1) Faculty of Agriculture, University of Peradeniya, Perandeniya, Sri Lanka
2) International Rice Research Institute, PO Box 933, Manila Philippines
3) Department of Genetics, University of Birmingham, Birmingham, B15 2TT U.K.
The genetic designs, Triple Test Cross (Kearsey and Jinks 1968; Mather and
Jinks 1982) and Basic generations (Mather and Jinks 1982) were used to identify
the cause of heterosis in six yield characters of three crosses of rice.
Two situations can arise when heterosis is measured as the superiority of the F\1\ over its better parent: positive and negative heterosis (Jinks 1983).
Results of tests for heterosis are given in Table 1. Only the characters that showed significant heterosis are indicated. Analysis of Basic generations revealed positive heterosis for three characters. The cause of beterosis was studied by using the model-fitting technique of obtaining genetic parameters that cause the differences in means of the Basic generations. In number of panicles per plant, overdominance seemed to be the cause since [h] was greater than [d]. Tests for linkage using Triple Test Cross progeny were not significant for this character. For days to heading and 1000 grain weight, however, the value of [h] was large but negative, indicating that dominance is for early heading and light grains. But F\1\ showed heterosis because [l] was positive and large. A negative [h] and positive [l] indicate duplicate type of epistasis. In addition, tests for linkage were highly significant. In the presence of interactions, contribution of the interacting genes to the means of segregating generations depends on the magnitude of the linkages. Therefore, heterosis in these two characters appear to be due to interacting genes in linkage disdquilibrium. In these circumstances it should be possible to extract recombinant inbred lines performing better than the F\1\.
Table 1. Mean performance of parents and F\1\, heterosis, genetic components
and test for linkage of three crosses
=============================================================================
Estimate Days to heading 1000 grain No. of panicles
weight per plant
=============================================================================
Cross Bg380-2/82-662 Bg380-2/Bg34-6 Ob2029/Bg34-6
Mean Performance
P\1\ 100.8+/-0.84 31.1+/-0.16 25.8+/-1.26
P\2\ 99.6+/-2.05 30.7+/-0.43 18.2+/-1.31
F\1\ 105.7+/-0.38 33.8+/-0.15 28.6+/-0.25
Heterosis Positive Positive Positive
*** *** ***
Genetic Components
^
m 119.4 40.1 21.3
^
[d] 1.9 - 3.6
^
[h] -58.3 -37.1 7.0
^
[i] -19.7 -9.1 -
^
[j] - - -
^
[l] 44.6 30.7 -
== == == ==
Linkage L3-F2*** L1-B1*** n.s.
=============================================================================
*** = Significant at 0.1% level.
References
Jinks, J.L., 1983. Biometrical Genetics of Heterosis. In: Heterosis. Ed. R. Frankel, Monograph of Theoretical and Applied Genetics, Springer Verlag.
Kearsey, M.J. and J.L. Jinks, 1968. A general method of detecting additive, dominance and epistatic variation for metrical traits. I. Theory. Heredity, 23: 403-409.
Mather, K. and J.L. Jinks, 1982. Biometrical Genetics, 3rd Edition, London: Chapman and Hall.