JOURNAL OF THE ARNOLD ARBORETUM THE REDUCTION OF RUSBYANTHUS AND THE TRIBE RUSBYANTHEAE (GENTIANACEAE) RICHARD E. WEAVER, JR. THE SYSTEM FOLLOWED by most taxonomists for the classification of the Gentianaceae is that of Gilg (1895). In this system five tribes of the Gentianoideae are recognized, all distinguished on the basis of pollen grain characters: Gentianeae Gilg (40 genera), Rusbyantheae Gilg (1 genus), Helieae Gilg (15 genera), Voyrieae Gilg (1 genus), and Leiphai-meae Gilg (2 genera). With more adequate collections since Gilg's time, better equipment for the study of pollen grains, and utilization of a broader spectrum of characters, several of these tribes have been shown to be either unnatural or unnecessary. For example, Voyriella Miq., one of the two genera in Gilg's Leiphaimeae is certainly more correctly classi-fied among the Gentianeae; the other genus, Leiphaimos Cham. & Schlecht., is probably congeneric with Voyria Aublet, therefore completely eliminat-ing the tribe. In this paper, the elimination of the Rusbyantheae is pro-posed in addition. Macrocarpaea cinchonifolia (Gilg) Weaver, comb. nov. Rusbyanthus cinchonifolius Gilg in Engler & Prantl. Nat. Pflanzenfam. 4(2): 95. 1895. LECTOTYPE: Bolivia: Mapiri, Rusby 1173 (NY). M. pachystyla sensu Ewan, Contr. U.S. Natl. Herb. 29: 245. 1948. To judge from the few available collections, the distribution of thisspecies ranges from 1000 to 1800 meters in the Andes of the Peruvian De-partments of Huanuco and Puno and the Bolivian Department of LaPaz. It is closely related to Macrocarpaea viscosa (R. & P.) Gilg, M.corymbosa (R. & P.) Ewan, and M. pachystyla Gilg, the three speciesfound by Nilsson (1968) to have pollen of what he called the "M. corym-bosa-type." Gilg proposed the tribe Rusbyantheae to accommodate Rusbyanthuscinchonifolius Gilg, the pollen grains of which were supposedly unique inthe family. He described them as lacking furrows but with three equa-torial pores and with the exine beset with numerous large processes stand-ing apart from each other at regular intervals. However, Nilsson (1968,1970) recently described the pollen of Rusbyanthus as being 3-colporate(with 3 aperturate furrows), the normal condition in the family, and henoted a fine reticulum between the large, wartlike processes. In addition,he found that three species of Macrocarpaea (Griseb.) Gilg, the genusclosest to Rusbyanthus in Gilg's treatment, but in the Gentianeae-Tachii-nae, possessed pollen of the same type. In Macrocarpaea the exine pattern of the pollen grains is typically acoarse reticulum, the muri of which are 0.7-2 pm. broad. Nilsson noted[voL. 55
BioStor
