The identity of this species is well demonstrated, but its systematic relationships and generic placement are not. Diodia was long circumscribed to include a set of species with a schizocarpous fruit that divides into two dry mericarps of slightly varied form, along with (generally) a herbaceous to suffrutescent habit, fimbriate stipules, sessile and glomerulate axillary inflorescences, small, 4-merous, homostylous flowers, and often transfersally grooved seeds. The species of Diodia s. lat. are weedy and rather commonly encountered, although humble in stature and often overlooked, and their characters are subtle and often enigmatic to find and analyze, and the circumscription of this genus was quite generalized.
In recent decades this genus has been, similarly to many other genera of Tribe Spermacoceae, reviewed in more detail with new techniques and separated into several genera. The American species of this genus were studied in detail morphologically first in the careful review by Bacigalupo & Cabral (1999), who separated them into several genera. They retained five species in Diodia, all of them with indehiscent, firm-textured or often corky mericarps and a general habitat in periodically wet sites. They transferred nine species to Galianthe; referred 13 other species to Borreria s.lat., another group with poorly understood systematics; and referred 16 other species to a long-overlooked genus, Diodella. Diodia sarmentosa was one of the species they referrred to Diodella; they listed the species they thought belonged in this group but did not make nomenclatural combinations to transfer them formally.
Based on this work, Delprete (2004) published new combinations in Diodella for a couple of the species that Bacigalupo & Cabral (1999) referrred to that group. Later, Bacigalupo & Cabral (2006) published several additional combinations in Diodella, including Diodella sarmentosa. They did not comment much about its affiinites there, only noting that it seemed similar in particular to Diodia radula. Subsequently additional combinations in Diodella were published by Delprete, Cabral, Cabaña Faber, and others to move the other species listed by Bacigalupo & Cabral (1999) into that genus, without overall study of the group as a genus. Later Cabaña Faber, Salas, Dessein, and colleagues did undertake study of the genus, and analyzed it morphologically and with molecular data (see next paragraph). They also found that the long-overlooked genus Hexasepalum was an older name for Diodella as it was then circumscribed.They proposed formal nomenclatural conservation of Diodella and rejection of Hexasepalum due to the change being unduly destabilizing, especially shortly after Diodella had just been revived. The proposal was rejected and as soon as that report was published, Kirkbride & Delprete (2015) published combinations in Hexasepalum for the Bacigapulo & Cabral's (1999) species referred to Diodella, including Diodia sarmentosa.
Along with these nomenclatural reviews, this general group of Spermacoceae genera was studied with molecular data. Groeninckx et al. (2009) published a first broad analysis, which found that the dozen or so species of Diodella/Hexasepalum were not all closely related, and other analyses were not formally published but found similar results. In contrast, Kirkbride & Delprete (2015) noted in their review of Hexasepalum that although the available molecular data showed that Diodia sarmentosa was not closely related to their other species of Hexaspalum, they were transferring it to this genus anyway because they did not consider the phylogenetic studies "definitive". Separately during these period, Cabaña Faber et al. (2016) studied this group with additional molecular and morphological data, and narrowed the circumscription of the genus to a group of five species. They excluded the other species of Kirkbride & Delprete's Hexasepalum, including Diodia sarmentosa. They noted that their molecular analysis found Diodia sarmentosa again grouped with several other species of unclear generic identity, in a well supported and separated clade. In another molecular study focussed on this separated clade, Dutra et al. (2021) found a similar result.
Salas and collaborators have studied this group of leftover species in more detail, and noted that Ernodea is placed here; separated several of these species into a new genus Planaltina; separated two others into revived overlooked genus, Tessiera; found that two other species of Kirkbride & Delprete's Hexasepalum are placed here; and left Diodia sarmentosa with that name until its systematic identity can be further clarified. Tessiera is notable in its fruits with the mericarps separating and falling individually and then opening via a smoth, thin-textured adaxial wall, and these fruits also have a small basal portion with a strengthtened lower part of the septum that remains on the fruit after the mericarps fall. This structure seems to be present on many specimens of Diodia sarmentosa also, though it is absent or not well developed in some (pers. obs.). The fruits of "core" Hexasepalum also have two indehiscent, firm-textured mericarps that disperse separately from the plant, but in those they are fully indehiscent and the adaxial face of those mericarps is similarly textured to the abaxial surfaces and grooved.
Diodia scandens of Hispaniola has been regarded as a separate species distinguished by relatively smaller leaves (e.g., Liogier, 1995), but their separation is not well supported. These have been separated just by leaf size, but Swartz's specimens of Diodia scandens show variation in leaf size even on individual stems, its larger leaves overlap with the size of the smaller leaves of traditional Diodia sarmentosa. The reproductive characters of specimens assigned to these species are not separable. These species have been treated as separate by various 20th century authors, especially in Africa where their separation is based on confidence in Swartz's taxonomy rather than modern taxonomic review (Verdcourt, 1976). However, these names are synonymized here because no clearly distinguishing characters have been stated or found here, and two species cannot be confidently separated among modern collections from Hispaniola. Diodia scandens has sometimes been cited as a "sensu" synonym according to Bentham's identification of some French Guianan specimens [J. Bot. (Hook.) 3: 236, 1841, often mis-cited as p. 263] that were later reidentified as Diodia sarmentosa (e.g., Verdcourt, 1976), but this name is regardedin the review here as an actual synonym. These two names were published simultaneously, and perhaps have not yet been synonymized formally but Diodia sarmentosa is by far the more widely used name. Kirkbride & Delprete (2015) also treated these as two different species but without clarification of their identities.
The variety Diodia sarmentosa var. bisepala has been published twice independently, once by Schumann in 1881 as a new variety, and again by Bremekamp in 1934 as a synonymization and change in nomenclatural status for Diodia rudis. These two varietal names have been confused, for example in annotations on the holotype of Diodia rudis where various annotators were unaware of both names.
Bacigalupo & Cabral (1999) cited two supposedly synonymous names under Diodia sarmentosa, but neither of those is a validly published name. The first of these is a "sensu" bilbiographic reference to a misidentification of this species. The second is an unpublished name used on a herbarium annotation, which was cited as a nom. nud. to clarify which specimen was being cited not as a synonymous name.
Diodia sarmentosa is here treated under this name for now following Cabaña Faber et al. (2016), until its relationships and generic identity (and morphology) are better understood.
Author: C.M. Taylor
The content of this web page was last revised on 13 June 2024.
Taylor web page: http://www.mobot.org/MOBOT/Research/curators/taylor.