Ruled from above, grafting success rates are more controlled by the scion than the rootstock

Abstract

Grafting is essential for the cultivation of many horticultural crops. Consequently, nurseries play a crucial role in assuring horticultural crop multiplication. However, grafting success can vary considerably from year-to-year and between different scion/rootstock combinations. The objective of this study was to determine the effect of rootstock geographical origin, the rootstock and scion genotype, and their interaction on grafting success rates after one season of growth in a field nursery.

We used a dataset of grafting success rates obtained over a 10 year period in IFV nursery, a large grapevine nursery in France. This dataset included 17 rootstocks, 36 rootstock clones, 276 scions and 648 scion/rootstock combinations. A generalized linear mixed model was applied to quantify the effects of each factor, despite the over-representation of the scions relative to the rootstocks.

Rootstocks, scions, and their interactions had a significant impact on grafting success rate. The pseudo-R² of the model with the rootstock alone was 5.6%. Adding the scion increased the pseudo-R² to 26.2%. Including the interaction between rootstock and scion further raised the pseudo-R² to 31.8%. This indicates that the scion has a greater influence on grafting success than the rootstock. For the rootstocks SO 4, 1103 P, 140 Ru, and 3309 C, clone-level differences in grafting success were statistically significant, though they accounted for only 2-7% of the variance. Year and rootstock origin also significantly influenced grafting success.

This is the first analysis of graft success rates based on such a large dataset for any species. These results suggest that the scion has a greater role in determining grafting success rates than the rootstock in grapevine. Consequently, all new scion varieties must be evaluated with a set of rootstocks to be sure that grafting success rates will not be a limiting factor of their deployment.