The aim of DNA barcoding is to enable fast and accurate species identification. However, universal plant DNA barcodes often do not provide species-level discrimination, especially in taxonomically complex groups. Here we use Lauraceae for the design and evaluation of DNA barcoding strategies, considering: (1) the efficacy of taxon-specific DNA barcode regions compared with universal barcodes for species discrimination; and (2) how the extent of intra- and interspecific sampling affects species discrimination rates. To address these areas, we targeted the highly polymorphic, taxon-specific barcode regions ycf1 + ndhH–rps15 + trnL–ycf2 for Lauraceae and compared them against the suite of standard plastid loci used for DNA barcoding (rbcL + matK + trnH–psbA) and the standard nuclear barcode ITS. The highest discrimination success came from nrDNA ITS, whereas the plastid regions (rbcL + matK + trnH–psbA) and the taxon-specific regions (ycf1 + ndhH–rps15 + trnL–ycf2) showed limited and inconsistent resolution. These results highlight that taxon-specific plastid barcodes may provide limited gains in discriminatory power in complex, closely related groups like Lauraceae. Moreover, our study showed that species discrimination greatly depends on the taxon sampling scheme, with relatively lower species discrimination observed where there is more comprehensive intra- and interspecific sampling. The outstanding challenge for plant DNA barcoding is the development of assays that allow routine low-cost access to large numbers of nuclear markers to facilitate the sequencing of large numbers of individuals.
