Genetic variation of Norway spruce (Picea abies [L.]Karst.) populations in Austria. I. Digenic disequilibrium and microspatial patterns derived from allozymes

1998

S.221-230

2 Abb., 3 Tab., zahlr. Lit.

Unselbständiges Werk

200055355

Forest Genetics 1998 : International Journal of Forest Genetics ; 5(4), S.221-230
; 5(4)

Allozyme loci were used to study digenic disequlibrium and the spatial pattern within three putatively autochthonous, naturally regenerated, high elevational Picea abies stands in Austria. In aLL populations more cases of digenic disequilibria were found than were to be expected stochastically alone. For all populations together, 24 two-locus combinations resulted in significant deviations from expected random multilocus structure. The measure of disequlilbrium, common normalized correlation, was significant for loci Fest2 - Mdh2 in all populations and for Aat3 - 6Pgd3 and Pepca - Mdh3 significant disequilibria were found in two out of three populations. In all other cases disequilibria were population specific. Non-random mating may be the most likely explanation of disequilibrium for unlinked or weakly linked loci. Fourteen two-locus combinations, for which no data on linkage are currently available, were in disequilibria. For this group it remains open whether linkage has been the major deterministic force. In most cases a random distribution of genotypic scores was found within the three populations. Spatial sutocorrelation parameters were not significant for nearest neighbors based on the Gabriel-connected net. Spatial correlograms up to a distance of 100 m were computed. In total, 45 out of 700 (6.4%) standard normal variates (SND) were found to be significant. Extensive pollen and seed dispersal may have accounted for the randomness of genes in space. In single loci, such as Aat3-2 in POP-2, MORAN's index (I) was significant and positive in short distance classes and decreased to vallues close to zero in medium and long distances. On the other hand, significant negative MORAN's I values were also found in short distance classes for some loci (e.g., 6-Pgd2-2 in POP-3). While variation in the breeding system (e.g., nonrandom mating, limited seed and pollen dispersal) cannot be ruled out with certainty to have acted differently at the sites, lack of consistency in significant spatial patterns over the three populations maybe better explained by microselection or rare significant values have resulted by chance alone.