Harvested yields of sour cherry (Prunus cerasus) cultivar 'Stevnsbaer' grown in Denmark have been highly variable over the years, yet some propagated derived trees selected from within the two widely grown clones of 'Stevnsbaer' cl. 'Birgitte' and cl. 'Viki' have consistently produced higher yields. This
observed variability within the clones raises the question as to whether more suitable plant material for the future might be found through similar selection.
In this study, a number of the selected individuals were evaluated based on their morphological (harvested yield, number of buds/flowers/fruits, leaf chlorophyll content, cold hardiness and tree structure) and genomic (DNA fingerprinting, nuclear DNA content, DNA methylation) differences to
better understand the differences that produced the improved individuals within the existing clones of sour cherry. In addition, it was investigated whether dormant buds of the clones could be preserved and recovered from cryopreservation.
Variation in yield was mainly found between groupings of the selected individuals from cl. 'Birgitte' and cl. 'Viki'. The higher yield can partly be explained by a higher number of buds/flowers/fruits and higher chlorophyll content in the leaves. Some of the derived selections showed a significant higher ratio of buds and flowers, compared to cl. 'Birgitte' but this increase did not seem to affect the production of cherries. The higher fruit production can also partly be explained by the ability to avoid lethal winter freezing in the buds. Measurements of hardiness in dormant winter buds suggested that
potentially lethal freezing events occur in buds of cl. 'Birgitte' before cl. 'Viki' or any of the derived selections. No genetic differences were detected in the fingerprinting allele profiles or in genome sizes between the clones and their derived selections. It can be concluded that observed variance in the
investigated material is probably not a result of any genetic difference which may affect instability.
Only small differences in DNA cytosine methylation pattern was found in cl. 'Viki' and some of the derived selections compared to cl. 'Birgitte'. However, such epigenetic differences may be enough to change gene expression and result in phenotypic variability.
This study demonstrates that morphological differences exist both between and within the 'Stevnsbaer' clones 'Birgitte' and 'Viki'. However, the genetic and epigenetic studies strongly suggest that little further genetic variation exists within the narrow base spanning the two clones. Consequently, there
would appear to be little value in continuing to pursue an improvement strategy for sour cherry based on non-mutant selection within the ommercial clones of 'Stevnsbaer'. If the potential for future improvement of sour cherries is to be raised to a satisfactory level, it will be necessary to establish an outward-looking breeding program involving the introduction of new ermplasm, with efforts devoted
to hybridization.