Statistical Analysis.

Data were analyzed by using one-way ANOVA and linear regression analysis with the SPSS version 16.0 software for Windows. Unless otherwise stated, we chose the sample size n = 4 of plant individuals per genotype and treatment. Analysis of large sample size experiments revealed that normal distribution occurred with a SD, σ, of 10%. The sample size was computed with the G*Power 3.1.9.2 software program under the assumption that a difference of 30% (3 × σ) between RCAR-overexpressing lines and the wild type would occur, with presumed values for α and β of 0.05 and 0.1, respectively. The cohort size of four individuals is recommended based on these parameters for an unpaired, two-sided t test.