查詢結果分析
相關文獻
- Soybean Yields Response to Planting Dates and Climatic Factors
- 灌溉水中不同鹽分濃度對大豆生育及產量的影響(1)--滲漏與不滲漏處理下灌溉頻度及鹽液濃度對大豆農藝性狀及產量之影響
- 浸水及天然災害對大豆生長與產量的影響
- 不同播種期對有限型及無限型大豆生育與產量的影響
- Effects of Corn Residues on Soybean Yield
- Growth and Yield Responses of Soybean to Direct Incorporation of Corn Residues
- Effects of Corn Residues on the Subsequent Soybean Crops
- 大豆子粒充實期及充實速率對產量關係之研究
- 水稻期作與地區間產量及產量構成要素之品種間差異
- 大豆生殖生長期葉片光合作用能力與產量關係之研究
頁籤選單縮合
| 題 名 | 大豆產量構成因素的分析與檢討=A Correlation and Path Analysis of Yield Components of Soybean |
|---|---|
| 作 者 | 戴喬治; | 書刊名 | 中華農學會報 |
| 卷 期 | 46 民53.06 |
| 頁 次 | 頁9-18 |
| 關鍵詞 | 大豆; 產量; |
| 語 文 | 中文(Chinese) |
| 英文摘要 | 20 F8 lines of a soybean cross, Taichung Green ×Miyo, in 10 plant plots replicated 3 times were used to study the component structure of seed yield. Data of seed yield and six yield component characters, i.e. unit seed weight, no. of seeds per pod, no. of nodes on main stem, no. of branches, no. of nodes per branch and no. of pods per node, were analyzed to determine their heritability values, phenotypic and genotypic correlations and relative importance of six yield component characters to seed yield. Heritability estimates of seed yield and six component characters are presented in Table 1. High values were obtained by the method of variance components excepting that of number of seeds per pod. Since the effects of years, locations and non-additive genetic interactions cannot be removed from the variance due to lines, these estimates were quiet uncredible. Another estimation by F2-F3 regression in standard unit (WF2F3/√VF2VF3) showed that heritability values of seed yield, unit seed weight, number of seeds per pod, number of nodes on main stem and number of branches were in general not high. Unit seed weight and number of nodes on main stem had moderate degrees of heritability (56% and 43% respectively) and seed yield was lowest one (27%). Phenotypic and genotypic correlation coefficients between all possible pairs of seven traits are given in Table 2. In general, genotypic coefficients were higher than their respective phenotypic values. This is in agreement to the reports of JOHNSON et al (1955), ANAND and TORRIE (1963) and GOTOH (1963). It can be observed from Table 2 an obvious fact that six component characters can be divided into two groups by their relationships one another, i.e. (1) Unit seed weight, number of nodes on main stem and number of nodes per branch; and (2) Number of branches, number of seeds per pod and number of pods per node. Positive correlations existed between each pair of characters within a group negative correlations between characters of different groups, which had only one negligible exception (r=0.1194 between number of branches and number of nodes on main stem). The close associations between seed yield and both number of nodes per branch (r=0.7459) and number of nodes on main stem (r=0.5793) and between the later two traits (r=0.8053) indicate a fact that in this hybrid population more nodes of a plant tended to have higher seed yield. Significant negative correlations were observed between number of pods per node and each of the following characters: unit seed weight, number of nodes on main stem and number of nodes per brach. The negative association between seed yield and number of pods per node, however, was low (r=-0.1430). JOHNSON et al (1955), GOTOH(1963), and ANAND and TORRIE (1963) had reported phenotypic and genotypic correlations of several soybean crosses between seed yield, number of seeds per pod, number of pods per plant and unit seed weight. Their data are cited in Table 3 for the purpose of comparison with the result of this study. Seed weight was closely associated genotypically with high seed yield in the results of JOHNSON et al (1955) and GOTOH(1963). ANAND and TORRIE (1963), on the contrary, reported that number of pods per plant and number of seeds per pod were more associated phenotypically with high seed yield. This investigation revealed that high seed yield was significantly correlated with number of pods per plant. Negative phenotypic correlation between seed weight and both number of seeds per pod and number of pods per plant was in agreement among the different studies. The relation between number of seeds per pod and number of pods per plant was inconsistent as shown in Table 3. Since correlation coefficients discussed above simply measured relationships among characters, an information about the relative importance of different components to seed yield could not be given. The method of path coefficient analysis, however, provides an effictive tool for arriving this purpose. Under the condition of contructing a path diagram based upon a prior knowledge, the direct and indirect effects of each of the component traits on seed yield can be specified and compaired. The path diagram used in this study is shown in Figure 1. Path coefficient estimate of six components and phenotypic correlation coefficients between them are registered on their proper paths in the diagram. The residual variable X in the diagram is assumed to be an independent factor. By the use of these path and correlation coefficients into normal equations (I), each correlation coefficient between seed yield and one of the six components (rij, j=2,3...7) could be partitioned into six parts in which one was the direct effect from the component (j) to seed yield (l)(i.e. Plj) and others were indirect effects via each of other five components (j') to seed yield. (i.e. rjj' Plj'). The results are presented in Table 4. It is apparent from Figure 1 and Table 4 that number of nodes on main stem, number of nodes per branch and number of pods per node exerted greater influences on seed yield than other factors. The greatest direct effect from number of pods per node indicated a fact that, with other components held constant, increasing number of pods per node will increase seed yield. However, large negative indirect effects from this character via unit seed weight, number of nodes on main stem and number of nodes per branch to seed yield masked this direct effect and made a final negative association with seed yield (r=-0.1430). The influences of number of nodes on main stem and number of nodes per branch on seed, on the other hand, were strong and cooperative. Both indirect effects via each of them reciprocally, i.e. r46P14 and r46P16, were positive and comparable. Besides, unit seed weight had a smaller effect in seed yield and other two component charaters were unimportant. A clearer explanation of above discussion can be presented with the aid of the coefficients of determination, which are components of following equation: 1=ΣP2lj+2ΣPljPljrjj'+P2lj=Σdlj+Σdljj'+dlx. where dlj measures the proportion of the variability of seed yield which is directly attributable to component character j and dljj' is the joint determination of seed yield by components j and j'. Table 5 gives the results of the analysis of determination. It can be pointed out clearly from Table 5 that although number of pods per node had greatest coefficients of determination (=0.4133), it's joint effects with unit seed weight, number of nodes on main stem and number of nodes per branch prevented it to be a good determiner. On the contraty, both individual and joint determinations by number of nodes on main stem and number of nodes per branch were positive and consequently profitable so seed yield. Seed weight was also observed having positive effects, individual or joint with upper two component, to seed yield and might be a better determiner. In conclusion, selection of more number of pods per node in this hybrid progeny may increase seed yield but also may introduce a reduction of number of nodes per plant as well as unit seed weight, which may be more effective for determining seed yield. A compromise, however, should be praticed in the selection program among these characters. Number of branches and number of seeds per pod are relatively unimportant for future selection in the breeding program. |
本系統中英文摘要資訊取自各篇刊載內容。