To one unfamiliar with the variation exhibited by 
flowers of Stelisand unmindful of the changes that take 
place in fleshy organs when they are dried for preservation 
in herbaria, the analytical drawings of the lip of a flower 
taken from Schlechter’s type of S. ovypeta/a may seem 
to be quite different from the lips of the Honduran plant. 
With regard to this point it should be borne in mind that 
dried flowers of the Honduran plants agree almost per- 
fectly with the Guatemalan type. In two flowers taken 
from the same raceme of Edwards no. 545, one of the 
lips showed the form exhibited in Schlechter’s type of 
S. ovypetala, while the other agreed in every way with 
the lip represented in the plate of Edwards no. 545. 
In Costa Rica there are several species that are close- 
ly related to Stelis rubens Schltr. They are hardly in the 
‘ategory of geographical varieties. Subtle differences in 
the structure of the labellum and in the gynostemium 
serve to distinguish them, although these differences are 
difficult to describe. ‘The Costa Rican species are stouter 
and larger-flowered than S. rubens and when placed side 
by side with it are readily separated from it. The first 
of these Costa Rican species to be described was Stelis 
Tonduziana Schitr. This is the largest flowered species 
of the group, with the lateral sepals exceeding 4 mm. 
in length and at least twice larger than the sepals of 
S. rubens. Furthermore, SS. Tonduziana is characterized 
very markedly by its 5—nerved sepals. This is the same 
species as the one described by Schlechter as S. mirabilis. 
In 1924, Paul C. Standley collected several specimens 
of a remarkable species that I published under the name 
of Stelis Standleyi. The chief difference between this 
species and S. Tonduziana is found in the sepals having 
three instead of five nerves, although the flowers of the 
type are considerably smaller, and in this respect easily 
differentiated. 
[13 ] 
