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


     

Published in Agron J 100:344-351 (2008)
DOI: 10.2134/agrojnl2007.0189
© 2008 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 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 Google Scholar
Google Scholar
Right arrow Articles by Gélinas, B.
Right arrow Articles by Seguin, P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Gélinas, B.
Right arrow Articles by Seguin, P.
Agricola
Right arrow Articles by Gélinas, B.
Right arrow Articles by Seguin, P.
Related Collections
Right arrow Best Management Practices
Right arrow Grazing Management
Right arrow Turfgrass

NEW CROPS

Evaluation of Management Practices for Grain Amaranth Production in Eastern Canada

Bruce Gélinas and Philippe Seguin*

Dep. Plant Sci., McGill Univ., Macdonald Campus, 21111 Lakeshore Rd., Sainte-Anne-de-Bellevue, QC H9X 3V9, Canada

* Corresponding author (philippe.seguin{at}mcgill.ca).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Grain amaranth (Amaranthus spp.) is a C4 dicotyledonous pseudocereal crop that was widely cultivated in pre-Columbian America. It was successfully introduced in many regions with contrasting environmental conditions. The introduction of grain amaranth in eastern Canada would represent an opportunity for diversification. A study was conducted to evaluate management practices for grain amaranth grown in this region. Three field experiments replicated in three environments were conducted to evaluate the following factors: (i) seeding date (mid-May, early-June, and mid-June) and cultivar (K432, K593, and Plainsman); (ii) row spacing (38, 58, and 76 cm) and seeding rate (1, 2, and 4 kg ha–1), and (iii) N fertilization rate (0, 50, 100, 150, and 200 kg N ha–1) and cultivar (D136 and Plainsman). Seeding date affected grain yield in only one out of three environments, with the earlier date resulting in the highest yields. Cultivars differed in yield in only one of three environments, with Plainsman resulting in highest yields. Later seeding dates resulted in higher seed moisture at harvest in all environments. Seeding rate and row spacing did not affect grain yield, but row spacing affected grain moisture at harvest, with narrower rows resulting in grains with lower moisture content. Nitrogen fertilization increased yield and lodging in only one environment. Seed moisture and plant height were positively related to N fertilization in all environments. Cultivar D136 yielded more than Plainsman in 2005 and less in 2006. Therefore, grain amaranth production in eastern Canada seems possible, management practices having limited impact on grain yield, which averaged 923 kg ha–1 across all experiments and environments.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.

Received for publication June 3, 2007.
   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
GRAIN AMARANTH is a C4 dicotyledonous pseudocereal crop that was widely cultivated in pre-Columbian America (Sauer, 1950). Its cultivation decreased to a point of near extinction in the early 20th century due to pressure from the conquistadors, who disliked its use in native ceremonies (Sauer, 1950). Interest in grain amaranth was revived in the 1970s, after reports of its high protein and lysine content (Downton, 1973). This plant is adaptable to a wide range of environments and has been successfully introduced in several regions with contrasting environmental conditions (National Academy of Science, 1984). In the USA, range of cultivation of the crop extends as far north as North Dakota (Henderson et al., 1998), with most production being concentrated in Nebraska.

Nutritionally, grain amaranth is interesting due to its high protein (i.e., 15 to 18%), lysine, and calcium concentrations and its lack of gluten (Petr et al., 2003). It is estimated that over 2.1 million people in the USA alone are affected by the celiac disease, which renders affected people intolerant to gluten (Celiac Sprue Association, 2004). This makes grain amaranth a crop with a great market potential. Grain amaranth is currently a niche market crop, with most sales coming from health food stores. Several commercial products can be prepared with grain amaranth, including snacks, bars, breakfast cereals, breads, and pasta. Hackman and Myers (2003) identified three key reasons that motivate consumers to purchase grain amaranth: (i) people affected by celiac disease need gluten-free food, (ii) the favorable nutritional profile of the grain, and (iii) the desire for more exotic food.

Seeding of grain amaranth in the northern USA usually takes place from mid-May to mid-June (Henderson et al., 1998). Due to the requirements for sufficient moisture, high soil temperature, and shallow seeding depth, the time where optimal emergence conditions are met is more limited for grain amaranth than for larger seeded crops (Webb et al., 1987). It has been suggested that seeding should be done approximately 2 wk after the last spring frost, seeding in early-June being recommended for the Northern Great Plains (Henderson et al., 1998). Under northern latitudes in the United States, the usual harvesting method for grain amaranth consists in harvesting approximately 10 d after the first killing frost, which allows for a good dry-down of the plant (Sooby et al., 2005). Freezing therefore acts as a desiccant.

Plant population density studies conducted with grain amaranth often report conflicting results; environmental factors most likely accounting for a great part of the differences observed. It has been suggested that water availability should determine optimal plant population, higher populations requiring greater amounts of water (Weber, 1987). In North Dakota, Henderson et al. (2000) found a significant environment x plant density effect on grain yield, suggesting that different population densities should be adopted in different environments. The plasticity of the plant's morphology may limit its response to seeding rate and row spacing (Henderson et al., 2000). As no herbicide is currently available for weed control in grain amaranth, mechanical weed control must be practiced, which limits the options for row width (Sooby et al., 2005). Narrower row spacing has been suggested as a means of reducing weed pressure, due to the resulting earlier canopy closure (Peiretti and Gesumaria, 1998); this, however, makes mechanical weed control difficult. Others have argued that wider rows result in increased competition in the row, reduced plant height, later maturity (Myers, 1996), and increased lodging (Henderson et al., 2000).

Soil fertility requirements of grain amaranth must be defined for the different environments where it is cultivated. It was suggested that more N may be required in higher rainfall areas (Stallknecht and Schulz-Schaffer, 1993). In Missouri, Myers (1998) evaluated N fertilization rates between 0 and 180 kg ha–1; yields increased with N fertilization up to 90 kg ha–1. Nitrogen fertilization, however, increased seed moisture at harvest and number of days to anthesis. In Minnesota, Elbehri et al. (1993) obtained yield response with N fertilization up to 180 kg ha–1; no responses were observed for P and K. Yields increased from 1094 kg ha–1 without N application to 1428 kg ha–1 with N applied at 180 kg ha–1.

Grain amaranth belongs to a different family than all of the main crops grown in eastern Canada and could thus be a good addition to existing crop rotations. It is from the same genus as redroot pigweed (Amaranthus retroflexus L.), a problematic weed in the area; however, grain amaranth is unlikely to become a weed problem as it usually only produces a very small proportion of hard or dormant seeds (Gélinas, personal observations). Grain amaranth seems adapted to eastern Canada, as demonstrated by preliminary trials in southwestern Québec and southern Ontario, where yields comparable to those reported from the northern USA were obtained (Gélinas and Seguin, unpublished results, 2007). However, management strategies for the region have not been researched. The objectives of this research were therefore to determine appropriate (i) seeding dates, (ii) seeding rate and row spacing, and (iii) N fertilization rate for grain amaranth when grown in eastern Canada.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Seeding Date and Cultivar Experiment
This experiment was conducted at two sites in 2005 and one site in 2006 (Table 1 ). A randomized complete block design with a split-plot arrangement and four blocks was used. The main plot consisted of three seeding dates (mid-May, early-June, and mid-June); the subplot consisted of three cultivars [K432 (Amaranthus hybrid), K593 (Amaranthus hybrid), and Plainsman (A. hypochondriacus) (obtained from D. Baltensperger, Univ. of Nebraska, 2005]. Seeding dates were targets and actual dates varied (Table 1). Seeding was done at a depth of approximately 2.5 cm and a rate of 2 kg ha–1 using a disk drill (Fabro, Swift Current, SK, Canada). Plots were 5 m long, with four rows spaced at 76 cm. The middle two rows of each plot were harvested mechanically using a self-propelled harvester (Wintersteiger, Saskatoon, SK, Canada). Weed control was done manually. Nitrogen, P, and K fertilization was done based on soil tests according to local recommendations for sorghum (Centre de Référence en Agriculture et Agroalimentaire du Québec, 2003) because no recommendations were available locally for grain amaranth. Fertilizer was broadcasted and incorporated before seeding. Plots were harvested on average 10 d after the first killing frost (Table 1). All harvested seeds were dried at 50°C until moisture content stabilized to express yields on a dry matter basis and determine grain moisture at harvest. Weather data were retrieved from a nearby station (Table 2 ).


View this table:
[in this window]
[in a new window]

  		Table 1. Description of field conditions and seeding and harvest dates of three grain amaranth field experiments conducted in Sainte-Anne-de-Bellevue, QC, Canada, in 2005 and 2006.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Average temperature and precipitation in Montreal, QC, Canada, in 2005 and 2006 and 30-yr averages.{dagger}

 
Effects of Row Spacing and Seeding Rate
This experiment was conducted at one site in 2005 and two sites in 2006 (Table 1). A randomized complete block design with a split-plot arrangement and four blocks was used. The main plot consisted of three row spacings (38, 58, and 76 cm); the subplot consisted of three seeding rates (1, 2, and 4 kg ha–1). Plots were 5 m long and the number of rows differed in the different row spacing treatments. Plots at 38, 58, and 76 cm row spacing had four, six, and four rows, respectively. Harvested rows were the middle two rows in the 76-cm plots, three rows in the 58-cm plots, and all four rows in the 38-cm plots. Seeding, harvesting, seed drying methods, and fertilization were as previously described for the seeding date and cultivar experiment. The cultivar Plainsman was used.

Nitrogen Fertilization Rate and Cultivar Experiment
This experiment was conducted at two sites in 2005 and one in 2006 (Table 1). The experiment was laid out as a randomized complete block design with a split-plot arrangement and four blocks. The main plot consisted of five N fertilization rates (0, 50, 100, 150, and 200 kg ha–1); the subplot consisted of two cultivars (D136 [Amaranthus hybrid] and Plainsman). Nitrogen fertilizer in the form of calcium ammonium nitrate (27.7% N, 4.6% Ca, and 2.4% Mg) was broadcasted and incorporated before seeding. Calcium ammonium nitrate was used due to local legislation currently limiting the availability of ammonium nitrate. Seeding, harvesting, seed drying methods, and P and K fertilization were as previously described for the seeding date and cultivar experiment.

Data Collection
The number of days to anthesis was defined as the number of days between seeding and the day when 50% of the plants were showing first signs of anthesis. The remaining data were collected shortly before harvesting. Plant height was measured as the distance between soil level and the top of the main inflorescence. Number of branches per plant was the number of side shoots having at least one visible node. Measurements were made on 10 plants per plot. Plant population was determined by doing three random counts in the middle two rows of each plot on a 36-cm length. Lodging was recorded on the day of harvest and was determined as the fraction of the total number of plants in the plot bending at an angle greater than 30°. A visual evaluation was done using a 0 to 5 scale, 0 referring to absence of lodging and 5 to all plants in the plot being lodged.

Statistical Analyses
All statistical analyses were done using SAS 9.1 (SAS Institute, 2003). Statistical models and appropriate F-tests were elaborated after McIntosh (1983). Every site–year was considered an environment and each experiment was replicated in three environments (Table 1). When the model and main effects or interactions were significant, multiple comparisons were performed with Scheffe's test. Regression analyses were performed for the N rate experiment; when a N rate x environment interaction was significant, analysis was done separately for each environment. Linear and quadratic regression coefficients were tested. When the quadratic coefficient was significant, both linear and quadratic were kept in the equation. Statistical significance level was set at 0.05 for all tests.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY AND CONCLUSIONS
 REFERENCES
 
Climate Data and General Observations
The 2 yr of experimentation differed in terms of climatic conditions (Table 2). Compared with 2006, 2005 was warmer; amounts of precipitation were generally comparable, except for May, which was dryer, and September, which was wetter, in 2005. Amounts of precipitation in both years were greater than the 30-yr average in October, the month during which harvest took place. The last spring frosts were on 13 May and 28 April in 2005 and 2006, respectively, and the first fall frosts were on 21 and 22 October in 2005 and 2006, respectively. In both years and for both experiments, plants from all cultivars under evaluation matured before the first killing frost, indicating that grain amaranth production in eastern Canada is possible.

No precise data were recorded on the diversity and incidence of insects among the different cultivars. However, aphids were present in some cultivars and tarnished plant bugs were observed in all cultivars, some appearing to be more tolerant than others. Tarnished plant bugs were more prevalent in 2005 than 2006. In the wetter environments, K593 was sometimes affected by a fungal disease hypothesized to be Pythium aphanidermatum stem canker (C. Block, personal communication, 2005); other cultivars were not affected. The disease caused dry, dark-colored cankers to develop, which spread 15 to 45 cm on the bottom of the stem. The infected plants eventually lodged and died. These symptoms are characteristic of Pythium aphanidermatum stem canker (Block et al., 2002)

Seeding Date and Cultivar Experiment
Environment
Environment main effects were observed for grain yield, grain moisture, branch number, and days to anthesis; however, environment affected variables mainly through interactions with seeding date and genotype (Table 3 ). Grain yield was highest in 2006 at Site A (1113 kg ha–1), intermediate at Site B in 2005 (876 kg ha –1), and lowest at Site A in 2005 (647 kg ha–1). Grain moisture at harvest also differed significantly across environments. The difference between Site A and Site B in 2005 is due to the later harvest date of Site B, which allowed for four extra days of drying (Table 1). Fitterer et al. (1996) reported decreasing seed moisture with later harvesting dates, the difference being the greatest after a killing frost. They also reported yield losses with later seeding dates, which they attributed to seed shattering. Yield losses with later harvesting, however, have to be weighed against reduced drying costs.


View this table:
[in this window]
[in a new window]

  		Table 3. Effects of seeding date and cultivar on grain yield and several agronomic traits of grain amaranth grown in Sainte-Anne-de-Bellevue, QC, Canada, in 2005 (two sites) and 2006 (one site). Results illustrate main effects, with means averaged over other factors.{dagger}

 
Seeding Date
The effect of seeding date varied depending on the environment for all variables except plant height, which was not affected by seeding date (Table 3). Grain yield was only affected by seeding date at Site A in 2006 (Table 4 ), where the highest yield occurred with the earliest seeding date (mid-May, 1398 kg ha–1). Henderson et al. (1998) in North Dakota and Aufhammer et al. (1995) in southern Germany observed an effect of seeding date on grain yield in all of the five environments they evaluated; the seeding date resulting in greatest yield was not always the same in different environments.


View this table:
[in this window]
[in a new window]

  		Table 4. Effects of seeding date on grain yield and agronomic traits of grain amaranth grown in three different environments in Sainte-Anne-de-Bellevue, QC, Canada, in 2005 and 2006.

 
In all environments, later seeding dates resulted in higher grain moisture at harvest (Table 4). Wetter grain at harvest can decrease grain quality by favoring microbial growth. Gimplinger et al. (2007) reported a positive correlation between grain amaranth moisture at harvest and number of colony forming units of aerobic mesophilic bacteria. They also observed differences between cultivars, which they attributed to differing environmental conditions during maturation and differences in inflorescence architecture. Since the climate in eastern Canada is particularly humid during the fall, it would be important to conduct studies on the microbial status of the grain produced. Mycotoxin contamination should also be examined because toxinogenic species of Fusarium have been isolated from grain amaranth (Bresler et al., 1995). Fusarium is known to have a greater incidence in more humid environments (Soriano and Dragacci, 2004).

In all three environments, we noted that unlike plants seeded in mid-May, plants seeded in mid-June had not started to dry-down by the date of first frost (Table 4). The first and second seeding dates at Site B in 2005 are the only ones that resulted in grain moisture lower than 20%. Interestingly, a significant negative correlation was observed between grain yield and grain moisture at harvest (r = –0.39, P < 0.0001), indicating that cultivars and seeding dates that favor high yields might also result in lower drying costs.

Plant height was not affected by seeding date in any environment. This suggests that plants seeded in mid-June grew at a much faster rate than plants sown in mid-May. It appeared that this faster growth rate of later seeded plants resulted in weaker stems, as later seeding dates resulted in increased lodging in all environments (Table 4). Another indicator of this was the significant negative correlation found between days to anthesis and lodging (r = –0.50, P < 0.0001). Aufhammer et al. (1995) reported a higher N concentration in vegetative parts with later seeding. Grain N concentration was also lower with later seeding dates, thus reducing seed quality. They attributed this effect to a lack of N translocation with later seeding dates. The increased lodging with later seeding dates we observed was probably due to the fact that the stems were less mature and lignified, and thus were weaker than with earlier seeding dates. Henderson et al. (1998) also reported increased lodging with later seeding date in one out of two environments in North Dakota. Later seeding dates also markedly reduced number of days to anthesis in all environments (Table 4). Similar results were observed in North Dakota by Henderson et al. (1998).

There was a strong seeding date x environment crossover interaction for plant population at harvest (Table 4). At both sites in 2005, the early-June seeding date resulted in much better stand establishment and hence higher plant population. However, in 2006, the reverse was observed; late-June and mid-May seedings resulted in the best stand establishment (Table 4). Branch number was also affected by a seeding date x environment crossover interaction. This may indirectly reflect a strong negative correlation between branch number and plant population (r = –0.59, P < 0.0001), which suggests that the greater interplant space at lower populations promoted branching. Henderson et al. (1998) also observed a seeding date x environment interaction for plant population, with later dates consistently resulting in higher plant population at harvest.

Soil temperature for optimal grain amaranth germination has been estimated to range between 18.5 and 24°C, while seeding depth should in general not exceed 2.5 cm (Webb et al., 1987). Ideal germinating conditions are therefore a warm and moist soil at the surface. In spring in eastern Canada, such conditions typically happen on a few occasions, lasting only a few days each time. This suggests that it is preferable to wait for the proper field conditions, rather than trying to seed on a particular recommended date. Nevertheless, when possible, earlier seeding is preferable in that it allows for better seed drying, minimizes lodging, and may result in higher grain yields.

Cultivar
There was a significant difference in yield among cultivars in only one environment, with Plainsman yielding more than either K432 or K593 at Site A in 2005 (Table 5 ). The absence of difference in grain yield between cultivars has been reported in other regions (Pospisil et al., 2006; Gimplinger et al., 2007), however, several studies reported yield differences among cultivars or environment x cultivar interactions (Elbehri et al., 1993; Henderson et al., 1998; Henderson et al., 2000). Cultivars differed in grain moisture at harvest, with Plainsman producing the seeds with highest moisture, followed by K432 and K593. Interestingly, K593 had the lowest grain moisture in all three environments and also had the highest number of days to anthesis (Table 5). Cultivar K432 was shorter and hence easier to harvest mechanically than Plainsman and K593 at both sites in 2005; differences in height in 2006 were not significant. There were, however, no differences in lodging between the cultivars. Cultivars K432 and K593 produced significantly more branches than Plainsman and this difference was more pronounced at Site A in 2006.


View this table:
[in this window]
[in a new window]

  		Table 5. Effects of cultivar on grain yield and on agronomic traits of grain amaranth grown in three different environments in Sainte-Anne-de-Bellevue, QC, Canada, in 2005 and 2006.{dagger}

 
Row Spacing and Seeding Rate Experiment
Environment
Environment was the factor affecting the most variables in this experiment (Table 6 ). Lower grain yields at Site A in 2005 and at Site C in 2006 were probably due to compaction and poor drainage in these two environments. Poor drainage at Site C in 2006, combined with heavy rainfall in October, also delayed harvesting (Table 2). The delayed harvest resulted in significantly lower grain moisture at Site C in 2006 compared to the other two environments (Table 6). Lower yielding environments also produced the shortest plants. The higher yielding environment, Site A in 2006, also had the highest incidence of lodging, but lodging was low overall. Branch number was affected by a significant environment x row spacing interaction. This interaction reflected that branches were observed in only one environment, Site A in 2006, where row spacing was positively related to branch number (Table 6). The number of days to anthesis was also considerably affected by environment. Experiments seeded later had fewer number of days to anthesis (Tables 2 and 6). This is in agreement with results from our seeding date experiment. Finally, Site A in 2006 had a plant population approximately half of that observed in the other two environments.


View this table:
[in this window]
[in a new window]

  		Table 6. Effects of row spacing and seeding rate on grain yield and several agronomic traits of grain amaranth grown in Sainte-Anne-de-Bellevue, QC, Canada, in 2005 (one site) and 2006 (two sites). Results illustrate main effects, with means averaged over other factors.{dagger}

 
Row Spacing and Seeding Rate
Row spacing and seeding rate did not affect grain yield (Table 6). Such results are in accordance with Aufhammer et al. (1995), Myers (1996), Henderson et al. (2000), and Gimplinger et al. (2007), who also all failed to observe a yield response to row spacing or seeding rate. Myers (1996) evaluated seeding rates ranging from 0.28 to 4.4 kg ha–1 at a row spacing of 76 cm and attributed the lack of grain yield response to the fact that plants responded to high densities by self-thinning and lowering seed production per plant. On the other hand, Bhaskar et al. (1996) and Malligawad and Patil (2001) in India reported grain yield increases with increasing established plant population. Malligawad and Patil (2001) reported yield increased from 1858 to 3242 kg ha–1 with an increase in plant population from 55,000 to 222,222 plant ha–1. Such densities are extremely low and would be difficult to obtain using mechanical seeding under our conditions. For example, the lowest plant population observed in our experiments was 778,000 plants ha–1, obtained with a seeding rate of 1 kg ha–1.

Doubling seeding rate also nearly doubled plant population (Table 6). At a seeding rate of 4 kg ha–1, plant population was slightly above 2 million plants ha–1, which probably promoted interplant competition and thus might explain the resulting reduced plant height (Table 6). Henderson et al. (2000) observed a similar height reduction when manipulating plant population up to 272,000 plants ha–1. Seeding rate also affected the number of days to anthesis (Table 6), although differences were biologically insignificant.

Row spacing affected grain moisture at harvest, with narrower rows resulting in drier grains (Table 6). This might be due to a more even plant distribution at narrower row spacings, allowing for a better airflow. The row spacing x seeding rate interaction observed for plant population reflects that with a 38-cm row spacing and a seeding rate of 4 kg ha–1; plant population was higher than at the same seeding rate for the other two row spacings. Interestingly, plant population was significantly higher with narrower rows, suggesting a weaker self-thinning effect at a row spacing of 38 cm than with spacings of 58 or 76 cm (Table 6). Self-thinning at narrow row spacings has been reported by Myers (1996).

Nitrogen Rate and Cultivar Experiment
Environment
As in the seeding rate and row spacing experiment, environment affected all variables studied, through environment main effects and N rate x environment interactions, but mostly through cultivar x environment interactions (Table 7 ). Site A yielded more than s Site B, which had a more fertile but coarsely textured soil (Table 1). The difference in grain moisture between the two sites in 2005 is probably due to local topography; Site A was more severely subjected to frost, which probably allowed for faster dry-down. Plant height differed among environments, with plants at Site A in 2006 being significantly taller. Lodging was low overall and was completely absent at Site B in 2005. There was a slight, biologically insignificant difference in number of days to anthesis between environments (Table 7).


View this table:
[in this window]
[in a new window]

  		Table 7. Effects of N fertilizer rate and cultivar on grain yield and agronomic traits of grain amaranth grown in Sainte-Anne-de-Bellevue, QC, Canada, in 2005 (two sites) and 2006 (one site). Results illustrate main effects, with means averaged over other factors.{dagger}

 
Nitrogen Rate
Regression analyses were performed with all environments pooled only for those variables not affected by a significant interaction involving environment (i.e., grain moisture at harvest, plant height, and number of days to anthesis). For grain yield, lodging, and branch number, which were affected by a N rate x environment interaction, regression analysis was run for each environment separately. The regression coefficient for all environments pooled was only significant for plant height (Fig. 1 ), with increasing N fertilization rate resulting in increased plant height in a linear fashion. A similar relation between N fertilization and plant height was also reported by Myers (1998) and Elbehri et al. (1993).


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 1. Effects of N fertilization on grain yield, lodging score and plant height of grain amaranth grown in Sainte-Anne-de-Bellevue, QC, Canada. Grain yield and lodging are for one environment in 2006, whereas plant height represents plant response averaged across three environments in 2005 and 2006. Lodging score, 0 refers to absence of lodging and 5 to all plants in the plot being lodged. Vertical bar represent one standard error.

 
For the variables affected by a N rate x environment interaction, significant regression coefficients were found for grain yield (linear) and lodging (quadratic) at Site A in 2006 only (Fig. 1). Elbehri et al. (1993) also reported a linear effect of N fertilization on grain yield and a quadratic effect on lodging, while Myers (1998) reported a linear effect of N fertilization on grain yield and lodging. The increased lodging caused by N fertilization makes mechanical harvesting less efficient and could decrease seed quality when inflorescences are in direct contact with the soil. Our results are in contradiction with Erley et al. (2005) who observed no effects of N fertilization on lodging; they, however, only evaluated N fertilization rate up to 120 kg N ha–1. The N rate x environment interaction for branch number illustrates that branch number was not affected by N fertilization at either site in 2005, but was positively affected by N at Site A in 2006 (data not shown). Finally, increasing N fertilization resulted in slightly greater grain moisture (Table 7). This has economic importance because the extra costs of drying and fertilizer have to be weighed against yield gain.

Cultivar
Cultivar main effects and/or cultivar x environment interactions affected several variables (Table 7). Cultivar D136 had lower yields than Plainsman at Site A in 2006. The reverse was observed in 2005 (Table 8 ). Cultivar D136 is later maturing than Plainsman, has superior grain moisture at harvest, greater number of days to anthesis, and exhibited less dry-down after frost. Its lower yield at Site A in 2006 could therefore be explained by the cooler temperature prevailing in most months of that year (Table 2). D136 also has more branches and leaves, which can interfere with mechanical harvesting when plants are not properly dried down. Therefore, this cultivar should perhaps not be recommended for use in eastern Canada due to the short growing season and prevailing humid conditions.


View this table:
[in this window]
[in a new window]

  		Table 8. Effects of cultivar on grain yield and agronomic traits of grain amaranth grown in Sainte-Anne-de-Bellevue, QC, Canada, in 2005 and 2006.{dagger}

 
The cultivar x N rate x environment interaction for grain yield illustrates the fact that at Site A in 2005 and Site B in 2005 there were no effects of N fertilization on yield and that at Site A in 2006 there was a positive effect of N on yield (Fig. 1). Also at Site A in 2006, D136 responded to N only up to 150 kg N ha–1, whereas Plainsman kept responding up to 200 kg N ha–1.


   SUMMARY AND CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
REFERENCES
 
According to our results, earlier seeding dates seem preferable, as they resulted in higher grain yield in one environment, and lower seed moisture at harvest and minimized lodging in all three environments. Seeding rate and row spacing did not affect grain yield. However, row spacing affected grain moisture at harvest, with narrower rows resulting in drier grains. Seeding in narrow rows might thus be preferable when better weed management strategies become available; however, as no herbicide is currently available for grain amaranth, the use of wider rows combined with mechanical weeding remains the only practical choice. Nitrogen fertilization increased grain yield in only one environment; however, it also increased lodging. Seed moisture and plant height were also increased by N fertilization in all environments.

Optimal management practices for grain amaranth production in eastern Canada are mainly dictated by the relatively short growing season and prevailing humid conditions. Grain yield averaged 923 kg ha–1 across experiments and environments, which is comparable to yields obtained in North Dakota (Henderson et al., 1998). Eastern Canada could, therefore, be considered a potential area for grain amaranth production. Given that we have seen little response to N fertilization and that no herbicides are currently registered for use in grain amaranth, organic production should be considered. Indeed producers usually receive a premium for organically produced grain amaranth that may be substantial. It is, however, difficult at the moment to reach conclusions regarding the profitability of grain amaranth production. Markets, and hence demand, for grain amaranth are currently limited and grain price is subject to large variations depending on production volume. Unless markets can be developed, with major food processors integrating grain amaranth in widely distributed products, the adoption of grain amaranth will remain problematic.


   ACKNOWLEDGMENTS
 
The authors would like to thank D. Baltensperger (Univ. of Nebraska) for providing seeds of the cultivars used in this project. This work was financially supported by the Fédération des Producteurs de Cultures Commerciales du Québec (FPCCQ) and the Fonds Québécois de la Recherche sur la Nature et les Technologies (FQRNT).

All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.


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





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 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 Google Scholar
Google Scholar
Right arrow Articles by Gélinas, B.
Right arrow Articles by Seguin, P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Gélinas, B.
Right arrow Articles by Seguin, P.
Agricola
Right arrow Articles by Gélinas, B.
Right arrow Articles by Seguin, P.
Related Collections
Right arrow Best Management Practices
Right arrow Grazing Management
Right arrow Turfgrass

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