-
Heterosis, genetic and correlation analysis of main agronomic traits in F1 population derived from crossing between Miscanthus sacchariflorus and M. lutarioriparius
- AI Xin, JIANG Jian-Xiong, CHEN Zhi-Yong, QIN Jing-Ping, YI Zi-Li
-
2017, 26(2):
110-122.
DOI: 10.11686/cyxb2016370
-
Asbtract
(
345 )
HTML
(
1)
PDF (1145KB)
(
750
)
-
References |
Related Articles |
Metrics
A F1 population with 232 individuals was developed from a cross between Miscanthus sacchariflorus and M. lutarioriparius, two Miscanthus species significantly different in the main agronomic traits. A total of 12 traits including the number of nod (Non) , the length of leaf (Lol), the width of leaf (Wol), the length of main stem (Lms), the length of flower stalk (Lfs), the length of inflorescence (Loi), the plant height (Ph), the basal stem diameter (Sd), the number of tillers (Not), the average dry weight of single tiller (Wsl), the average dry weight of the biggest single tiller (Wbt), the weight of individual plant (Wop) were determined, and then subjected to genetic and correlation analysis using a mixed major gene plus polygene inheritance model. The results showed that each of trait produced a continuous, single peak and skewed distribution pattern, suggesting that they are quantitative traits. Except for the stem diameter (Sd), the other 11 traits exhibited a mid-parent heterosis, especially in which the Wop, Wsl, Loi, Not, Lms, Ph and Wbt even exhibited a heterobeltiosis, which indicates that the hybrid heterosis would be a major breeding method for Miscanthus. In addition, the mixed factors genetic analysis indicated that the performance of Lfs and Wbt might be controlled by a pair of major genes, and those of Wol, Lms, Ph, Not, Wsl and Wop by 2 pairs of major genes. The orders of the heritability of major genes for each trait are as follows: Wol (87.76%)>Wsl (81.48%)>Wop (65.12%)>Not (59.20%)>Lms (49.87%)>Ph (48.01%)>Lfs (47.75%) >Wbt (37.19%). Some yield-related traits comprising ‘Wop’, ‘Wsl’ and ‘Not’ are suitable to be selected in early generation because they are with a higher major gene effect. Correlation analysis among traits revealed that Wop is positively correlated with Non, Lms, Ph, Sd, Not, Wsl, and Wbt. Of those, Not, Wsl, Lms and Ph gave the highest correlation with Wop, and thus these traits could be selected indirectly according to their related trait performance in breeding practice.