YIELD COMPONENT PATH-ANALYSIS OF OKANAGAN RIESLING VINES CONVENTIONALLY PRUNED OR SUBJECTED TO SIMULATED MECHANICAL PRUNING

Okanagan Riesling grapevines (Vitis spp.; parentage unknown) were subj ected over five seasons (1984-1988) to three training systems (mid-wir e bilateral cordon; Hudson River umbrella; Lenz Moser) in combination with two pruning strategies 18 nodes per meter of row; simulated mech anical pruning (SMP)!. SMP consistently reduced weight of cane pruning s, cluster weight, berries per cluster, clusters per shoot, berry weig ht, and degrees-Brix and increased shoots per vine, yield, and cluster s per vine. Yields of SMP vines were not significantly higher than tho se of their manually pruned counterparts in the final year of the tria l. Path analysis showed large direct effects of clusters per vine and cluster weight on yield. Negative direct effects on yield came from be rries per cluster and berry weight. Shoots per vine had a large indire ct effect on yield via clusters per vine. Training system had minimal effects on all aspects of vine performance. A lack of a yield increase in the final year of the trial suggests that Okanagan Riesling vines subjected to mechanized pruning may require occasional cordon renewal to regenerate large volumes of fruitful nodes.