Morphological phylogeny of Tradescantia L. (Commelinaceae) sheds light on a new infrageneric classification for the genus and novelties on the systematics of subtribe Tradescantiinae Author Pellegrini, Marco O. O. https://orcid.org/0000-0002-8783-1362 Universidade de Sao Paulo, Departamento de Botanica, Rua do Matao 277, CEP 05508 - 900, Sao Paulo, SP, Brazil & Jardim Botanico do Rio de Janeiro, Rua Pacheco Leao 915, CEP 22460 - 030, Rio de Janeiro, RJ, Brazil & Current address: Smithsonian Institution, NMNH, Department of Botany, MRC 166, P. O. Box 37012, Washington D. C. 20013 - 7012, USA marcooctavio.pellegrini@gmail.com text PhytoKeys 2017 2017-10-26 89 11 72 http://dx.doi.org/10.3897/phytokeys.89.20388 journal article http://dx.doi.org/10.3897/phytokeys.89.20388 1314-2003-89-11 9C1BFFADD6208818FFDAFFD9FC60390A 1138297 1.1. Elasis guatemalensis (C.B.Clarke ex Donn.Sm.) M.Pell. comb. nov. Fig. 6L Tradescantia guatemalensis C.B.Clarke ex Donn.Sm., Bot. Gaz. 18(6): 210. 1893. Lectotype (designated here). GUATEMALA. Jalapa: Laguna de Ayarza, fl., fr., Sep 1892, Heyde & Lux 3886 (US barcode US00045211!; isolectotypes: NY barcode NY00039636!, P barcode P02173850!) Nomenclatural notes. Hunt (1994) designated the specimen Heyde & Lux 3515 (K) as the lectotype for T. guatemalensis . This specimen was indeed examined by Clarke, being annotated as a new species and presenting drawings with diagnostic features for the new species. Nonetheless, after carefully analyzing the protologue and the collections of K, NY and P, I noticed that the collector's number for the specimen at K is actually "3519" , which is annotated in the specimen by the original collectors and by Clarke, instead of "3515" as cited by Smith (1893). This lead me to conclude that Smith (1893) had limited access to this specimen, and probably did not base his diagnosis on it. Thus, the lectotype designated by Hunt (1994) must be disregarded, since the collector number is incorrect, and the specimen chosen by him does not correspond to a specimen of T. guatemalensis and was not cited in the protologue. On the other hand, the collection Heyde & Lux 3886 was clearly available to Smith, being housed at the NY, P and US herbaria, and was most probably studied by him. The US specimen is greatly preserved, presenting flowers and fruits, and is a good option for a lectotype. Thus, it is here designated as the lectotype of E. guatemalensis .