Ten Epithelantha species. Areoles, roots and species boundaries

One stem form, many characters to compare

Epithelantha forms small tuberculate stems whose surface is largely concealed by pale spines. Similarity between whole plants long favoured differing classifications. In 2019, Aquino and coauthors began with eleven named groups treated as hypotheses of taxon boundaries. Their final revision recognized ten species. POWO accepts the same number in October 2026. [Aquino et al. 2019, pp. 600–601, 611–612] [2-WWW]

The characters distinguished included the structure of the areole, the small spine-producing zone on a tubercle, and the structure of reproductive organs and underground parts. Species identification relied on a combination of characters. Areole width could become significant when compared with flower length, and spine number when root type was considered. This approach separates variation in a single character from consistent differences between entire groups. [Aquino et al. 2019, pp. 602–606, 608–612]

How 195 adult plants were compared

Photographs of five adult plants per site were obtained from 39 sites in Coahuila, Nuevo León and San Luis Potosí. Adulthood was assessed from the presence of reproductive structures or their traces. Each photograph included a scale for reading dimensions. Nine quantitative and six qualitative characters were initially analysed. Highly correlated characters were reduced, and stem and areole proportions were expressed as dimensional ratios. [Aquino et al. 2019, pp. 601–603, 615]

Areoles and spines were assessed in the middle of the stem to reduce the effect of position and developmental stage. The flower was measured from the base of the pericarpel, the part surrounding the ovary, to the tip of the longest tepal. Reported flower length refers to this entire axis. Scanning electron microscopy also revealed surface characters of spines and seeds that are difficult to identify in an ordinary photograph of the whole plant. [Aquino et al. 2019, pp. 601–602, 610, 615]

Characters useful for identification

The spine number of E. potosina partly overlaps the range of E. pulchra. An initial count of 40 or 41 spines should therefore lead to comparison of areole shape and dimensions, roots and the remaining characters. The table organizes selected differences, and its numbers retain the scope of the 2019 sample. A species is identified from the entire combination of structures. [Aquino et al. 2019, pp. 606, 611–612]

Two major plastid lineages

Four plastid DNA regions were sequenced, petL-psbE, psbA-trnH, trnL-F and trnQ-rps16. The tree included 39 Epithelantha samples and 23 outgroup samples. The genus itself and its division into two major lineages received high support. The first included current E. bokei, E. cryptica, E. ilariae, E. micromeris, E. potosina and E. spinosior. The second grouped E. greggii, E. pachyrhiza, E. polycephala and E. pulchra. [Aquino et al. 2019, pp. 603–609; Fig. 3–4]

Plants with thickened roots predominated in the second lineage, apart from slender-rooted E. greggii. However, E. potosina in the first lineage also had thickened roots. The distribution of this character therefore does not correspond to a single exclusive division of the genus. The hypothesis of a close relationship between Epithelantha and Turbinicarpus received much weaker support than the distinctness of the genus itself and does not resolve relationships between genera with the same confidence. [Aquino et al. 2019, pp. 604, 606–607; Fig. 3]

Micromeris and cryptica distinguished by structure

E. micromeris and E. cryptica had very similar sequences for the markers used. They differed, however, in flowers and areoles. In the material studied, flowers of E. micromeris were 6–7 mm long and those of E. cryptica 11–12 mm. The latter also had broader oval areoles. Analysis combining DNA with ten coded morphological characters allowed the two lineages to be separated. [Aquino et al. 2019, pp. 605, 608–609, 612]

Morphological classification analysis assigned all studied plants of these two groups to their own identifications. The result concerned the same dataset from which the identification was developed. A later 2021 study also indicated environmental differences between the sites of the two species. The result thus combines morphological distinctness with differences in site environments. [Aquino et al. 2019, p. 605, Table 5] [Aquino et al. 2021, pp. 4525–4527]

Former subspecies, current species

E. potosina was formerly described as a subspecies of E. greggii. On the plastid tree, they fell into two major lineages, and morphometrics confirmed their different character sets. For E. pulchra, formerly a subspecies of E. pachyrhiza, numerous spines and round areoles, together with a distinct DNA lineage, were significant. New combinations of these two names at species rank were published in 2019. [Aquino et al. 2019, pp. 606–609, 611–612]

Material of E. pachyrhiza was sequenced under the name subsp. parvula. The authors did not obtain plants of the nominate subspecies or subsp. elongata for this purpose. Inclusion of these names in one species also relied on earlier descriptions. E. pulchra and the studied E. pachyrhiza shared uneven multicellular seed-coat sculpture and expansions near the hilum region, the scar left by seed attachment. Their shared seed structure was considered alongside differences in stems and DNA. [Aquino et al. 2019, pp. 601, 608, 610–611]

Changing historical names and the limits of the sample

Samples provisionally identified as E. densispina and E. rufispina fell within the variation of E. greggii. Morphological and plastid data did not justify separating them in this sample. Two groups formerly named E. unguispina and its subsp. huastecana were included under the older name E. spinosior. POWO currently retains two accepted subspecies within this species, subsp. spinosior and subsp. huastecana. [Aquino et al. 2019, pp. 606, 608, 611–612] [2-WWW]

Plant age remains part of a valid comparison. In 2026, Gorelick described the similarity of juvenile E. micromeris to E. bokei and considered retention of juvenile characters as a possible explanation for the latter's appearance. This was a developmental hypothesis. The 2019 adult-plant dataset revealed distinctive spine characters and a separate plastid lineage in E. bokei. The two types of observation can be compared while retaining their different material and developmental stage. [Gorelick 2026, pp. 1–7] [Aquino et al. 2019, pp. 601, 606–609]