本試驗於1998年二期作及1999年一期作,於本分所溪口試驗農場,以水稻高級世代產量比較試驗的參試品系為材料,共調查134個小區樣品,每小區97株。每小區稻穀以手割方式收穫及脫粒後,先調查穀物收穫量測定器之測定值(Y)。而後這些稻穀經乾燥及風選後調查其重量、含水率及碾糙率,以計算各小區之稻穀產量(g/97株)(X3)及單位面積糙米產量(kg/10公畝)(X6)。將此134組數據之3/5劃分作為模式開發數據,用來估算模式之參數;另2/5數據則作為模式驗證之用。由模式開發數據所估算的參數估值經反複估算,結果顯示參數估值之穩定性甚高。繼而以模式驗證數據測試其估計能力,結果顯示開發出令人滿意之小區產量及單位面積糙米產量的估測模式: X3=(Y+54.5)/0.1973(小區產量) X6=(Y+52.3)/1.0586(糙米產量) 試驗結果顯示利用穀物收獲量測定器可作為水稻雜交後代產量比較時之應用,以減輕育種工作之負擔,增加水稻育種工作之效率。
This research was conducted from the second crop season of 1998 to the first crop season of 1999, a total of 134 samples for advanced yield trial were concluded. First, we evaluated the cropton value(Y) of the 134 advanced lines. Then, samples were investigated of their weight, moisture content and milling ratio of brown rice for estimating the plot-yield (g/97 plants)(X3) and milling yield of brown rice (kg/10 acre)(X6). The whole data was split into two portions. One of which contained 3/5 of the 134 samples (as the estimation data) was used to estimate the model parameters. The other (as the validation data) was reserved to check the prediction accuracy of the developed model. The sampling variation of the estimated parameter was examined empirically by a first-order jackknifing from the estimation data. It shows that the estimates of the parameters of a prediction model were quite stable. This model was then tested and updated by the validation data, with a satisfactory result for each yield by cropton value: X3=(Y+54.5)/0.1973 (for plot-yield) X6=(Y+52.3)/1.0586 (for milling yield of brown rice) Results show that the cropton value can be applied for rice yield trial to promote the efficiency of rice breeding procedure.