| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dep. of Agronomy, Kansas State Univ., Manhattan, KS 66506
* Corresponding author(gmpaul{at}ksu.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Gains in wheat yields were reported for a number of regions. Mean annual genetic gains for experimental plot yields of spring wheat were 1.5% from 1950 to 1982 and 1.3% from 1968 to 1990 in Mexico (Waddington et al., 1986; Bell et al., 1995) and 0.5% from 1982 to 1985 in Canada (Hucl and Baker, 1987). Yields of winter wheat increased annually 0.5% from 1830 to 1986 in England (Austin et al., 1989), 0.4% from 1912 to 1980 and 1.0% from 1920 to 1989 in Argentina (Slafer and Andrade, 1989; Calderini et al., 1995), and 0.5% from 1900 to 1983 in Italy (Canevara et al., 1994). In the Great Plains, annual genetic gains in experimental yields of hard red winter wheat were 0.7% from 1958 to 1980 in regional nurseries (Schmidt, 1984), 0 to 1.4% depending on injury from drought and diseases from 1874 to 1987 in Kansas (Cox et al., 1988), and 0.2% from 1969 to 1993 in Oklahoma (Khalil et al., 1995). Actual advances in growers' yields in the region ranged from about 0.2 to 1.1% per year from 1954 to 1979 (Feyerherm et al., 1984). Yield potential of spring wheat increased about 0.5% per year from 1911 to 1979 in North Dakota (Deckerd et al., 1985).
Many changes in the wheat plant contributed to the genetic gains in grain yields. Most studies reported no change in biomass (e.g., Austin et al., 1980; Waddington et al., 1986), although a few found that increased biomass paralleled release of improved cultivars (Perry and D'Antuono, 1989; Siddique et al., 1989a). Instead, most researchers attributed higher yields to an enhanced harvest index (HI) from more favorable partitioning of assimilates from the vegetation to the grain (Austin et al., 1980; Siddique et al., 1989b; Slafer et al., 1990; Sayre et al., 1997). High HI was associated with more kernels per spikelet and spike (Siddique et al., 1989b; Slafer and Andrade, 1991; Sayre et al., 1997). The increased number of kernels per spike and, to a lesser extent, increased number of tillers, resulted in a close relationship between kernel numbers per unit of soil area and grain yield (Austin et al., 1989; Perry and D'Antuono, 1989; Calderini et al., 1995; Sayre et al., 1997).
Other traits that were associated with higher yields include early maturity (Cox et al., 1988; Khalil et al., 1995), resistance to lodging (Allan, 1989), and long leaf area duration after heading (Borojevic, 1986). Resistance to fungal and viral diseases and, in some cases to insects, was important in many regions (Hucl and Graf, 1994). Resistance to winter injury, drought, and heat also was undoubtedly important in yield gains of modern cultivars but was not evaluated.
Unlike those in other regions, the plant changes that contributed to wheat yield gains in the Great Plains have never been documented adequately. Schmidt (1984) and Feyerherm et al. (1984) considered only advances in grain yields. Khalil et al. (1995) measured plant heading date and grain volume weight as well as yield but none of the other important traits. The most extensive study, that of Cox et al. (1988), included a number of plant characteristics but not HI or kernels m-2, the traits most associated with increases in wheat grain yields in other regions. Their study was compromised further by severe, uniform infection of tan spot (caused by Helminthosporium tritici-repentis Died.), which reduced the reported yields of even the newest cultivars to less than one half of their productivity in performance trials (Roozeboom, 1999).
Our study was prompted by the need to assess the genetic changes in traits related to grain yields of winter wheat in the Great Plains, the major wheat region in the USA. Periodic evaluation of genetic improvement is essential for understanding yield-limiting factors, illustrating the importance of plant breeding to the public, and identifying traits that might require increased effort by breeders (Cox et al., 1988; Evans, 1993). The objective of our study was to determine plant traits that were associated with advances in grain yields of benchmark wheat cultivars in the past, in order to ascertain traits that might be important in the future.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
The genotypes were planted on 18 Oct. 1996 and 21 Oct. 1998 at Hutchinson and 16 Oct. 1996 and 18 Oct. 1998 at Manhattan. Seed with a minimum germination of 85% was planted at 90 kg ha-1 in 1.5 x 3.3 m-plots containing six rows 25 cm apart. Glean herbicide {-2-chloro-N-[(4-methoxy-6-methyl-1,3,5-triazin-2-yl) aminocarbonyl] benzenesulfonomide} was applied at the rate of 24.5 g ha-1 at both locations in both years during the last week of January. Other conventional practices for production of wheat were followed.
One half of each plot was encircled with 10- by 10-cm fine plastic netting at the jointing stage (Feekes 6 to 7) (Large, 1954) to prevent plants from lodging and sprayed with 288 g ha-1 of Folicur (a-{-2(4-chlorophenyl)ethyl}-a-1-1-dimethylethyl) 1H-1,2,4-triazole-1-ethanol) to protect them from leaf rust. The other half of each plot was left untreated to assess genetic gains in lodging and leaf rust resistance.
Observations
Winter survival, grain yield, and total biomass were measured during both years. Additional observations were taken on plant height, leaf rust infection, lodging, heading date, and grain yield components during 1999. Winter survival was estimated by rating plant stands after growth resumed in 1997 and 1999. Heading date (Feekes 10.5) was recorded when spikes were extruded from the boots of one half of the plants in the plots. Plant height was measured from the soil surface to the top of the main spike at physiological maturity (Feekes 11.3). Leaf rust infection was rated on a scale of 1 (no infection) to 9 (completely infected) at 2 wk after anthesis (Feekes 11.1) on plants in plots that were not protected with Folicur or netting. Lodging was estimated on a scale of 1 (no lodging) to 9 (completely lodged) in the same plots immediately before they were harvested.
One 0.5-m-long row of plants was cut from each nonprotected and protected plot after the grain ripened. The number of spikes in the sample was counted, and the mean number of spikelets and spike length were measured in a 50-spike subsample. Plots were harvested on 24 June 1997 and 27 June 1999 at Hutchinson and on 18 June 1997 and 24 June 1999 at Manhattan. All plants in a 1-m2-area of each end of the plots were cut with a sickle at the soil level, dried at 50°C for 72 hr, and weighed. The grain was threshed with a Vogel-type plot thresher (Bill's Welding Co., Pullman, WA) and weighed. Kernel weight was determined on a 1000-kernel subsample. Grain yields and kernel weights were adjusted to 120 g kg-1 moisture content. Harvest indices, kernels spikelet-1, and kernels m-2 were calculated.
Experimental Design and Statistical Analyses
Treatments were arranged in a split-plot design with the genotypes as the main plots and no protection vs. protection against lodging and leaf rust as subplots. Four replications were used at both locations during both years. Data were analyzed by SAS proc mixed procedures (SAS Institute, Cary, NC). Individual genetic gains in traits in each genotype were calculated as the mean values for the genotypes relative to the cultivar Turkey over the differences in years of release. Mean genetic gains in traits of all genotypes were determined as the quotients of the b values from linear regression of the traits on years of release of the genotypes over the mean values of the traits for all genotypes. Phenotypic relationships among measurements were determined by Pearson correlation analysis.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Grain Yield and Above-Ground Biomass
Grain yield and above-ground biomass were statistically similar (P 
0.05) in nonprotected plots and plots protected against lodging and leaf rust, and data were expressed as means of the two treatments (Table 2). Susceptible plants lodged and were infected by leaf rust in nonprotected plots, which enabled rating their severity. However, both lodging and leaf rust occurred too late during maturation to significantly affect yield, and hand-harvesting recovered the grain from all plants in the plots.
|
|
|
Agronomic Traits
Heading date, a relative measure of maturity, was earlier in most improved genotypes than in Turkey (Table 5). Individual genotypes gained 0.01 to 0.04% yr-1 (0.3 to 1.2 hr yr-1), with an overall mean gain of 0.03% yr-1 (Table 3). Heading date was highly negatively correlated with grain yield and HI (Table 4).
|
Spike length remained constant or increased as the height of new genotypes decreased (Table 5). Two genotypes, Newton and 2137, had significant genetic gain in spike length relative to Turkey, but the mean change for all genotypes was nonsignificant (Table 3). Spike length was positively correlated with grain yield, above-ground biomass, and HI and inversely correlated with heading date (Table 4).
Lodging and Leaf Rust
Lodging occurred in all of the tall genotypes but none of the semidwarf genotypes at the two locations (Table 6). Some genetic gain for reduced lodging was expressed by the tall genotypes through the 1960s, but the semidwarf genotypes were resistant and the mean genetic gain was high, 2.15% yr-1 (Table 3). Lodging was negatively correlated with yield, biomass, HI, and spike length and positively correlated with plant height (Table 4).
|
Grain Yield Components
All genotypes survived winter with little or no injury and developed excellent stands during spring (data not shown). Semidwarf genotypes generally formed more spikes m-2 than tall genotypes (Table 7). They also had greater genetic gain for spike density, 0.17 yr-1, vs. no gain for tall genotypes (Table 3). The mean genetic change in all genotypes was nonsignificant. The spike density was positively correlated with grain yield and above-ground biomass and negatively correlated with plant height, lodging, and leaf rust severity (Table 4).
|
Kernel number spike-1 increased unevenly as new genotypes were released (Table 7). One tall genotype and five semidwarf genotypes had significant genetic gain in kernel number relative to Turkey, and the mean of all genotypes increased 0.24% yr-1 (Table 3). Kernel number was positively correlated with grain yield, biomass, HI, spike length, and spikelet number and negatively correlated with heading date, lodging, and leaf rust rating (Table 4).
Kernel number spikelet-1, calculated from the ratio of kernels spike-1: spikelets spike-1, increased more in the semidwarf genotypes than in the tall genotypes (Table 7). Two tall genotypes and four semidwarf genotypes had 0.15 to 0.29% yr-1 genetic gain for the trait, and the mean gain for all genotypes was 0.19% yr-1 (Table 3). Kernel number spikelet-1 was positively correlated with grain yield, biomass, HI, spike length, and kernels spike-1 and negatively correlated with heading date (Table 4).
Kernel density (number m-2), the product of kernel number spike-1 and spikes m-2, changed little in the tall genotypes but increased markedly in the semidwarf genotypes (Table 7). None of the tall genotypes had significant genetic gain, but kernel density increased 0.22 to 0.51% yr-1 in the eight semidwarf genotypes (Table 3). The mean increase for all genotypes was 0.43% yr-1. Kernel density was positively correlated with yield, biomass, HI, spike length, spike density, spikelet number, kernels spike-1, and kernels spikelet-1 and negatively correlated with heading date, height, lodging, and leaf rust ratings (Table 4).
Kernel weight differed little among the genotypes (Table 7). The trait increased 0.15 to 0.16% yr-1 in Wichita and Triumph, but the mean change in all the genotypes was nonsignificant (Table 3). Kernel weight was not correlated with any of the other plant characteristics (Table 4).
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The mean annual rate of genetic gain in yield of all genotypes in the study, 0.44%, was similar to values found by other studies in the Great Plains (Feyerherm et al., 1984; Schmidt, 1984; Cox et al., 1988). However, the mean rate obscured a steady increase in the pace of genetic gain over time from 0.15% yr-1 for early genotypes to 0.63% yr-1 or the newest genotypes. Many factors undoubtedly contributed to the accelerating gain in yield. Specific traits such as reduced plant height and the wider diversity of germplasm for developing new cultivars were probably most important (Cox et al., 1988; Allan, 1989; Kronstad, 1996). The steady gain appeared to contradict any barrier to yield increases in the near future, as was suggested in other regions (Reynolds et al., 1996).
The increase in above-ground biomass that coincided with the gain in grain yield differed from results of most other studies, in which total dry matter stayed constant as partitioning of photosynthate to the grain and HI increased (Austin et al., 1980; Waddington et al., 1986; Calderini et al., 1999). Nevertheless, the gain in grain yields in the Great Plains genotypes was accompanied by a small decrease in vegetative mass and a large increase in HI. The mass of vegetative parts decreased at the same rates in tall and semidwarf genotypes, 0.05 and 0.04% yr-1 respectively, and the greater rate of increase in HI of the latter explained the accelerating gain in yield.
Two yield components, spike density and kernels spike-1, were associated with the gain in grain yields by most of the genotypes. Both traits contributed to the increase in kernel number m-2 as yields advanced, particularly in semidwarf genotypes. Tiller survival likely was limited by availability of assimilates and nutrient resources (Miralles and Slafer, 1999). The concurrent increase in spike number and decrease in vegetative mass indicated that new genotypes utilized assimilates more efficiently for growth of tillers than early genotypes (Fischer, 1983). The gain in kernels spike-1 in the Great Plains wheats came partly from an increase in spikelets spike-1 and mostly from an increase in kernels spikelet-1. Survival of florets to form kernels was determined primarily by competition for assimilates for stem elongation during the 3-wk period before anthesis (Frederick and Bauer, 1999).
The small contribution of kernel weight to the gain in grain yield suggested that the yield component was changed little by breeding during the twentieth century (Calderini et al., 1999). However, that is not the case. Low kernel weight and low grain test weight were persistent problems of early semidwarf genotypes under dryland conditions in the Great Plains (Dalrymple, 1980). Remedying that defect in modern genotypes was a substantial achievement, particularly considering the strong compensation of the trait with kernel number (Frederick and Bauer, 1999).
The greater genetic gain in yield, HI, lodging resistance, spikes m-2, kernels spikelet-1, and kernels m-2 in semidwarf genotypes than in tall genotypes in our study reflected many of the attributes associated with the Rht genes in previous studies (Dalrymple, 1980; Allan, 1989). Changes within the semidwarf genotypes since the release of Newton in 1977 may be more relevant to improvement of modern cultivars but were less distinct. In addition to the accelerating gain in yield, resistance to leaf rust, kernels spikelet-1, and kernels m-2 of soil area increased in most of the semidwarf genotypes in our study. Apparently, both an increase in yield potential and protection of the plants against diseases contributed to the improvement in productivity of most recent genotypes.
Development of components that contributed most to the gain in yields in Great Plains genotypes was related to the climate in the region. Moisture and temperature in the region are most often favorable for vegetative growth of winter wheat during autumn and early spring (Thompson, 1962). The yield components that increased most rapidly in the past, spikes m-2 and kernels spike-1, are determined during early vegetative stages (Frederick and Bauer, 1999). In contrast, kernel weight, the yield component that changed least, is determined during late spring and early summer, when stress from drought and heat is more common (Thompson, 1962) and the crop is most susceptible to injury (Paulsen, 1994; Saini and Westgate, 2000). The greater improvement in components that develop during stages when conditions are favorable than when they are unfavorable may be related to the faster pace of yield increases in regions with benevolent environments than in regions with stressful environments (Feyerherm et al., 1984).
In addition to climate, other factors might influence the improvement of yield in the future. The highest HI in the study, 0.36, was well below the purported limit of 0.62 (Austin et al., 1980). A considerable increase in HI of future genotypes might be possible, but a marked rise would not be desirable if plant height decreased simultaneously (Calderini et al., 1999). Similarly, the advance in heading date that was noted undoubtedly benefitted yield. It likely reduced kernel loss from abortion and increased kernel growth by enabling fertilization and maturation to occur before stress from inadequate moisture and high temperature became severe. However, further advance in development might injure plants by subjecting them to late spring frosts at critical stages (Paulsen and Heyne, 1983).
Breeding for problems that detract from yield likely will be as important in the future as it has in the past. Changes in races of leaf rust require new sources of resistance in improved cultivars (Reis et al., 1999). The excellent lodging resistance noted in the semidwarf genotypes in our study might suggest that that problem has been significantly alleviated. However, a continuation of past changes—higher densities of smaller tillers with greater HI—might increase susceptibility to lodging in the future.
Breeding for higher yield potential can take three approaches: increasing the HI, increasing total biomass, or both (Slafer et al., 1999). Some increase in HI might be feasible but, in addition to the adverse relationship with plant height, the Great Plains climate of relatively favorable conditions for vegetative growth and stressful conditions for grain filling might limit future gains. Total biomass might be increased by raising the photosynthetic rate (Slafer et al., 1999). However, enhancement of radiation use efficiency of wheat is limited by high temperature, the low CO2:O2 concentration of the atmosphere, and light saturation of photosynthesis (Loomis and Amthor, 1996). Some combination of increases in both HI and biomass that is expressed as kernels spike-1 might be most productive. Additional kernels m-2, the major determinant of increased yield in semidwarf genotypes, results from better allocation of assimilates from the elongating stem to the developing spike immediately before anthesis (Fischer, 1983; Slafer et al., 1999). Many floret primordia that are initiated cease developing before anthesis (Bonnett, 1936). Further improvement in partitioning of assimilates to the spike during the critical preanthesis period might increase kernels spikelet-1 and kernels m-2.
The small genetic gain and conserved nature of the last yield component to form, kernel weight, suggest that the potential for future improvement in the trait is limited. However, the mean kernel weight of all entries in the study, 29.1 mg, while typical for the region, was substantially lower than values for wheat in more favorable climates (Frederick and Bauer, 1999). Thus, there might be more room for improvement in kernel weight of genotypes for the Great Plains than for other regions. Drought and high temperature during maturation of wheat are common in the Great Plains (Thompson, 1962) and are major weather variables reducing the yield potential (Fischer, 1983). Winter wheat cultivars grown under typical high temperatures suffered a mean 6% loss in kernel number but a mean 21% decline in kernel weight (Al-Khatib and Paulsen, 1990). The range in response of kernel weight to high temperature of 7 to 32% suggested that the yield component is elastic and might be increased by improving resistance of wheat to stress during maturation. Resistance to stress would lengthen the leaf area duration during maturation and prolong the grain-filling period to increase the kernel weight (Borojevic, 1986).
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Received for publication June 21, 2000.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Royo, V. Martos, A. Ramdani, D. Villegas, Y. Rharrabti, and L. F. Garcia del Moral Changes in Yield and Carbon Isotope Discrimination of Italian and Spanish Durum Wheat during the 20th Century Agron. J., February 26, 2008; 100(2): 352 - 360. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Underdahl, M. Mergoum, Joel. K. Ransom, and B. G. Schatz Agronomic Traits Improvement and Associations in Hard Red Spring Wheat Cultivars Released in North Dakota from 1968 to 2006 Crop Sci., January 16, 2008; 48(1): 158 - 166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Dhungana, K. M. Eskridge, P. S. Baenziger, B. T. Campbell, K. S. Gill, and I. Dweikat Analysis of Genotype-by-Environment Interaction in Wheat Using a Structural Equation Model and Chromosome Substitution Lines Crop Sci., March 1, 2007; 47(2): 477 - 484. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Zhou, Z. H. He, X. X. Sui, X. C. Xia, X. K. Zhang, and G. S. Zhang Genetic Improvement of Grain Yield and Associated Traits in the Northern China Winter Wheat Region from 1960 to 2000 Crop Sci., January 22, 2007; 47(1): 245 - 253. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Hansen, J. M. Martin, S. P. Lanning, and L. E. Talbert Correlation of Genotype Performance for Agronomic and Physiological Traits in Space-Planted versus Densely Seeded Conditions Crop Sci., May 6, 2005; 45(3): 1023 - 1028. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. J. Shearman, R. Sylvester-Bradley, R. K. Scott, and M. J. Foulkes Physiological Processes Associated with Wheat Yield Progress in the UK Crop Sci., January 1, 2005; 45(1): 175 - 185. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Marque, A. K. Fritz, T. J. Martin, and G. M. Paulsen Agronomic and Quality Attributes of Winter Durum Wheat in the Central Great Plains Crop Sci., May 1, 2004; 44(3): 878 - 883. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. T. Campbell, P. S. Baenziger, K. M. Eskridge, H. Budak, N. A. Streck, A. Weiss, K. S. Gill, and M. Erayman Using Environmental Covariates to Explain Genotype x Environment and QTL x Environment Interactions for Agronomic Traits on Chromosome 3A of Wheat Crop Sci., March 1, 2004; 44(2): 620 - 627. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | 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 | |||
