It is not common for a taxonomist to contradict his/herself or to admit his/ her faults or misinterpretations of data. Usually, taxonomic decisions depend upon the amount of evidence (data) available and will reflect the knowledge available for a group of organisms and by a taxonomist at a particular time. Nevertheless, as more data is gathered it is possible to revise your own decisions and conclusions. Currently, there is a trend to treat species as a scientific hypothesis (Pante et al. 2015, Assis et al. 2018), which can be tested in different moments, confirmed, rejected or revised. This means that, as with any scientific hypotheses, it may be corroborated by data or not, and even better, may be re-evaluated or re-presented in face of new evidence. In this paper, I will describe a new species (or species hypothesis) of Alcantarea (Bromeliaceae, Tillandsioideae), based on specimens that I have previously examined and cited in recent revisions (Versieux & Wanderley 2010, 2015) as belonging to Alcantarea nahoumii (Leme) J.R. Grant.

In our monograph of Alcantarea, fruit of my PhD thesis advised by Maria das Gracas Wanderley, in Brazil, we stated that we would delimit species based on the morphological species concept, in which a species must have its own set of morphological traits, ideally found in several individuals of different populations. Also, we required that this assemblage of morphological features would have to clearly separate the different species (Versieux and Wanderley, 2015). However, when based on herbarium specimens alone, interpretation of morphological features in bromeliads may be tricky due to losses in color information or in the fragile flower parts. Limitations to the recognition of important characters when using bromeliad herbarium specimens have been discussed (e.g., Brown and Gilmartin 1984), and nowadays most taxonomists take advantage of living collections as well as studies in the field, to understand better the living plant morphology or, especially, to study fresh flowers. In some cases, horticulturists are decisive, since plants are frequently introduced into cultivation by collectors prior to attracting the attention of taxonomists (see the case of Alcantarea australiana Versieux & Smythe, [Versieux & Wanderely 2015]).

Recently, botanist Cassio Van den Berg, from State University of Feira de Santana, photographed living and blooming specimens of Alcantarea around Feira de Santana municipality, which caught my attention. Firstly, due to the color of the peduncle and primary bracts of the inflorescence completely yellow or yellowish green. Also, these same bracts were narrower and more erect, compared to the more inflated and arcuate bracts of A. nahoumii, the binomial I used for specimens from Feira de Santana municipality. This made me question my own delimitation, and check more carefully the herbarium material. In fact, both specimens, that I will treat herein as type material, mention that the inflorescence has yellowish bracts, which turned out to be the key feature of the new species I am proposing below.

These same specimens, however, were limited in providing enough information about their uniqueness. After seeing some photographs of blooming individuals...