Blake — A Revision of the Genus Viguiera 7 
range — from San Luis Potosi to Colombia and western Vene- 
zuela — a fact of considerable significance in any conjectures as to 
their relationships. In such a case as is presented by the ‘plants 
under discussion — the occurrence in a common area of several 
closely related groups — it is natural to suppose that the one 
most closely allied to some group of much wider range represents 
most nearly the ancestral stock, especially when as in this case the 
distinctive features of the other more localized groups can be con- 
sidered in the light of reductions. The group of Gymnolomia above 
referred to presents this most clearly in its total lack of pappus, a 
feature of common occurrence in the Verbesininae among varieties, 
species, and genera which show the most certain evidence of 
derivation from: pappus-bearing forms. The relation of three 
species of Gymnolomia (G. microcephala Less., G. guatemalensis 
(Rob. & Greenm.) Greenm., and G. costaricensis Benth.) to the 
Mexican and Central Ameriodn species of Hymenostephium is so 
intimate that they can only be considered as derivatives of the 
latter or of a very closely allied ancestral stock. 
The genus Hymenostephium Benth., as at present interpreted, 
is composed of three herbaceous species ranging from Mexico to 
Colombia, all closely related, with a single much more distinct 
frutescent representative in Colombia. They are distinguished 
from the section Diplostichis of Viguiera, which they closely re- 
semble in all other features, solely by their pappus, which is made 
up of several subequal membranaceous-scarious scales of about 
equal length, with no distinction into awns and squamellae of dif- 
ferent form such as is found in Viguiera. This type of pappus is 
of rather uncommon occurrence in the Verbesininae, and is cer- 
tainly not to be looked upon as primitive. As an approach to the 
same condition is shown by various species of Viguiera — notably 
by the subgenus Yerbalesia of South America, and by V. brevifolia 
and V. Brandegei of Mexico, the latter a member of the section 
Diplostichis itself — it is evident that a tendency to it is of some- 
what wide distribution in the genus Viguiera, and its occurrence in 
V. Brandegei is sufficient to make the suggested origin of Hymeno- 
stephium from some past or present members of the section Diplo- 
stichis a matter of the greatest probability. Obviously related to 
the fruticose Hymenostephium angustifolium Benth. of Colombia 
is Gymnolomia Goebelii Klatt of Venezuela, an apparently frutescent 
