D. candida, the shorter corolla and corolla teeth, and 
especially the ovoid fruit, are indicative of hybridization 
involving D. sanguinea. Danert (5) has not questioned 
the validity of D. arborea and has suggested that D. 
candida, D. affinis, and even D. suaveolens may be syn- 
onymous with the Linnaean species. 
D. cornigera Hooker in Curt. Bot. Mag. pl. 4252 
(1846); Lagerheim in Bot. Jahrb. Engl. 20: 668 (1895) 
is also similar to D. candida, but it differs in the very 
short corolla, in the more emarginate corolla lobes, in 
the capitate stigma, and especially in the calyx which 
tapers to a long subulate point. Though such a calyx is 
very rarely seen in the tree Daturas, several collections 
show intermediates between it and that of D. candida 
(Bristol 738, 1213, 1387; GH). The ovoid fruit, recorded 
by Lagerheim (Joc. cit.), the size and form of the corolla, 
and the capitate stigma are all characters associated with 
D. sanguinea. 
D. rubella Safford tbid. 185 (1921), with acuminately 
tipped leaves and a spathe-like, caudate-acuminate calyx 
again suggests hybridization between D. candida and D. 
sanguinea, though the red corolla (18-14 cm.) and the 
pistil and stamens are much shorter even than in D. 
sanguinea. 
SPECIMEN EXAMINED: 
Ecuapor: Rose, Pachano & Rose 22828 (Tyre; US). 
D. mollis Safford ibid. 183 (1921) resembles D. candida 
but for the much longer calyx, stamens and pistil and its 
light pink corolla. 
SPECIMEN EXAMINED: 
Ecuapvor: Rose & Rose 23418 (Type; US). 
A uniquely attractive hybrid in highland central Peru 
(Ochoa 601; NY) has a very small (18 cm.) white flower 
[ 243 ] 
