| To understand genetic mechanism and develop fast testing methods of grain quality are important for maize quality improvement. The objectives of this study were (1) to investigate the feasibility and methods of measuring quality of intact maize kernel with near infrared reflectance spectroscopy (NIRS) by using bulk-kernel samples of maize inbred lines and hybrids. (2)to analyze the genetic effects and GE interaction of grain quality for 10 inbred lines and their 90 crosses of FI, F2 generation in a 10 X 10 complete dial lei mating design planted at two sites. A genetic model proposed for analysis of quantitative traits by Zhu Jun was used (3) to investigate the relationship of parents with their offspring, and behavior of quality traits in F2. (4) to determine quality of 89 normal inbred lines widely used in China maize production by chemical methods. The results were as follows:1. Different spectral data preprocessing methods and spectral regions have influences on the performance of model established. Calibration equations for quantitatively determination of crude protein(CP), crude starch(CS), and crude fat(CF) content were developed by partial least square regression (PLS) of chemical values on NIRS data and tested through both cross and external validation. In the external validation, coefficients of determination (R2vai) between NIRS and chemical values ranged from 0.87 for CS to 0.96 for CP, while the root mean square error of prediction (RMSEP) from 0.31 for protein to 0.88 for starch. In addition, 40 progenies of FI, F? not included in calibration and validation sets were verified to further evaluate their performance. Coefficients of correlation(R) of 0.98,0.93 and 0.98 between NIRS predicted and actual protein, starch and oil content were obtained in these materials, respectively. However, the greatest relative errors were 2.7% (protein), 2.46% (starch) and 7% (oil). Thus, the accuracy of prediction could be comparable to chemical methods.2. It was indicated by analysis of variance that genetic effects of seed, maternal and cytoplasm as well as GE interaction simultaneously controlled grain quality in different ways.Seed and maternal effects, which accounted for 52.49%, 46.96 of the total genetic variance, play a predominant role in protein inheritance, while cytoplasmic effect was relatively small. Further analysis indicated that dominant gene effects were the major contributors of genetic variation of protein content. Significant maternal negative heterosis was observed. There existed significant genotype x environment interaction for direct and maternal effects. The interaction heritability was larger than general heritability.For starch content, the maternal genetic variance(65.48%)was larger than seed (27.61%) and cytoplasm variance(6.91%), indicating that starch content was mainly under the genetic control of maternal plant. Additive gene effect was found further to be more important. There existed significant interaction of maternal, seed effects by environments. Significantly positive maternal heterosis was found. The general heritability was larger than interaction heritability.For oil content, both maternal (44.05%) and seed direct genetic variance (40.87%) appear equally important. Cytoplasm effect, which accounted for 15.08% of total genetic variance, could not be neglected. In the main genetic effects, additive effects (including maternal and direct additive) were more important than direct effect. No significant positive or negative heterosis was found. Interaction of maternal and seed effects by environment was significant. Its general heritability was larger than interaction heritability.3. The performance of quality traits in F2 generation is complicated and showed much difference between traits, crosses and locations (environments). In F2 generation, the heterosis of protein content over parental means, high parent and low parent were all negative, this indicates protein content of F2 was close to low parent and showed negative heterosis. Starch content con...
|
PDF Full Text Request