Skip to main content
ARS Home » Research » Publications at this Location » Publication #75437


item McClung, Anna
item McClung, Anna
item VAWTER, J

Submitted to: Rice Technical Working Group Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 2/1/1996
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: The traditional method for harvesting rice experimental yield plots has been to hand-cut the 2 center rows of a 6 row plot that is generally 6 m long. With the availability of small plot combines which cut 4 or 6 rows, we were interested in comparing hand versus combine harvest methods for precision in yield estimates. In one study, we determined the contribution of each row of the 6 rows to the yield of the whole plot. It was observed that the outer 2 rows were an over estimate of the yield while the next 2 interior rows were an under estimate. Harvesting the center 2 rows was a good predictor of yield of the whole plot but a lower error (standard deviation) was observed when all 6 rows were harvested. There was a high correlation between yield of the 6 row plot and yield using fewer rows, but some varieties were ranked differently for yield depending on how they were harvested. In a another study, yield estimates on an area basis were observed to be the same regardless of the length of plot used. Thus, it wa not necessary to increase the length of our plots to increase precision. We also evaluated what level of plot to plot seed mixture might occur by using a small plot combine. After harvesting plots of a gold hull variety in tandem with a straw hulled variety, we evaluated the percent of seed mixture. It was found that one mixed seed occurred in about 1800-1900 kernels. This level of purity was achieved by stopping the combine after harvesting each plot, allowing it to run for 2-3 minutes, and cleaning the header and sickel bar area of remaining seed with compressed air. The small plot combine harvest method is very efficient and offers a great savings in labor. Data from these studies show that using this equipment will not compromise the precision of yield estimates in research plots.