Dyckia sordida Baker (1889: 132)

Büneker, Henrique Mallmann, Soares, Kelen Pureza & Assis, Lucas Coelho De, 2016, The Dyckia sordida complex (Bromeliaceae, Pitcairnioideae) and a new species from Minas Gerais, Brazil, Phytotaxa 244 (1), pp. 57-68 : 63-65

3. Dyckia sordida Baker (1889: 132) , Figs. 4A–G

Type: — BRAZIL. Minas Gerais: Itambé, Saint-Hilaire 402 (holotype P!). = Dyckia duarteana Smith (1967: 480) . Type:— BRAZIL. Minas Gerais: Serra do Cipó, Estrada para Conceição do Mato Dentro, km 137,

21 April 1950, A. P. Duarte 2749 (holotype RB!).

Etymology: —The brownish-ferruginous indument presented in the peduncle and sepals probably originated the Latin epithet “ sordida ” (sordidus = dirty).

Observations: —According Forzza & Wanderley (1998),the inflorescence indument similarity between D. sordida and D. ursina has provided erroneous identifications of this species in herbaria collections. Despite the similar habit, in D. sordida clump formation (caespitous plant) is observed, while D. ursina generally presents solitary rosettes. The peduncle, inflorescence, bract and sepal indument, although brownish-ferruginous in both species, shows distinct kind of trichomes, being tomentose, dendritic and sparse in D. sordida and lanate, long, simple or branched (in the terminal portion) and denser in D. ursina ( Forzza & Wanderley 1998) .

The specimens of the D. sordida from type collection, deposited in the herbarium P, were probably collected in Pico do Itambé area ( Forzza 1997) by Saint-Hilaire ( Fig. 5C ). However, in a Bromeliaceae survey in this locality, Versieux (2008) did not report the occurrence of D. sordida there, but only D. glandulosa L.B.Sm. & Reitz ( Smith 1967: 484) . In search for D. sordida in that locality, we also did not find the species, confirming only the presence of D. glandulosa . In this case the type population may have been extinguished or the holotype specimen was collected in some adjacent locality. Searching for historical literature to discovering the exact location where Saint-Hilaire collected his number 402 ( D. sordida type) was not successful because this naturalist often resumed his collection numbering and his field notebooks have not been accessed, so more than one time the corresponding number (402) was collected from Minas Gerais; the date of original collection is also impossible to be checked.

The synonymization of D. duarteana under D. sordida by Forzza & Wanderley (1998) was not explained there. However, Forzza (1997) clarified through studies of protologue and specimens from type-locality that D. duarteana was described by Smith from only one specimen collected in the Serra do Cipó at MG 010 road on km 137, which is ( D. duarteana holotype, Duarte 2749 RB!). Forzza (1997) argued that the specimen has all characteristics observed in D. sordida , except for the pedicels size, which are longer (up to 1.5 cm long), and floral bracts that exceeding the flowers. However, the latter characteristic can be seen in the flowers of the inflorescence base in other species. We could not verify the occurrence of such high lengths in the pedicels and bracts of D. sordida in any analyzed specimen, only in the holotype of D. duarteana . Thus, there is no overlap nor a gradient, but there is a gap in the pedicel length variation (3–5 mm long in D. sordida and 7–15 mm long in D. duarteana ). We have not located D. duarteana in situ (or specimens with D. duarteana characteristics) but E. Guarçoni (personal communication) claims to have located populations with D. duarteana characteristics. Further studies will be necessary to support D. duarteana as a distinct species or keep it as a synonym of D. sordida .

Conservation Status: —The species occurs discontinuously with an extension (EOO) of ca. 110 km in the Espinhaço Mountain Range, mainly in the Serra do Cipó. According to criterion B1 a(i, iii) of IUCN (2012), it is considered an Endangered species (EN).

Additional specimens examined:— BRAZIL. Minas Gerais: Conceição do Mato Dentro, 27 April 1978, G. Martinelli 4415 ( RB) ; Santana do Riacho, Km 119 da rodovia Belo Horizonte-Conceição do Mato Dentro, próximo ao córrego Alto do Palácio , 30 April 1981, A. M. Giulietti s.n. et al. ( RB 236422 ) ; Km 142 da rodovia Belo Horizonte-Conceição do Mato Dentro , 11 October 1986, M. G. Wanderley s.n. et al. ( SPF 44891 ) ; trilha para Cachoeira da Capivara , 25 October 1996, R. C. Forzza 241 ( SPF) ; Serro, 18 June 2014, H. M. Büneker 300 et al. ( HDCF) ; Vaccaria, Serra do Cipó, Km 138 da estrada da Conceição, 6 December 1949, Duarte 2106 ( RB) .

Figures

FIGURE 4. A–G. Dyckia sordida Baker (A–C. Saint-Hilaire 402) A. Holotype. B. Flower detail in holotype. C. Label detail probably written by Saint-Hilaire where there has been “Itambe” (Image credits: herbarium P, Barcode: P00438701). (D. Büneker 300 et al.) D. Inflorescence detail with flowers in bud. (E–G. Soares & Assis s.n.). E. Flower in lateral view. F. Fertile specimen in habitat. G. Rosette with lateral peduncle.

FIGURE 5.A–D. Dyckia ursina L.B. Sm.(Büneker 294 et al.). A. Rocky slope habitat with fire in the background.B. Simple inflorescence detail. C. Branched inflorescence detail. D. Specimen in habitat.