637 Calonje & al.  Typification of Zamia erosa TAXON 59 (2)  April 2010: 637–642 INTRODUCTION Orator Fuller Cook and Guy Collins, then botanists at the United States Department of Agriculture (USDA), co-authored a paper on economic plants of Puerto Rico (Cook & Collins, 1903) which included the following description for Zamia erosa O.F. Cook & G.N. Collins: Zamia erosa. MARUNGUAY. What appears to be a previ- ously undescribed cycad was found in the forests cover- ing the rough limestone hills to the south of Vega Baja. It is peculiar in having but one or two fronds with distant, broad, apically erose-dentate leaflets. The large fleshy root is hidden in crevices of the jagged rocks, but is dug out by the natives and used in the manufacture of starch. In his review of the taxonomy of West Indian cycads, Eckenwalder (1980) designated a specimen collected in south central Puerto Rico as the lectotype of Zamia erosa (Ponce: Coamo Springs, 14 Jun–22 Jul 1901, Cook s.n. , lectotype: NY!, Fig. 1; isolectotype: NY!), placing the name into synonymy under Z. pumila L. subsp. pumila. The specimen Eckenwalder chose as a lectotype for Zamia erosa was collected on the much drier south side of the island, at least 30 km away from the type locality on the northern side and separated by the Cordillera Central mountain range. Fur- thermore, it does not have the “distant, broad, apically erose- dentate leaflets” mentioned in Cook & Collin’s protologue, but rather has closely spaced, narrowly oblong leaflets with smooth leaflet tips consistent with Puerto Rican Z. pumila. as circum- scribed by Acevedo-Rodríguez & Strong (2005). Moreover, the specimen itself was not labeled as Z. erosa by either Cook or Collins. Instead the specimen bears a printed label from L.M. Underwood and R.F. Griggs, with hand-writing that identifies it as Zamia media Jacq. Zamia media was not listed in Cook & Collins’s (1903) treatment, nor was the locality where the specimen was collected mentioned as falling within the geo- graphic range of either Z. portoricensis Urban or Z. erosa, the two species they treated. However, Britton & Wilson (1926) assigned the name Z. media to plants with narrowly lanceolate or linear lanceo- late leaflets from southern Puerto Rico, differentiating these from the wider leaflet plants from northern Puerto Rico which they treated as Z. latifoliolata Prenlel. (with Z. erosa listed as a synonym), and from the narrower leaflet plants from south- western Puerto Rico which they treated as Z. portoricensis. Zamia latifoliolata was described as a plant with broad leaflets to 3 cm wide (Prenleloup, 1872). The type locality is in the Dominican Republic, within present day Santo Domingo. However, all recent workers who have studied Zamia from that country, e.g., Zanoni (1982) and Stevenson (1987), have treated them as belonging to a single species, Z. pumila. Zamia pumila not only has priority over all other names assigned to Zamia in Dominican Republic, it also provides the type for the generic name. On the basis of Eckenwalder’s (1980) selection of Commelijn’s drawing as the lectotype, the type locality of Z. pumila is Hispaniola where the species is only known to occur in present-day Dominican Republic. Other Zamia popu- lations occurring in southern Puerto Rico and Cuba are also currently treated as belonging to this species, e.g., by Stevenson (1987) and Sabato (1990). Fieldwork in the Dominican Republic by the first author in July of 2009 confirmed that wild Zamia pumila populations throughout the island were indeed morphologically similar and should be treated as a single species. Leaflet widths on plants within Santo Domingo and throughout the country typically ranged from 1–3 cm, confirming that the leaflet width attributed to Prenleloup’s Zamia latifoliolata is not exceptional and falls within the natural range of variation for Zamia pumila. In con- trast to the situation in Dominican Republic where individuals The typification of Zamia erosa and the priority of that name over Z. amblyphyllidia Michael Calonje, 1 Alan W. Meerow 2 & Dennis W. Stevenson 3 1 Montgomery Botanical Center, Miami, Florida 33156, U.S.A. 2 USDA-ARS-SHRS, National Germplasm Repository, Miami, Florida 33158, U.S.A. 3 New York Botanical Garden, Bronx, New York 10458, U.S.A. Author for correspondence: Michael Calonje, michaelc@montgomerybotanical.org Abstract The lectotype designation of Zamia erosa O.F. Cook & G.N. Collins by Eckenwalder (1980) is corrected to a neotype as it does not constitute original material on which Cook & Collins based their description. The neotype designation is itself superseded, as the specimen selected by Eckenwalder is in serious conflict with the protologue, a conclusion based on over- looked descriptive and geographic information in the protologue, and new collections recently made near the type locality. A new neotype is designated. We have further determined that Z. erosa is an earlier validly published name for the species currently known as Zamia amblyphyllidia D.W. Stev. Keywords nomenclature; Puerto Rico; Zamiaceae