论文部分内容阅读
多因素多水平回归设计的田间试验,固然具有处理少,信息量丰富,和可不设重复的优点。但得到非典型回归方程有相当的比例,结果事倍功半。本试验采用回归设计设置重复结合拉丁方排列技巧,通过对建立的数学模型进行择优,可有效地提高试验的成功率。1材料与方法 试验于1997年晚季在溪柄农科所农场进行,前作早稻。供试土壤肥力中等,供试品种“汕优63”。
The multi-factor and multi-level regression design of field experiments, although with less processing, informative, and can be set without duplication of advantages. But there is a considerable proportion of atypical regression equation, the result is less effective. In this study, the regression design was used to set the repetitive combination of Latin square arrangement skills. By choosing the optimal mathematical model, the success rate of the experiment can be effectively improved. 1 Materials and Methods The experiment was carried out on the farmer’s farm in Sik Sang Agricultural Institute in the late 1997 season, with early rice being used. Medium soil fertility test, test varieties “Shanyou 63”.