COCKERELL : NORTH AMERICAN SPECIES OF HyMENoxys 487 
H. floribunda type. H. floribunda utilis. H, floribunda Hl, Earlei. 
arizonica 
Height of welds: Heise of welt ae Hei eight of well-| Height of well-de- 
veloped plant 30 cm. veloped pen 26 cm. developed plant 25 veloped plant 32 cm. 
Lower stem-leaves oliage eaniiitiatly “"Foli liage essentially | Lower stem-leaves 
55 mm. long, cme type. | ae ire ees ype. about 125 mm. long; 
their divisions very their segments much 
narrow, r mm. broad broader, up to 2,5 
or less. mm, 
Stems numerous, Stems st fie nume Stems several, not. Ste fe 
not very woolly at ous, ve oolly a very woolly at base ; am ‘ all fics 
base; the bases of base; the! casi of the the bases of the at base; base of stems 
the-old leaves ferru- old leaves pallid. leaves pallid. reddened, which is 
ginous. not the case in the 
; other three. 
Heads (excl. bya) Heads (excl. rays) Heads(excl. rays) Heads doops rare) 
about 7 mm, dia about6mm. diam. about 8 mm. diam. about 7 mm. dia 
| thew th e 
others, except per- 
haps artzonz 
Bracts pale, outer os pale yellow Bracts sgt asin Bracts formed 
bracts more resinous ish; Fier a about as in Aori- 
than in wézilis, — united psa bunda, but outer 
nach Se so muc bracts conspicuously 
aah with dark 
Pappus and achene- Pappus andachene- Pappus ferruginous Yates and achene- 
— ee ferrugin- hairs shining silvery at base, the rest hairs shining ferru- 
_ white. whitish;  achene- Bi the pa ppus 
— pale ferrugin- cae tk. Achene 
sor 
Pappus- one a pl tarde — P appus-sca ne es Pa i 
pointed (ar e) long- poin ot broad, ‘8 ho ae sett poseted, 
about 24 seen of quite so rue as in pointed. as in artzonica. 
disc-corolla. typical floribunda. 
Rays yellow. ig pee yellow Rays rather light Rays orange. 
orange. 
Type from NM. Type from vicinity. Type from near | | Type from Mancos, 
of hea g Fé, N. es bens Visio, Colo,, Flagstaff, Ariz.,7000 Colo. 
ft. 000 ft. _ ft., ft., MacDougal. 
macrantha the receptacle is hemispherical and somewhat fimbril- 
ose, these species having in part the characters of Gadlardia ; in 
floribunda and wzilis the receptacle is conical and bare.” 
When I published this description, I was half inclined to treat 
utilis as a species, and did not in the least doubt that it should 
have at all events subspecific rank. Further studies, however, 
have shown that the characters of the bracts and the color of the 
Pappus are not to be relied upon to the extent I had supposed ; 
and moreover, the real type of A. floribunda, Fendler’s plant, has 
the bracts nearly as in wéi/is. The characters of wéd/is are therefore 
largely evanescent, and until further studies can be made, especially 
