|
|
||||||||
a Plant Science Dep., Rutgers Univ. Philip Marucci Center, 125 Lake Oswego Rd., Chatsworth, NJ 08055
b Office of Research, Univ. of Guelph/Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA), 1 Stone Road West, Guelph, ON N1G 4Y2 Canada
c Dep. of Plant Agriculture, Univ. of Guelph, Guelph, ON, N1G 2W1 Canada
* Corresponding author (kumudini{at}aesop.rutgers.edu)
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: N, nitrogen NHI, Nitrogen harvest index LAI, leaf area index LAD, leaf area duration SFP, seed-filling period
| INTRODUCTION |
|---|
|
|
|---|
Nitrogen is considered an important yield-determining factor in soybean production (Sinclair and deWit, 1976; Frederick and Hesketh, 1994; Sinclair 1998). Sinclair and de Wit (1976) postulated that the high demand for N by the growing seed of high protein grain crops, such as soybean, cannot be satisfied by daily N accumulation rates and therefore N must be remobilized from vegetative tissue. The depletion of N from vegetative tissue was expected to accelerate leaf senescence and thus lower the photosynthetic capacity of the canopy which would ultimately lead to yield reduction by a shortening of the seed filling period. The proposed role of N in soybean yield limitation was based on a number of underlying assumptions, popularly referred to as the self-destruct hypothesis. Inherent to the self-destruct hypothesis is a close association between leaf N status, leaf senescence, potentially remobilizable N, and yield. If this model accurately describes the role of N in soybean yield limitation, yield improvement must have occurred by overcoming the proposed yield limitation. Consistent with this hypothesis, it was recently reported that genetic improvement in short-season soybean cultivars was associated with longer leaf area duration (Kumudini et al., 2001).
Most grain crops remobilize some vegetative N to the seed. In soybean, remobilized N has been estimated to contribute 20 to 60% of the N to the seed (Hanway and Weber, 1971; Egli et al., 1978; Zeiher et al., 1982). However, N in the seed can come from either (i) N accumulated prior to the SFP and remobilized to the seed or (ii) N that is accumulated during the SFP. Pazdernick et al. (1997) illustrated that genetic variability in N accumulation strategies and seed yield exist in the germplasm. Considering the importance of N to soybean yield, genetic improvement in yield should show evidence of overcoming N limitation. The objective of the experiment was to determine the relationship between N accumulation, remobilization, or partitioning and genetic yield improvement of short-season soybean cultivars.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Four genotypes were selected to represent two maturity groups of old and new cultivars (Table 1). Each old cultivar (prior to 1950) was paired with a newer (post-1980) counterpart with similar lodging score, seed quality and maturity, to reduce the confounding impact of these variables on yield gain. The seeds of the two historical genotypes (Pagoda and Mandarin Ottawa) were obtained from Agriculture and AgriFood Canada (Ottawa, Canada). The two newer cultivars (Maple Glen and OAC Bayfield) were certified seeds from a local supplier (First Line Seeds, Guelph, ON, Canada).
|
Data Analysis
The experiment was designed as a split-plot arrangement of treatments in a randomized complete block design with four replications (blocks). The main plots were the four genotypes, with seven sampling dates as the subplots. Sampling date plots were allocated to the genotype main plots at random. The data were analyzed by Proc Mixed (Tables 2 and 3) or Proc Corr (Table 4) (Windows SAS v. 6.12, SAS Institute Inc., Cary, NC). Analyses were conducted on individual year data and then on the two years combined. In the combined year analysis, year was treated as the main unit with replication nested within year, the cultivars were the subunits, and the sampling dates were the sub-subunits. Year and replication were considered random effects and all other effects were considered fixed effects. The sample dates were combined across year based on phenological staging at sample time. Combined year data were presented unless significant year x treatment interactions occurred, in which case the data presented were for each year. When significant treatment effects (P < 0.1) were found, contrast statements (95% confidence limits) were constructed to compare old versus new cultivars utilizing the correct error terms for the specific split-plot comparison (Steel and Torrie, 1980).
|
|
|
Within the 0.5-m2 area, all plants were dug with a spade to a 15-cm depth and a subset of 10 plants were separated into leaves, stems (+petioles), roots, pod walls, and seeds. Roots of the 10 subsample plants were placed in screen boxes, washed, and nodules removed before drying. The green leaves of the 10 plants were used for measurements of leaf area (model LI-3000, LI-COR, Inc., Lincoln NE). Separated plant materials were dried for at least 48 h in a forced-air oven (80C). The dry weights of the green leaves were used to calculate specific leaf weight. Senesced leaves were not collected. Dry weights of the plants parts as well as the total dry weight of the 0.5-m2 harvested area was measured. The dry weight of plant tissue was calculated on the basis of the dry weight of the total harvested area and the percentage of dry matter in the plant part (dry weight of plant part divided by total weight of the 10 subsample plants). Leaf area index (LAI) was calculated on the basis of specific leaf weight and total dry weight of leaves. The separated and dried tissue samples were ground with a Wiley Mill (1-mm mesh screen). A 1-g sample was removed and N concentration was determined with a Leco model FP-248 N analyzer (Leco, St. Joseph, MI). The N concentrations of the individual tissue was multiplied by the calculated weight of the tissue to determine tissue N content. Total N accumulated was the sum of the N content of all plant parts including the root tissue. Remobilized N was calculated as the difference in N content of vegetative tissue between R8 and R4 (vegetative tissue N at R4-vegetative tissue N at R8). A 100-g sample of whole seeds collected at maturity in 1996 was passed through a near infrared grain analyzer (Foss Electric, Wheldrake, York, UK) to measure seed protein and oil concentration.
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
Voldeng et al. (1997), in their study of 41 short-season soybean cultivars, reported that the reduction in protein concentration of 4 g kg-1 yr-1 was concomitant with an increase in yield of 11 kg ha-1 yr-1. Consistent with Voldeng et al. (1997), the reduction of protein concentration in the modern cultivars was accompanied by higher seed yields (Table 2), resulting in greater seed N content (Table 3).
N Partitioning
Greater seed N content may be due to either increased total N accumulation or a greater partitioning of the accumulated N to the seeds, or both. Nitrogen harvest index (NHI), defined as the amount of N partitioned to the seed relative to total above ground N at maturity, of the older cultivars was either the same or lower than their newer counterparts (Table 3). An equation (Eq. [1]), modified from Tollenaar et al. (1994), was used to separate the contribution of N accumulation and partitioning to seed N content.
![]() | [1] |
Using Eq. [1] and the treatment means, we calculated N partitioning (NHI) to account for 0 to 16% of the difference in seed N content between high and low yielding cultivars. This suggests that NHI is not an important contributor to the greater seed N content characteristic of genetically improved cultivars.
Other researchers have also questioned the role of NHI in genetic yield improvement (Jeppson et al., 1978; Cregan and Yaklich, 1986; and Shiraiwa and Hashikawa, 1995). Cregan and Yaklich (1986) were unable to detect a significant correlation between NHI and seed yield. Jeppson et al. (1978) reported a positive correlation between NHI and seed yield; however, the genotypes with the higher yield did not necessarily have the highest NHI. Thus they concluded that genetic yield improvement was not associated with improved NHI. Shiraiwa and Hashikawa (1995) also reported that although the modern cultivars exhibited a significantly higher NHI, all values were close (from 0.910.95), suggesting the contribution of NHI to yield gain was limited.
N Accumulation and Remobilization
Relative to N partitioning, N accumulation was calculated (Eq. [1]) to account for 84 to 100% of the gain in seed N content between low and high yielding cultivars. Both old and new cultivars accumulated about the same amount of total N content by the R4 stage of development. The difference in total N accumulated between old and new cultivars was not apparent until after the R4 growth stage (Table 3). After the R4 growth stage, the newer cultivars accumulated significantly more N than the old cultivars.
Seed N content is the sum of the N accumulated during the SFP and the N remobilized to the seed during seed development. Calculated values for remobilized N (vegetative N at R4 - vegetative N at R8) together with values of N accumulated during the SFP agreed fairly well with seed N content (Table 3) suggesting that the method of estimating remobilized N was realistic. There were no consistent differences between old and new cultivars for the amount of N remobilized from vegetative tissue to seed. However, there were significant differences between old and newer cultivars in their ability to accumulate N during the SFP (Table 3). A comparison of the Pearson correlation coefficients between the amount of N remobilized from the vegetative tissue, or N accumulated during the SFP, and seed yield suggests that the latter is the greater contributor to yield improvement (Table 4). Notwithstanding the fact that remobilized N was the largest contributor of N to the seed, the amount of N remobilized was similar for both the old and new cultivars (Table 3). Although it was found that the N accumulated during the SFP is only a small fraction of N present in the seed; that fraction was found to be significantly greater for the new cultivars (2325%) than for the old cultivars (56%). Therefore, the results of the current experiment suggest that an increase in the capacity to accumulate N during the SFP rather than differences in remobilization of N from vegetative tissue is associated with genetic improvement in yield. A study of genetic improvement of Japanese cultivars grown at low plant density also reported that the newer cultivars accumulated more N during the SFP (Shiraiwa et al., 1994). These researchers reported that daily N2 fixation generally decreased during the SFP. However, the older Japanese soybean cultivars showed a more rapid decline in daily N2 fixation compared with newer cultivars.
N Limitation and Soybean Yield
The results of the current study suggest some inconsistencies with previously defined models of the role of N in soybean yield limitation. Researchers have argued for a close association between leaf N status, LAD, and yield (Sinclair and deWit, 1976; Shibles and Sundberg, 1998). The closest correlations between leaf characteristics and seed yield were predominantly associated with the R6 growth stage (data not shown); therefore, the correlation coefficients at R6 were reported (Table 4). Consistent with the self-destruct model, longer leaf area duration (as measured by LAI at R6) was correlated with higher yields. Final seed yield was correlated significantly with leaf area index at R6 in both 1996 and 1997. However, the leaf N parameters; leaf N content per unit leaf area (g N m-2), and leaf N concentration (g N g-1 leaf dry weight) were only correlated to yield in one of the two years (Table 4). Sinclair and Horie (1989) modeled a curvilinear relationship between N content per unit leaf area and the production of plant biomass. Considering that greater biomass production was reported to be the main contributor to the higher yields of newer short-season soybeans cultivars (Kumudini et al., 2001), it may be speculated that leaf N content per unit area would be an important contributor to genetic yield improvement. However, in the current study, leaf N content per unit area of (non-senescent) leaves was not correlated consistently with yield.
The self-destruct model describes leaf N status as a driving force in determining LAD and seed yield. However, the correlation coefficients for the relationships between leaf N values and both seed yield and LAD (LAI at R6) were significant only for leaf N content (Table 4). A caveat in interpreting the correlation between leaf N content and LAI, is a possibility of auto-correlation. Leaf N content is the product of leaf N concentration and leaf weight, and leaf weight is also a function of LAI. Shibles and Sundberg (1998), consistent with the current study, reported low correlation coefficients between leaf N concentration (at R5) and seed yield and a positive correlation between leaf N content during the SFP (R5) and yield. These authors proposed that leaf N content at R5 represented the pool of potentially remobilizable N. They maintained that their results were consistent with the self-destruct model because the higher leaf N content represents a greater pool of remobilizable N available to meet seed N demand. Leaf N content was found to be correlated with N remobilized from vegetative tissue (data not shown). However, in the current experiment, we found no differences in the ability of the new and old cultivars to accumulate either more N prior to the SFP or remobilize the N accumulated prior to the SFP (Table 3). The difference between old and new cultivars arose from differences in their ability to accumulate N during the SFP. Therefore, the correlation between LAD and genetic yield improvement may in fact be distinct from the role of leaf N content as a source of potentially remobilizable N.
Alternatively, assimilate supply rather than nitrogen supply may be the driving force for LAD. Research on maize (Zea mays L.) has suggested that the ratio of source (i.e., supply of carbon) and sink (i.e., demand of carbon) during the grain filling period was important in maintaining leaf area (Tollenaar and Daynard, 1982; Rajcan and Tollenaar, 1999). Tollenaar and Daynard (1982) reported that leaf longevity of maize hybrids was greatest when source-to-sink ratios, during the grain-filling period, were balanced. Also, Rajcan and Tollenaar (1999) showed that the increased leaf longevity of modern maize hybrids was associated with an increase in the source-sink ratio. Soybean research has also indicated an association between source supply and leaf area: Jones et al. (1984) reported that the leaf area index of Bragg was on average 30% greater in canopies grown under high CO2 levels. Hayati et al. (1995) reported that increasing photosynthesis (by shade removal) delayed the rate of leaf senescence (McCall) under available N conditions. Abdin et al. (1998), showed that increasing the source supply by sucrose stem infusion doubled leaf area per plant (Maple Glen).
Contrary to previous models of the role of N in soybean yield limitation, mechanisms other than leaf N status and the remobilization of N from vegetative tissue are involved in the genetic improvement of the short-season soybean cultivars tested. The newer genotypes accumulated more N in the seed than their older counterparts. Greater total N accumulation contributed more than greater N partitioning (NHI) to the ability of the modern cultivars to accumulate more N in the seed. Most of the N found in the seed was accumulated prior to the SFP and remobilized into the seed after the beginning of the SFP. Although, N accumulated during the SFP was a smaller contributor to the overall N content of the seed, genetic improvement of short-season soybean genotypes tested was associated with the greater ability of the newer genotypes to accumulate N during the SFP. The accumulated evidence suggests that yield improvement of the short-season soybeans tested in the current study is neither related to higher N status in the leaves nor remobilization of N from vegetative tissue but rather with greater N accumulation during the SFP.
| NOTES |
|---|
|
|
|---|
Received for publication December 8, 2000.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. L. Naeve, T. A. O'Neill, and J. E. Miller-Garvin Canopy Nitrogen Reserves: Impact on Soybean Yield and Seed Quality Traits in Northern Latitudes Agron. J., May 7, 2008; 100(3): 681 - 689. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Kumudini, J. Omielan, and D. J. Hume Soybean Genetic Improvement in Yield and the Effect of Late-Season Shading and Nitrogen Source and Supply Agron. J., February 29, 2008; 100(2): 400 - 405. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
||||