Path analysis is a statistical method for determining the magnitude and direction of multiple effects on a complex process. We used path analysis to determine the direct effects of nightshade density on yield components (number of green fruit per plant, rotted fruit per plant, total fruit per plant, and weight per fruit) of the processing tomato cv. Heinz 6004. In addition, the analysis indicated the direct and indirect effects of yield components on total yield per ha and marketable yield per ha. The greatest direct effects of eastern black nightshade and black nightshade were on green fruit per plant and total fruit per plant. Effects other than density (density-independent factors) were more important in determining the number of rotted fruit per plant and weight per fruit. Path analysis showed that the total number of fruit per plant was the most important yield component determining total yield and marketable yield per ha.
Milton E. McGiffen Jr. and Dan J. Pantone
Milton E. McGiffen Jr., Dan James Pantone, and John B. Masiunas
Path analysis is a statistical method for determining the magnitude and direction of multiple effects on a complex process. We used path analysis to assess 1) the impact of black nightshade(Solarium nigrum L.) or eastern black nightshade(Solarium ptycanthum Dun.) competition on the yield components of `Heinz 6004' processing tomato (Lycopersicon esculentum Mill.) and 2) the relationship between tomato yield components and total and marketable yield. Either black or eastern black nightshade was interplanted with tomatoes at population densities from 0 to 4.8/m2. Path analysis revealed that increasing weed population density led directly to fewer green and total fruit per plant, two components of marketable yield. However, the percentage of culls per plant and fruit weight were not affected by nightshade population density. Using correlation coefficients alone would have lead to the erroneous conclusion that the percentage of culls did not affect marketable yield; our path analysis demonstrated that decreasing the percentage of culls through breeding or cultural practices will strongly affect marketable yield. The total number of fruit was the most important yield component in determining total and marketable yields per plant. Breeding and management practices that maximize fruit set, increase maturity at harvest, and decrease the percentage of culls would be expected to increase marketable yield.
Christopher S. Cramer and Todd C. Wehner
The relationships between fruit yield and yield components in several cucumber (Cucumis sativus L.) populations were investigated as well as how those relationships changed with selection for improved fruit yield. In addition, the correlations between fruit yield and yield components were partitioned into partial regression coefficients (path coefficients and indirect effects). Eight genetically distinct pickling and slicing cucumber populations, differing in fruit yield and quality, were previously subjected to modified half-sib family recurrent selection. Eight families from three selection cycles (early, intermediate, late) of each population were evaluated for yield components and fruit number per plant in four replications in each of two testing methods, seasons, and years. Since no statistical test for comparing the magnitudes of two correlations was available, a correlation (r) of 0.7 to 1.0 or –0.7 to –1.0 (r 2 ≥ 0.49) was considered strong, while a correlation of –0.69 to 0.69 was considered weak. The number of branches per plant had a direct positive effect on, and was correlated (r = 0.7) with the number of total fruit per plant over all populations, cycles, seasons, years, plant densities, and replications. The number of nodes per branch, the percentage of pistillate nodes, and the percentage of fruit set were less correlated (r < |0.7|) with total fruit number per plant (fruit yield) than the number of branches per plant. Weak correlations between yield components and fruit yield often resulted from weak correlations among yield components. The correlations among fruit number traits were generally strong and positive (r ≥ 0.7). Recurrent selection for improved fruit number per plant maintained weak path coefficients and correlations between yield components and total fruit number per plant. Selection also maintained weak correlations among yield components. However, the correlations and path coefficients of branch number per plant on the total fruit number became more positive (r = 0.67, 0.75, and 0.82 for early, intermediate, and late cycles, respectively) with selection. Future breeding should focus on selecting for the number of branches per plant to improve total fruit number per plant.
Mehdi Mohebodini, Naser Sabaghnia, and Mohsen Janmohammadi
its extracts for various medicinal uses and industrial uses for edible oil ( Rehman et al., 2012 ). The common path analysis approach might result in multicollinearity for variables, particularly when associations among some of the traits are high
Reza Amiri, Kourosh Vahdati, Somayeh Mohsenipoor, Mohammad Reza Mozaffari, and Charles Leslie
–and-effect relationships in the variables, because the association between two variables may depend on a third variable. The use of path analysis provides a plausible explanation of observed correlations by modeling the cause-and-effect relations between the variables
Peng Shi, Yong Wang, Dapeng Zhang, Yin Min Htwe, and Leonard Osayande Ihase
estimate the interrelation among the agronomic traits (Boo et al., 1990). Hierarchical cluster analysis shows the similarity of the samples ( Peeters and Martinelli, 1989 ). Path analysis serves as a great tool to evaluate the relationship among agronomic
Benjamin D. Toft, Mobashwer M. Alam, John D. Wilkie, and Bruce L. Topp
genotypes for all traits. Path coefficient analysis was used to analyze the standardized phenotypic means of measurements as outlined by Akintunde (2012) . Path analysis has been used for other crops to understand the complex interactions that underlie
Yun Kong, Xiangyue Kong, and Youbin Zheng
. For the model construction data, the correlation between all the variables was analyzed first. Then, path analysis was carried out to determine the direct and indirect effects of the six variables (i.e., SMD, SML, LL, LW, SEL, and SEW) on the FW of pea
JoAnn Robbins and Patrick P. Moore
Weight and morphological characteristics of red raspberry (Rubus idaeus L.) fruit, including drupelets (height, diameter, number), receptacle cavities (depth, diameter), and pits (individual weight), were measured on 78 seedlings from the cross `Chief' × `Chilliwack'. Fruit strength, as measured by compression, correlated with fruit weight, drupelet number, receptacle cavity depth, and individual pit weight. Fruit weight was positively correlated with all morphological characteristics. Individual pit weight, drupelet height, and drupelet number provided the largest component contributions to fruit strength as measured by path analysis.
Contribution no. 730. The assistance of J.W. Hall, Agr. Canada, Res. Sta., Vancouver, B. C., for his comments related to the statistical analysis is gratefully acknowledged. The cost of publishing this paper was defrayed in part by the payment of