Mazaea is characterized by its xerophytic shrub habit with the stems comprising numerous closely set nodes and the leaves clustered at their ends; relatively small narrow leaves; short, interpetiolar, triangular, persistent stipules; axillary 4-merous- flowers that are solitary or 2-3 in small cymes and borne near the stem apex; rather well developed, shallowly lobed calyx limbs; salverform, pink to red corollas with a yellow to orange fleshy ring in the throat and imbricate rounded lobes; fusiform capsular fruits that open septicidally and then loculicidally and have the calyx limb deciduous; and small to medium-sized, flattened seeds with an elongated wing on each end. The corollas are rather small, and their lobes are similar in size and general shape to the calyx lobes. The anthers and stigmas are situated at the same level and positioned in the corolla throat, so sometimes they are partially exserted on the specimens but whether this is an artifact of drying is not clear. This species was illustrated by Delprete (1999: 222, fig. 1A-F), and this is generally accurate but appears to show the stipules united around the stem rather than interpetiolar. It is found in a limited range and apparently is a specialist on serpentine. 

Mazaea was initially established with one species, Mazaea phialanthoides. Subsequently Standley added another species, Mazaea shaferi; Urban added another, Mazaea pungens; and Alain added one more. Mazaea tinifolia. Subequently Borhid et al. (1980, 1981) began to review this overall group with new material and additional evidence from their observations and other studies. They separated the species of Mazaea into three genera: Mazaea phialanthus, Acunaeanthus tinifolia, and Suberanthus pungens. The leaf epidermal study of Vales (1983)  supported their classification. The molecular analysis of Torres-Montúfar et al. (2020) found these three genera on separated lineages and recognized them as distinct, although the type species of Mazaea was not included in their sampling. 

Urban also later described another, similar new species from Cuba, and separated it into in a new genus, Ariadne. He soon noted that his new species was the same as Neomazaea shaferi, and so transferred that species out of his circumscription of Mazaea, which left Mazaea monotypic. Delprete (1999) noted that Ariadne was separated based on, basically, an incomplete septum in the ovary and fruit, vs. complete in Mazaea, but his direct observations of the plants found this character to be varably incomplete to complete and he synonymized these genera to circumscribe Mazaea as having two species. Torres-Montúfar et al. (2020) concurred with that synonymy, although they only studied one of the species of Delprete's Mazaea, Mazaea shaferi, which corresponds to the Ariadne element of Delprete's genus. In contrast, Borhidi et al. (2017) continued to recognize Ariadne and Mazaea as separate genera based on inflorescence arrangement  and several corolla characters. The characters given for the two species that correspond to these two genera by Borhidi et al. and in their protologues do not completely agree with the characters cited for these groups by Delprete and Torres-Montúfar et al. Torres-Montúfar et al. also noted that previously Rova et al. (2002) and Manns & Bremer (2010) did include both of these species in molecular stuides, and found them arranged in a polytomy or paraphyletic position with Phyllomelia. They noted that those results suggested separating all three into monotypic genera, and it seems likely that Torres-Montúfar et al. would have separated Ariadne from Mazaea based on those results if its corolla form and perhaps other characters had been correctly noted.  

For these two species, as illustrated by Delprete (1999: 222, fig. 1C, I) the corollas differ in the presence in Mazaea vs. absence in Ariadne of a thickened ring in the corolla throat, and as detailed by Borhidi et al. (2017) in color: the corollas of Mazaea phialanthoides are pink to red with an orange to yellow throat ring, and white or cream throughout with the throat not ornamented in Ariadne. Torres-Montúfar et al. (2020) characterized Delprete's expanded genus Mazaea as having a thickened ring in the corolla throat, so thier morphological analysis was incomplete, and they included in their analysis only Mazaea shaferi, which lacks such a ring. Torres-Montúfar et al. considered presence vs. absence of the fleshy ring in the corolla throat to be diagnostic for genera in their group, so the inclusion of both of these species in Mazaea is problematic within their taxonomic scheme. Additionally, Torres-Montúfar et al. considered the ovary and seed features of both of these species, with 2-6 ovules per locule and seeds, to be distinctive among the genera of the Rondeletia assemblage. Delprete (1999: 220) also characterized his expanded circumscription of Mazaea as having 2-6 ovules per locule in his overview of this group (1999: 220), but his technical description for this expanded Mazaea stated it had "(1-)2 seeds per locule". Borhidi et al. (2017) and the protologue descriptions detailed Mazaea phialanthoides as having 4-6 ovules per locule and Ariadne shaferi as having 2 or rarely 1 ovule per locule. Delprete additionally documented differences in these species between the anthers, with these relatively long and included in Mazaea phialanthoides but quite short and exserted in Ariadne shaferi, but he described the stamens as inserted near the top of the corolla in both species while Borhidi et al. and the protologue description of Ariadne shaferi specifies the stamens insertion as near the corolla base. The protologues of these two species followed by Borhidi et al. additionally contrasted the ovules of Mazaea phialanthoides as borne on a central axile placenta while those of Ariadne shaferi are borne on a basal elongated placenta; however Delprete described both species as having a "central placenta", and stated that this placenta is "apically free or extending throughout the length of the ovaryI. Delprete's technical description of the free to perhaps fused placenta appears to conflict with his genus discussion, which referred to a complete to incomplete septum, or perhaps his formal genus description includes a typographical error. Overall, the arrangement of the gynoecia of these two species appears to need further study, which is beyond the scope of this review. Based on the characters that were clearly documented by the specimens studied here, Borhidi et al., and the protologue descriptions along with the molecular results of  Rova et al. (2002) and Manns & Bremer (2010), Mazaea and Ariadne are here considered closely related but separate genera based on corolla form, ovule number, the stamens inserted in the upper part (Mazaea) vs. near the base (Ariadne) of the corolla, and (not previously noted) the stipules persistent after the leaves fall in Mazaea vs. deciduous with the leaves in Ariadne.

Mazaea is related to and similar in aspect to Rachicallis, with smaller thickly fleshy leaves and smaller loculicidal capsules. Mazaea is also similar to Ariadne, with white corollas that lack a fleshy ring in the throat. The well developed calyx limb is similar to that of Phyllomelia, but this is deciduous after flowering in Mazaea while the calyx of Phyllomelia is persistent and functions as a wing on the fruits of Phyllomelia. Phyllomelia also differs in its mult-flowered, pedunculate inflorescences, ovary with one ovule in each locule, and indehiscent fruits.   

The name Neomazaea was published as a replacement name for Mazaea Krug & Urb., because that genus name had previously been used, in 1881, so Krug & Urban's name was thought to be an illegitimate later homonym. However, the earlier Mazaea name was a genus of nostoc algae, and the starting date for nomenclature in that group is 1886 (Art. 13.1) This means the first Mazaea name is not effectively published so it does not block Krug & Urban's name. This point has been noted several times (Sandwith in sched.; Robbrecht & Bridson, 1993; Delprete, 1999) but the illegitimate name Neomazaea still generates some confusion. 

Author: C.M. Taylor.
The content of this web page was last revised on 20 August 2021.
Taylor web page: http://www.mobot.org/MOBOT/Research/curators/taylor.shtml

• Borhidi, A. L., M. Fernández Zequeira & R. Oviedo Prieto. 2017. Rubiác. Cuba 1–494.
• Delprete, P. G. 1999. The Cuban endemic genera Ariadne, Mazaea, Acunaeanthus, Phyllomelia (Rondeletieae), and Eosanthe (Rubiaceae). Brittonia 51(2): 217–230.
• Robbrecht, E. & D.M. Bridson. 1993. Nomenclatural notes on three Rubiaceae genera. Opera Bot. Belg. 6: 197–200.
• Borhidi, A. L., M. Járai-Komlódi & M. Moncada Ferrera. 1980 [1981]. Acunaeanthus, a new genus of Rubiaceae. Acta Bot. Acad. Sci. Hung. 26(3–4): 277–287.
• Borhidi, A. L. & M. Fernández Zequeira. 1981. Studies in the Rondeletieae (Rubiaceae) II. A new genus: Suberanthus. Acta Bot. Acad. Sci. Hung. 27(3–4): 313–316.
• Torres-Montúfar, A., T. Borsch, S. Fuentes-Bazán, J. Gutiérrez & H. Ochoterena-Booth. 2020. It is not a disaster: molecular and morphologically based phylogenetic analysis of Rondeletieae and the Rondeletia complex (Cinchonoideae, Rubiaceae). Pl. Syst. Evol. 306(26): 1–25.
• Rova, J. H. E., P. G. Delprete, L. Andersson & V. A. Albert. 2002. A trnL-F cpDNA sequence study of the Condamineeae-Rondeletieae-Sipaneeae complex with implications on the phylogeny of the Rubiaceae. Amer. J. Bot. 89(1): 145–159.
• Manns, U. & B. Bremer. 2010. Towards a better understanding of intertribal relationships and stable tribal delimitations within Cinchonoideae s.s. (Rubiaceae). Molec. Phylogen. Evol. 56(1): 21–39.
• Vales, M. A. 1983. Studies in Rondeletieae (Rubiaceae), VII. The significance of leaf epidermis for taxonomy in Neomazaea sensu lato. Acta Bot. Hung. 29(1-4): 43-53.