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


     

Published in Agron J 98:930-937 (2006)
DOI: 10.2134/agronj2005.0336
© 2006 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
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 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 Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tollenaar, M.
Right arrow Articles by Liu, W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Tollenaar, M.
Right arrow Articles by Liu, W.
Agricola
Right arrow Articles by Tollenaar, M.
Right arrow Articles by Liu, W.
Related Collections
Right arrow Maize
Right arrow Spatial Distribution
Right arrow Maize Management

Production Papers

Effect of Crowding Stress on Dry Matter Accumulation and Harvest Index in Maize

Matthijs Tollenaar*, William Deen, Laura Echarte and Weidong Liu

Department of Plant Agriculture, Crop Science Building, Univ. of Guelph, Guelph, ON, Canada, N1G 2W1

* Corresponding author (mtollena{at}uoguelph.ca)

Received for publication December 14, 2005.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Conflicting results have been reported on the effects of spacing and emergence variability on grain yield in maize (Zea mays L.). Effects of spacing and emergence variability on maize grain yield are the net result of the responses of all plants within the stand. The objective of this study was to quantify effects of spacing and emergence variability on crop yield in terms of increased or decreased crowding stress on resource capture (i.e., dry matter accumulation) and resource utilization (i.e., dry matter partitioning) of the individual plants within the crop canopy. Results of previously reported studies were analyzed in terms of plant dry matter accumulation, leaf area, plant growth rate during the critical period for kernel set bracketed by silking (PGRs), grain yield, and harvest index, that is, the proportion of dry matter partitioned to the grain at maturity. Results show that a moderate increase in plant-spacing variability does not influence maize grain yield at the canopy level because reductions in grain yield of plants that experience enhanced crowding stress is compensated, in part, by increased yield of plants that experience reduced crowding stress; crowding stress affected dry matter accumulation but did not affect harvest index. In contrast, plant-emergence variability reduced grain yield at the canopy level because the reduction in grain yield was attributable, in part, to a reduction in harvest index of plants with PGRs less than the threshold for kernel set. Hence, plants can compensate for factors that influence resource capture, but cannot compensate for a reduction in factors that influence resource utilization.

Abbreviations: Ican, irradiance absorbed by a crop canopy • k, the extinction coefficient of irradiance in the crop canopy • LAI, leaf area index • PGRS, plant growth rate during the critical period for kernel set bracketed by silking • SD, standard deviation


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
CROP management and genetic improvement in maize have been profoundly influenced by intra-specific competition or crowding stress during the past seven decades. Plant density for commercial maize production in the U.S. Corn Belt has increased from about three plants m–2 in the 1930s to about seven plants m–2 in the 1990s (Duvick et al., 2004). Increased plant density results in enhanced crowding stress for all plants within the plant stand. Other changes in maize management have included reductions in the distance between rows and increased precision of plant spacing within the row and plant emergence, which result in reducing the variability in crowding stress among plants within the crop stand. The response of a maize canopy to increased spacing or emergence variability is the net effect of the impacts of increased and reduced crowding stress on resource capture and resource utilization of the individual plants within the crop canopy. Although several studies have shown no significant effects of spacing variability on grain yield in maize (Edmeades and Daynard, 1979; Daynard and Muldoon, 1983; Lauer and Rankin, 2004; Liu et al., 2004b), other reports have shown that grain yield increased with enhanced plant-spacing precision (Doerge et al., 2002; Nielsen, 2004, 2006; Liu et al., 2004c). Uneven plant emergence reduces grain yield in maize and the reduction is associated with increased barrenness, that is, plants that do not have a grain-bearing ear (Nafziger et al., 1991; Ford and Hicks, 1992). Genetic improvement in maize has been closely associated with an increase in plant density. Mean rate of genetic improvement for grain yield in maize in the U.S. Corn Belt during the past seven decades have been about 75 kg ha–1 yr–1 (Duvick et al., 2004), but rate of improvement varies with the plant density at which the genetic gain is measured. Duvick (1997) compared maize hybrids representing each decade from the 1930s to the 1990s at plant densities ranging from 1.0 to 7.9 plants m–2 and genetic gain was small or absent at 1.0 plant m–2 and was largest at 7.9 plants m–2. Despite the importance of crowding stress on maize production, little is known about the processes at the plant level that affect crowding stress at the crop canopy level.

Crowding stress can result in a reduction of resource capture and/or in a reduction of resource utilization of maize plants. Resource capture includes processes such as the absorption of solar irradiance by the crop canopy, and water and nutrients absorption by the roots. Resource capture per plant will be reduced when crowding results in enhanced mutual shading. If crowding is a result of increased plant-to-plant variability, due to for example gaps in the row, plants that experience reduced crowding stress may offset the reduction in absorption of irradiance by crowded plants. If crowding is a result of an increase in plant density, total resource capture by the crop canopy will increase or remain equal, because irradiance absorbed by a crop canopy (Ican) is equal to the irradiance absorbed per plant times the number of plants per unit area. Assuming a uniform distribution of leaf area in horizontal plane, Ican can be quantified as follows (cf., Tollenaar and Dwyer, 1998):

Formula 1[1]
where LAI is the leaf area index and k the extinction coefficient of irradiance in the crop canopy. Although plant distribution within a commercial maize canopy is non-uniform (e.g., at a plant density of 6.6 plants m–2 and a row width of 0.76 m, plants are spaced 0.2 m within the row), the non-uniform plant distribution is compensated, in part, by a preferential across-row leaf orientation (Girardin and Tollenaar, 1994; Maddonni et al., 2002). However, if crowding is associated with large gaps within the row (e.g., Lauer and Rankin, 2004), the increased non-uniform distribution of LAI could result in a reduction of Ican.

Resource utilization includes processes such as those that convert absorbed solar irradiance into dry matter (i.e., photosynthetic rate per unit absorbed irradiance or photosynthetic efficiency) and those that partition dry matter to economically important plant components (e.g., harvest index). Canopy photosynthetic efficiency will not be affected by crowding if absorbed irradiance per unit sunlit leaf area (i.e., sunlit leaf area =Ican k–1) is not changed and if the relationship between leaf photosynthesis and absorbed irradiance is not affected by crowding (cf., Tollenaar and Dwyer, 1998). Canopy photosynthetic efficiency will be reduced by crowding if the relationship between leaf photosynthesis and absorbed irradiance is affected by water or nutrient stress, for instance, because plant water and nutrient uptake is reduced more than plant sunlit leaf area. Crowding can also influence dry matter partitioning. Harvest index is stable across a wide range of plant densities in maize, but harvest index declines when plant density is raised above the optimum plant density for grain yield (e.g., Tollenaar, 1992) and when dry matter per plant at maturity is very low (Echarte and Andrade, 2003). Similarly, increased barrenness has been reported for late-emerging plants in a maize canopy (e.g., Nafziger et al., 1991; Ford and Hicks, 1992; Liu et al., 2004a). Genetic improvement in maize during the past seven decades in North America has been attributed, in part, to an increase in the tolerance of newer hybrids to the effects of crowding stress on resource utilization (Tollenaar and Lee, 2002).

The objective of this study was to quantify changes in dry matter accumulation and harvest index in individual plants within a maize canopy that had been exposed to either enhanced or decreased crowding stress during their life cycle. We analyzed the responses of dry matter accumulation and harvest index in individual plants to changes in crowding stress in a study by Liu et al. (2004a) on the effects of plant-spacing and plant-emergence variability on grain yield in maize. Effects of crowding stress on harvest index were also analyzed in terms of the relationship between harvest index and plant growth rate during the critical period for kernel set bracketed by silking (PGRS) that has been reported in another study (Echarte and Andrade, 2003; Echarte et al., 2004). Finally, we will discuss results of studies that have shown grain-yield reductions in response to moderate increases in plant-spacing variability in the context of the mechanisms presented in this paper.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Experimental procedures of the studies on the effects of plant-spacing and plant-emergence variability on maize grain yield have been described in detail in Liu et al. (2004a). In short, field experiments were conducted in 2000 and 2001 at the Elora Research Station (43°39' N, 80°25' W, elevation 376 m) and the Woodstock Research Station (43°08' N, 79°06' W, elevation 317 m) in south central Ontario, Canada. In each experiment, plots were 23-m long and consisted of four 0.76-m rows. Three of the rows were machine planted using a vacuum planter (John Deere 1750, Moline, IL). One of the two center rows was hand planted to achieve the treatments described below. Each hand-planted row consisted of 19 repeated sequences of six-plant units. Both hand and machine planted rows were arranged at the target plant density of 66 000 plants ha–1. Nine treatments were arranged in a 3 x 3 factorial experiment (Fig. 1 ) and replicated four times in a randomized complete block design. The first factor was plant-emergence delay. No plant-emergence delay (Treatment control), a two-leaf emergence delay (Treatment 2L), and a four-leaf emergence delay (Treatment 4L) were established for one of the six plants in each unit. The position number of each plant in the repeatable six-plant unit was marked as Position 1 through Position 6 (Fig. 1). The plant in Position 3 was the only plant that had a delay in emergence. The second factor was within-row plant spacing variability (Fig. 1). The three plant-spacing treatments were uniform 20-cm plant spacing (Treatment 20), a 40-cm gap associated with a double in each six-plant unit (Treatment 40), and a 60-cm gap associated with a triple in each six-plant unit (Treatment 60). The doubles and triples were planted side by side within 3 cm in the row. Plant samples were taken at 5 wk after the plant emergence and 2 wk after the silking of the plant emergence control treatment. At each sampling date, the aboveground biomass of plants in two six-plant units was sampled from the pre-marked sampling areas. Green leaf area of all harvested plants was measured with a LI-3000 leaf area meter (LI-COR, Lincoln, NE). The leaves and stems of sample plants were dried at 80°C for 72 h before measurement of plant dry matter. At maturity, aboveground dry matter and grain was measured for five six-plant units in every plot and the number of barren plants (i.e., plants without a grain-bearing ear) was recorded. Harvest index was calculated from grain dry matter and total aboveground dry matter at maturity. All measurements were done by plant positions and the data for each plant were kept separate for analysis.


Figure 1
View larger version (48K):
[in this window]
[in a new window]
  		Fig. 1. Plant spacing and emergence treatments are composed of repeatable sequences and each sequence unit consists of six maize plants: C, plant-emergence control; 2L, two-leaf emergence delay for plant in Position 3; 4L, four-leaf emergence delay for plant in Position 3; 20, plants spaced uniformly at 20-cm intervals (plant-spacing control); 40, 40-cm gap between plants in Positions 2 and Position 4, and double plants at Position 4; 60, 60-cm gap between plants in Position 2 and Position 5, and triple plants at Position 5. Plant positions from Position 1 to Position 6 are indicated under diagram (adapted from Liu et al., 2004a).

 
Experimental procedures on relationships between harvest index and final shoot biomass per plant or plant growth rate during the critical period for kernel set bracketed by silking (PGRS) in four Argentinean maize hybrids have been described in detail in Echarte and Andrade (2003) and Echarte et al. (2004). In short, the four maize hybrids M400, DK4F36, DK664, and DK752 were grown at five plant densities (2, 4, 8, 16, and 30 plants m–2) at Balcarce, Argentina (37°45' S, 58°18' W, elevation 130 m), during the 1998–1999 growing season. The experiment was arranged in a split plot randomized complete-block design with three replications, and with plant densities as main plots and hybrids as subplots. Subplots consisted of four 7-m rows at the lower plant densities and six to seven 7-m rows at the higher plant densities, with a row width of 0.7 m. Between six plants per subplot at the lowest plant density to 25 to 30 plants per subplot at the highest plant density were labeled at approximately 20 d before silking. Labeled plants were harvested at maturity and grain yield, total plant dry matter, and harvest index were determined for individual plants. Plant dry matter was estimated for each labeled plant at approximately 10 d before and 15 d after silking using the allometric relationships described in Echarte et al. (2004) to estimate PGRS for each labeled plant.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Impact of Inter-Row Spacing Variability on Grain Yield and Dry Matter Accumulation
A change in crowding stress associated with increased inter-row spacing variability (Fig. 1) affected grain yield and dry matter accumulation of plants within a six-plant sequence but did not influence overall grain yield or dry matter accumulation of the six-plant sequence (Table 1). Reductions in grain yield and dry matter accumulation at maturity for plants in Positions 3 and 4 in the 40-cm spacing treatment and plants in Positions 3, 4, 5 in the 60-cm spacing treatment, relative to their respective controls, were associated with enhanced crowding stress (Table 1). Increases in grain yield and dry matter accumulation at maturity for plants in Positions 2 for the 40-cm spacing treatments and plants in Positions 1 and 2 for the 60-cm spacing treatment, relative to their respective controls, were associated with reduced crowding stress (Table 1). The overall impact of spacing variability within the row on grain yield and dry matter accumulation at maturity of the six-plant sequence was not statistically significant because effects of enhanced crowding stress were largely compensated by effects of reduced crowding stress on grain yield and dry matter accumulation at maturity (Table 1): Mean compensation was 63% for grain yield (i.e., sum of all increases for the two six-plant sequences was 1.85 Mg ha–1 which is equivalent to a mean yield increase of 0.15 Mg ha–1, and sum of all decreases was 2.90 Mg ha–1 which is equivalent to a mean yield decrease of 0.24 Mg ha–1) and 82% for dry matter at maturity (i.e., mean dry matter increase for the two six-plant sequences was 0.28 Mg ha–1 and mean yield decrease was 0.34 Mg ha–1).


View this table:
[in this window]
[in a new window]
  		Table 1. Effects of plant-spacing variability within a six-plant sequence on grain yield and aboveground dry matter at maturity. Means across plant-emergence treatments.

 
Effects of crowding stress on grain yield of the plant adjacent to the plants in Position 3 in the six-plant sequence were attributable to effects on dry matter accumulation, as increased spacing variability did not influence harvest index in plants that were adjacent to Position 3 (Liu et al., 2004a). The mean reduction in dry matter at maturity of the three plant positions that were significantly affected by enhanced crowding stress, that is, plants in Position 4 for the 40- and 60-cm spacing treatments and plants in Position 5 for the 60-cm spacing treatment, was approximately 6%. The mean increase in dry matter at maturity of the three plant positions that were significantly affected by reduced crowding stress, that is, plants in Position 2 for the 40- and 60-cm spacing treatments and plants in Position 1 for the 60-cm spacing treatment, was approximately 7% (Table 1).

During the life cycle, the temporal response of leaf area and dry matter accumulation to increased within-row plant-to-plant variability was different for enhanced than reduced crowding stress. Leaf area and dry matter accumulation of plants exposed to enhanced crowding stress were reduced by approximately 16% relative to the control at 5 wk after emergence, but differences in leaf area had disappeared and differences in dry matter were much smaller at 2 wk after silking (Table 2). Plants were still small at 5 wk after emergence and mutual shading was likely limited and, consequently, the reduction in leaf area per plant at 5 wk after emergence may have been a result of a low red to far-red ratio of the light reflected of nearby plants (Ballaré et al., 1990). In contrast to our results, it has been reported that leaf area of maize seedling increased when plants were exposed to a low red to far-red ratio of incident radiation due to either increased plant density (Kasperbauer and Karlen, 1994) or by placing a grass sod in the rows between the maize plants (Rajcan et al., 2004). Unlike plants exposed to enhanced crowding stress, leaf area and dry matter at 5 wk after plant emergence were not affected by a reduced crowding stress (Table 2). Increases in dry matter accumulation at 2 wk post-silking and from 2 wk post-silking to maturity in plants exposed to reduced crowding stress were similar to the increase at maturity in this treatment (Table 2). Changes in dry matter accumulation of plants exposed to either enhanced or reduced crowding stress at 2 wk post-silking and between 2 wk post-silking and maturity (Table 2) were probably associated with changes in mutual shading of crowded plants (i.e., resource capture), as the stress effect was probably too small to influence photosynthetic efficiency (i.e., resource utilization).


View this table:
[in this window]
[in a new window]
  		Table 2. Evolution of leaf area and aboveground dry matter during the life cycle of plants that were exposed to either enhanced or reduced crowding stress due to increased inter-row spacing variability.

 
Effects of enhanced crowding stress on plants in Position 3 were generally similar to those on plants in Positions 4 and 5, but means across plant-emergence treatments could not be evaluated due to significant plant-spacing x plant-emergence treatment interactions for grain yield, dry matter accumulation, harvest index, and leaf area (Liu et al., 2004a). Treatment interactions for grain yield were attributable, in part, to a difference in harvest index between the 20-cm spacing and the 40- and 60-cm spacing treatments (Liu et al., 2004a).

Impact of Plant-Emergence Variability on Grain Yield and Dry Matter Accumulation
In contrast to effects on grain yield of the increase in inter-row spacing variability, increasing plant-emergence variability reduced overall grain yield of the six-plant sequence. Grain yield of the six-plant sequence was reduced by 4.2 and 8.4% when emergence of plants in Position 3 was delayed by two or four leaf stages, respectively (Table 3). Grain yield of plants in Position 3 was 39 and 79% lower in the two-leaf and four-leaf emergence delay treatments, respectively, than in early-emergence treatment and aboveground dry matter at maturity of plants in Position 3 was 28 and 55% lower in the two-leaf and four-leaf emergence delay treatments, respectively, than in the early-emergence treatment (Table 3). Competitive ability of plants in Position 3 was reduced because of lower leaf area and a lower plant height: Leaf area per plant at 2 wk post-silking was 4210 cm2 for the control emergence treatment, 3420 cm2 for the two-leaf emergence delay treatment, and 2490 cm2 for the four-leaf emergence delay treatment (P < 0.01) and plant heights were 246 cm for the control emergence treatment vs. 214 cm for the four-leaf emergence delay treatment in Position 3 (P < 0.01; Liu et al., 2004a).


View this table:
[in this window]
[in a new window]
  		Table 3. Effects of plant-emergence variability within a six-plant sequence on grain yield and aboveground dry matter at maturity. Means across plant-spacing treatments.

 
The greater impact of plant-emergence variability relative to increased plant-spacing variability on overall grain yield was attributable, in part, to a smaller compensation by plants that experienced reduced crowding stress (Table 3). Mean compensation for grain yield by plants bordering the late-emerging plants was 34% (i.e., mean yield increase for the two treatments was 0.24 Mg ha–1 and mean yield decrease for plants in Position 3 in the two treatments was 0.70 Mg ha–1). The relatively low compensation was attributable, in part, to effects of emergence-variability on harvest index. Means for harvest index of plants in Position 3 were 0.50, 0.43, and 0.23 for the control emergence, the two-leaf emergence delay, and the four-leaf emergence delay treatments, respectively (differences significant at P < 0.01), and harvest index of plants in the other positions of the six-plant sequence were not affected by the treatments (Liu et al., 2004a). Approximately one-third of the effect of plant-emergence delay on grain yield of plants in Position 3 can be attributed to reductions in harvest index, that is, 59% reduction in grain yield vs. 42% reduction in dry matter (Table 3). Plants adjacent to Position 3 could not compensate for the reduction in harvest index of late-emerged plants (Table 3).

Harvest index was associated with dry matter per plant at maturity when plant emergence was delayed for plants in Position 3. The slope of the relationship between dry matter per plant at maturity vs. harvest index was not different from zero for the control plant emergence treatment with plot means >140 g per plant (y = –0.00009x + 0.52; r2 = 0.0058), but harvest index declined when dry matter at maturity declined below 140 g per plant in the two-leaf emergence delay and the four-leaf emergence delay treatments (Fig. 2 ). Harvest index declined linearly with dry matter per plant at maturity for the late-emergence treatment with plot means <140 g per plant (y = –0.0047x + 0.1378; r2 = 0.8376) and the threshold dry matter for harvest index (i.e., the x intercept) was 29 g per plant. Harvest index was highly correlated with number of barren plants per plot for plants in Position 3 (r2 = 0.84, n = 144) and, consequently, the decline in harvest index with dry matter at maturity was associated with an increasing proportion of barren plants as plot dry matter declined from 140 to 29 g per plant. Echarte and Andrade (2003) also showed that the harvest index per plant of four Argentinean maize hybrids was relatively stable across a large range of total plant dry matter at maturity. In contrast to results depicted in Fig. 2, they reported a sharp decline in harvest index when plant dry matter at maturity approached the threshold value for harvest index, possibly, because they recorded individual plant data rather than plot means. The threshold dry matter at maturity for harvest index ranged from 27 to 30 g per plant for the two newer hybrids in their study (Echarte and Andrade, 2003). Results reported in this study show that the decline in harvest index is highly associated with the proportion of barren plants and barren plants are plants with dry matter at maturity less than or equal to the threshold dry matter at maturity for harvest index.


Figure 2
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 2. Relationships between harvest index and dry matter at maturity for plants in Position 3 (cf., Fig. 1).

 
The lack of kernel set in barren plants, that is, full-grown plants that do not have a grain-bearing ear, is frequently the result of a low plant growth rate during the critical period for kernel set bracketed by silking (PGRS): that is, PGRS ≤ threshold PGRS for kernel set (Tollenaar et al., 1992). Harvest index and number of barren plants were closely associated for the plants in Position 3 depicted in Fig. 2 and, consequently, harvest index should be examined in terms of PGRS.

Relationship between Harvest Index and PGRS
The relationship between harvest index and PGRS was examined in four Argentinean maize hybrids that were included in a study by Echarte and Andrade (2003) and Echarte et al. (2004). Results show that the response of harvest index to PGRS is an almost "Blackman-rate-limiting-type" response: harvest index is stable for high PGRS values and declines steeply to zero at PGRS values close to the threshold for harvest index (Fig. 3a ). This response differs from the rectangular hyperbole that characterizes the relationship between kernel number per plant and PGRS (Tollenaar et al., 1992). Among the four hybrids, mean threshold PGRS for kernel set or harvest index was higher for the two older hybrids (PGRS = 1.18 g plant–1 d–1) than for the two newer hybrids (PGRS = 0.59 g plant–1 d–1) (P < 0.05) (Echarte et al., 2004). A large variation in harvest index was apparent among plants with PGRS equal to threshold ±0.5 g plant–1 d–1 (Fig. 3a). Maddonni and Otegui (2004) also reported a large variation in kernel number per plant for plants with PGRS values close to the threshold PGRS for kernel set and they attributed this variation to differences between dominant plants (i.e., those ranked within the upper 30% in shoot dry matter at physiological maturity) vs. dominated plants (i.e., those ranked within the lower 30% in shoot dry matter at physiological maturity). The scatter in harvest index of plants near the threshold PGRS in Fig. 3a, however, was not associated with plant dry matter at the onset of the critical period for kernel set bracketing silking (Fig. 3b). As indicated by Echarte et al. (2004), the threshold PGRS for harvest index for the four Argentinean hybrids depicted in Fig. 3 was associated (r2 = 0.95, n = 4, P < 0.05) with the threshold dry matter at maturity for 'harvest index reported by Echarte and Andrade (2003).


Figure 3
View larger version (32K):
[in this window]
[in a new window]
  		Fig. 3. Relationships between (a) harvest index and plant growth rate during a critical period for kernel set bracketed by silking (PGRS) and (b) kernel number and plant dry matter before silking for plants with PGRS equal to threshold for kernel set ±0.5 g plant–1 d–1 for four Argentinean maize hybrids. Equation harvest index (HI) = a + bx PGRS, if PGRS > PGRS threshold for kernel number was fitted to the relationship between HI and PGRS and R2 of the fitted equations ranked from 0.42 to 0.56 (Fig. 3a). Closed symbols represent plants with PGRS equal to threshold for kernel set ±0.5 g plant–1 d–1.

 
Circumstantial evidence supports the contention that the reduction in harvest index due to plant-emergence variability in the Liu et al. (2004a) study was associated with PGRS of plants in Position 3. Dry matter accumulation before 2 wk post-silking, that is, 2 wk post-silking for plants in the early-emergence treatment, had a large influence on differences in grain yield among plant-emergence treatments. Dry matter of plants in Position 3 at 2 wk post-silking were 7.6 Mg ha–1 for the early-emergence treatment, 4.4 Mg ha–1 for the two-leaf emergence delay treatment, and 2.1 Mg ha–1 for the four-leaf emergence delay treatment. Relative to the control emergence treatment, reductions in grain yield were similar to reductions in dry matter accumulation at 2 wk post-silking, that is, 39 vs. 42% for the two-leaf emergence delay treatment and 79 vs. 72% for the four-leaf emergence delay treatment (cf., Table 3). In contrast, reductions in grain yield due to delayed emergence of plants in Position 3 were more than two times greater than reductions in dry matter accumulation during the period from 2 wk post-silking to maturity, that is, 39 vs. 13% for the two-leaf emergence delay treatment plants and 79 vs. 36% for the four-leaf emergence delay treatment plants. In addition, harvest index was highly correlated with plant dry matter at 2 wk post-silking for two-leaf and four-leaf emergence delay treatment plants in Position 3 in each of the four location yr–1 (Fig. 4 ).


Figure 4
View larger version (25K):
[in this window]
[in a new window]
  		Fig. 4. Relationship between harvest index (HI) of plants in Position 3 and total plant dry weight (DM) at 2 wk postsilking for plots of the two-leaf delayed and four-leaf delayed treatments at two locations and 2 yr.

 
Differences in the response of grain yield to spacing and emergence variability among studies reported in the literature may sometimes be attributable to variation in yield stability characteristics among maize hybrids. When the threshold value of PGRS for harvest index varies between maize hybrids (e.g., Fig. 3a), differences in the response of grain yield to plant-spacing variability at high levels of crowding stress would be anticipated (Andrade and Abbate, 2005). We have shown that large differences exist in kernel-set dynamics between maize hybrids and their parental inbred lines, and among parental inbred lines (Echarte and Tollenaar, 2006).

Other Causes for Reductions in Grain Yield which Are Associated with Increased Plant-Spacing Variability
Although we have shown that moderate levels of plant-spacing variability in maize may not affect grain yield, various studies have reported significant reductions in grain yield in response to increased plant-spacing variability without a large increase in plant-emergence variability (Vanderlip et al., 1988; Nielsen, 2004, 2006; Liu et al., 2004c). In some cases a significant grain yield response to spacing variability may be attributed to a low LAI or very large gaps in the row. Crop LAI has to be sufficiently large in order for compensation to be possible (e.g., Vanderlip et al., 1988) and there is a limit to the length of the gap that can be compensated by neighboring plants, for example, crop canopy yield was reduced by four- or eight-plant gaps in the study by Lauer and Rankin (2004). Liu et al. (2004b) showed that spacing variability did not affect grain yield for values of plant-spacing standard deviation (SD) <17 cm, but grain yield reduction could occur if plant-spacing variability exceeds variability for which grain-yield compensation cannot nullify reductions in grain yield of plants that experienced increased crowding stress. If spacing variability is sufficiently large to result in grain-yield reductions at the canopy level, care should be taken to only express yield reductions in terms of kg ha–1 yield loss per cm SD when the yield loss due to spacing variability is linear across the whole range of plant spacing variability encountered in the study. Grain yield reductions associated with spacing variability in the range from SD < 5 cm to SD > 20 cm have been reported in large-scale field studies (Nielsen, 2004, 2006), but yield differences in these studies were apparent among treatments with SD values <15 cm, indicating that the grain yield response did not result exclusively from treatments with very high spacing variability.

The association between grain yield and spacing variability may sometimes be attributable to factors that are associated with plant spacing variability, rather than to spacing variability per se. For instance, the spacing-variability treatment could influence planting depth, surface crop residue distribution, and microsite variation in seedbed conditions that may affect plant growth and development after plant emergence. Liu et al. (2004c) showed a significant effect of spacing variability on grain yield in a study involving three different planter types and two planting speeds, for maize grown under conventional and no-tillage conditions, but significant yield differences were apparent between treatments with similar spacing variability. (i) Plant-spacing variability increased from about 8 to 19 cm when maize was planted with a vacuum-meter planter vs. an air seeder, respectively, under both tillage systems, but grain yield reductions associated with increased plant-spacing variability were less under conventional-tillage than under no-till conditions (i.e., 4 vs. 13%; LSD for grain-yield differences was <4%). (ii) For the air seeder under no-tillage conditions, spacing variability was slightly greater for the high planting speed (SD = 21.8 cm) than for the low planting speed (SD = 19.3 cm), but the difference in grain yield between the two treatments was substantial (7%). Differences in days to emergence among the treatments in this study were probably too small to result in differences in harvest index as depicted in Fig. 2, but these differences may be indicative of treatment effects other than spacing variability. Indeed, differences in grain yield between the planter-type and planting-speed treatments were associated with reductions in leaf area and plant dry matter, in particular, during early phases of development (Liu et al., 2004c). Hence, treatments in the Liu et al. (2004c) study probably influenced seedbed conditions in addition to their effect on plant spacing variability, which may have resulted in grain yield differences.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
A moderate increase in plant-spacing variability may not influence maize grain yield at the canopy level because of compensation by plants within the population. The plant-spacing treatments within the six-plant sequence in the study reported by Liu et al. (2004a) resulted in reductions in grain yield of plants that experienced enhanced crowding stress, but this was compensated, in part, by increased yield of plants that experience reduced crowding stress. Moderate levels of crowding stress did not influence harvest index. Grain-yield compensation under those conditions can be attributed to changes in dry matter accumulation that resulted predominantly from either enhanced or reduced mutual shading (i.e., resource capture). Gaps within the row of 40 to 60 cm would not be expected to result in reduced light interception by the crop canopy, as row width of most commercially grown maize is around 76 cm. The lack of response of maize grain yield to moderate non-uniform spacing within the row reported in the literature (Edmeades and Daynard, 1979; Daynard and Muldoon, 1983; Lauer and Rankin, 2004; Liu et al., 2004b) is consistent with this contention. Several studies have reported grain-yield reductions in response to plant-spacing variability (Doerge et al., 2002; Nielsen, 2004, 2006; Liu et al., 2004c) and we speculate that yield reduction could sometimes be attributable to factors that are associated with plant spacing variability, rather than to spacing variability per se.

A relatively large increase in plant-to-plant variability due to plant-emergence delay reduced grain yield at the canopy level because grain-yield compensation by plants that experienced reduced crowding stress was relatively small. Reduction in grain yield due to plant-emergence delays of two- and four-leaf stages in the study by Liu et al. (2004a) was attributable, in part, to a reduction in harvest index of plants that emerged late. In contrast to compensation for factors that influence resource capture such as leaf area per plant and changes in mutual shading by leaves in the canopy, plants cannot compensate for reductions in factors that influence resource utilization such as harvest index. Barrenness can result from a low PGRS due to enhanced crowding stress, that is, when PGRS less than or equal to threshold PGRS for harvest index (Fig. 3a).

The impact of increased plant-spacing and plant-emergence variability on maize grain yield will depend on the proportion of incident solar irradiance that is intercepted by the canopy and on plant growth rate during the silking period. Although gaps within the row of ≤60 cm may not reduce light interception by a maize canopy when light interception is >95%, canopy light interception could be reduced when gaps are large and/or when LAI is substantially lower than that required for full-light interception. In addition, any reduction in harvest index due to crowding stress will reduce grain yield of the canopy and, consequently, PGRS of plants exposed to crowding stress should not drop below the threshold value. The impact of crowding stress on grain yield is, therefore, related to climatic conditions that influence PGRS, (e.g., solar irradiance, precipitation, temperature), agronomic practices that influence PGRS (e.g., plant density, because PGRS is inversely related to plant density at high LAI), and stress tolerance of maize hybrid (e.g., threshold PGRS for kernel number or harvest index).


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Crop Sci.Home page
W. Liu and M. Tollenaar
Response of Yield Heterosis to Increasing Plant Density in Maize
Crop Sci., September 1, 2009; 49(5): 1807 - 1816.
[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 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 Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tollenaar, M.
Right arrow Articles by Liu, W.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Tollenaar, M.
Right arrow Articles by Liu, W.
Agricola
Right arrow Articles by Tollenaar, M.
Right arrow Articles by Liu, W.
Related Collections
Right arrow Maize
Right arrow Spatial Distribution
Right arrow Maize Management

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