Euphytica 29: 115-120 (Feb 1980)
Breeding research on rose pigments. II. Combining ability analyses of variance of four flavonoids in F1 populations
D. P. de Vries, Frida Garretsen, Lidwien A. M. Dubois, H. A. Keulen

SUMMARY

18 Hybrid Tea-rose varieties were crossed and selfed in an incomplete diallel. In F1 seedlings the relative contents of the flavonoids: pelargonidin, cyanidin, kaempferol and quercetin, were determined by means of paper chromatography. Inheritance of each pigment was mainly controlled by additive gene action. Combining ability analyses of variance of these pigments showed highly significant differences between GCA's of the varieties. The pigment content of parent varieties can be a useful basis for breeding.

Table 2. Differences (x 10) between the yearly means and the overall
means of 4 flavonoids on the basis of 20 HT-rose populations.

Year Pelargonidin Cyanidin Kaempferol Quercetin
1972 +5.8 +4.5 -6.7 -5.7
1973 +2.3 +2.4 +0.5 +0.9
1974 -1.5 -1.9 -0.0 +1.2
1975 -5.0 -3.6 +2.3 +0.9
1976 -1.6 -1.4 +4.0 +2.7

RESULTS

Table 3. Combining ability analyses of variance of four (corrected) flavonoid components (x 10) in HT-roses.

Source Dimension Pelargonidin Cyanidin Kaempferol Quercetin
M.S. F MS. F M.S. F M.S. F
GCA 17 90.0 12.3++ 92.9 l5.2++ 285.8 32.8++ ++ 26.6 8.1++
SCA 77 20.9 2.9++ 17.1 2.8++ 16.4 1.9+ 2.8 0.8
Reciprocals 19 8.3   5.7   11.4   4.2  
Error (1) 17 6.1   6.6   5.7   2.3  
Error (11) 36 7,3   6.1   8.7   3.3  

For each pigment highly significant GCA-mean squares were found (Table 3). Although for pelargonidin, cyanidin and kaempferol also significant SCA-mean squares were obtained, they appeared to be less important than the GCA-mean squares. This demonstrates that the inheritance of each pigment was mainly controlled by additive gene action.

Table 4. GCA's (x 10) for 14 HT-rose varieties for pelargonidin (1), cyanidin (2), kaempferol (3), quercetin (4).

Parents Code GCA's
  (1) (2) (3) (4)
Duke of Windsor 14 3.5 1.9 -2.9 -0.2
Muria 4 3.3** 1.7 1.3 0.6
Numéro Un 18 2.8 1.7 4.8* 2.8*
IVT 48 16 2.7 2.9 -0.5 0.9
Arnica 10 2.5* 1.7 1.2 1.3
Constanze 3 2.4** 3.3** 2.0 1.2
Wizo 13 1.7 -3.3 -2.2 -1.6
Super Star 7 0.9 0.2 0.9 0.7
Ann Cocker 11 0.8 -2.4* -8.3** -1.0
Korp 2 0.5 3.8** -0.9 -0.3
Interview 17 0.1 -1.4 3.9** 0.2
Sofia 8 -1.1 -0.6 -4.9** -2.1**
Fresco 1 -1.6 -2.3* 5.7* 0.5
Zorina 12 -2.5* -4.7* -1.7 1.5*
Polynesian Sunset 6 -3.0** 2.0 5.9** 1.2
Königin der Rosen 9 -3.1** -1.4 -10.0** -3.2**
Blessings 15 -3.8** -1.8 2.3 -0.1
Hawaii 5 -5.9** -1.6 3.3** 0.5
Overall mean:   31.9 17.0 41.7 16.3
S.E. (GCA)   1.3–0.6 1.2–0.5 1.4–0.6 0.9–0.4

In Table 4 the estimated GCA's of the individual parents are presented in the order of decreasing GCA of pelargonidin. Each GCA was tested against its standard error (SE) to show positive or negative deviations from zero.

'Muria', 'Amica' and 'Constanze' appeared to be parents with a significant positive effect on the pelargonidin content in their offspring, whereas 'Hawaii', 'Blessings', 'Konigin der Rosen' and 'Zorina' affected pelargonidin content negatively. However, 'Constanze', like 'Korp', also influenced cyanidin positively, while 'Zorina', 'Ann Cocker' and 'Fresco' exerted a negative effect.

For kaempferol 'Polynesian Sunset', 'Fresco', 'Numéro Un', 'Interview' and 'Hawaii' were positive parents, 'Konigin der Rosen', 'Ann Cocker' and 'Sonia' negative ones.

As regards quercetin, 'Numéro Un' was a positive parent, while 'Sonia' and 'Konigin der Rosen' were negative ones.

DISCUSSIONS AND CONCLUSIONS

Our results showed that with regard to the flavonoids: pelargonidin, cyanidin, kaempferol and quercetin, mainly additive genetic variation was present in 18 selected Hybrid Tea-rose varieties. Although these varieties are a small sample, there is no reason to assume that in other Hybrid Tea-roses the same pigments are inherited in a different way. Consequently, a straightforward breeding strategy for pigment content and hence for flower colour can be followed. This means that breeding with F1's containing less cyanidin and more pelargonidin than their parents, can positively contribute to our breeding aim, i.e. genotypes with pelargonidin as the only red pigment. However, with regard to the quantitative inheritance of these anthocyanidins in a crop that, moreover, is tetraploid the realization of such genotypes may takes several generations of breeding.

It is worth mentioning that the combining ability analyses of variance of the pigments, which were made without the populations arisen from selling, gave no other results than when the selfings were included. This indicates that, with regard to the inheritance of the flavonoids, the varieties behaved similarly in self breeding and in crossbreeding.