Elsevier

Field Crops Research

Volume 248, 1 March 2020, 107717
Field Crops Research

Allometric analysis reveals enhanced reproductive allocation in historical set of soybean varieties

https://doi.org/10.1016/j.fcr.2020.107717Get rights and content

Highlights

  • Harvest index is inconsistent across literature to describe genetic gain in reproductive allocation at maturity.

  • Allometric analysis was used to characterize genetic gain in reproductive allocation in R5 in seven soybean varieties.

  • Allometric analysis at the end of the seed set revealed an increase in relative allocation into pods.

  • Allocation to seed must have increased, but HI did not capture this necessary change in allocation.

Abstract

Seed yield is commonly expressed as the product of shoot biomass and harvest index (HI) at physiological maturity. However, HI is a size-dependent ratio unsuitable to describe shifts in reproductive partitioning in historical soybean [Glycine max (L.) Merr.] studies where selection has enhanced shoot biomass. The aim of this work was to analyze allocation of biomass to reproductive organs in the onset of seed filling (R5) using allometric analysis in a set of historical soybean varieties. Seven varieties released between 1980 and 2013 were evaluated in field trials in Rossville (Kansas, United States) under two N fertilizer rates (0 and 670 kg N ha−1) in 2016 and 2017. Seed yield increased at a rate of 0.74 % yr−1 (p < 0.001); while the shoot biomass increased at a rate of 0.41 % yr−1 (p < 0.001), suggesting an increase in allocation to reproduction. However, the rate of increase in HI at maturity was not different from zero highlighting the inadequacy of this trait. Allometric exponents (slopes of the log-log relationships) relating pod and shoot biomass, pod and leaf, and pod and stem plant fractions increased linearly with the year of release (p < 0.05). Allometric analyses thus revealed genetic gains in reproductive allocation not detected by HI at maturity. The latter outcome highlights the contribution of improved reproductive partitioning to soybean yield gains, and the need for allometric analyses to account for size-dependence in allocation of shoot biomass.

Introduction

Seed yield in crops is commonly expressed as the product between shoot biomass and the reproductive allocation through the harvest index (HI; i.e., the ratio of seed to shoot (aboveground) biomass; Donald and Hamblin, 1976) both determined at physiological maturity. These two traits have been commonly used to dissect management and genotypic effects on yield improvement.

Reproductive allocation differs among cultivated species defining physiological strategies between shoot biomass accumulation and its allocation to reproductive organs partially shaped by natural and agronomic selection and plant morphology (Vega et al., 2000, 2001a; Weiner et al., 2009). For example, the relationship between yield and crop or plant growth rate is linear in indeterminate crops such as soybean [Glycine max (L.) Merr.] and chickpea (Cicer arietinum L.) (Lake and Sadras, 2016; Vega et al., 2001a) and non-linear in maize (Zea mays L.) and sunflower (Helianthus annus L.) where morphological constraints impose a ceiling in grain set (Vega et al., 2001a).

In soybean, the contribution of shoot biomass production to seed yield improvement has been large and consistent relative to the influence of HI in retrospective studies comparing historic sets of genotypes (Balboa et al., 2018; Rowntree et al., 2014; Specht et al., 1999). In contrast, the contribution of HI to genetic yield gain has been large and consistent in cereals (Hay, 1995; Slafer, 1994), with the exception of maize in North America, with HI presenting minimal changes over time, one percentage unit per decade (Duvick et al., 2004).

There are two main reasons for inconsistencies in shifts of soybean HI in comparisons of historic varieties. The first one is related to errors and different approaches in the estimation of HI, including intensity and timing of plant sampling (e.g., phenological stages), treatments applied, environmental conditions and genotypic background (Kumudini, 2002; Rowntree et al., 2014). In particular, the loss of leaves in mature soybean is a major source of variation in HI (Hay, 1995; Schapaugh and Wilcox, 1980; Unkovich et al., 2010). The second is that the HI of soybeans is a size-dependent plant trait (Vega et al., 2000), and ratios are a biased measure of reproductive allocation when plant size varies. Thus, allometric analysis is needed to account for size-dependent variation in shoot biomass allocation (Jasienski and Bazzaz, 1999; Niklas, 1994; Pearsall, 1927; Poorter and Sack, 2012). Of special relevance for this paper, Qin et al. (2013) highlighted the bias and instability of HI in comparison to allometric exponents relating grain and vegetative biomass on log-log scales.

Soybean yield improvements are usually associated with seed number rather than seed weight (de Felipe et al., 2016). Seed set occurs between the beginning of flowering (R1) and the onset of seed filling (R5) where the number of reproductive sinks reaches a maximum (Egli and Zhen-wen, 1991). During this period, the rate of reproductive partitioning peaks and the relationship between crop growth rate and seed number is linear (Vega et al., 2001b) but with significant variation among environments and genotypes (Masino et al., 2018; Rotundo et al., 2012). Furthermore, other physiological frameworks include partitioning to reproductive structures and the seed set efficiency at R5 to explain seed number determination (Charles-Edwards, 1984). However, comparison of older and modern soybean varieties have shown similar dynamics in shoot biomass accumulation until R5 where newer cultivars portray greater rates of shoot biomass accumulation during the seed filling period (Shiraiwa and Hashikawa, 1995; Specht et al., 1999).

Historical shifts in biomass allocation patterns in soybean would benefit from an allometric perspective to account for size-dependent allocation. We propose to analyze allocation of biomass to reproductive organs in R5 using allometric analyses in a set of historical soybean varieties. This evaluation will help to better understand the effects of genetic gain in reproductive partitioning as a more accurate alternative to HI at physiological maturity.

Section snippets

Crop husbandry

Field experiments were conducted in Kansas River Valley research station in Rossville, Kansas, United States (39°07′N; 95°55′W). Table 1 shows weather and soil details. Crops were sown after maize on May 12, 2016 and May 18, 2017 with seed inoculated using a commercial strain of Bradyrhizobium japonicum and commercially treated with insecticide and fungicide. Target plant density was 25.5 plants m−2. Crops were rainfed and supplemented with irrigation when needed to avoid water stress. Pest and

Seed yield genetic gain

Experimental seasons showed similar growing conditions for soil characteristics, temperature, water supply, and accumulated evapotranspiration (Table 1). Growing conditions were reflected in similar average seed yield of 2988 ± 100 kg ha−1 in 2016 and 2969 ± 89 kg ha−1 in 2017.

Genetic gain for seed yield was 33.1 kg ha−1 yr−1 (Fig. 1) or 0.74 % yr−1 (Fig. 3) irrespective of N rate. The seed yield vs. year of release regressions were parallel, with an offset of 461 kg ha−1 associated with

Discussion

Seed yield increased linearly for a set of seven commercial varieties released between 1980 and 2013. The overall increase of 33.1 kg ha−1 year−1 in both N conditions of this study is in agreement with those present in the literature (de Felipe et al., 2016; Rowntree et al., 2013; Rincker et al., 2014; Specht et al., 1999). In common with these studies, genetic gain in seed number was the driver of genetic crop yield improvement.

Wilson et al. (2014) found the rate of genetic gain in 57

Conclusion

For our set of varieties and environments, seed yield in soybean increased at a rate of 0.74 % yr−1 between 1980 and 2013 whereas the rate of increase in shoot biomass was only 0.41 % yr−1. New analysis of independent studies reinforced the conclusion that improvement in biomass explains only part of the genetic gain in yield. This means allocation to seed must have increased, but HI did not capture this necessary change in allocation. Allometric analysis at the end of the seed set revealed

Declaration of Competing Interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Santiago Tamagno: Writing - original draft, Conceptualization, Visualization, Formal analysis, Investigation. Victor O. Sadras: Writing - review & editing, Conceptualization. Osler A. Ortez: Writing - review & editing, Investigation. Ignacio A. Ciampitti: Writing - review & editing, Supervision, Conceptualization.

Acknowledgments

Authors would like to acknowledge the support of the Fluid Fertilizer Foundation, and the International Plant Nutritional Institute–Global Soybean Project (IPNI GBL-62). This is contribution no. 20-090-J from Kansas Agricultural Extension Station.

References (56)

  • S. Tamagno et al.

    Nutrient partitioning and stoichiometry in soybean: a synthesis-analysis

    Field Crops Res.

    (2017)
  • S. Tamagno et al.

    Physiological differences in yield related traits between flint and dent Argentinean commercial maize genotypes

    Eur. J. Agron.

    (2015)
  • M. Unkovich et al.

    Variability in harvest index of grain crops and potential significance for carbon accounting: examples from Australian agriculture

    Advances in Agronomy

    (2010)
  • C.R.C. Vega et al.

    Reproductive partitioning and seed set efficiency in soybean, sunflower and maize

    Field Crops Res.

    (2001)
  • C.R.C. Vega et al.

    Reproductive allometry in soybean, maize and sunflower

    Ann. Bot.

    (2000)
  • J. Weiner

    Allocation, plasticity and allometry in plants

    Perspect. Plant Ecol. Evol. Syst.

    (2004)
  • G.R. Balboa et al.

    Shifts in soybean yield, nutrient uptake, and nutrient stoichiometry: a historical synthesis-analysis

    Crop Sci.

    (2018)
  • D. Bates et al.

    Fitting linear mixed-effects models using lme4

    J. Stat. Software

    (2015)
  • R.R. Bender et al.

    Nutrient uptake, partitioning, and remobilization in modern soybean varieties

    Agron. J.

    (2015)
  • M.T. Brett

    When is a correlation between non-independent variables “spurious”?

    Oikos

    (2004)
  • D.A. Charles-Edwards

    On the ordered development of plants. I. An hypothesis

    Ann. Bot.

    (1984)
  • J.S. Coleman et al.

    Elevated CO2 and plant nitrogen-use: is reduced tissue nitrogen concentration size-dependent?

    Oecologia

    (1993)
  • P.B. Cregan et al.

    Dry matter and nitrogen accumulation and partitioning in selected soybean genotypes of different derivation

    Theor. Appl. Genet.

    (1986)
  • J.L. De Bruin et al.

    Growth, yield, and yield component changes among old and new soybean cultivars

    Agron. J.

    (2009)
  • M. de Felipe et al.

    Soybean genetic gain in maturity groups III to V in Argentina from 1980 to 2015

    Crop Sci.

    (2016)
  • D.N. Duvick et al.

    Long-term selection in a commercial hybrid maize breeding program

    Plant Breed. Rev.

    (2004)
  • D.B. Egli et al.

    Crop growth rate and seeds per unit area in soybean

    Crop Sci.

    (1991)
  • A.P. Gaspar et al.

    Dry matter and nitrogen uptake, partitioning, and removal across a wide range of soybean seed yield levels

    Crop Sci.

    (2017)
  • Cited by (13)

    • A strong negative trade-off between seed number and 100-seed weight stalls genetic yield gains in northern Japanese soybean cultivars in comparison with Midwestern US cultivars

      2022, Field Crops Research
      Citation Excerpt :

      This suggests that the contribution of the increase in seed number to increased seed yield is higher in US cultivars. Tamagno et al. (2020) found that seed number and seed yield increased with increasing YOR, but that 100-seed weight did not change in the US MGIII and MGIV cultivars. Morrison et al. (2000) studied a group of cultivars released in Canada from 1934 to 1992 and showed that the yield gain during this period was due to an increase in seed number rather than an increase in seed mass.

    • Selection for yield shifted the proportion of oil and protein in favor of low-energy seed fractions in soybean

      2022, Field Crops Research
      Citation Excerpt :

      This leads to a tradeoff between N remobilization and photosynthesis during seed filling, particularly strong in energy-rich seed crops (Sadras et al., 1993). Newer soybean varieties have increased their ability to accumulate N during seed filling (Kumudini et al., 2002) and improved the net C balance via longer leaf area duration (Kumudini et al., 2001), both processes permitted to increase the overall plant growth and biomass allocation to pods (Tamagno et al., 2020b). Yet, resource availability to sustain more seeds has been offset by changes in their composition leading to seed protein dilution in modern varieties.

    • Both biomass accumulation and harvest index drive the yield improvements in soybean at high and low phosphorus in south-west China

      2022, Field Crops Research
      Citation Excerpt :

      Empirical soybean breeding has been very effective in improving yield and this has been demonstrated in numerous studies (Koester et al. 2014; Wu et al. 2015; Wang et al. 2016; Todeschini et al. 2019; de Felipe et al. 2020). Previous studies have shown that crop biomass has increased as a result of breeding for seed yield and this has contributed to yield improvements in soybean in China ( Jin et al., 2010), USA (Tamagno et al. 2020), Argentina (de Felipe and Prado, 2021) and Brazil (Todeschini et al., 2019). In north-east China, soybean breeding has increased the seed yield by 0.58% per year while the biomass increased 0.19% per year (Jin et al. 2010).

    • Field pea

      2020, Crop Physiology Case Histories for Major Crops
    View all citing articles on Scopus
    View full text