Winter Wheat Seeding Decisions for Improved Grain Yield and Yield Components

C.F


Introduction
In recent decades, much of the improvement in wheat production has largely been driven by nutrient management and advances in crop genetics [1][2][3][4]. In addition, other agronomic practices have long been investigated as a means of enhancing the productivity of food, forage, and energy crops [5][6][7][8]. This is expected to continue, and as new cultivars are introduced, it becomes paramount that refinement of agronomic practices is regularly conducted. In particular, dryland cropping systems may require sound agronomic practices to make the best use of limited precipitation [9]. This could help to bridge the yield gap between what is realized, and the yield potential or maximum yield associated with a particular crop cultivar. One aspect of agronomy that can be further explored is the seeding rate for different available genotypes.
Previous research has shown that modern winter wheat cultivars (Triticum aestivum L.) may do well under a high seeding rate [10]. As a result, an optimum plant density can lead to efficient use of resources resulting in improved crop productivity [11]. One factor that influences plant density is the tillering potential of a given cultivar [12]. Lower plant density is likely to be used for cultivars with high tillering potential [13]. Conversely, All fields were under no-till production and planted with a no-till drill that has single disk John Deere openers. Nitrogen (N) fertilizer was streamed pre-plant as urea ammonium nitrate (32-0-0) based on soil recommendations for each field belonging to each of the participating growers. On average, growers applied 62 kg N ha −1 preplant followed by 11 kg N ha −1 as top-dressed urea ammonium nitrate (32-0-0) in the spring during spring weed control at tillering/spring greenup. No phosphorus (P) and potassium (K) fertilizers were applied as the test of soil supply of P (34 mg kg −1 ) and K (132 mg kg −1 ) was considered to be at a sufficient level for winter wheat production. Pre-plant herbicides included burndown with glyphosate (N-(phosphonomethyl)glycine) at 2.3 L ha −1 with ammonium sulfate (AMS) and 2,4-D LV-6 (2,4-Dichlorophenoxy acetic acid, 2-ethylhexyl ester) at 0.6 L ha −1 applied two weeks before the first planting date. The second herbicide application in the spring was achieved with 2,4-D LV-6 at a rate of 0.6 L ha −1 applied during the spring top-dressing of N. Spot weeding to clean alleys was occasionally done with a hand application of 2,4-D or manual hoeing. Crops were produced under a dryland production system without supplemental irrigation. Tillers were counted at maturity (prior to harvest) from a 30.5 cm row length at two different locations within a plot. From the same samples used for counting tillers, biomass weight and grain yield were measured. From the resulting grains, a 1000-kernel weight for wheat was determined from 1000 randomly selected seeds from each experimental unit. The moisture content of the grain yield was adjusted to 125 g kg −1 .
Statistical analysis was achieved using PROC GLIMMIX in SAS 9.4 [28]. Since the treatment design was a nested factorial, but the row spacing treatment necessitated the use of strips for the row spacing by seed rate effects, a split plot topographical design was considered in the analysis. Whole plot effect was the planting date, split plot effect included the row spacing within planting date. Split-split plot and nested effect was the seed rate within row spacing and the combination of variety and nested seed rate term. A linear mixed model was employed using the GLIMMIX Procedure of SAS 9.4 where year and replication within year were treated as random effects. Mean square errors and F-statistics were calculated using the appropriate error terms and degrees of freedom that accounted for both the treatment structure and topographical design aspects. Restricted maximum likelihood (REML) and Gaussian assumptions were used to estimate the variance components of the model. Least squares means (LS-means) were computed using LSMeans statement with a standard error option provided to show the precision of the estimate. Visualization was accomplished in R [29]. Tidyverse, a collection of several R packages, was deployed in the visualization [30].

Grain Yield
At Hemingford, the interaction between planting date, variety, and seed rate nested within row spacing had no significant effect on winter wheat grain yield (p > 0.05, Table 2). Apart from planting date × variety, row spacing × variety, and planting date × seed rate nested within row spacing interactions (p ≤ 0.01), there were no significant effects of two-way, and three-way interactions (p > 0.05) on the grain yield. As main effects, variety and row spacing influenced grain yield (p ≤ 0.01). The interaction between variety and row spacing ( Figure 1) meant that the choice of a variety was conditioned on the row spacing used. For instance, Robidoux had 5.4% substantially more grain yield with a row spacing of 19 cm than Goodstreak (5.2 Mg ha −1 ) planted at the same spacing. However, at the spacing of 25 cm, a grower could choose to plant any of the two varieties as there was no significant yield difference between them. Additionally, even though all the varieties yielded more at a spacing of 19 cm, Wesley planted at 25 cm generated a yield (5.5 mg ha −1 ) that significantly exceeded that of Goodstreak planted at a spacing of 19 cm by 5.6%. This illustrates the dependence of variety on row spacing, that is, if planting was to be done at 25 cm, then the choice of a cultivar was likely to be Wesley. Planting cultivars at narrow row spacing led to more yield than wide spacing possibly because close spacing improves the utilization of soil nutrients [31]. This may be because narrow row spacing leads to more spikes per unit area and together with the correct choice of a variety may lead to an improvement in grain yield [32]. Further, the yield tends to be higher for narrowly spaced varieties because of an increase in the number of tillers per unit area [33]. Planting date × variety interaction showed that grain yield was greatest when Wesley was planted at an optimum date (Figure 2a). This yield (6.0 Mg ha −1 ) was significantly greater than the yield of Wesley planted before the optimum planting date (early) and all the cultivars planted during the late planting dates by at least 16.1%.      Like results at Hemingford, planting date × variety × seeding rate nested within row spacing did not influence the final grain yield at McCook experimental site (p > 0.05, Table 2). Planting date and variety interacted to significantly affect the final winter wheat grain yield (p < 0.01). The rest of the interaction terms had no major effect on grain yield. For planting date × variety interaction, the greatest grain yield (5.3 Mg ha −1 ) was obtained with early planting date and Wesley ( Figure 2). This grain yield was similar for all the cultivars planted early, on-time or late except for late-planted Wesley and Robidoux whose average yield was 58.3% lower than that of early planted Wesley. Row spacing, as the main effect, did not substantially affect the final grain yield (p > 0.05). This meant that both spacings of 19 and 25 cm resulted in a similar grain yield.
The grain yield obtained from Sidney was not substantially affected by the interaction between the different variables considered in this study (p > 0.05, Table 2). Among the main effects, it was the variety that affected the grain yield for winter wheat (p < 0.05). This result indicated that Goodstreak and Robidoux had similar grain yields averaging 2.9 Mg ha −1 (Figure 3). At this site, Wesley had the lowest grain yield which was statistically different from the other two cultivars by at least 4.4%.
Although not in all cases, grain yield appeared not to vary widely among cultivars planted at optimum and early planting dates. However, the choice of a variety becomes important when planting late. Wesley and Goodstreak are apparently better choices when producers contemplate planting late as they yielded greater than Robidoux at Hemingford. Late planting possibly did not allow the crop to accumulate enough growth before the winter dormancy period [34]. Additionally, late planting dates possibly reduced grain yields because of the reduction in kernel weight and the number of kernels per unit area [35,36]. However, as shown in Figure 2a, some varieties will tolerate delayed planting more than others. In this case, Wesley yielded more than Robidoux when both were planted late. Late planting possibly subjects the crop to low temperature during the vegetative growth stage and high temperature during the critical grain filling time [37]. Optimum planting dates increase the possibility of accumulating enough growing degree days before vernalization to promote the transition from the vegetative to reproductive phase [34,38]. This may explain why grain yield was more for varieties planted on time. The height of varieties may also influence the final grain yield [4]. Wesley might have performed better Agronomy 2022, 12, 3061 7 of 16 than other cultivars as it accumulates less biomass due to its short height and directed more resources towards kernels per head. did not substantially affect the final grain yield (p > 0.05). This meant that bot of 19 and 25 cm resulted in a similar grain yield.
The grain yield obtained from Sidney was not substantially affected by t tion between the different variables considered in this study (p > 0.05, Table the main effects, it was the variety that affected the grain yield for winter whea This result indicated that Goodstreak and Robidoux had similar grain yields 2.9 Mg ha −1 (Figure 3). At this site, Wesley had the lowest grain yield which w cally different from the other two cultivars by at least 4.4%. Although not in all cases, grain yield appeared not to vary widely amon planted at optimum and early planting dates. However, the choice of a variet important when planting late. Wesley and Goodstreak are apparently better cho producers contemplate planting late as they yielded greater than Robidoux a ford. Late planting possibly did not allow the crop to accumulate enough gro the winter dormancy period [34]. Additionally, late planting dates possibly red yields because of the reduction in kernel weight and the number of kernels pe [35,36]. However, as shown in Figure 2a, some varieties will tolerate delaye more than others. In this case, Wesley yielded more than Robidoux when planted late. Late planting possibly subjects the crop to low temperature durin etative growth stage and high temperature during the critical grain filling tim timum planting dates increase the possibility of accumulating enough grow days before vernalization to promote the transition from the vegetative to rep phase [34,38]. This may explain why grain yield was more for varieties plante The height of varieties may also influence the final grain yield [4]. Wesley m

Biomass
The interaction between planting date and crop variety significantly affected the quantity of biomass produced by wheat grown at Hemingford (p < 0.05; Table 3). Row spacing which had no significant interaction effect with other variables had a substantial effect on wheat biomass when considered as the main effect (p ≤ 0.01). Apart from the significant interrelationship above, the other interaction terms did not result in biomass differences among treatments (p > 0.05). Substantially more biomass (16.2 Mg ha −1 ) was produced by planting winter wheat at a row spacing of 19 cm than at 25 cm ( Figure 4). Relative to 19 cm row spacing, the biomass was 19% lower for wheat planted at 25 cm row spacing. A similar result was observed by May et al. [39] who found biomass to be lower at wider spacing. This might be because wider spacing may be less efficient at suppressing weeds when compared to narrow row spacing [40]. This is because wide row spacing provides less ground cover, particularly during early growth, allowing weeds to grow as well as increased evaporation [41]. Alternatively, narrow row spacing led to more efficient use of soil nutrients that translated into more plant biomass [32]. Planting date × variety interaction revealed that the greatest biomass (16.6 Mg ha −1 ) was attained with Robidoux planted at optimum planting dates (Figure 5a). This biomass yield was substantially higher than that of other varieties planted early, on time or late. However, late planting appeared to favor Goodstreak since it had 10% more biomass than Wesley. This drastic difference in biomass was not observed between late-planted Robidoux and Goodstreak. performed better than other cultivars as it accumulates less biomass due to its sh and directed more resources towards kernels per head.

Biomass
The interaction between planting date and crop variety significantly af quantity of biomass produced by wheat grown at Hemingford (p < 0.05; Tab spacing which had no significant interaction effect with other variables had a s effect on wheat biomass when considered as the main effect (p ≤ 0.01). Apar significant interrelationship above, the other interaction terms did not result i differences among treatments (p > 0.05). Substantially more biomass (16.2 Mg produced by planting winter wheat at a row spacing of 19 cm than at 25 cm Relative to 19 cm row spacing, the biomass was 19% lower for wheat planted at spacing. A similar result was observed by May et al. [39] who found biomass t at wider spacing. This might be because wider spacing may be less efficient at su weeds when compared to narrow row spacing [40]. This is because wide ro provides less ground cover, particularly during early growth, allowing weeds well as increased evaporation [41]. Alternatively, narrow row spacing led to mo use of soil nutrients that translated into more plant biomass [32]. Planting dat interaction revealed that the greatest biomass (16.6 Mg ha −1 ) was attained with planted at optimum planting dates (Figure 5a). This biomass yield was su higher than that of other varieties planted early, on time or late. However, lat appeared to favor Goodstreak since it had 10% more biomass than Wesley. T difference in biomass was not observed between late-planted Robidoux and G   With the exception of variety, none of the interactions nor main effects resulted in differences of winter wheat biomass at McCook (p < 0.05; Table 4). For varietal differences, the biomass of 12.5 Mg ha −1 produced by Goodstreak exceeded the quantity produced by Wesley and Robidoux by an average of 17.9% ( Figure 6). The tall height of Goodstreak may be contributing to more biomass accumulated by this variety when compared to Wesley and Robidoux which are short and medium in height, respectively [20]. In addition, With the exception of variety, none of the interactions nor main effects resulted in differences of winter wheat biomass at McCook (p < 0.05; Table 4). For varietal differences, the biomass of 12.5 Mg ha −1 produced by Goodstreak exceeded the quantity produced by Wesley and Robidoux by an average of 17.9% ( Figure 6). The tall height of Goodstreak may be contributing to more biomass accumulated by this variety when compared to Wesley and Robidoux which are short and medium in height, respectively [20]. In addition, Goodstreak is often a mixed purpose wheat that can be used for grain or for hay production. Since the effect of row spacing was not significant, the significant interaction for seed rate nested within row spacing suggests that seed rates at each level of row spacing drove the significant differences in wheat biomass.
Biomass at Sidney experimental site was affected by planting date × variety and row spacing × variety interactions (p < 0.05; Table 3). These were the only interaction terms that were significant with the rest of the interactions and main effects of planting date, row spacing, and seed rate nested within row spacing having no significant effect on biomass yield (p > 0.05). Results from planting date × variety interaction revealed early planting date and Robidoux as the treatment combination with the greatest biomass of 42.1 Mg ha −1 (Figure 5b). All the cultivars performed poorly with optimum planting dates with an average biomass yield of 11.8 Mg ha −1 . The reason for this phenomenon is unclear as this was expected to mainly affect the late-planted varieties because of a reduction in spikes per unit area [42]. In terms of variety × row spacing, Goodstreak planted at 19 cm had a biomass of 36.6 Mg ha −1 , a biomass yield that significantly exceeded the biomass produced by Robidoux and 19 cm row spacing treatment combination by 4.3% (Figure 7). Similarly, Goodstreak and Robidoux planted at narrow row spacing (19 cm) had on average 34.8% more biomass than all the varieties planted at a wide row spacing (25 cm). This study showed that a producer wishing to produce more biomass may choose Goodstreak planted at 19 cm because of statistical difference in biomass but at 25 cm, the choice could be either Goodstreak or Robidoux since they had similar biomass.  3 Seeding rate. It is nested within row spacing, meaning that every interaction term involving seed rate was nested within row spacing since seed rates were different for each spacing. 4 Variety. NS, not significant at p > 0.05. Mean square error of biomass was analyzed in kg ha −1 while a 1000-kernel weight was treated as mg 1000 seeds −1 .

Agronomy 2022, 12, x FOR PEER REVIEW
Goodstreak is often a mixed purpose wheat that can be used for grain or for h tion. Since the effect of row spacing was not significant, the significant interacti rate nested within row spacing suggests that seed rates at each level of row spa the significant differences in wheat biomass.   Goodstreak and Robidoux planted at narrow row spacing (19 cm) had on average more biomass than all the varieties planted at a wide row spacing (25 cm). Thi showed that a producer wishing to produce more biomass may choose Goo planted at 19 cm because of statistical difference in biomass but at 25 cm, the choic be either Goodstreak or Robidoux since they had similar biomass.

1000-Kernel Weight
A 1000-kernel weight was not influenced by both the main effect and the inter among several variables considered in this study at Hemingford (p > 0.05; Table showed that a 1000-kernel weight at this site did not depend on planting date, row ing, variety, seed rate, and/ or the interaction among them at this environment. Variety and the interrelationship between planting date and variety affec 1000-kernel weight at McCook experimental site (p < 0.05; Table 4). There was n effect of other main and interaction effects on the 1000-kernel weight at the study s 0.05). Wesley planted at an optimum date resulted in the highest weight of 42.1 g p seeds ( Figure 8). This result suggests that planting Wesley at the optimum time w to more weight of 1000 seeds and yet if the choice was to plant Robidoux, then early ing time may be more appropriate. Planting at an optimum time particularly for could lead to an adequate but not excessive number of tillers resulting in mor weight [43]. Shahzad et al. [44] reported a consistent decrease in a 1000 kernel we each delay in planting date and that varieties responded differently. As varieties bred to improve a particular crop trait, a 1000 kernel weight may vary among v potentially making them respond differently to planting dates. Delayed plantin largely responsible for the lower 1000-kernel weight associated with most varieti As grain starch content plays an important role in the weight attained by the gra layed planting likely reduced its rate of accumulation, and the magnitude varies f variety [45]. The reduction may be more significant in a warm and dry winter and [26]. Therefore, the selection of a variety also needs to consider the time of plan both interact to affect a 1000-kernel weight.

1000-Kernel Weight
A 1000-kernel weight was not influenced by both the main effect and the interactions among several variables considered in this study at Hemingford (p > 0.05; Table 4). This showed that a 1000-kernel weight at this site did not depend on planting date, row spacing, variety, seed rate, and/ or the interaction among them at this environment.
Variety and the interrelationship between planting date and variety affected the 1000-kernel weight at McCook experimental site (p < 0.05; Table 4). There was no such effect of other main and interaction effects on the 1000-kernel weight at the study site (p > 0.05). Wesley planted at an optimum date resulted in the highest weight of 42.1 g per 1000 seeds ( Figure 8). This result suggests that planting Wesley at the optimum time will lead to more weight of 1000 seeds and yet if the choice was to plant Robidoux, then early planting time may be more appropriate. Planting at an optimum time particularly for Wesley could lead to an adequate but not excessive number of tillers resulting in more grain weight [43]. Shahzad et al. [44] reported a consistent decrease in a 1000 kernel weight for each delay in planting date and that varieties responded differently. As varieties may be bred to improve a particular crop trait, a 1000 kernel weight may vary among varieties potentially making them respond differently to planting dates. Delayed planting was largely responsible for the lower 1000-kernel weight associated with most varieties [35]. As grain starch content plays an important role in the weight attained by the grain, delayed planting likely reduced its rate of accumulation, and the magnitude varies for each variety [45]. The reduction may be more significant in a warm and dry winter and spring [26]. Therefore, the selection of a variety also needs to consider the time of planting as both interact to affect a 1000-kernel weight. Agronomy 2022, 12, x FOR PEER REVIEW 11 of 15 At Sidney, none of the factors considered influenced a 1000-kernel weight either as main effects or the interaction among them (p > 0.05; Table 4). This is similar to observations made at Hemingford. Liu et al. [46] reported that planting date did not affect a 1000kernel weight. Although a 1000-kernel weight was less influenced by the planting date at the two sites, it is important to view the broader context that grain yield may be reduced due to the reduction in the number of spikes and the production of dry matter and nitrogen [46].

Tillers
A significant effect of the winter wheat variety was observed on the number of tillers produced at Hemingford (p < 0.01; Table 4). However, the number of tillers was not influenced by planting date, row spacing, seed rate, and the interaction among the variables. Wesley produced on average 10% fewer tillers when compared to Goodstreak and Robidoux which had an average number of tillers that equaled 6.4 M ha −1 (Figure 9).  At Sidney, none of the factors considered influenced a 1000-kernel weight either as main effects or the interaction among them (p > 0.05; Table 4). This is similar to observations made at Hemingford. Liu et al. [46] reported that planting date did not affect a 1000-kernel weight. Although a 1000-kernel weight was less influenced by the planting date at the two sites, it is important to view the broader context that grain yield may be reduced due to the reduction in the number of spikes and the production of dry matter and nitrogen [46].

Tillers
A significant effect of the winter wheat variety was observed on the number of tillers produced at Hemingford (p < 0.01; Table 4). However, the number of tillers was not influenced by planting date, row spacing, seed rate, and the interaction among the variables. Wesley produced on average 10% fewer tillers when compared to Goodstreak and Robidoux which had an average number of tillers that equaled 6.4 M ha −1 (Figure 9). At Sidney, none of the factors considered influenced a 1000-kernel weight either as main effects or the interaction among them (p > 0.05; Table 4). This is similar to observations made at Hemingford. Liu et al. [46] reported that planting date did not affect a 1000kernel weight. Although a 1000-kernel weight was less influenced by the planting date at the two sites, it is important to view the broader context that grain yield may be reduced due to the reduction in the number of spikes and the production of dry matter and nitrogen [46].

Tillers
A significant effect of the winter wheat variety was observed on the number of tillers produced at Hemingford (p < 0.01; Table 4). However, the number of tillers was not influenced by planting date, row spacing, seed rate, and the interaction among the variables. Wesley produced on average 10% fewer tillers when compared to Goodstreak and Robidoux which had an average number of tillers that equaled 6.4 M ha −1 (Figure 9).   Table 4). However, at McCook, Goodstreak had significantly more tillers than either Robidoux or Wesley (data not shown).
In Sidney, row spacing and variety as well as the interactions between planting date and variety, and the individual main effects of row spacing and variety affected the total number of tillers recorded (p < 0.03; Table 4). The rest of the treatment combinations and planting date (main effect) only differed in the number of tillers because of a random experimental error (p > 0.05). Early planting of Robidoux led to more tillers (7.2 M ha −1 ) and was comparable to that of early planted Goodstreak ( Figure 10). Late planting dates for all the varieties led to an average of 50.9% lower number of tillers when compared to early planted Robidoux (Figure 10. However, late-planted Robidoux and Goodstreak still outperformed late-planted Wesley since it had only 4.8 M tillers ha −1 . At McCook, the main effect of variety led to a significant difference in wheat tillers, while none of the interaction effects had significant impacts on tillering, similar to the conclusion found in Hemingford for the same trait (p = 0.05; Table 4). However, at McCook, Goodstreak had significantly more tillers than either Robidoux or Wesley (data not shown).
In Sidney, row spacing and variety as well as the interactions between planting date and variety, and the individual main effects of row spacing and variety affected the total number of tillers recorded (p < 0.03; Table 4). The rest of the treatment combinations and planting date (main effect) only differed in the number of tillers because of a random experimental error (p > 0.05). Early planting of Robidoux led to more tillers (7.2 M ha −1 ) and was comparable to that of early planted Goodstreak ( Figure 10). Late planting dates for all the varieties led to an average of 50.9% lower number of tillers when compared to early planted Robidoux (Figure 10. However, late-planted Robidoux and Goodstreak still outperformed late-planted Wesley since it had only 4.8 M tillers ha −1 . Row spacing × variety interaction showed a higher number of tillers for closely spaced (19 cm) Goodstreak and Robidoux with an average of 6.9 M tillers ha −1 (Figure 10). The two varieties also had an average of 5.3 M tillers ha −1 when spaced at 25 cm, a number substantially lower than for a row spacing of 19 cm. Wesley planted at 25 cm resulted in 52.1% lowest number of tillers relative to Goodstreak spaced at 19 cm. Efficient utilization of nutrients, space, and solar radiation may be responsible for why Goodstreak had more tillers under narrow row spacing [47]. Plants attempt to compensate for the wide spacing by producing more tillers but as this study demonstrates, the production of tillers is not adequate to match those produced under narrow spacing. The choice of the right variety and spacing can reduce soil evaporation to ensure more efficient use of the moisture and this may translate into more tillers [41]. In our study, narrow row spacing used with Goodstreak led to a greater number of tillers. Wide spacing is possible to use with all the varieties as they give a related number of tillers.
Early planting of Robidoux and Goodstreak led to more tillers when compared to wheat planted at an optimum time or late in the season. This enhanced tiller formation with these early planted varieties is expected but can also lead to competition and depletion of soil moisture and may not translate to more grain yield [43]. Row spacing × variety interaction showed a higher number of tillers for closely spaced (19 cm) Goodstreak and Robidoux with an average of 6.9 M tillers ha −1 (Figure 10). The two varieties also had an average of 5.3 M tillers ha −1 when spaced at 25 cm, a number substantially lower than for a row spacing of 19 cm. Wesley planted at 25 cm resulted in 52.1% lowest number of tillers relative to Goodstreak spaced at 19 cm. Efficient utilization of nutrients, space, and solar radiation may be responsible for why Goodstreak had more tillers under narrow row spacing [47]. Plants attempt to compensate for the wide spacing by producing more tillers but as this study demonstrates, the production of tillers is not adequate to match those produced under narrow spacing. The choice of the right variety and spacing can reduce soil evaporation to ensure more efficient use of the moisture and this may translate into more tillers [41]. In our study, narrow row spacing used with Goodstreak led to a greater number of tillers. Wide spacing is possible to use with all the varieties as they give a related number of tillers.
Early planting of Robidoux and Goodstreak led to more tillers when compared to wheat planted at an optimum time or late in the season. This enhanced tiller formation with these early planted varieties is expected but can also lead to competition and depletion of soil moisture and may not translate to more grain yield [43].

Conclusions
Overall, evidence from this study suggests that early and optimum planting dates may improve grain yield and that this is often driven by the variety planted. In particular, optimum planting dates may offer some insurance against volatile conditions that may affect early or late-planted crops. Although early planting dates also improved yield, tillers, 1000-kernel weight, and biomass depending on the variety grown, further studies may be necessary to unearth evidence that reinforces its effectiveness in the region. The choice of a variety also depended on the row spacing or planting date and this affected a range of responses including yield, biomass, and tillers. For example, Robidoux attained 5.4% significantly more grain yield at 19 cm row spacing than Goodstreak (5.2 Mg ha −1 ) planted at the same spacing. However, at 25 cm row spacing, any of the two varieties could be grown since they produced identical yields. At two of the three sites, 1000-kernel weight exhibited no response to the main and interaction effects. At the site where 1000-kernel weight was affected by the interaction between variety and planting date, Wesley planted at the optimum time produced the largest 1000-kernel weight (42.1 g per 1000 seeds). However, for early planting, the advantage gained by Wesley planted on-time becomes nonsignificant relative to early planted Robidoux. A disadvantage of planting early, however, is the increased risk of diseases such as wheat streak mosaic virus.
This illustrates the need for considering the interrelationships among these variables in decisions as to which variety to grow at a particular location. Generally, narrow row spacing was more influential on grain yield, biomass, and tillers, suggesting that wider row spacing may have decreased solar radiation utilization efficiency, increased weed competitiveness, or lower soil nutrient and water efficiency. In some instances, the interaction effect was not observed but with significant main effects. This was demonstrated by varietal differences where Goodstreak produced a biomass of 12.5 Mg ha −1 and this was 17.9% more biomass than the average quantity produced by Wesley and Robidoux. This could be associated with the tall height of Goodstreak and the short and medium heights of Wesley and Robidoux, respectively. Generally, because of the interrelationship exhibited in this study, the decision about which variety to grow should also consider the planting date, and/or row spacing used at a particular site. Although early planting can have a positive impact on grain yield, early planting can increase the risk of exposure to insect and disease pressure.