7 
5 ENGELMANN—CUSCUTA. es ; 
in the merican Cue scute I have often found them interme- 
cha It is scarcely necessary to add that only ri 
seeds ou be examin unripe ones, especially when 
r hard, have led to the strangest mistakes; winged 
margined seeds, described by authors, are such unripe seeds. 
shag Spe seeds are smoother and larger, when soaked, than 
ripe 0 
The cakes has been supposed to offer good characters; 
but I have reason to believe that those embryos with one or 
lacece proper, where only one es ‘ ilsoni a ls gamosepa- 
The specific characters of Cuscute are found in the thick- 
ness of the stem, but principally in the P yubanecuid and in 
the ere organs of the flower and 
inflorescence ae lage with the hears or absence of 
hifncke within it offers good characters, less so the presence or 
proportion of pedicels. 
The shape and proportion of calyx and corolla and of their 
parts (tube and lobes) furnish important but not unchangeable 
characters. Their texture must also be ee and often gives 
an important clue to the ee of specie 
It is unnecessary to re what has ery said by former 
monographers about these coco but it may not be useless 
to indicate a few facts not so clearly stated by them. 
he tube of the a generally more or less campanulate 
or ae is angular in some species, the angles corre- 
sponding to © commissure, or to the midrib of the sepals. 
Its lobes are col or less deeply divided and are often auri- 
cled at — rei overlapping these characters, however, are 
not very con able, as they not rarely depend on 
the rich Remerverse ve consequent vigo growth of the 
parast ite. 
species, and either fleshy or membranaceous, (often very thin, 
