Taxonomy and Phylogeny
Although B. racemosa can sometimes be separated from B charlesworthii, the large numbers of intermediate forms make some taxonomists think that they might be better treated as a single species.The firstBifrenaria species to be described was in 1824 by English Botanist William Jackson Hooker, under the name Dendrobium harrisoniae. Three years later, he also described the first small-flowered species, B. racemosa, but placed it in Maxillaria. With these two publications began a long series of species descriptions and confusing genera creation which generated much doubt for the next two centuries. The Royal Botanic Garden registers the submission of 69 species or subspecific taxa under Bifrenaria since the description of the first species. Among these, twenty are generally accepted but only seventeen are truly well established, with no doubts about their limits and classification. Thirteen other species are still accepted but now placed in other genera, and four or five, due to deficiencies in their descriptions, might never be definitely identified.
In 1832, John Lindley proposed the genus Bifrenaria and described its type species, Bifrenaria atropurpurea, previously named by Conrad Loddiges as Maxillaria atropurpurea. The name Bifrenaria comes from bi, two, and freno, brake, a reference to the shape of the two pairs of pollinia hold by separated caudicles presented by its flowers.
In 1837, Constantine Samuel Rafinesque, considering the noticeable vegetative difference between the few Bifrenaria known at the time, proposed the genus Adipe, based on B. racemosa morphology, described by Hooker few years before, to which he added the description of a supposed new species, Adipe fulva (today treated as a synonym of B. racemosa). The next year Lindley received a specimen from Amazonia, which was morphologically even more distant from the known species, but nonetheless described it as Bifrenaria longicornis'. Five years later, apparently not aware of Rafinesque's previous genus Adipe, Lindley changed his mind and suggested that this species should be classified under a new genus, Stenocoryne. Six species were subsequently attributed to Stenocoryne by various taxonomists, but the genus Rafinesque proposed remained unused until 1990.
Two species similar to Bifrenaria, but that showed a highly salient claw at the base of the labellum and lateral lobes abruptly divided were then classified under this genus. In 1914, Rudolf Schlechter suggested they should be classified under the genus Lindleyella, with Lindley's Bifrenaria aurantiaca (which presented the mentioned differences) as the type. However this genus name was already occupied (It is a synonym of Lindleya, in the Rosaceae). Just thirty years later, in 1944, Frederico Carlos Hoehne, working on the first revision of genus Bifrenaria, corrected the suggestion of Schlechter. Hoehne initially proposed the genus Schlechterella for these species but, coincidentally, this name was also taken, this time by African Asclepiadaceae. Finally a genus with an available name in homage to Schlechter was erected in the very next issue of the journal, Rudolfiella, by which time time the number of species had increased to seven. On this revision, besides Rudolfiella, Hoehne divided Bifrenaria into two genera, accepting Lindley's Stenocoryne but calling attention to the existence of Rafinesque's Adipe, which should have nomenclatural priority, while also raising doubts about the identity of several described species. In 1990, Manfred Wolff formally resurrected the genus Adipe and transferred to it ten Bifrenaria species, besides the two already described by Rafinesque; his change was purely nomenclatural and he did not revisit the species.
Making the picture even more complex, in 1994, Karheinz Senghas, based on several characteristics shared only by B. tetragona and B. wittigii, described the genus Cydoniorchis to accommodate them. In 1996, Gustavo Romero and Germán Carnevali transferred to Bifrenaria a species originally described by Schlechter as Maxillaria petiolaris and now classified as Hylaeorchis petiolaris. On the same year, Vitorino Castro Neto published a revision of Bifrenaria, with five sections, which is the classification generally used today.
Bifrenaria has traditionally been classified in subtribe Bifrenariinae of tribe Maxillariae (Epidendroideae), however, the relationships among the several genera within this tribe are not well defined and changes are expected in the upcoming years. The genus closest to Bifrenaria is Rudolfiella. Other related genera are Teuscheria, Guanchezia, Hylaeorchis and Horvatia, in addition to the more distant Scuticaria and Xylobium. The unification of subtribes Lycastinae, Maxillariinae and Bifrenariinae has recently been suggested. However, there is no consensus on the path to be followed. Contrary to what was previously thought, the relationship among Bifrenaria and all these genera from Central America seems to indicate a primitive origin of Bifrenaria in Central America and its posterior dissemination towards the Southeast of Brazil, where it found fertile grounds to its more recent evolution.
In 2000, the first relatively complete molecular analysis on Bifrenaria species were made. Sixteen species from it and six of close genera were studied searching for confirmation of their phylogenetic relations, besides the delimitation of each species and each of Bifrenaria's groups. The results did not allow for the acceptance of Adipe as a separate genus and, although they confirmed the monophyly of Cydoniorchis (B. tetragona and B. wittigii), they dissuade its recognition because six other genera would then be required to accommodate the remaining species. The study also expounded on convenience of splitting two species that are similar to each other and variable among themselves, but with many hard intermediate forms hard to delinetate as B. charlesworthii and B. racemosa. It also confirmed the position of B. steyermarkii outside of Bifrenaria, but without suggesting a new name.
Read more about this topic: Bifrenaria