Pods of oilseed rape (Brassica napus) stronger than those in current cultivars are needed to reduce the considerable loss of seed caused when the fragile, mature pods open prior to or during harvest. But threshing of pods by a combine harvester must be complete and must cause minimum damage to the seed thus released. This study addressed the question of where the limits to threshability lay. Using a small but realistic threshing apparatus, the e!ect of threshing rotor speed and concave clearance on the degree of threshing of pods and the resulting seed damage were measured for small samples of pods; two commercial cultivars, a resynthesized line of B. napus with stronger pods, and a sample of pods harvested early so as to be stronger than when allowed to reach maturity.
Threshing of 98% of the pods of the two commercial cultivars was achieved at a threshing rotor speed of 600 min, at which speed 1)0% of the seed was damaged. The synthetic line required 800 min\ for threshing at which speed it su!ered 2)2 percentage points more damage than at the lowest speed. The immature pods were not completely threshed even at 1520 min, at which 14% seed damage was recorded. These results show that pods as strong at maturity as the immature sample used in this study would be likely to result in an unacceptable level of seed damage due to the severity of the threshing action needed. The response of each cultivar to threshing rotor speed was very well correlated (variance accounted for '99%) with half-life of pods in an established &random impact' test, the half-life being the treatment time needed to open 50% of the pods. Using the relationships developed, an upper limit half-life of 57 s (standard error 2)95 s) was determined which de"nes a target maximum pod performance in the random impact test for which breeding lines of oilseed rape can readily be screened.