Seed Yield and Quality of Desmodium uncinatum (Jacq.) DC. as Affected by Method and Time of Harvesting Seed at Wondo-Genet

Abstract

A three year experiment was conducted at Wondo-Genet Agricultural Research Center to determine the appropriate time and efficient method of seed harvesting to produce large quantity of better quality Desmodium uncinatum (Jacq.) DC seed. The experiment was laid out in factorial combinations of Randomized Complete Block Design with four replications. The first harvest was made when the early pods started to disintegrate; and the second, third and fourth harvests were made one week, two weeks and three weeks after the first harvest, respectively. The harvesting methods used include direct mowing of the whole sward by sickle, hand picking of matured seed pods and sweeping of shattered seed pods from the ground. The results revealed that seed yield of D. uncinatum (Jacq.) DC. was significantly (P<0.001) affected by harvesting methods. However, there was no significant (P>0.05) interaction effects between harvesting method and time on seed yield. On average, the highest seed yield (281.9±11.8 kg ha−1) was obtained in the hand picking method, while the lowest (27.2±11.8 kg ha−1) was obtained from the sweeping method. On the other hand, the sweeping method resulted in higher 1000-seed weight (8.2±0.4) than the hand picking method (6.3±0.4) (P<0.01). Harvesting method and time did not have significant effect on germination percentage, number of inflorescences per meter square, number of spikelet per raceme and plant height (P>0.05). Thus it can be concluded that harvesting by hand picking is promising for producing large quantity seed from D. uncinatum (Jacq.) DC. Moreover, D. uncinatum (Jacq.) DC. seeds should be harvested within one week time following the beginning of fragmentation of the early pods.