Analysis of Covariation of Grain Yield and Dry Matter Yield for Breeding Dual Use Hybrid Rye

  • PDF / 381,131 Bytes
  • 6 Pages / 595.276 x 790.866 pts Page_size
  • 74 Downloads / 253 Views

DOWNLOAD

REPORT


Analysis of Covariation of Grain Yield and Dry Matter Yield for Breeding Dual Use Hybrid Rye Stefan Haffke . Barbara Kusterer . F. Joachim Fromme . Steffen Roux . Bernd Hackauf . Thomas Miedaner

Published online: 26 October 2013 © Springer Science+Business Media New York 2013

Abstract Winter rye (Secale cereale L.) is becoming increasingly important as substrate for biogas production in Central Europe. Dry matter yield has evolved as a breeding goal comparably important to the traditional grain yield. We analyzed the covariation between both traits and tested other agronomic traits for their correlation to dry matter yield that could be used for prediction of biomass yield. A set of 258 experimental hybrids were tested for dry matter yield harvested at late milk stage and grain yield harvested at full ripening at three to four locations in Germany in 2011 and 2012. We observed a wide range of dry matter yield (10– 24 Mg ha−1) and grain yield (6–15 Mg ha−1) among testcross progenies. Genetic variances were significantly (P  0.05) difference among experiments was found. Statistical Analyses Statistical analyses were based on plot data of 258 testcross progenies. Checks were calculated seperately. All statistical computations were performed with the PLABSTAT [14] software package in a two-step procedure. Analyses of variance [15] were firstly performed for all traits in each environment separately. The adjusted entry means from each location was used in a second step to estimate variance components based on the following linear model:



y  G  E  G  E,



where G and E denote genotype and environment, respectively. Both factors were treated as random effects. Simple phenotypic correlation coefficients (rP) were calculated between the traits and their significance was tested using tabulated values of z transformation [16]. Coefficients of genotypic correlation (rG) and their standard errors were calculated based on the procedures described by Mode and Robinson [17]. Response to selection (R) of each trait was calculated according to Falconer [18] as:



R  hyy ,



where hy refers to the square root of heritability and σy to the genotypic standard deviation of the trait under consideration, assuming equal selection intensity. The correlated response (CR) for dry matter yield using a secondary trait was calculated according to Falconer [18] as:



CR  hy rGx ,

where hy denotes the square root of heritability of the secondary traits, rG is the genotypic correlation between secondary traits (e.g. GY, PH, etc.) and DMY, and σx is the genotypic standard deviation of DMY. The relative selection efficiency was calculated as the ratio of CR of secondary traits to direct R of DMY [18]. If this parameter is >1, indirect selection is more efficient than direct selection.

Results A highly differing productivity among location × year combinations (environments) was detected (Table 1). DMY ranged from 11.9 Mg ha−1 in Wulfsode (2011) to 21.3 Mg ha−1 in Groß Lüsewitz (2012) and GY ranged from 6.2 Mg ha−1 in Bornhof (