2 Contributions from the Gray Herbarium 
close relationship existing between all the species now referred to 4 
Clibadium, as well as the gradations observable in these two fea- ~ 

tures themselves, renders such a course as artificial as it would be - 
unsatisfactory. In all the species the outer (female and fertile) a 
florets, 3-37 in number, are subtended by chaff which may be con- 
sidered either as receptacular pales (as they are taken by writers — 
generally) or as inner scales of the 1-5-scaled genuine involucte, 
to which they are very similar in every character; the difference is ie 
after all one of words only and of no great consequence. In the five 
species of the section Trizidium now known the inner (hermaph — 
rodite and sterile) florets of the disk are also subtended by pales, 
considerably reduced it is true but easily found; but in the twenty — 
species of Euclibadium they vary greatly, being sometimes entirely — 
absent, sometimes present only at the base of a few florets and a 
it were sporadically, and sometimes at the base of all but the inne 
most disk-florets. The gap between such species as C. strigilloswm 
with 16-flowered heads, only the five female florets provided with 
pales, and C. Eggersii, with 47-flowered heads paleaceous through — 
out, is bridged by such species as C. polygynum with 37-flowered 
heads of which only the innermost of the disk are without pales. < 
The amount of pubescence on the sterile achenes of the disk 6 2 
also not without variation in some of the species of each sectioR 
and is of much less importance as a possible generic character. 
: Of the some twenty-five species of Clibadium now known (in- 
cluding six here first published) only three are in agreement with : 
the other genera referred to the Millerinae in the possession of : 
less than ten flowers in the head. Hoffmann’s statement in 1 
(Nat. Pfl. iv. pt. 4, 214) of the number of female florets as “ 1-2 : a 
was very incorrect even for the species then known, no species — 
having less than 3, while C. erosum has about 18 and C. fragiferum : 
about 28. Hoffmann’s error may have derived from Bentham’ — 
description of the female florets as “ 1-2-seriati,” but surely neve — 
from dissection of a specimen. | 
From what has been written above it seems beyond dispute that 
Clibadium should be removed from the Millerinae and referred 10 — 
the Melampodinae next to Ichthyothere, from which it differs chiefly — 
in habit and in the more or less complete reduction of the receP- _ 
tacular pales. The typical or original group would seem to be — 
Trixidium, with many-flowered heads and a receptacle paleaceous — 
Ree es 


