Previous PageTable Of ContentsNext Page

Genotypic variation for nitrogen efficiency in winter rapeseed cultivars

Christian Möllers, Maria Kahlmeyer, Bettina Kessel, Ariane Ossenkop and Heiko C. Becker

Institut für Pflanzenbau und Pflanzenzüchtung, Georg-August-Universität, Von-Siebold-Str. 8, D-37075 Göttingen, Germany, Email: cmoelle2@gwdg.de

Abstract

Breeding for improved N-efficiency in rapeseed is gaining increasing interest for ecological and economical reasons. 18 different German winter rapeseed cultivars and breeding lines were tested for two years in multi-locational field trials for their yield under three different N-fertilizer regimes, namely nil, medium and high. Significant genotypic differences were found for grain, oil and protein yield as well as for total N-uptake. Genotypes giving a superior yield under limiting N-supply were identified. The tested hybrid cultivars gave superior yields not only at the high but also at the medium and low N-fertilizer levels. Genotype x N-fertilizer interactions were significant for grain yield (P<0.05) and for total N-uptake (P<0.01). A close correlation between total N-uptake and grain yield was found at low and high N-fertilizer supply, indicating the importance of N-uptake efficiency.

Keywords

Brassica napus, N-efficiency, Nmin, hybrid, nutrient efficiency

Introduction

Economic rapeseed production requires high rates of N fertilizer application. An increasing number of reports about nitrate leaching into the groundwater after rapeseed cultivation and decreasing profit margins of the rapeseed growers demands the breeding and use of cultivars with improved N efficiency. However, to date there have been only few investigations about the genetic variability for N-uptake and N-utilization efficiency in oilseed rape (Brassica napus L.). The present study was performed to analyse the genotypic variability for N-efficiency among 18 winter rapeseed cultivars in field experiments under German conditions.

Methods and materials

18 different winter rapeseed genotypes were tested during 1996/97 and 1997/98 growing seasons in field experiments at five different environments in Central and Northern Germany. The experiments were performed with three different N-fertilizer levels, namely 0, 120 and 240 kg N/ha, including soil Nmin. The tested genotypes consisted of 14 line varieties (Bristol, Capitol, Express, Falcon, Licondor, Licord, Limbo, Lirajet, Lisabeth, Lizard, Maplus, Marathon, Rasmus, Wotan), 3 hybrid varieties (Life, Panther, Synergy) and one experimental doubled haploid line (DH Mansholts) derived from the old cv. Mansholts Hamburger Raps. A randomized split plot design with three replications at the locations Bonn, Hohenlieth and Kritzkow and four replications at the two other locations close to Göttingen was used. Plot sizes ranged from 11.25 to 17.50 m2. 2 weeks before threshing plants from an area of 0.375 m2 of the N0 and the N2 variant were harvested from each replication at the two experimental sites near Göttingen to determine the N-content in the seeds, the straw and the silique walls. Nitrogen concentration was determined in a sample of about 20 mg flour by the Dumas combustion method using an automated CN analyser (Heraeus CN-Rapid, Hanau, Germany) and standard procedures (Anonymous 1990); part of the samples were also analysed by NIRS using a calibration developed at the department (Velasco and Möllers 1999, submitted). At maturity, seed yield was determined and after NIRS analysis (Tkachuk 1981) of seed oil and protein content, the oil and protein yield was calculated. Total N-uptake/ha was calculated from the seed protein yield plus N-yield of the straw. Immediately after harvest and about 6-8 weeks later the soil Nmin (0-90cm) was determined in each replication of the N0 and the N2 variant of the two locations close to Göttingen. The statistical analyses were performed using the programme PLABSTAT (Version 2F; Utz 1991).

Results and discussion

The mean grain yield increased from 26 dt/ha at N0 over 36 dt/ha at N1 to 41 dt/ha at N2. Significant differences (P<0.01) between the genotypes were found for grain yield, seed oil and protein content, and for oil and protein yield at the 3 N-fertilizer levels (data not shown). 'Genotype x N-fertilizer' (GxN) interactions were significant for grain yield (P<0.05), seed oil (P<0.05) and protein (P<0.01) content and for oil (P<0.05) and protein yield (P<0.01). The interactions for grain yield are demonstrated in Figs. 1 and 2 by comparing the yield obtained at N0 and N1 with the yield at N2. The experimental line DH Mansholts gave the lowest yields; however, with no N-fertilizer the difference to other cultivars was much less compared to N1 and N2. The highest grain yields at all fertilizer levels were obtained by the hybrid cvs. Life, Panther and Synergy and by the line cv. Rasmus.

Fig. 1: Grain yield of 18 winter rapeseed genoytpes obtained with no (N0) and with high levels of N-fertilizer (N2)

There were a number of line cvs. with similar yields at N2, but with varying yields at N0 and N1, indicating differences in N-efficiency. The cv. Bristol reached at the N1 fertilizer level already 98% of its N2-yield. Significant genotypic differences (P<0.01) were found for total N-uptake at N0 and N2 fertilizer levels. 'GxN'-interaction was signifcant (P<0.01) for total N-uptake. The results in Figs. 3 show a close correlation of r=0.92** between the total-N-uptake and the grain yield at limiting N supply (N0). This shows the relative importance of N-uptake efficiency. However, yield

Fig. 2: Grain yield of 18 winter rapeseed genoytpes obtained with medium (N1) and high levels of N-fertilizer (N2)

Fig. 3: Total N-uptake at harvest and grain yield at N0 of 18 winter rapeseed genotypes

differences e.g. between cvs. Marathon and Wotan with similiar total N-uptake indicate that there are also differences in N-utilization efficiency among the tested genoytpes. Interestingly, the correlation between total N-uptake and grain yield is also high under non-limiting N-supply (Fig. 4). According to Fig. 4 cvs. Life and Limbo show a slightly superior N-utilization efficiency, whereas cvs. Maplus and Falcon show a below average N-utilization efficiency.

Fig. 4: Total-N uptake at harvest and grain yield at N2 of 18 winter rapeseed genotypes

At harvest significant genotypic differences (P<0.01) were found for soil Nmin at 0-30, 30-60, 60-90 and 0-90cm at both N-fertilizer levels. Significant 'GxN'-interactions were found for soil Nmin at all depths and were highly significant at 60-90 cm. Correlations between soil Nmin and grain yield, oil yield, total N-uptake and grain N-yield at N2 were all negative, but significant only for soil Nmin at 60-90 cm (Fig. 5). No significant differences could be found for Nmin at N0 and for the soil samples taken 6-8 weeks after harvest.

Conclusions

The genetic basis of the rapeseed material tested in the present study was relatively narrow; no non-adapted material was included. Nevertheless considerable variation was observed for grain yield and N-efficiency related traits at the different N fertilizer levels. In particular the grain yield differences at N1 are intriguing and of importance for rapeseed growers and breeders. The cv. Bristol is one outstanding example of the present experiments, showing that not all cultivars demand high N-fertilizer levels to give high grain yields. For breeding of rapeseed with improved N-efficiency, genotypes with a high grain yield at low N supply could be combined to select superior recombinants. The present data also give an indication that N-uptake effciency is more important than N-utilization efficiency and that high yielding genotypes more efficiently deplete the soil from N, leading to lower soil Nmin contents at harvest.

Fig. 5: Soil Nmin content and grain N-yield of 18 winter rapeseed genoytpes at harvest

Acknowledgements

We thank Isabel Bürgel and Raffaela Gerlach for excellent technical assistance. We also gratefully acknowledge the financial support of the Deutsche Bundesstiftung Umwelt (DBU), Osnabrück, and the support of the Deutsche Saatveredlung (DSV), Lippstadt, Norddeutsche Pflanzenzucht (NPZ), Hohenlieth, and of the Institute of Agronomy and Plant Breeding, Universität Bonn.

References

Anonymous, 1990. Crude protein in animal feed combustion method. In: Association of Official Analytical Chemistry International, AOAC (ed.), Official Methods of Analysis, 15th Edition, 1st Supplement, Section 990.03.

Tkachuk, R., 1981. Oil and protein analysis of whole rapeseed kernels by near infrared reflectance spectroscopy. Journal of the American Oil Chemists Society 58, 819-822.

Utz, H.F., 1991: PLABSTAT. A computer program for statistical analysis of plant breeding experiments, Version 2F. Institut für Pflanzenzüchtung, Saatgutforschung und Populationsgenetik, Universität Hohenheim.

Previous PageTop Of PageNext Page