Comarostaphylis is an endemic genus to North America, with a geographical distribution ranging from California to Panama. Ten species of Comarostaphylis are recognized, seven of which are endemic to Mexico. Comarostaphylis discolor comprises four subspecies, including C. discolor subsp. macvaughii, a perennial dwarf shrub endemic to isolated populations in the mountains near Tapalpa, Jalisco, Mexico. Based only on a morphology study, this subspecies was elevated to species rank but not recognized by some authors. In this study, we tested the taxonomic rank hypothesis of this subspecies using morphological data as well as molecular data within a phylogenetic framework. A total of 38 accessions of the Internal Transcribed Spacer (ITS) were included in the analysis: three as the ingroup and 35 as the outgroup. Six new individuals were analyzed—three belonging to Comarostaphylis macvaughii and the other three to C. discolor subsp. discolor, all from the same locality. We performed Maximum Likelihood (ML) analyses using RAxML. Our phylogenetic results support the recognition of C. macvaughii as a monophyletic lineage, sister to the clade comprising C. arbutoides and C. discolor species. Additionally, our preliminary conservation status assessment indicates that it is a critically endangered species.

Sampling: A total of 38 accessions were included in the analysis: three as the ingroup and 35 as the outgroup. Six new individuals were added, three of them belonging to Comarostaphylis subsp. macvaughii (Fig. 1) and the other three to C. discolor subsp. discolor. Both species grow in sympatry at the type locality of C. subsp. macvaughii (Fig. 2; Table 1), where both were collected.

Morphological characters: We examined live and herbaria specimens of Cormarostaphylis discolor subsp. discolor and C. discolor subsp. macvaughii from the following herbaria ENCB, IBUG, IEB, MEXU, MICH, NY, WIS, and XAL (acronyms based on Thiers 2025). The comparative morphological analysis adopts the character set defined by Diggs (1988) and González-Villarreal (1990a).

DNA extraction, PCR, and sequencing: DNA was isolated with the CTAB method (Doyle & Doyle 1987), modified by Cota-Sánchez et al. (2006), from 100 mg of dry leaf tissue preserved in silica gel-dried. Plant tissue was pulverized using a TissueLyser LT (QIAGEN, Hilden, Germany). The DNA was dissolved in 100 µL of Tris-EDTA buffer solution (TE). DNA concentration and purity was measured using a NanoDrop 2000TM Spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). We amplified the nrDNA region that includes the ITS1, 5.8S, and ITS2, using the universal primers ITS4 and ITS5 (White et al. 1990). Polymerase chain reaction (PCR) was carried out using an AERIS™ thermal cycler (Esco Healthcare, Singapore) under the following conditions: initial denaturation at 95 °C for 4 min, followed by 35 cycles of denaturation at 94 °C for 30 sec, annealing at 45-50 °C for 30 sec, and extension at 72 °C for 1.5 min, with a final extension step at 72 °C for 10 min. The PCR products were visualized on a 1% agarose gel stained with GelRed and examined under a UV Transilluminator (UVP). Subsequently, the amplified products were analyzed on a SeqStudio-232000826 (Applied Biosystems®) at the Laboratorio Nacional de Identificación y Caracterización Vegetal (LaniVeg, University of Guadalajara). The nrDNA sequences were edited using Sequencher v. 4.1.2 (Gene Codes Corporation, Ann Arbor, MI, USA), and aligned using Phyde v. 0.92 (Müller et al. 2005). GenBank accession numbers: ITS: PQ329515-PQ329517. Genomic DNA extraction, product purification, amplification, and sequencing were performed at the National Laboratory of Plant Identification and Characterization (LaniVeg), University of Guadalajara.

Phylogenetic analysis: Our sampling included 38 DNA sequences; six of them were newly generated for this study. Three of them from Comarostaphylis subsp. macvaughii and three from C. discolor subsp. discolor the rest were used previously in the phylogenetic analysis of Arbutoideae by Hileman et al. (2001) or uploaded to GenBank (Table 1). Pleuricospora fimbriolata A. Gray was used as the functional root. First, we used ModelTest-NG (Darriba et al. 2020) to identify the molecular evolution model that best fits our two different matrices. Next, we performed Maximum Likelihood (ML) analyses using RAxML v. 8 (Stamatakis 2014) and implemented through raxmlGUI v 2.0.10 (Edler et al. 2021). Nodal support was estimated with a parametric bootstrap with 1000 replicates.