (i) decreasing root density at increasing distances from the plant row (the most frequently described pattern) and (ii) constant density at increasing distances from the plant row—usually related to special circumstances, such as mulching, tillage, crop residue management (Nelson and Allmaras, 1969; Allmaras and Nelson, 1971), or the placement of irrigation pipes in the interrow (Phene et al., 1991).

There are many references to depth in studies of root distribution because of its assumed importance for an adequate water supply (Phene et al., 1991; Sharp and Davies, 1985). However, variations in root distribution in the horizontal plane may also influence water availability to plants (Tardieu, 1988). From the agronomic point of view, variations in the horizontal distribution of roots are important, especially as far as wheel traffic and fertilizer placement are concerned (Kaspar et al., 1991).

The minirhizotron technique has eased the investigation of spatial and temporal patterns of root growth compared with the classical soil-core sampling. Minirhizotrons are transparent interfaces for the nondestructive observation of soil processes. Characteristics, usage, advantages, and disadvantages of the minirhizotron technique have been described in detail in the various chapters of the book edited by Taylor (1987). Various scientists have used the minirhizotron technique to study the maize root system. The influence of drought (Box et al., 1989), irrigation (Merrill et al., 1987), fertilizer use (Ferguson and Smucker, 1989), and crop rotation (Nickel et al., 1995) on the distribution of maize roots in the soil profile has been reported. However, only Schröder et al. (1996) used minirhizotrons to systematically describe the spatial distribution of maize roots.

The objective of the present study was to characterize the spatial distribution of the maize root system in the soil. Horizontally placed minirhizotrons located at different depths in drainage lysimeters were used to describe patterns and variation in maize root density at varying distances from the plant row and at five developmental stages in 3 yr.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
The experiment was conducted in a drainage lysimeter facility with four replications at the Experimental Field Station, Institute of Plant Sciences of the Swiss Federal Institute of Technology (ETH Zürich), Lindau, Switzerland, at an elevation of 550 m above sea level. The lysimeters had a crop area of 1 m2 and a soil height of 1.5 m. They were uniformly repacked with the topsoil of a slightly alkaline (pH of 7.2–7.5) sandy loam soil with a low organic matter content. Horizontal minirhizotrons, aligned perpendicular to the maize row, were installed successively during soil filling at soil depths of 5, 10, 15, 20, 25, 30, 45, 60, 80, and 100 cm. By means of a slight compacting of the soil during filling and the natural settling of the soil thereafter, realistic values of soil bulk density and a good soil contact of the minirhizotrons were achieved. Physical characteristics of the soil profile in the lysimeters are given in Table 1. Information on lysimeter characteristics, measurement equipment, and functioning are described in detail by Liedgens et al. (2000).


View this table:
[in this window]
[in a new window]
  		Table 1. Physical characteristics of the repacked soil in the lysimeters.

 
‘Atlet’ maize (FAO maturity rating 250; KWS, Einbeck, Germany) was grown from May to mid-September (1994) or early October (1995 and 1996). One maize row was planted in the center of each lysimeter. The distance between adjacent plants was 0.135 m. Three seeds per hill were sown manually, 30 to 40 mm deep. After emergence, the plants were thinned to one per hill. To simulate the presence of neighboring plant rows, shading screens were mounted and adjusted continuously to actual plant height. To avoid carryover effects, lysimeters were kept without plant cover in the year before experimentation.

Shoot development (Ledent et al., 1990) and growth [leaf area (LA)] were screened on all plants two or three times a week. A developmental stage was considered to be reached when at least 50% of the plants in all replications had reached this stage. Leaf area was estimated nondestructively from leaf length (LL) and maximum width (LW) measurements:

This equation was shown to be adequate (R2 = 0.92) for the investigated cultivar by means of destructive LA measurements in a parallel field experiment.

Root observations were made every 7 d using a video camera system (Bartz Technol. Corp., Santa Barbara, CA). Starting at the maize row, images (13.5 mm long by 18 mm wide) were recorded in a 405-mm-long window along the upper side of the 60-mm-diam. minirhizotrons. Using a TV monitor in the lab, the number of root members, as defined by Upchurch and Ritchie (1983), was counted on each image and averaged across five successive images to reduce the influence of the high image-to-image variability. The results were expressed as root density, which is the number of root members per unit interface area of the minirhizotron tube. Root density was studied at five specified developmental stages of the shoot: leaf developmental stages (LDS; number of fully developed leaves) 3, 6, 9, and 12 and pollen shed (PS).

To provide additional information, the years were characterized in agrometeorological terms: mean air temperature (°C) and growing degree days (°C days with a base temperature of 8°C) between successive developmental stages, precipitation, soil water storage [SWS; calculated from three horizontal time domain reflectometry (TDR) probes located at 30, 60, and 90 cm in the soil profile], deep percolation (DP; water flowing out of the lysimeters), and potential evapotranspiration (ETP; calculations based on the Jensen–Haise method—detailed by Burman and Pochop, 1994).

The statistical analysis of the root density comprised the following effects: horizontal position (6), soil depth (10), developmental stage (5), year (3), and their interactions. The SAS PROC MIXED (Littell et al., 1996) was used to perform the statistical analysis. All effects were defined as fixed while replication was defined as a random effect within years. The ANOVA model specified included all single effects, single-effect interactions, as well as quadratic and cubic effects of positions and depths and their interactions with the other factors. Higher order interactions were not included into the model due to computational constraints (memory, processing time, and computing stability) and interpretation difficulties. For taking the multiple correlation of root densities across positions, depths, and developmental stages into account, a spatial covariance structure based on the exponential covariance model has been identified as most suitable among the various available alternatives, based on model-fit criteria computed by PROC MIXED.

The effect of years on shoot phenology was estimated with the nonparametric Wilcoxon rank sum test (Gibbons and Chakraborti, 1992). A nonparametric test has been used to avoid the influence of outliers, which were evident from the investigation of the data distribution. The effect of years on LA was analyzed using the LSD test for means (Gomez and Gomez, 1984).


   RESULTS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Agrometeorological Characterization of the Experimental Years
Mean air temperatures in 1994 and 1995 were similar over the whole maize growing season while in 1996, it was warmer early in the season (up to LDS 6) and cooler during mid-season (between LDS 9 and PS). Growing degree days were fairly similar in all 3 yr at all developmental stages (Table 2). The greatest difference (17%) was found at LDS 3 while at all other developmental stages, the differences were <10%.


View this table:
[in this window]
[in a new window]
  		Table 2. Agrometeorological parameters measured across years and developmental stages.

 
Absolute quantities and proportions of precipitation between successive developmental stages were high early in the season in 1994 and 1995 and lower and more evenly distributed up to PS in 1996. Soil water storage was generally high early in the season and decreased towards PS. Variations in SWS were strongly influenced by precipitation, but the decreasing SWS trend indicates the increasing water demand of the crop. High DP values were observed up to LDS 9 and were related to the time of precipitation (earlier in 1994 and 1995 and later in 1996) and to the low water demand of the young maize plants. The DP was low beyond LDS 9, showing that there was not a surplus of water in the system. As expected, the calculated ETP was similar across the years for all developmental stages because the calculated values essentially depended on global irradiance, which did not vary greatly across years. Cumulative precipitation was markedly higher than ETP at all developmental stages during the three experimental years, mainly due to high early season precipitation, suggesting that the water supply was adequate throughout the season. On the other hand, the combination of low precipitation (especially in 1994 and 1995), reduction in DP, and the depletion of SWS at later developmental stages indicates an increased uptake of water. The calculation of ETP assumes closed, continuous crop canopies, which was not the case in the present study. Hence, the calculated ETP may underestimate the true evapotranspiration in this study. Overall, differences in hydric conditions were small between 1994 and 1995, but 1996 was drier although water distribution was more uniform across the season.

Shoot Development and Growth
Data related to shoot development of maize plants (time at which developmental stages were observed and LA measured) in the three experimental years are shown in Table 3. The time at which developmental stages were observed varied more during early growth (LDS 3 and 6 were anticipated in 1996) than late in the season. Formation of LA was similar in all years, with very low values at the early developmental stages compared with PS. Although the relative differences in LA across years at all developmental stages varied from 14 to 41%, only large differences at LDS 3 and PS were significantly different.


View this table:
[in this window]
[in a new window]
  		Table 3. Maize shoot growth and development across years.

 
Root Characteristics
The effects of the main factors year, developmental stage, depth, and position on root density are shown in Table 4. Table 5 contains the ANOVA results of these effects. Year had a weak effect on root density, which was practically the same in 1994 and 1995 and only 6% lower in 1996 (statistically not significant). Developmental stage had the greatest effect on root density. The smallest value was 25% of the maximum. Root density nearly doubled between LDS 3 and 6 and again between LDS 6 and 9 while there were only small increases at later developmental stages. However, comparisons of root densities among developmental stages were always significant. The effect of soil depth on root density was somewhat smaller compared with developmental stage. The smallest values (39% of the maximum) were found for the deepest layer (100 cm). Root density increased in the 5- to 25-cm depth interval and then decreased as depth increased. The complexity of the change in root density with soil depth is reflected by the significant linear, quadratic, and cubic components of the sum of squares of the soil depth effect in the ANOVA model (Table 5). Comparisons of root density at different soil depths were mostly significant (31 of 36 cases); nonsignificant comparisons of root density were usually observed for adjacent soil depths. Compared with the other factors, horizontal position relative to the plant row had an intermediate effect on root density. The lowest value was about half of the maximum value. Root density decreased continuously with increasing distance from the plant row. This decrease was greater near the plant row (0–270 mm) compared with positions farther away. The observed pattern of the change in root density with distance from the plant row is reflected by the partitioning of the sum of square of this effect into its polynomial components (Table 5): linear (reflecting the continuous decrease), quadratic (reflecting the change in rate of decrease), and cubic. The linear and quadratic components were significant but not the cubic one. The horizontal pattern (distance from the plant row) of change in root density is less complex than the vertical pattern (soil depth). Comparisons of root densities among different horizontal positions were always significant.


View this table:
[in this window]
[in a new window]
  		Table 4. Maize root density as influenced by year, developmental stage, soil depth, and distance from the plant row (position).

 

View this table:
[in this window]
[in a new window]
  		Table 5. Analysis of variance of maize root density (roots cm-2) as influenced by year, developmental stage, soil depth, distance from the plant row (position), and their interactions. Sums of squares of significant effects and interactions were partitioned into their linear, quadratic, and cubic components.

 
The interaction of year and developmental stage on root density was highly significant (Table 5). Details of these results are shown in Table 6. Comparisons of root densities at different developmental stages within years were significant in most cases. A steady increase in root density at subsequent developmental stages (as described above) was only observed in 1994. Occasionally, root density did not vary between early (LDS 3 and 6 in 1995) or late developmental stages (LDS 12 and PS in 1994 and LDS 9, 12, and PS in 1996). At most developmental stages, there was at least 1 yr with significantly higher root densities than the other 2 yr.


View this table:
[in this window]
[in a new window]
  		Table 6. Maize root density as influenced by year and developmental stage.{dagger}

 
The characteristics of root distribution with soil depth for different years and selected developmental stages are shown in Fig. 1A and 1B, respectively. The interaction of soil depth with developmental stage was highly significant while years did not change maize root distribution in the soil profile (Table 5). The significant interaction between soil depth and developmental stage on root density reflects the change in the root distribution with soil depth from a continuous decrease early in the season (LDS 3) to a pattern in which root density first increased in the 5- to 25-cm depth interval and then decreased as soil depth increased (LDS 9 to PS). The change in the root distribution with soil depth was accompanied by an increase from 32 to 40 significant differences in root density among depths. Developmental stages influenced the variation (interaction with the linear component) and the rate of change in root density with soil depth (interaction with the quadratic component).



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 1. Maize root density observed with minirhizotrons as influenced by soil depth: (A) year effect, (B) developmental stage effect (3 and 9 correspond to three and nine fully expanded leaves, respectively; PS is pollen shed), and (C) position effect (distance from the plant row). Presented data are main-effect means; error bars represent overall standard errors of the mean.

 
The characteristics of root density at varying distances from the plant row (position) for different years and selected developmental stages are shown in Fig. 2A and 2B, respectively. The interaction of position with developmental stage was highly significant while years did not affect maize root density in relation to the plant row (Table 5). A decreasing root density at increasing distances from the maize row was observed at all developmental stages. At LDS 3, the differences in root density among positions were small, compared with the later developmental stages, and not significant between adjacent positions. At all other developmental stages, differences in root density among positions were significant (with only three exceptions) and greater between positions near the plant row compared with those farther away. The significant interaction of developmental stage with the linear and quadratic effects of the position effect on root density indicates that developmental stage affected the magnitude and the rate of change in root density across positions. The interaction of soil depth and distance from the plant row (position) on root density was highly significant (Table 5). The root distribution with soil depth is shown in Fig. 1C for selected positions. Near the plant row (0–135 mm), root density is rather constant in the topsoil (5–25 cm) and decreases with increasing soil depth below. At greater distances from the plant row (135–405 mm), root density increased in the 5- to 25-cm depth interval and then decreased as depth increased. This rooting pattern is reflected by the location of significant differences in root densities in the soil profile. The total number of significant differences did not change much among positions (34–37 of 45 comparisons), but the proportion of nonsignificant differences decreased from 90% near the plant row (0–67.5 mm) to only 30% at greater distances from the plant row (270–405 mm). The significant interactions of position with the linear and the quadratic components of the soil depth effect indicates that position influenced the variation and the rate of change in root density with soil depth.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 2. Maize root density observed with minirhizotrons as influenced by the distance from the plant row (position): (A) year effect, (B) developmental stage effect (3 and 9 correspond to three and nine fully expanded leaves, respectively; PS is pollen shed), and (C) soil depth effect. Presented data are main-effect means; error bars represent overall standard errors of the mean.

 
The pattern of root distribution at increasing distances from the plant row is shown for selected soil depths in Fig. 2C. The concentration of roots near the plant row was observed at all depths. In the topsoil (5–30 cm), there was always a clear pattern of continuously decreasing root densities, with a decreasing rate of change, at increasing distances from the plant row. In this soil depth range, differences in root densities among positions were mostly significant (1–3 of 15 comparisons). In deeper soil layers, the pattern of horizontal root distribution is less clear, as differences in root density among positions were smaller and variability was greater. The number of nonsignificant differences in root density among positions increased markedly (6–8 of 15 comparisons). The significant interactions of soil depth with the linear, quadratic, and cubic components of the position effect indicates that soil depth affected the horizontal root distribution in a highly complex way.


   DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The presented data enable us to evaluate the effects of four factors (year, developmental stage, soil depth, and horizontal distance from the plant row) as well as their interactions on root density of maize as observed with minirhizotrons.

The most remarkable result is the small difference in root density observed in each of the three experimental years: same root density in 1994 and 1995 and a difference of 5% in 1996. Relatively small effects of year on root density were observed at various developmental stages, depths, and positions in spite of large variations in maximum LA among years (as high as 36%). The weather in 1996 (lower precipitation and an early warmer and late colder crop season) explains the smaller LA in this year. In 1994, unsatisfactory plant establishment was responsible for the reduced LA. The small variation in root density among years in spite of large annual differences in climatic conditions and shoot size indicated that root development was insensitive to these variations and may have been influenced by root recolonization of old root channels along the minirhizotron tubes (Rasse and Smucker, 1998). These results of our 3-yr study suggest that the integrated temporal and spatial dimension of the maize root system varies within narrow limits; an unexpected finding, as such variations have been observed for field data (Allmaras and Nelson, 1973; Allmaras and Logsdon, 1990; Voorhees, 1989).

Although the observed average effect of years on root density was small, the differences among years are in agreement with climatological and shoot data. Root development in 1996 was almost complete at LDS 9 (reflecting the warm, early crop season), in contrast to the other 2 yr in which root growth continued until PS. The absence of further root growth beyond LDS 9 in 1996 may be the consequence of an adaptive response, favoring shoot growth, and may reflect the colder, late crop season. Similar effects were reported for shading (Lambers and Posthumus, 1980).

Root distribution in the soil profile was characterized by first increasing and then decreasing root density with depth. This pattern was observed in all years, at most developmental stages, and at all positions relative to the plant row. This pattern was also found in other studies (Box et al., 1989; Nicoullaud et al., 1994), but it is not typical of the maize crop (Nakamoto et al., 1992). Minirhizotrons have been reported to underestimate root density in the topsoil and to overestimate it in deeper soil layers (Wiesler and Horst, 1994), probably arising from insufficient soil–minirhizotron contact (allowing for preferential root growth in gaps), tracking of roots along vertical or angled minirhizotrons, or alterations in the immediate environment of the minirhizotron. Of these possibilities, only the third one may have influenced the data obtained in the present experiment because minirhizotrons have been installed horizontally (avoiding root tracking) during soil filling (providing a good soil–minirhizotron contact). However, the rooting profile of various species {rice (Oryza sativa L.), soybean [Glycine max (L.) Merr.], and wheat (Triticum aestivum L.)}, described according to data obtained with minirhizotrons, has also been shown to first increase and then decrease with depth (Yamaguchi and Tanaka, 1990).

Morphological aspects can account for the observed root distribution profiles. Axile maize roots do not branch at the root base or at the tip (Pagès and Pellerin, 1994). This may lead to most root branching occurring at intermediate depths. Nonrandom root orientation, combined with the observation of roots only at the upper side of the minirhizotrons, may also have influenced the final distribution of roots in the soil profile. The growth of axile roots is preferentially vertical (Nakamoto, 1994), but little is know about the orientation of their lateral roots. When the preferred orientation of roots is vertical, the observed root density in a horizontal soil cross section (top view of the minirhizotron) will probably be higher than in a vertical soil cross section (side view of the minirhizotron). If the proportions of vertically and horizontally growing roots change with depth, this will affect the observed distribution of roots with depth (Merrill et al., 1994).

An increasing verticality in root orientation would explain the underestimation of roots in top soil layers and overestimation in deeper soil layers and would also favor the observation of most intense rooting at intermediate depths. However, in the present experiment conducted in uniformly filled lysimeters (see Table 1), maximum root density at intermediate soil depths can not be explained by an abrupt change in soil density, which has been observed at plough pans (Fiskell et al., 1968). The location of maximum maize root density in deeper soil layers is supported by some agronomic studies (Box et al., 1989). Irrigation pipes at a depth of 30 cm, compared with surface irrigation, was considered to enhance nutrient availability at the center of the root system, justifying the enhanced productivity of sweet maize observed by Martinez-Hernandez et al. (1992). Deep (10 cm), but not shallow (4–7 cm), interrow cultivation reduced maize yield (van der Werf et al., 1991).

Root density decreased at LDS 3 along the whole soil profile, and it was constantly high in the topsoil at LDS 6. These patterns of root distribution with depth are probably steps towards the typical pattern observed from LDS 9 on, as it takes time for roots to penetrate deeper soil layers. At later developmental stages, the smallest root density is not found in the deepest soil layer but at 80 cm. This is probably an experimental artifact due to the repacked soil or the water-saturated conditions at the base of the lysimeters.

Roots were concentrated near the plants. The observed distribution of roots at increasing distances from the plant row has been related to branching and orientation of axile roots (Logsdon and Allmaras, 1991). Maize root branching along axile roots is characterized by unbranched regions at the root base and the root tip, a high frequency of short laterals, and only a small proportion of longer roots that continue to branch (Pagès and Pellerin, 1994). Orientation of axile roots is preferentially vertical (Nakamoto, 1994), a characteristic that is enhanced by higher temperatures and decreasing water availability (Nakamoto, 1989) and root age (Foth, 1962). The small root gradients between positions near the plant row and at increasing distances from the plant row observed at early developmental stages were not found in other studies and may be an artifact arising from the relatively small soil volume that can be observed with minirhizotrons. The small change in root density with increasing distance from the plant row at greater soil depths has also been observed by other researchers (Nakamoto, 1989; Schröder et al., 1996; Yamaguchi and Tanaka, 1990), possibly due to the trajectory of successively formed axile roots: Early roots have a more horizontal orientation in the topsoil but are just as vertical as later roots in deeper soil layers (Tardieu and Pellerin, 1990).

The concentration of maize roots near the plant row has important implications for the utilization of resources and for agricultural management practices: (i) There is less uptake of water when roots are clustered compared with a uniform distribution (Tardieu, 1988); (ii) the probability of fertilizer uptake becomes less at greater distances from the plant row, which may help to explain leaching losses; and (iii) the horizontal root distribution found represents an opportunity to grow maize in intercropping systems and in combination with under-sown catch crops or in living mulches because the risk of competition with the other species in the interrow is assumed to be smaller than in the row. However, the consequences of a concentration of roots near the plant row must be considered in relative terms. Estimates of root length density (Melhuish and Lang, 1968) for the present experiment at all locations and most developmental stages are reported to be high enough to ensure unrestricted water and N uptake (Kage and Ehlers, 1996).

From the observation of the spatial distribution of maize roots with minirhizotrons, clear patterns of root density with soil depth and distance from the plant row were identified. The effect of soil depth on root density was stronger and its pattern more complex. There was a significant interaction between soil depth and horizontal position. The spatial components interacted significantly with maize developmental stage, which showed the strongest single-factor effect, but not with years. These results and the significant interaction of developmental stage and year on root density suggest that there is a basic pattern of maize root distribution in the soil, which is modified, but not fundamentally changed by the plastic response of the roots to specific environmental and developmental constraints. On the other hand, the large differences between actual and maximum (or potential) root density observed at all developmental stages, depths, and positions relative to the plant row, suggest that the root system available for maize plants to capture soil resources is characterized by large variations in time and space. Under these conditions, the resource capture ability of the maize root system, integrated over time and space, may not be able to completely cope with the resource availability of the soil.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
This research was supported by the Swiss National Science Foundation, Project no. 31-39498.93.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Crop Sci.Home page
D. B. Myers, N. R. Kitchen, K. A. Sudduth, R. E. Sharp, and R. J. Miles
Soybean Root Distribution Related to Claypan Soil Properties and Apparent Soil Electrical Conductivity
Crop Sci., July 30, 2007; 47(4): 1498 - 1509.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
R. Qin, P. Stamp, and W. Richner
Impact of Tillage and Banded Starter Fertilizer on Maize Root Growth in the Top 25 Centimeters of the Soil
Agron. J., April 27, 2005; 97(3): 674 - 683.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (11)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Liedgens, M.
Right arrow Articles by Richner, W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Liedgens, M.
Right arrow Articles by Richner, W.
Agricola
Right arrow Articles by Liedgens, M.
Right arrow Articles by Richner, W.
Related Collections
Right arrow Spatial Distribution
Right arrow Maize
Right arrow Plant Analysis
Right arrow Root Development

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Crop Science Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome