Proceedings of the XLV Italian Society of Agricultural Genetics - SIGA Annual Congress

Salsomaggiore Terme, Italy - 26/29 September, 2001

ISBN 88-900622-1-5

Poster Abstract

SELECTION OF GERMPLASM WITH GOOD COMBINING ABILITY IN SUNFLOWER (HELIANTHUS ANNUS L.)

LAURETI D., DEL GATTO A.

Istituto Sperimentale per le Colture Industriali,Via 5 Torri, 26, 60027, Osimo-Italy

Tel, Fax + 0717132345

isciosim@libero.it

Maintaining a germplasm bank for and using the genetic variability useful in breeding is costly and impossible for a small research institution. To by-pass this, it may be possible to work with populations from commercial selfed hybrids. F₂ populations may have the advantage of being the accumulation of selection efforts made by breeders and hence require less work than needed in using ancestral or wild races. Each year by seed companies release many hybrids, but their direct agronomic value is not enough to allow them to be used as germplasm source. Testing their suitability, for use in breeding programs, could be done by evaluating their agronomic performance in combination with an appropriate tester wich may have a wide or narrow genetic base. The former would be preferable, giving information useful for the whole breeding program, but the results would not be immediately usable for the breeder who consequently prefers to use a narrow genetic base tester. From a hybrid made with genetic cytoplasmic male sterility, the most immediate breeding lines that can be extracted are restorers; therefore the suitable narrow testers are cms lines.

To evaluate if the direct agronomic value of a hybrid could be taken into account (as a meter) for its suitability to give good combinations in a breeding program, many F2 populations, coming from hybrids of divergent agronomic value, were test crossed in 1998 with a cms line with good combining ability. The test cross progenies were grown in 1999 in Osimo in a field trial under irrigated and rainfed conditions with two replications. Progenies coming from good hybrids gave better test cross values, whereas those of the bud always had poor mean values. The progenies of a very good hybrid showed very poor test cross values with reduced variability among them. Poor hybrids always had high variance among test crosses, whereas the best could have high or low mean squares. As a consequence the bad performance of a hybrid was enough to discard it, whereas for the good performing hybrid it is necessary to check their suitability for use as a germplasm source.

The possibility of combining well with the tester depends on the degree of heterosis of the hybrid: the smaller it is, the higher is the possibility of having good test cross values and hence being suitable as a source of breeding lines. The hybrid that gave bad test crosses could therefore have had high heterosis and require a lot of work to find a suitable combiner to be used in breeding.