Combining Ability in Summer Squash (Cucurbita pepo L.)

Cucurbit Genetics Cooperative Report 21:40-42 (article14) 1998

F. Lopez Anido¹, I.T. Firpo², S.M. Garcia², and E. Cointry¹
Catedras de Genetica¹ y Horticultura², Fac. de Cs. Agrarias, Universidad Nacional de Rosario, CC 14, Zavalla 2123, Argentina

Introduction. Diallel analyses ave been used extensively in plant species for genetic interpretation, which requires certain assumptions, and in a more practical way in the estimation of general combining ability (GCA) and specific combining ability (SCA), which are useful in devising breeding strategies (1, 6). In the cucurbitaceae GCA and SCA estimates have been reported mainly in Cucumis sativus L. and Cucumis melo L. (2,3,5) among others. In Cucurbia pepo L., which has been adapted also to hybrid production (9), information regarding combining abilities is far from complete. The objective of this report is to evaluate GCA and SCA in lines of summer squash, and to use this information when planning breeding programs.

The experimental material consisted of ten selectd inbred lines (UR 4, 7, 18, 19, 30, 33, 34, 38, 52 y 53) derived from the highly variable summer squash popuolation ‘Tupungato Magnif INTA’ (caserta type) and the diallel crosses includeing reciprocxals. The lines were selected at an early stage for yield (one cucle top-cross) and light colored fruits. The experiment was established on 18 October 1996 at the Experimental Field of the Fac, de Cs AGrarias, UNR, located at Zavalla (33 ˚ 01’S;60 ˚ 53′ W), Argentina, in a randomized complete block design with three replications of eight plants, in a plant spacing of 1.4m between rows and 0.80 m within plants in the row. The following variables were evaluated on a plot basis: Total fruit number (TFN) (12 harvests); precocious fruit numbner (PFN) (first three harvests), days from sowing to first harvest (DFH), number (LN) and diameter (LD) (cm) of expanded leaves and height of plants (PH) (cm). These last three vegetative characters were evaluated in three plants 30 days after sowing. The analyses were performed considering Griffing’s model, method 1, model 1(4) using the DIALLEL computer program (6). The weight of 100 seeds was used as a covariable. the relative importance of CGA and SCA was evaluated following (1). In order to meet normal distribution of residuals TFN, PFN and DFH were transformed by square root and LN by loge.

The GCA mean square was highly significant (p<0.01) for all variables except plant height, for which it was significant (p<0.05). The SCA and reciprocal mean squares were highly significant for all variables except for days to first harvest for which they were significant (Table 1). The ratio 2 φ g /(2 φ g+ φ s) which is a measured of the relative importance of GCA (values close to unity) or SCA (values close to zero) was 0.45, 0.36, 0.50, 0.17, 0.37 and 0.19 for TFN, PFN, DFH, LN LD and PH respectively. Non-additive gene actions were of major relative importance in the vegetative characters and precocious production. For total fruit number and days to first harvest additive and non-additive genetic variation accounted for nearly the same amount of the total genetic variation. In other cucurbits the GCA seems to play a major role (2,3,5). The best hybrids cannot be predicted simply on the basis of GCA of the parents alone. Breeding strategies should consider both selection per se of the lines (maybe in an early stage) and the testing of a wide range of hybrid combinations.

Differences between reciprocals were attributed in early reports (7,8) to maternal seed size. In our study some other maternal and/or non-maternal (specific reciprocal effect would be responsible for such differences since the covariable weight of 100 seeds had failed to explain any significant variation of traits. The ratio φ r / 2 φ g + φ s) which is a measure of the relative importance of the reciproval quadratic component in relation to genetic variation was 0.28, 0.23, 0.31, 0.30, 0.38, and 0.35 for TFN, PFN, DFH, LN, LD and PH respectively. With the same resources, not considering reciprocals, the number of lines to evaluate in diallel crosses can be increased approximately 37-41%. It seems reasonable to increase the chance of better combinations, leaving the evaluation of reciprocals to a later stage of the breeding program, only in the outstanding experimental hybrids. Finally the qudratic components estimated in this report may be biased upward due to genetic x environment interaction. In this species the importance of this interaction is yet to be proven.

Table 1.1 ANOVA Griffing’s analysis for total fruit number (TFN), precocious fruit number (PFN), days to first harvest (DFH) in a 10 lines complete diallel crosses.

		Total fruit number		Precocious fruit number		Days to first harvest
Source of variation	df	M.S.	Sig	M.S.	Sig	M.S.	Sig
Genotypes	99	1.8258	**	1.3511	**	0.04015	**
G.C.A.	9	5.2287	**	3.4905	**	0.10885	**
S.C.A.	45	1.4697	**	1.2052	**	0.03231	*
Reciprocal	45	1.5009	**	1.0691	**	0.03443	*
Covariablez	1	0.4503	ns	0.3417	ns	0.01498	ns
Error	197	0.9561		0.7041		0.02382

Table 1.2 ANOVA Griffing’s analysis for leaf number (LN) diameter (cm) and plant height (PH) (cm) in a 10 lines complete diallel crosses.

		Leaf number		Leaf diameter		Plant height
Source of variation	df	M.S.	Sig	M.S.	Sig	M.S.	Sig
Genotypes	99	0.034836	**	14.6830	**	24.2072	**
G.C.A.	9	0.053583	**	34.5032	**	35.5655	*
S.C.A.	45	0.035160	**	12.2111	**	23.6690	**
Reciprocal	45	0.030767	**	13.1938	**	22.4807	**
Covariablez	1	0.001208	ns	4.4192	ns	20.8769	ns
Error	197	0.018717		7.6084		14.7638

** p<0.01
* p<0.05
^z 100-seed weight

Acknowledgment: This research was supported in part by PID 202 UNR Grant.

Literature Cited

1. Baker, R.J.1978. Issues in diallel analysis. Crop Sci. 18:533-536.
2. El-Shawarf, I.I.S. and L.R. Baker. 1981. Combining ability and genetic variances of GxHF1 hybrids for parthenocarpic yield in gynoecious pickling cucumber for once-over mechanical harvest. J. amer. Soc. Hort. Sci. 106:365-370.
3. Frederick, L.R. and J.E. Staub. 1989. combining ability analyses of fruit yield and quality in near-homozygous lines derived from cucumber. J. Amer. Soc. Hort. Sci. 114:332-338.
4. Griffing, B. 1956. Concept of general and specific combining ability in relation to diallel crossing systems. Aust. J. Biol. Sci. 9:463-493.
5. Lippert, L.F. and P.D. Legg. 1992. Appearance and quality characters in muskmelon fruit evaluated by ten-cultivar diallel cross. J. amer. Soc. Hort. Sci. 97:84-87.
6. Magari, R. and M.S. Kang. 1994. Interactive BASIC program for Griffing’s Diallel analysis. J. Heredity 85:336.
7. Passmore, S.F. 1934. Hybrid vigour in reciprocal crosses in Cucurbita pepo. Ann. Bot. 48:1029-1030.
8. Von Maltzahn, K.E. 1957. A study of size differences in two strains of Cucurbita pepo L. Can. J. Botany 35:809-830.
9. Wehner, T.C. 1997. Heterosis in important US vegetable crops. In: The genetics and Exploitation of Heterosis in Crops. An International symposium, pg. 272-273. CIMMYT, Mexico, D.F., Mexico.