Blake — A Revision of the Genus Viguiera 19 
G. sericea Klatt (no. 31) corresponds to the series Grammatoglos- 
sae of Viguiera in all its essential characters, and is in fact so 
closely similar to V. grammatoglossa DC. that the only known 
collection (Purpus 5615) beside the type was distributed under a 
name which is synonymous with V. grammatoglossa. The last two 
species of Robinson and Greenman’s revision (G. scaberrima 
(Benth.) Greenm. (G. platylepis Gray) and G. calva (Sch. Bip.) 
Gray), with the later G. Pittiert Greenm. and G. auriculata Brande- 
gee, go naturally into the genus Tithonia, of which they have the 
characteristic fruiting peduncle and the other important features.' 
The first two were in fact originally described as species of Tithonia, 
and were included in that genus by no less an authority than O. 
Hoffmann, as is evident from the synonymy in the Pflanzenfamilien. 
It thus appears that the group to which the name Heliomeris of 
Nuttall applies contains, after the exclusion of extraneous forms 
from Robinson & Greenman’s revision, some 33 species (G. cineras- 
cens and G. Kunthiana, not available for study, being omitted), 
which fall readily into three main groups whose direct affinities are 
with the genera Viguiera (25 species), Hymenostephium (4 species), 
and Tithonia (4 species), from which they differ only in the absence 
of a pair of linked characters (pappus and pubescence of achene) 
which are shown by numerous species of not remotely related 
genera to be of no more than varietal or specific worth, and the 
artificiality of which is shown conclusively by the fact that as now 
applied they separate into different genera plants growing side by 
side and distinguished in other features literally not by the dif- 
ference of a hair. It seems impossible on any grounds to defend 
the reference of Gymnolomia flava and Viguiera Ghiesbreghtii, G. 
megacephala var. simulans and V. pachycephala, and G. costaricensis 
and Hymenostephium cordatum to different genera when no specific 
difference can be shown between them, while the relationships of 
the other species of Heliomeris (Gymnolomia auct.) to other groups 
of Viguiera, Tithonia, and Hymenostephium are so close that any 
loss in ease of definition caused by their union should be more 
than compensated by the increased approach to a natural classifi- 
cation thereby secured. 
As regards the evolution of the typical pappus-bearing species of 
Viguiera, it is difficult to speak with much assurance. The most 
1 See footnote, p. 9. 
