DNA extraction and chloroplast genome determination
Seedlings were raised from seeds in a greenhouse, harvested, and quickly dried in a convection oven. Total genomic DNA (~30 mg) was extracted from dry leaves using the mCTAB method ( Li, Wang, Yu, Wang, & Zhou, 2013). A library was constructed and sequenced for each sample at Beijing Novogene Bioinformatics Technology Co., Ltd, Beijing, using an Illumina HiSeq X Ten platform. Chloroplast genome reads were sorted out and the genomes were assembled de novo using SPAdes 3.9 (Bankevich et al., 2012). The generated contigs were mapped to the closest references by blastn 2.8.10 (Altschul, Gish, Miller, Myers, & Lipman, 1990), assembled with Sequencher 5.4 (Corperation)and gaps were filled by Sanger sequencing using primers reported by Dong, Xu, Cheng, Lin, and Zhou (2013).