With the advent of new rotary-head mechanical harvesters, it is now possible to machine-harvest a wider raspberry canopy. In Spring 1996, a trial was established in a grower's field in Lynden, Wash., comparing raspberries trained to two top wires with a 70-cm spread (split trellis) to the conventional single top wire system. Within the split trellis, increases in leaf number per cane (97%), and leaf area per cane (55%) were not reflected in a concomitant increase in total leaf dry weight per cane (35%). Leaf dry weight per fruit weight was 11% less within the split trellis. These data suggest that the canopy is more efficient with this type of trellis. Increases in estimated yield per cane (49%) and projected yield per acre (50%) associated with the split trellis were due to increases in berry number per cane (47%). Fruit number per meter of lateral was 35% greater within the split trellis. Greatest enhancements to yield components were in the upper parts of the canopy where canes were tied over. Since there were no differences in lateral numbers or lateral lengths between the two systems, this increased productivity was due to increased floral expression, enhanced fruit set, increases through Spring bud initiation or any combination thereof. In both trellis systems, the longest laterals occurred on the middle third of the cane and decreased in length progressively towards the tip of the cane. Primocane lengths were shorter (20%) and diameters were smaller (10%) and more uniform in the split trellis system.