Variability and correlation analysis between area of single seedling occupancy (ASSO) and rice individual characters under non-uniformity spatial distribution

摘要

In order to study the individual characters under non-uniformity spatial distribution, the variation characteristics of the area of single seedling occupancy (ASSO) were used as a control group to study morphological variability of rice. Different mathematical models were applied to combine the yield components of rice and some of its morphological index. Manfred Hühn's mathematical statistical model was used to analyze the possibility of increasing yield by non-regular spatial distribution. The plot models were constructed by inlaying individual unit area designed by computer simulation methods. A different ASSO was designed under non-regular horizontal distribution, Thiessen polygons were used to determine the ASSO. The result showed that, coefficients of variation of yield components were all less than those of ASSO, the effects of ASSO variability on effective panicles per plant and filled grain number of per panicle were the greatest; except flag leaf, the variation coefficient of most of the individual plant characters were greater than that of ASSO, and the variation coefficient of empty-shrivel grain number (0.3637) was the greatest. Most of individual plant characters suited the power function model, the setting percentage and filled grain number of per panicle kept steady after a quick increase as ASSO was increased, the spikelet number of per panicle peaked slowly and decreased slowly. The spikelet number of individual plants had linear relationships with individual bio-mass of single plant and dry matter of stem leaf in different degrees. The economic coefficient had linear relationships with setting percentage, had non-linear relationships with ASSO, yield of unit area, grain number of per panicle, filled grain number of per panicle, and average length of flag leaf. The effects of ASSO variability on 1000-grain weight and average length of flag leaf were small. It was possible to increase rice grain yield under non-regular spatial distribution which was proven by the mathematic statistical model with the condition of the ratio between the variation coefficient of ASSO and yield was greater than the correlation coefficient of the two (vA/vS> rSA).