narrow winged and glabrous, at least some leaves are triplinerved, and stamens of whorl III tend 
to be only with two sporangia, while the former has twigs more or less pubescent, ridged, leaves 
always pinninerved, and anthers of whorl III with four sporangia. 
I have not found duplicates of the Schiede 87 collection. Thus, following 
recommendation 9A.4 of the International Code of Nomenclature (Greuter et al., 1994), since no 
other collections were mentioned in the original description, I accept the specimen housed at B as 
the holotype. That Cerro Colorado is in this part of the state of Veracruz was found reading 
Schiede's (1828) reports on his travels along with Deppe in Mexico. 
No further collections of the species have been made for almost 150 years.The second 
collection (Ventura 11880) is just a few miles northwest of the original locality. Schiede's and 
Ventura's collections (both with fruits) are amazingly alike, and the fact that they come from the 
same region in Mexico supports the idea that they belong to the same species. However, the other 
material I place under this name comes from the Pacific region and presents a wide 
morphological variation. Two reasons made me put them together (at least temporarily): first, the 
presence of some specimens from the Pacific region that closely match the material from the 
Atlantic area, pointing out they are conspecific; secondly, with only two individual collections 
from the Atlantic area I cannot determine how variable the populations are in this region and 
assess their overlap with populations in the Pacific region. However, the great separation 
between them without any other population somewhere in the middle, rather supports the idea of 
two different species. 
Specimens examined. MEXICO. Guerrero. San Luis Acatlan, Mojonera de Atotonilco a 
50 
