Estimation of general and specific combining ability variances and their implications on hybrid rye breeding.

P. Wilde

wilde@lochow-petkus.de
Lochow-Petkus GmbH, D-29296 Bergen (Germany)

J. Menzel

Lochow-Petkus GmbH, D-29296 Bergen (Germany)

B. Schmiedchen

Lochow-Petkus GmbH, 29303 Bergen (Germany)

Abstrakt

Effects and variances of general (GCA) and specific (SCA) combining ability have been estimated from two experiments. For the first experiment (EXP 1) 19 homozygous lines originating from the Petkus pool were crossed with two CMS single cross testers derived from the Petkus pool (intra-pool test crosses) and with two testers from the Carsten-Danko pool (inter-pool test crosses), respectively. In the year 2000 grain yield was measured at 5 locations in Germany. The second experiment (EXP II) comprised a factorial of inter-pool crosses by using 4 homozygous CMS Petkus pool lines as seed parents and 6 Carsten-Danko lines as pol­len parents. The respective 24 inter-pool crosses were evaluated for grain yield on several locations in 1996 and 1998.In EXP I GCA-variance of candidate lines (4.7 —5.9 q/ha² was estimated to have a similar size as SCA-variance. Possible reasons for an overestimation of SCA-variance are extreme divergence between testers, insufficient removal of inbreeding effects in the intra-pool test crosses and bias due to SCA × year interaction-variance. The use of intra-pool testers should be restricted to breeding material where inbreeding due to consanguinity between testers and can­didate lines is either absent or will evenly affect test cross performance of all candidates.In EXP II GCA-variances for female 9.2 q/ha² and male 2.4 q/ha² lines were much higher than SCA-variance. Repeatability for GCA-effects over different environments was medium (r=0.67), whereas for SCA-effects it was low (r=0.13). Selection intensity for yield should there­fore be moderate, if based on a single year of testing only. Selection of parental components for hybrids and prediction of hybrid performance should mainly base on GCA-effects. Alterna­tive to the use of factorial crosses test crosses with testers representing the opposite pool are recommended. Testing for SCA-effects and for SCA x environment interaction- effects can be a valuable diagnostic tool, to become aware of strengths and weaknesses of the elite material used to build new hybrids. SCA-tests should therefore be performed by using highly divergent environments.

Słowa kluczowe:

CMS cross testers, general combining ability (GCA), homozygous lines of rye, hybridrye breeding, specific combining ability (SCA)