Crop Science 42:1129-1134 (2002)
© 2002 Crop Science Society of America
CROP BREEDING, GENETICS & CYTOLOGY
Diallel Analysis of Winter Wheat at Two Nitrogen Levels
J. Le Gouis*,
D. Beghin,
E. Heumez and
P. Pluchard
INRA, Domaine de Brunehaut, 80200 Estrées-Mons, France
* Corresponding author (legouis{at}mons.inra.fr)
 |
ABSTRACT
|
|---|
Price reduction and environmental concerns advocate a lower use of nitrogen (N) fertilizer on the wheat (Triticum aestivum L.) crop. It is a common hypothesis that hybrids would be more valuable in stressed environments such as limited fertilizer conditions. The objective of this study was to assess heterosis and combining ability at two N levels. Seven winter wheat cultivars were used to produce a 7 x 7 diallel cross without the reciprocals. The 21 F1 hybrids and parental lines were tested in replicated plots over 2 yr without N fertilizer (N0) or with 150 kg N ha-1 (N+). The diallel analysis was conducted according to Griffing with year, genotype, and treatment as fixed effects. Mid-parent heterosis for grain yield was +12.2%** at N0 and +8.9%** at N+ in 1997 and +1.7%ns at N0 and -0.4%ns at N+ in 1998. This was directly related to high mid-parent heterosis for above-ground dry weight in 1997 (+11.2%** at N0 and +10.9%** at N+) and low heterosis in 1998 (+1.2%ns at N0 and +0.0%ns at N+). General (GCA) and specific (SCA) combining ability effects were always significant. The GCA/SCA ratio ranged from 3.6 to 14.8. The GCA x N level interaction was generally significant indicating different parental contributions at low or high N levels. The SCA x N level interaction was never significant. There was a tendency toward higher GCA/SCA ratio at N0 than at N+. The choice of parents will be dependent upon the N level under which the new hybrids will be grown.
Abbreviations: GCA, general combining ability • HI, harvest index • NHI, N harvest index • SCA, specific combining ability • TKW, thousand kernel weight
|
INTRODUCTION
|
|---|
MEAN NITROGEN APPLICATION to wheat crop was 169 kg N ha-1 yr-1 in a survey carried out in northern France between 1992 and 1997 (Quiévreux, 1997). There are, however, increasing environmental concerns about pollution related to agricultural practices. Because of overproduction, the new European Common Agricultural Policy also advocates lower crop prices. These two reasons tend to favor a decrease in crop inputs and particularly N fertilizer.
Since 1994, F1 hybrid wheat cultivars regularly have been registered in France and they now represent about 5% of the wheat growing area. They are produced thanks to efficient chemical hybridizing agents which, when applied at the right stage, induce male sterility. Because of the broader genetic basis of hybrids, compared with homozygous lines, it has been hypothesized that F1 hybrids would be more stable. Studies involving mostly the regression of genotype means over environment means to estimate stability parameters (Eberhart and Russell, 1966) have failed to reveal a higher stability for F1 hybrids (Carver et al., 1987; Borghi and Perezin, 1990; Pickett and Galwey, 1997; Bruns and Peterson, 1998). There was even a tendency toward higher regression slopes (b values) for hybrids than for pure lines, indicating an adaptation to favorable environments.
In plot trials conducted at normal seeding density, several authors have reported limited best-parent heterosis for grain yield under high yielding conditions, e.g., +4.3% for 10 hybrids (Borghi et al., 1988), +7.3% for 17 hybrids (Brears et al., 1988), +3.6% for 430 hybrids (Morgan et al., 1989), +3.1% for 34 hybrids (Oury et al., 1990a), from -5.0% to +7.9% depending on the location for 33 hybrids (Oury et al., 1994). Concerning N deficiency, it was shown with a set of 10 hybrids grown at one location (Le Gouis and Pluchard, 1996) that best-parent heterosis was higher at the low N level than at the high N level (9.9%* and 1.1%ns respectively). Moreover, Perezin et al. (1992) and Oury et al. (1994)(1995) have reported either a higher grain protein content of the hybrids or the same protein content despite a higher grain yield. These results tend to indicate a higher yield potential for hybrids compared with pure lines at low N level or a higher N uptake efficiency that could be of value in low-input management systems.
Diallel analysis are frequently used in plant breeding to assess general and combining abilities for traits (e.g., Parodi et al., 1983; Du et al., 1999). The few diallel studies that have been reported for replicated plot trials at normal densities for wheat (Borghi and Perenzin, 1994; Perenzin et al., 1998) were conducted at high N levels. The objectives of this study were therefore (i) to evaluate whether F1 wheat hybrids utilize nitrogen more efficiently than pure lines at low and high N levels, and (ii) to identify general and specific combining abilities for N utilization among a set of adapted cultivars.
|
MATERIALS AND METHODS
|
|---|
Eight winter wheat cultivars were used to produce a 8 x 8 diallel cross without the reciprocals. F1 hybrids were obtained in 1994 by means of a chemical hybridizing agent (Croisor), kindly provided by Dupont-Hybrinova (Paris, France). An experiment to evaluate the 28 F1 hybrids and the eight parental lines was established on 10 Oct. 1996 and 14 Oct. 1997 at Estrées-Mons INRA experimental station (Somme, northern France). Because of a very hard 1996-1997 winter, one of the parents and most of the hybrids derived from it sustained frost damage. The analysis was then conducted on only seven parents (Table 1) and the 21 corresponding hybrids. Six of the parents were registered cultivars (GEVES, 1993) and VM014 was a breeding line from INRA Estrées-Mons. They were mainly chosen for difference in nitrogen use efficiency at low and high N levels (Le Gouis et al., 2000).
View this table:
[in this window]
[in a new window]
|
Table 1. Main characteristics of the seven winter wheat varieties used as parental lines in the diallel cross. Means are reported by year or N level when the genotype x year or genotype x N level interaction was significant.
|
|
The soil was classified as a deep loam soil (Orthic Luvisol, FAO classification) contained an average of 190 g clay kg-1, 730 g silt kg-1, 52 g sand kg-1, and 19 g organic matter kg-1 with a pH of 8.1. Soil samples were found to have 76 kg ha-1 (10 Feb. 1997) and 80 kg ha-1 (3 Feb. 1998) mineral N in the upper 120-cm profile. Fertilized plots (N+) received 150 kg ha-1 N as a liquid solution (15% ureic acid, 7.5% ammonium, and 7.5% nitrate) in three applications (50 kg N at tillering, beginning of stem elongation, and 2nd node stage) while other plots (N0) received no N. The experimental design was a spit-plot with four replications. N levels and genotypes were assigned to main plots and subplots, respectively. Each plot consisted of six 4-m-long (1997) and 5-m-long (1998), rows 0.2 m apart and was sown at a density of 300 grains m-2. Fungicide, herbicide, and insecticide treatments were applied to achieve a total control of parasites. A growth regulator (1012 g ha-1 chlormequat chloride + 77 g ha-1 choline chloride + 22 g ha-1 imazaquine as 2.2 L ha-1 Cycocel CL) was sprayed on 3 Apr. 1997 and 19 Mar. 1998 to reduce the risk of lodging.
At flowering (anthers visible on 50% of the spikes), a six row by 1-m section was harvested at ground level and weighed. Shoot and ear dry matters (80°C for 48 h) were estimated on a 1-kg fresh weight basis. Before harvest, about 150 shoots were selected randomly from within each plot, cut at ground level and oven-dried at 80°C for 48 h. These shoots were used to estimate thousand kernel weight (TKW), number of grains per ear, harvest index (HI), grain N concentration and straw + chaff N concentration. Nitrogen harvest index (NHI) was calculated as grain N/total above-ground N. N concentrations were measured with a near infrared reflectance analyzer (Technicon InfraAlyzer 400, Technicon Instruments Corporation, Tarrytown, NY) calibrated against a Dumas procedure (Dumas, 1831).
Differences among parental lines was first tested by a standard analysis of variance with year, line, and treatment as fixed effects. The diallel analysis was conducted according to Griffing (1958) method 4 (excluding parents and reciprocal F1s) by a SAS (SAS, 1989) program modified from Zhang and Kang (1997) with year, genotype, and treatment as fixed effects. The response Cijkl of the cross between parent i and j was modeled as:
where µ is a general constant, yk is the main effect of year k, bl(k) is the main effect of block l within the year, 
i and, j are the general combining ability (GCA) of parent i and j, ßij is the specific combining ability (SCA) of hybrid ij, (y)ik, (y)jk, and (ßy)ijk are interactions of GCA and SCA with the year, (b)il(k), (b)jl(k) and (ßb)ijl(k) are interactions with the block, nm is the main effect of the N level m, (yn)km is the interaction between the year and the N level, (bn)l(k)m is the interaction between the block and the N level, (n)im, (n)jm and (ßn)ijm are interactions of GCA and SCA with N level, (yn)ikm, (yn)jkm and (ßyn)ijkm are interactions between of GCA and SCA with year and N level, and 
ijl(k)m is the residual effect. Each effect was tested using its interaction with the block as error term.
For each hybrid and each character, the difference between the hybrid and the mean of its two parents was computed. A t-test was used to test whether the mean of these differences was different from 0 and then if mid-parent heterosis was significant. For each trait, a contrast was used to test whether the best hybrid was superior to the best pure line. Linear correlation coefficient was calculated for each trait at each N level between the means of the parental lines (per se value) and their GCA.
|
RESULTS AND DISCUSSION
|
|---|
Significant differences were found among the seven parental lines for all characters except NHI (Table 1). The genotype x N level interaction was significant for the number of ears/m2, the number of grains/m2, and TKW. The genotype x year interaction was significant for total above-ground dry weight at anthesis and HI. The cultivar Trémie, a feed wheat, had the highest grain yield. This genotype was characterized by a high total above-ground dry weight at harvest, a high HI, and a low grain N content. The cultivar Renan, a very good bread-making quality variety, had the highest grain N yield, but it was not significantly different from VM014, Eurêka, and Soissons. Renan also had a high grain N concentration, but a low HI.
Grain yield and grain N yield at the N0 were lower in 1997 than in 1998 (Fig. 1)
. This probably was not related directly to soil mineral N in the upper 120-cm profile in February since they were similar for both years (76 and 80 kg N ha-1, respectively). Since mean total above-ground N at harvest exceeded soil mineral N for both years (85 kg N ha-1 in 1997 and 130 kg N ha-1 in 1998), appreciable amounts of N were assimilated from depths lower than 120 cm or provided by the soil after February. This N is likely to have originated mainly from mineralization during the growth period and was higher in 1998 than in 1997.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 1. Box-plot presentation of eight characters measured 2 yr at two N levels on seven parental lines and the 21 corresponding hybrids. The mean is indicated by a solid line. When the best hybrid is significantly higher than the best line, the best hybrid and the best line are represented by a black dot. The mid-parent heterosis is indicated at the bottom for each year and N level.
|
|
Significant positive mid-parent heterosis was observed for grain yield and for grain N yield at both N levels in 1997 but not in 1998 (Fig. 1). At the N+ in 1997, the best hybrid (Renan x Eurêka) was superior to the best line for grain yield and grain N yield. Negative heterosis was observed for grain N concentration but no significant differences were observed between the best hybrid and the best pure line. The data did not support enhanced performance at N0 of the hybrids over the parents. Values reported here for mid-parent heterosis are in accordance to previous data obtained in replicated plot trials at high N level sown at normal seeding rates (Borghi et al., 1988; Morgan et al., 1989; Oury et al., 1990a,b). Oury et al. (1994) found best-parent heterosis ranging from -1.9 to 7.9% for grain yield and from -8.0 to 5.5% for grain N yield in an experiment carried out at seven locations. Oury et al. (1990a) tested three hybrids in experiments over 2 to 4 yr. In each case, heterosis for grain yield varied according to year, from -8.0 to +21.0% for one hybrid tested over 4 yr. It is always difficult to explain genotype x year interaction. In our case, diseases must be ruled out since the trials were efficiently protected. Cold tolerance and lodging resistance could have affected genotype performance. The 1996-1997 winter was very cold. Renan is more cold resistant but more prone to lodging than other parents in this experiment. In spite of the use of a growth regulator, limited lodging occurred in 1998. Part of the interaction between year and heterosis may be due to the positive role of cold and the negative role of lodging in Renan and its progeny.
Significant positive heterosis was observed for the number of ears/m2 at the N0 in 1997 (data not shown), and for TKW both years at both N levels (Fig. 1). The best hybrid was superior to the best line at N0 and in 1998 at N+. At high N level and normal seeding rate, no mid-parent heterosis for ears/m2 or grains/m2 has been generally observed (Morgan et al., 1989; Oury et al., 1990a,b, 1993). On the contrary, heterosis measured on spaced plants was often reported for ears/m2 or grains/m2 (Pickett, 1993). Competition for light may be involved since it should be lower with spaced plants and at low N level.
Significant positive mid-parent heterosis for total above-ground dry weight and above-ground N at anthesis and at harvest was observed in 1997 (Fig. 1). At anthesis, the best hybrid was often superior to the best line and the best hybrid always involved the late flowering cultivar Renan. We observed in 1997 heterosis for dry matter post-anthesis assimilation but no heterosis for post-anthesis assimilation of N, for HI, and NHI (data not shown). Heterosis for total above-ground N at flowering and maturity may be related to a more efficient root system. Hybrids are generally taller than the mid-parent and near the highest parent (Morgan et al., 1989; Borghi et al., 1988, 1989; Oury et al., 1990b). Although the relationship between shoot and root development is not completely clarified (Clarke and McCaig, 1993) a correlation between shoot and root development has been reported in wheat (MacKey, 1973). Furthermore, heterosis has been shown for different root characteristics such as root length, root dry matter, and root area (Kraljevic-Babalic et al., 1988).
GCA effects were significant for all traits considered (Table 2). SCA effects were significant except for total above-ground N at anthesis. For several traits the GCA x year and SCA x year interactions were significant. The GCA x N level interaction was significant for five traits including grain yield and grain N yield but was not significant for grain N content. Absence of genotype x N level interaction was also reported for pure lines (Le Gouis et al., 2000). The SCA x N level interaction was never significant (Table 2). The GCA/SCA ratio ranged from 5.6 to 14.8 at the N0 and from 3.6 to 13.7 at N+. The ratio was higher at the N0 than at the N+ for grain yield, grain N yield, total above-ground N, and dry weight at harvest. Studies conducted in replicated plot trials have shown that GCA effects were larger than SCA effects in wheat (Morgan et al., 1989; Borghi and Perenzin, 1994; Perenzin et al., 1998; Oury et al., 2000). Considering grain yield, GCA effects were largely superior to SCA effects, 3.6 and 12.1 times higher in our study and about 5 times higher in Borghi and Perenzin (1994) and Perenzin et al. (1998). SCA effects were stronger at high N level, at least for grain yield and grain N yield. This is somewhat different from results on maize (Di Fonzo et al., 1982) who found non significant GCA effects at a low N level for grain yield and total dry matter while SCA effects were significant.
Récital and Thésée had negative GCA values for grain yield, while Eurêka, Trémie, and VM014 had positive values (Table 3). The significant GCA x N level interaction was due to Récital and Renan. Récital had a negative value for the GCA x N0 interaction, meaning that compared with the other genotypes its GCA was lower at N0 than at N+. On the contrary, Renan had a higher GCA at N0 than at N+. The GCA x year interaction was explained by Récital and Renan, which had better GCA values in 1997, and by Soissons and Trémie which had better ones in 1998. Significant SCA values for grain yield existed for Récital x Thésée (-28.3**), Récital x Trémie (+44.0***), Renan x Trémie (-30.9**), and Thésée x VM014 (+21.7*).
View this table:
[in this window]
[in a new window]
|
Table 3. Estimates of GCA, GCA x N level and GCA x Year effects when significant for agronomic characteristics in a seven winter wheat diallel. Since the experiment involved two N levels for 2 yr, the GCA x N0 and GCA x 97 values of the interaction are only presented. The GCA x N+ and the GCA x 98 are the opposite of the ones presented.
|
|
Récital was characterized by negative GCA for most traits, particularly at low N level (Table 3). Récital, Thésée, and Trémie had negative GCA for grain N yield, while Eurêka, Renan, and Soissons had a positive GCA. Renan was the only genotype with a positive significant GCA for grain N concentration. Trémie and Thésée had significantly negative ones. However, the GCA x year interaction was due to Renan which, in comparison with the other genotypes, had a low GCA in 1997. Significant SCA values for grain N yield existed for Eurêka x VM014 (-0.65*), Récital x Soissons (-0.62*), Récital x Thésée (-0.58*), Récital x Trémie (+1.02***), Renan x Soissons (+0.72*), and Thésée x VM014 (+0.79**).
Since GCA effects were largely superior to SCA effects, the correlation between per se value and GCA will give an indication about the possibility to use the means of the two parents to predict the value of the hybrid. At the N0, significant correlations were observed for grain N concentration on both years and for total above-ground dry weight at anthesis in 1997 (Table 4). At the N+, 13 correlations were significant. For three characters, grain N concentration, HI, and total above-ground dry weight at anthesis, the correlations were significant for both years. A very high correlation was observed for grain yield in 1998.
View this table:
[in this window]
[in a new window]
|
Table 4. Coefficients of correlation between GCA and per se values of seven winter wheat cultivars grown 2 yr at two N levels.
|
|
|
CONCLUSIONS
|
|---|
From a breeder point of view, the creation of good hybrids to be cultivated in low-input systems at low N level will be facilitated by the high GCA effects observed. The GCA x N level interaction indicates that results classically obtained at high N level would not be enough to identify parents and that specific experiments at low N level will be necessary. Among the seven parents, Eurêka showed the best GCA for grain yield at the N0. It has also a high GCA at the N+ but its low bread-making quality is certainly disadvantageous as the quality of the hybrid is often intermediate between the two parents (Borghi et al., 1988; Oury et al., 1994). Renan combines a high GCA at the N0 for grain yield, a high GCA for grain protein concentration, and a good bread-making quality.
|
ACKNOWLEDGMENTS
|
|---|
We thank D. Bouthors, D. Brasseur, L. Bodet, and M. Geloen for their valuable technical assistance. We also thank Dr. M. Brancourt-Hulmel, Dr. E. Hanocq, and Dr. F.X. Oury for helpful comments in the preparation of this paper.
Received for publication May 12, 2000.
|
REFERENCES
|
|---|
- Borghi, B., and M. Perenzin. 1990. Yield and yield stability of conventional varieties and F1 bread wheat hybrids. J. Genet. Breed. 44:307–310.
- Borghi, B., and M. Perenzin. 1994. Diallel analysis to predict heterosis and combining ability for grain yield, yield components and bread-making quality in bread wheat (T. aestivum). Theor. Appl. Genet. 89:975–981.
- Borghi, B., M. Perenzin, and R.J. Nash. 1988. Agronomic and qualitative characteristics of ten bread wheat hybrids produced using a chemical hybridizing agent. Euphytica 39:185–194.
- Borghi, B., M. Perenzin, and R.J. Nash. 1989. Combining ability estimates in bread wheat and performance of 100 F1 hybrids produced using a chemical hybridizing agent. J. Genet. Breed. 43:11–16.
- Brears, T., A.G. Hydon, and J. Bingham. 1988. An assessment of the feasibility of producing F1 and F2 hybrids for the UK. p. 1057–1062. In T.E. Miller and R.M.D. Koebner (ed.) Proceedings 7th International Wheat Genetics Symposium, Vol. 2. Institute of Plant Science Research, Cambridge, UK.
- Bruns, R., and C.J. Peterson. 1998. Yield stability and stability factors associated with hybrid wheat. Euphytica 100:1–5.
- Carver, B.F., E.L. Smith, and H.O. England. 1987. Regression and cluster analysis of environmental responses of hybrid and pure line winter wheat cultivars. Crop Sci. 27:659–664.[Abstract/Free Full Text]
- Clarke, J.M., and T.N. McCaig. 1993. Breeding for efficient root systems. p. 485–499. In M.D. Hayward et al. (ed.) Plant breeding, principles and prospects. Chapman & Hall, London.
- Di Fonzo, N., M. Motto, T. Maggiore, R. Sabatino, and F. Salamini. 1982. N-uptake, translocation and relationships among N-related traits in maize as affected by genotype. Agronomie 2:789–796.[Web of Science]
- Du, C.G., L.R. Nelson, and M.E. McDaniel. 1999. Diallel analysis of gene effects conditioning resistance to Stagonospora nodorum (Berk.) in wheat. Crop Sci. 39:686–690.[Abstract/Free Full Text]
- Dumas, J.B.A. 1831. Procédés de l'analyse organique. Ann. Chim. Physiol. 2:198–213.
- Eberhart, S.A., and W.A. Russell. 1966. Stability parameters for comparing varieties. Crop Sci. 6:36–40.[Abstract/Free Full Text]
- GEVES. 1993. Bulletin des variétés. Crales. Groupe d'Etude et de Contrôle des Variétés et des Semences, Guyancourt, France.
- Griffing, B. 1958. Concept of general and specific combining ability in relation to diallel crossing systems. Aust. J. Biol. Sci. 9:463–493.
- Kraljevic-Balalic, M., R. Kastori, M. Vojvodic. 1988. Inheritance of total root area, length and dry weight in F1 wheat crosses. Genetika (Zemun, Yugosl.) 20:229–234.
- Le Gouis, J., and P. Pluchard. 1996. Genetic variation for nitrogen use efficiency in winter wheat (Triticum aestivum L.). Euphytica 92:221–224.
- Le Gouis, J., D. Beghin, E. Heumez, and P. Pluchard. 2000. Genetic differences for nitrogen uptake and nitrogen utilisation efficiencies in winter wheat. Eur. J. Agron. 12:163–173.
- MacKey, J. 1973. The wheat root. p. 827–842. In E.R. Sears and L.M.S. Sears (ed.) Proc. 4th Int. Wheat Genet. Symp., University of Missouri, Columbia.
- Morgan, C.L., R.B. Austin, M.A. Ford, J. Bingham, W.J. Angus, and S. Chowdhury. 1989. An evaluation of F1 hybrid winter wheat genotypes produced using a chemical hybridizing agent. J. Agric. Sci. (Cambridge) 112:143–149.
- Oury, F.X., J. Koenig, P. Bérard, and M. Rousset. 1990a. Une comparaison entre des blés hybrides produits par voie chimique et leurs parents: niveau d'hétérosis et élaboration du rendement. Agronomie 10:291–304.
- Oury, F.X., P. Brabant, P. Pluchard, P. Bérard, and M. Rousset. 1990b. Etude multilocal de blés hybrides: niveaux d'hétérosis et élaboration du rendement. Agronomie 10:735–748.
- Oury, F.X., P. Brabant, P. Pluchard, P. Bérard, and M. Rousset. 1993. Une étude sur la supériorité des blés hybrides au niveau des capacités de remplissage du grain: résultats d'une expérimentation multilocale. Agronomie 13:381–393.
- Oury, F.X., P. Brabant, P. Pluchard, P. Bérard, and M. Rousset. 1994. Une étude de la qualité des blés hybrides à travers différents tests technologiques. Agronomie 14:377–385.
- Oury, F.X., E. Triboi, P. Bérard, J.L. Ollier, and M. Rousset. 1995. Etude des flux de carbone et d'azote chez des blés hybrides et leurs parents, pendant la période de remplissage du grain. Agronomie 15:193–204.
- Oury, F.X., P. Brabant, P. Bérard, and P. Pluchard. 2000. Predicting hybrid value in bread wheat: biometric modelling based on a "top-cross" design. Theor. Appl. Genet. 100:96–104.
- Parodi, P.C., W.E. Nyquist, F.L. Patterson, and H.F. Hodges. 1983. Traditional combining-ability and Gardner-Eberhart analyses of a diallel for cold resistance in winter wheat. Crop Sci. 23:314–318.
- Perenzin, M., N.E. Pogna, and B. Borghi. 1992. Combining ability for breadmaking in wheat. Can. J. Plant Sci. 72:743–754.
- Perenzin, M., M. Corbellini, M. Accerbi, P. Vaccino, and B. Borghi. 1998. Bread wheat: F1 hybrid performance and parental diversity estimates using molecular markers. Euphytica 100:273–279.
- Pickett, A.A. 1993. Hybrid wheat – results and problems. Plant Breed. 15(Suppl):1–259.
- Pickett, A.A., and N.W. Galwey. 1997. A further evaluation of hybrid wheat. Plant Varieties Seeds 10:15–32.
- Quiévreux, D. 1997. Réseau d'observation de parcelles de céréales. Résultats de la récolte 1997. Chambre d'Agriculture de la Somme, Amiens, France.
- SAS Institute Inc. 1989. SAS/STAT user's guide, Version 6. Cary, NC.
- Zhang, Y., and M.S. Kang. 1997. DIALLEL-SAS: A SAS program for Griffing's diallel analyses. Agron. J. 89:176–182.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
R. Sylvester-Bradley and D. R. Kindred
Analysing nitrogen responses of cereals to prioritize routes to the improvement of nitrogen use efficiency
J. Exp. Bot.,
May 1, 2009;
60(7):
1939 - 1951.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Y. Zhang, M. S. Kang, and K. R. Lamkey
DIALLEL-SAS05: A Comprehensive Program for Griffing's and Gardner-Eberhart Analyses
Agron. J.,
June 17, 2005;
97(4):
1097 - 1106.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Brancourt-Hulmel, E. Heumez, P. Pluchard, D. Beghin, C. Depatureaux, A. Giraud, and J. Le Gouis
Indirect versus Direct Selection of Winter Wheat for Low-Input or High-Input Levels
Crop Sci.,
May 27, 2005;
45(4):
1427 - 1431.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
