2.6 Data analysis
Data were presented as means ± standard error (SE) from four replications for each experiment and determination. The data were analyzed by full factorial ANOVAs followed by Tukey’s honestly significantly different (HSD) tests. Weed inhibition and allelochemical production were calculated as (1-T/C)×100, based on weed dry weights or allelochemical quantity in treatments (T) and in controls (C). Horizontal asymmetry in root length and biomass were calculated as response ratios Vn/Va (Yang & Kong, 2017). Vn is the response value for rice-weed interactions while Va is the response value for rice or weed alone. Vn/Va >1 represented root growth increased towards neighbors, resulting an intrusive pattern of root placement. Vn/Va<1 represented root growth decreased towards their neighbors and Vn/Va=1 represented an unaltered pattern of root growth. Principal component analysis (PCA) of PFLA data was performed with the STATISTICA software package, version 6.0 (StatSoft Inc., USA). Data were transformed using sample unit totals to represent the relative abundance of each PLFA (mole percentage of total PLFA). All data analyses were conducted with SPSS 25.0.
3 RESULTS