Crop Science Grow Your Career with CSSA
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 (14)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Angadi, S. V.
Right arrow Articles by Gan, Y.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Angadi, S. V.
Right arrow Articles by Gan, Y.
Agricola
Right arrow Articles by Angadi, S. V.
Right arrow Articles by Gan, Y.
Related Collections
Right arrow Best Management Practices
Right arrow Crop Growth and Development
Right arrow Canola
Crop Science 43:1358-1366 (2003)
© 2003 Crop Science Society of America

CROP ECOLOGY, MANAGEMENT & QUALITY

Yield Adjustment by Canola Grown at Different Plant Populations under Semiarid Conditions

S. V. Angadi*, H. W. Cutforth, B. G. McConkey and Y. Gan

Semiarid Prairie Agricultural Research Center, Agriculture and Agri-Food Canada, Swift Current, SK S9H 3X2, Canada

* Corresponding author (angadis{at}em.agr.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 SUMMARY
 REFERENCES
 
Establishing a good canola (rapeseed; Brassica napus L.) stand is difficult in the semiarid prairie region of Canada where low temperature, water stress, and soil crusting could result in poor seed bed conditions. A field study was conducted from 1999 to 2001 at Swift Current, SK, Canada, to determine the effect of a range of uniform (5 to 80 plants m-2) and nonuniform (seedlings from 1-m lengths from two adjoining rows were removed and retained alternatively; 10 to 40 plants m-2) plant populations on yield and yield components of canola. Canola adjusted seed yield across a wide range of plant populations, although it did not compensate completely for the decreasing populations. Environmental conditions played a significant role in the expression of plasticity of canola. For example, in 2000, with slightly above-normal growing season precipitation, canola maintained similar yield levels across a wide range of populations (20 to 80 plants m-2), while in 2001, with well below normal precipitation, seed yield declined as populations dropped below 40 plants m-2. Reducing plant population by half from 80 to 40 plants m-2 did not reduce seed yield when the reduced plant population was uniformly distributed, but reduced yield when the population was nonuniformly distributed. The primary response of canola to lower plant population was increased pods per plant through increased branching and increased pod retention at each node. The number of pods formed on primary and secondary branches increased as population decreased. Seeds per pod and seed weight were stable across populations.

Abbreviations: DM, dry matter • ES, early spring • HI, harvest index • LS, late spring


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 SUMMARY
 REFERENCES
 
IN THE SHORT GROWING SEASON of the Canadian prairie, canola has limited time to express potential plasticity compared with other regions of the world where the canola growing season is longer (Mendham and Salisbury, 1995). Therefore, optimum plant population of canola is higher in the Canadian prairie than other regions of the world and it is more critical to optimize plant populations. The plasticity of a plant to compensate for suboptimal plant populations depends on the availability of resources such as light, water, and nutrients (Sultan, 2000). In particular, the greater the availability of resources, the greater will be the expression of plasticity. Morrison et al. (1990a), using seeding rates instead of actual populations, focused on above-optimum plant population range. Thus, under the generally good growing season moisture in southern Manitoba, lowest seeding rates of 1.5 to 3.0 kg ha-1 (35 to 70 plant m-2) were enough to produce maximum grain yield. McGregor (1987) used actual plant populations; however, the main objective was to compare yield formation of very low populations (<22 plants ha-1) with the highest population during the season (144 to 200 plants ha-1). Canola yield plasticity in that study varied widely indicating the importance of weather conditions in determination of the optimum population. Because of this wide range of plasticity, a canola population of 80 to 180 plants m-2 has been recommended for canola production in the Canadian prairie (Thomas, 1984).

The production area of canola in the nontraditional regions of the northern great plains, similar to the Brown and Dark Brown soil zones of the Canadian semiarid prairie, has been gradually increasing (Johnston et al., 2002). Poor plant establishment of small seeded crops frequently occurs in these nontraditional areas because of poor seeding conditions. Factors reported to reduce plant populations of canola include inadequate or excessive soil moisture, soil crusting, low temperature, seeding equipment, late spring (LS) frost, and hail damage (Mendham and Salisbury, 1995). In addition, practices adopted by producers such as fall seeding before freeze-up, in which seeds remain dormant in the frozen soil and emerge in the spring, and seeding in the early spring (ES) have increased the challenge for obtaining good stand establishment.

Higher plant population has been recommended and adopted to ensure a competitive crop to check weeds in the early growth stages (Morrison et al., 1990a). All previous studies on canola plant population used herbicide-nontolerant cultivars (McGregor, 1987; Morrison et al., 1990a). Similarly, for the same reason seeding was recommended after killing weeds that emerged in the spring. However, compared with the traditional spring seeding dates, the benefits of either dormant seeding in the late fall or ES are often substantial and with the availability of herbicide-tolerant canola, weeds are easily removed from canola fields (Kirkland and Johnson, 2000).

Past studies have focused on optimum uniform plant stand for increasing seed yield. However, nonuniform plant spacing often occurs in practice. New agronomic practices, such as fall or ES seeding, are expected to increase nonuniformity in the population stand. Nonuniform plant spacing reduced seed yield in sunflower (Helianthus annuus L.; Wade, 1990), corn (Zea mays L.; Pommel and Bonhomme, 1998) and sorghum [Sorghum bicolor (L.) Moench; Larson and Vanderlip, 1994], but little is known about the effect of nonuniform plant population on spring canola. Increased variability in the stand was found to reduce seed yield in winter canola (Hühn, 1999). Reseeding canola late in the spring to correct perceived suboptimal stand densities exposes the canola crop to increasing seasonal temperatures, which have been reported to reduce seed yield of canola (Nuttal et al., 1992; Angadi et al., 2000).

Seed yield of canola is a function of population density, number of pods per plant, number of seeds per pod, and seed weight. However, yield structure is very plastic and adjustable across a wide range of populations. The number of pods per plant is the most responsive of all the yield components in canola (Diepenbrock, 2000) and is determined by the survival of branches, buds, flowers, and young pods rather than by the potential number of flowers and pods (McGregor, 1981). Canola branching and podding responses under a range of plant populations, especially at suboptimal plant populations, have not been studied. Similarly, proper understanding of the contribution of number of seeds per pod and seed weight at suboptimal plant populations is needed.

The adoption of late fall dormant and ES seeding, availability of herbicide-tolerant canola cultivars, and increase of canola production in nontraditional dryland regions warrants a new investigation into the effect of plant population on canola growth and yield. Therefore, a field study was conducted to determine the effect of reducing plant population on seed yield and yield components of canola under water-limited semiarid conditions. An additional objective was to assess the effect of uniform and nonuniform distributions of plant population on canola seed yield.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 SUMMARY
 REFERENCES
 
A field study was conducted from 1999 to 2001 at the Semiarid Prairie Agricultural Research Center, Swift Current, SK, Canada (50°17' N 107°48' W) located in the Brown soil climatic zone (Henry and Harder, 1991), a semiarid region generally considered marginal for canola production. The soil type was a Swinton silt loam (Orthic Brown Chernozem). Glyphosate-tolerant, open-pollinated spring canola cv. ‘Arrow’ was seeded under rainfed conditions on 6 May 1999, 25 Apr. 2000, and 24 Apr. 2001 (ES) using an air drill with 0.23-m row spacing. The field used for the 2001ES trial was extremely dry in ES. Therefore, the experimental site was irrigated ({approx}15 mm) after seeding using a portable sprinkler system. Plot area irrigated was small and only uniform population treatments were imposed in the trial. Therefore, to include the nonuniform plant population treatments, an additional trial was seeded under irrigated conditions on 8 June 2001 (LS) using a disc drill with 0.20-m row spacing. The 2001LS trial was irrigated four times, on 29 May (41 mm), 8 June (14 mm), 11 June (24 mm), and 29 June (29 mm).

All trials were conducted on fallowed fields having wheat (Triticum aestivum L.) previous to fallow. Vitavax RS (carbathiin + thiram + lindane) seed treatment was used to control seedling fungal diseases and provide protection against flea beetles (Phyllotreta cruciferae Goeze). Flea beetle, blister beetle (Lytta nuttalli Say and L. cyanipennis Leconte) and diamondback moth (Plutella xylostella L.) infestations, which were observed occasionally, were controlled with deltamethrin or carbofuran. In spring, a fertilizer mixture of 84-24-0-22 kg N, P2O5, K, and S ha-1 was uniformly broadcast over the experimental area. Postemergent glyphosate application was used to control weeds.

A higher-than-recommended seeding rate of 12 kg ha-1 was used to obtain a relatively high, uniform population density. At the 2-to-4 true leaf stage, seedlings were hand thinned to uniform plant stands of 80, 40, 20, 10, and 5 plants m-2 and nonuniform plant stands of 40, 20, and 10 plants m-2. Thinning ensured maximum distance between plants in adjacent rows and uniform distribution of population in uniform plant stand treatments. To obtain nonuniform plant stands, seedlings from alternate 1-m lengths from two adjoining rows were removed and when two adjacent rows had the seedlings removed, the next two rows had seedlings retained and vice versa. Thus, by removing half of the plant population from 80, 40, and 20 plants m-2 plots, we obtained 40, 20, and 10 plants m-2 in a nonuniform stand. Plot sizes ranged from 11 m2 (2001ES) to 26 m2 (2001LS).

At harvest, hand samples were collected from 0.66- to 1.62-m2 area in each plot. Samples were first air-dried and then oven dried at 80°C for 36 h. The dry weights and seed weights were used for estimating dry matter production per unit area (DM) and harvest index (HI). Six rows from the center of the plot were harvested using a plot combine. Before harvest, 25 mature pods were randomly collected (except in 2000, where pods produced on main (terminal raceme) inflorescence from three randomly selected plants were used) from the canopy and oven dried. They were used for assessing seeds per pod and thousand seed weights. Pods per plant were counted on three randomly selected plants (five in 1999). Weather parameters were collected from the Environment Canada Weather Station located within 300 m of the experimental locations.

For detailed analysis of yield components, three plants from the uniform population plots in 2000 and 2001ES were harvested just before swathing. The number of pods produced on the main raceme, on individual primary branches, and on all other higher order branches at each node were counted. Fertile pods were defined as pods which contained at least one seed. The nodes were counted based on when they initiated flowering branches, that is, top downward in canola (basipetal succession). Pods from the main raceme and each primary branch were oven dried separately, while the pods from secondary and higher order branches were pooled for drying. The dry weight of pods, seed weight, and seed number from each main raceme and primary branches were used to calculate seeds per pod, thousand seed weight and seed dry weight:pod dry weight ratio (seed:pod ratio) on each of those branches separately.

The experimental design for all trials was a randomized complete block design with three (2000 and 2001LS) or four (1999 and 2001ES) replicates. Whenever more than one subsample was taken from each plot, the mean of the plot subsamples was used for analysis. To gain further understanding of effect of plant population under different growing environments, all seed yield and yield-forming traits from each trial were analyzed separately using the analysis of variance technique; the Fisher protected LSD was used for mean separation (GLM procedure; SAS Institute, 1985). Because of large variations in environmental conditions, no effort was made to combine data. To understand the plant population effect under different yield potentials, seed yield from each environment, from each population, and the mean seed yield of four experiments were normalized to the seed yield at 80 plants m-2 and regressed against plant population. Canola yield formation response to plant population at different branch levels, that is, at main, primary, and secondary branch levels, was assessed by observing pod number and relative contribution to seed yield. Similarly, podding on primary and secondary branches and seeds per pod, seed weight, and seed:pod ratio on primary branches were compared statistically for population effect at each node level in 2000 and 2001ES. For all statistical comparisons, the P ≤ 0.05 level was used to signify statistical differences.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
SUMMARY
 REFERENCES
 
Weather Conditions
In general, growing conditions were typical for canola in 1999, favorable in 2000, and stressful in 2001 (Fig. 1). Maximum temperatures in May, June, and July of 1999 were 2.4 to 3.3°C cooler than normal (mean of 117 yr), while maximum temperatures in 2001 were 2.0°C and 3.8°C warmer than the normals for May and August. In 1999 and 2000, snowfall occurred a few days after seeding and minimum temperatures were below freezing. Since the crop was either not emerged or small, no adverse effect was observed. Precipitation was fairly well-distributed in 1999 and 2000, except for August. August precipitation was only 30 to 40% of the long-term average. In contrast, 2001 was dry during the winter months, which severely reduced soil water recharge. This was followed by a very dry growing season with only 40 to 50% of the normal rainfall amounts during May and June. Later growth stages, especially in the LS seeded trial, were also severely affected by drought when only 7% of the normal rainfall fell during August. Since 1885, 2001 was the second driest and the fifth warmest year on record at Swift Current (Judiesch and Cutforth, 2002). Extreme variations in the weather conditions were observed during the experimental period.



View larger version (48K):
[in this window]
[in a new window]
  		Fig. 1. Climatic conditions from April to August during 1999 to 2001 at Swift Current. Daily maximum and minimum temperature (°C) and rainfall (mm) are presented with solid line, dashed line, and solid bars, respectively. Open arrows on the horizontal axis indicates early spring seeding date, while solid arrow for 2001 indicates late spring seeding date.

 
Dry Matter Production
In general, canola maintained DM production across a wide range of populations (Table 1). Nonuniform plant stand had no effect on canola DM. The plasticity of the canola plant maintained biomass production across a wide range of uniform as well as nonuniform population densities. However, the tendency was for biomass to decrease as plant population decreased. In 2000, there was no effect of population on DM, while the largest effect of population on biomass production was in 2001LS. The leaf area index of 5 plants m-2 measured in 2001ES was 14 to 46% of 80 plants m-2 between 55 to 79 d after seeding (Angadi et al., 2002). Morrison et al. (1990b) suggested that the leaf area index of canola at lower population takes longer to cover the ground surface than that for higher plant populations. In general, DM accumulation takes place at a slower rate until the crop reaches full ground coverage and then increases at a much faster rate (Diepenbrock, 2000). A longer time to canopy closure results in a shorter period of more rapid increase in biomass. The delay in attaining full ground coverage with lower populations prevented canola from efficiently utilizing the solar radiation and was related to lower biomass.


View this table:
[in this window]
[in a new window]
  		Table 1. Biomass and seed production response of Argentine canola cv. Arrow to variations in population densities at Swift Current during 1999 to 2001.

 
Seed Yield
Similar to DM, canola maintained seed yield across a wide range of populations (Table 1). Reducing uniformly distributed plant population by half from 80 to 40 plants m-2 did not reduce seed yield in any trial. Seed yields of nonuniform plant populations were similar to uniform populations in all three environments. Reducing the population further to 20 plants m-2 reduced seed yield by 20% in years of normal precipitation (i.e., 1999) and by <36% in stressful environments (i.e., 2001). Seed yield decrease due to population reduction below 20 plants m-2 was severe and was highly dependent on environment. For example, decreases in population to 10 and 5 plants m-2 reduced seed yield by 42 to 92% of 80 plant m-2 in 1999 and 2001 trials. However, the unique observation in this study was when the 40 plants m-2 was nonuniformly distributed, seed yield reduced significantly from 80 plants m-2 in 2 of 3 observations (Table 1), indicating the importance of uniform plant distribution for higher seed yield in canola. Similarly, nonuniformity reduced seed yield in corn by reducing sink capacity (missing cobs) (Pommel and Bonhomme, 1998), and in sunflower by increasing lodging (heavier heads) (Robinson et al., 1982), although yield compensation was noticed in both crops. Similar comparisons at lower plant populations (20 and 10 nonuniform plants m-2), however, did not indicate the significance of plant distribution on seed yield.

Regression of normalized seed yield with population densities indicated a strong quadratic relationship in each environment (r2 > 0.80) (Fig. 2). The regression also indicated that in environments with higher yield potentials (e.g., 2000), yield plasticity by remaining plants compensated for lower plant population across a wider range of populations. The yield plasticity in environments with lower yield potentials (e.g., 2001LS) was less effective. These results suggest that canola seed yield plasticity is dependent on resource availability. Although the wide variation in weather conditions were unique to this study, McGregor (1987) also observed <20% seed yield reduction with population reductions from 100-to-200 to 40 plants m-2 under nonstressful environments, but >40% under stressful environments.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 2. Relationship between normalized seed yield (% of 80 plants m-2) of canola cv. Arrow and plant population in field trials during 1999 to 2001 early spring (ES) and 2001 late spring (LS) at Swift Current. The dotted line is the mean seed yield of 4 yr normalized to 80 plants m-2.

 
In general, the relative seed yield reduction was less than the relative plant population reduction due to the effect of interplant competition and plant plasticity. However, at very low population (<8 plants m-2), seed yield declines more than the population, indicating that interplant competition no longer limits plant growth at those plant densities (McGregor, 1987). Reducing population from 10 to 5 plants m-2 (i.e., reducing population by half), reduced seed yield by more than half in only the 2000LS trial. These results indicate that interplant competition existed at 10 plants m-2, except under extremely stressful conditions (such as occurred in 2001LS). Early spring seeding and use of a herbicide-tolerant cultivar to limit weed competition in the present study may have provided more favorable growing conditions for canola. Therefore, the canola in the present trials probably utilized the more favorable conditions and exhibited greater plasticity compared with the McGregor (1987) study.

Yield compensation in canola could occur by producing more seeds on the main shoot, by producing new inflorescences from the axils of the leaves (primary branches), and by producing secondary development of inflorescences on the existing inflorescences (secondary branches). Seed yield contributions from main, primary branches, and secondary branches were compared in 2000 and in 2001ES (Fig. 3). The major yield compensation in canola was through variations in contribution from main and secondary branches, while contribution from primary branches was relatively stable. For example, contribution to seed yield from the main shoot decreased from 38 to 9% and 40 to 16% with the population reduction from 80 to 5 plants m-2 in 2000 and 2001ES, respectively, while the contribution from secondary branches increased from 7 to 51% in 2000 and 6 to 29% in 2001LS for the same population reduction. The contribution from primary branches increased with initial levels of reductions in plant population before reducing to the lowest contribution with the lowest population stand of 5 plants m-2. Although none of the previous studies have attempted to determine the proportional contribution of primary and secondary branches to seed yield in canola, similar decreases in contribution to seed yield by the main branch at lower plant populations have been reported by others (Morrison et al., 1990a; Leach et al., 1999).



View larger version (50K):
[in this window]
[in a new window]
  		Fig. 3. Effect of reducing uniformly distributed plant population on the percentage contribution of main, primary, and secondary branches to the seed yield of individual canola plant cv. Arrow in 2000 and 2001 (early spring) trials at Swift Current.

 
Harvest index was stable across a range of populations in most environments except for the extremely stressful conditions of 2001LS (Table 2). Similarly, the effect of uniform and nonuniform populations on HI only occurred in 2001LS. Thus, seasonal environments were more significant in affecting HI than populations. This indicates that under good growing conditions, plant stand does not affect HI. However, under stressful conditions, seed yield decrease was accompanied by a reduction in HI. Similar to these results, Ball et al. (2000) reported a minor role of HI in yield compensation in a short season soybean. The decreasing HI in 2001LS suggests that the extra assimilate invested in vegetative structures such as primary and secondary branches failed to increase seed yields due to moisture stress.


View this table:
[in this window]
[in a new window]
  		Table 2. Harvest index and pods per plant of Argentine canola cv. Arrow in response to population densities at Swift Current during 1999 to 2001.

 
Yield Components
The number of pods per plant increased with decreasing plant population in all environments, although the magnitude was different (Table 2). However, increased pod number only partially compensated for the decreased population. Pod number compensation also depended on growing conditions. For example, reducing plant population from 80 to 40 plants m-2 in 1999 and 2000 increased the number of pods per plant by 81 and 74%, while the same population variation in 2001ES and 2001LS increased pod number by 28 and 36%, respectively. Thus, under the more favorable conditions of 2000, very low populations (5 plants m-2) produced six times more pods per plant than high populations (80 plants m-2). Similar partial compensation of pods were also observed by Morrison et al. (1990a). Nonuniform plant stands produced pod numbers similar to those produced by the corresponding uniform plant stand, except in 1999, where pods per plant were reduced by the nonuniform plant stand.

Contribution by main, primary, and secondary branches to pod compensation was compared in 2000 and 2001ES (Fig. 4). Pod formation on the main shoot did not follow any trend in response to population reductions. This is in agreement with previous studies, where stable pod numbers were observed on the main shoot across a wide range of populations (Diepenbrock, 2000). In contrast, pod numbers on primary and secondary branches increased significantly with reduction in population and followed quadratic relationships (r2 = 0.90 to 0.99 and P < 0.10). It was interesting to note that the contribution from secondary branches to fertile pods was observed only at very low populations. However, growing season environment had a significant influence on pod production by branches. Under the favorable growing conditions of 2000, secondary pods contributed 8% of the total pod number at 80 plants m-2, and 60% of the total pod number at 5 plants m-2. The same contribution under stressful growing conditions was 6% and 38% at 80 and 5 plants m-2, respectively.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 4. Effect of uniformly distributed plant population and environment on the number of pods produced on main shoot, primary branches, and secondary branches in canola cv. Arrow during 2000 (solid triangle) and 2001 early spring (open triangle) at Swift Current.

 
Other yield components (seeds per pod and thousand seed weight) were not affected by the population variation in any of the environments (Table 3). McGregor (1987) also found that seeds per pod and seed weight were not as strongly influenced by population as were pods per plant. Early-formed pods at the top of the canopy or on the main raceme have the developmental advantage (Mendham and Salisbury, 1995) that might have masked the smaller variations in seeds per pod or seed weight in the present study.


View this table:
[in this window]
[in a new window]
  		Table 3. Seeds per pod and thousand seed weight of Argentine canola cv. Arrow in response to population densities at Swift Current during 1999 to 2001.

 
Comparison of seeds per pod on main, primary, and secondary branches separately for two years indicated that main shoot had significantly higher seeds per pod than primary branches, which had significantly higher seeds per pod than secondary branches (Fig. 5). Seed number per pod in plants is a function of resource drawing ability of ovules (Ganeshaiah and Uma Shanker, 1992). However, the effect of plant population was not significant on any of the branch categories, although the trend indicated increase in seeds per pod with reducing population (data not presented). This contrasts with the observation of Morrison et al. (1990a), who reported significant increase in seeds per pod at lower populations. However, in that study, the highest seed yield was observed with the lowest seeding rate, suggesting higher interplant competition due to a better growing environment.



View larger version (39K):
[in this window]
[in a new window]
  		Fig. 5. Mean number of seeds produced by each main shoot, primary branch, and secondary branch pod in canola cv. Arrow in 2000 and 2001 (early spring) at Swift Current.

 
Nodal Analysis
Nodal analysis of yield formation under different plant populations, especially under two contrasting growing environments, was an unique contribution of this study. Seed yield formation observed at each node level in 2000 and 2001ES indicated that an increase in the number of pods per plant was achieved by both an increase in pods per node (both on primary and secondary branches) (Fig. 6 and 7) and an increase in branches (data not presented). Increased branching is the primary response of canola to modify the plant geometry to environment (Diepenbrock, 2000). Thus, the mean number of primary branches from four experiments in the present study increased from five to nine per plant with reduction in plant population from 80 to 5 plants m-2. However, the increase in pods per node was more significant than the increase in the number of primary branches, indicating pod production is the most important yield component to determine seed yield in canola. Increase in pod numbers in canola was due to greater retention of flower buds (McGregor, 1981). Since loss of floral structure depends on competition for assimilates (Diepenbrock, 2000), greater retention of pods indicates greater availability of assimilate. Better light penetration into the canopy at lower plant populations favor retention of leaf area (McWilliam et al., 1995). Improved photosynthetic source in the form of these leaves stimulates better retention of pods.



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 6. Distribution of yield components on different nodes (top to bottom) for canola cv. Arrow at different uniform plant populations in 2000 at Swift Current. Whenever the differences among plant populations were significant (P ≤ 0.05) for a yield component at a particular node, LSD values are presented as horizontal bars. For the sake of clarity, data from 40 and 10 plants m-2 are not presented.

 


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 7. Distribution of yield components on different nodes (top to bottom) for canola cv. Arrow at different uniform plant populations in 2001 (early spring) at Swift Current. Whenever the differences among plant populations were significant (P ≤ 0.05) for a yield component at a particular node, LSD values are presented as horizontal bars. For the sake of clarity, data from 40 and 10 plants m-2 are not presented.

 
There was a strong effect of population density on the distribution of pods on the primary and secondary branches (Fig. 6 and 7). The number of pods produced by the upper few nodes (also the main raceme; Fig. 4) did not differ with population variations. At 80 plants m-2, canola produced pods on the terminal branch and primary branches on the upper few nodes, and the number of pods decreased almost linearly with an increase in node number. At 5 plants m-2, peak pod production occurred a few nodes lower in the canopy before significant decrease in pod production was observed. Therefore, lower retention of pods on the upper nodes at lower plant populations should be related to lower assimilate availability. In this study, canopy closure in canola was delayed by lower plant populations (Angadi et al., 2002). Therefore, pods formed on upper branches and vegetative growth compete for the limited photosynthate. In addition, light reflection from canola flowers reduce photosynthetic efficiency during flowering (Diepenbrock, 2000). At lower nodes the plant can support more pods due to increased photosynthate availability caused by increased light interception due to increased pod area.

Comparing 2000 and 2001LS data indicated that canola rarely used secondary or higher order branching to increase pod number. First, only at very low population densities did secondary branching contribute to productivity. Second, the extent of using secondary branching was much lower under stressful conditions of 2001LS compared with 2000. Similar to pods produced on the primary branches at lower populations, peak secondary pod production occurred lower in the canopy. In general, later-formed branches are inefficient in producing seed yield in canola (Diepenbrock, 2000). In canola, flowering takes place in acropetal succession, while branching takes place in basipetal succession (Mendham and Salisbury, 1995). Racemes on the upper portion of the canopy are formed early and mature early, which provides a maturity advantage for high density canola (McGregor, 1981). We observed 3- to 4-d earlier maturity at 80 plants m-2 compared with 5 plants m-2 in the present study (data not shown).

Seeds per pod and thousand seed weight in the upper portion of the canopy were similar among different population treatments (Fig. 6 and 7). This again shows that the number of pods is more responsive to population than other yield components. Reductions in seeds per pod and seed weight started at upper nodes in denser canopies compared with sparser canopies. None of the previous studies have focused on seeds per pod and seed weight lower in the canola plant canopy. Results of this study indicated that seeds per pod and seed weight also played a role in yield compensation in canola at low plant populations. The early reduction in seeds per pod and seed weight again may be a result of lack of source (photosynthesis from lower leaves and lower pods) to support pod filling at higher plant densities (McWilliam et al., 1995). This is supported by the lower seed:pod ratio on lower nodes than those on upper nodes.


   SUMMARY
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 SUMMARY
REFERENCES
 
Canola exhibited plasticity to maintain seed yield across a wide range of population in the semiarid conditions. Reducing population by 50% from 80 to 40 uniformly distributed plants m-2 had no effect on seed yield. However, if the reduced population was distributed nonuniformly, then seed yield reductions were observed in 1999 and 2001. The plant structure adapted to the growing conditions by increasing branches and pods per plant as plant populations were reduced. However, the plant plasticity did not fully compensate for reduced plant populations as seed yields reduced curvilinearly as plant populations decreased. The number of pods per plant was the most important factor responsible for yield compensation, while seeds per pod and seed weight did not significantly contribute to yield compensation. Increase in pods per plant was achieved through both increased branching and increased pod retention at each node. Plant population affected seeds per pod and seed weight only for the lower nodes.


   ACKNOWLEDGMENTS
 
The research was financed by Saskatchewan Agriculture Development Fund, Saskatchewan Canola Development Commission, and AAFC-Matching Investment Initiative. We also thank Don Sluth, Evin Powell, Dean James, Dean Klassen, Rod Ljunggren, and a team of summer students for technical help. Authors also thank Dr. Hong Wang and Keith Hanson for reviewing the manuscript.

Received for publication April 22, 2002.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 SUMMARY
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
Y. Gan, S. S. Malhi, S. Brandt, F. Katepa-Mupondwa, and H. R. Kutcher
Brassica juncea Canola in the Northern Great Plains: Responses to Diverse Environments and Nitrogen Fertilization
Agron. J., August 10, 2007; 99(5): 1208 - 1218.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
J. Wang, Y. T. Gan, F. Clarke, and C. L. McDonald
Response of Chickpea Yield to High Temperature Stress during Reproductive Development
Crop Sci., September 8, 2006; 46(5): 2171 - 2178.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
L. M. Dosdall and F. C. Stevenson
Managing Flea Beetles (Phyllotreta spp.) (Coleoptera: Chrysomelidae) in Canola with Seeding Date, Plant Density, and Seed Treatment
Agron. J., October 19, 2005; 97(6): 1570 - 1578.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
C. Chen, G. Jackson, K. Neill, D. Wichman, G. Johnson, and D. Johnson
Determining the Feasibility of Early Seeding Canola in the Northern Great Plains
Agron. J., July 13, 2005; 97(4): 1252 - 1262.
[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 (14)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Angadi, S. V.
Right arrow Articles by Gan, Y.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Angadi, S. V.
Right arrow Articles by Gan, Y.
Agricola
Right arrow Articles by Angadi, S. V.
Right arrow Articles by Gan, Y.
Related Collections
Right arrow Best Management Practices
Right arrow Crop Growth and Development
Right arrow Canola

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