116 ROSA ABYSSINICA. 
most delicate flesh colour; the styles long, exserted, 
but not united. It has been found in the neighbour- 
hood of Wurtzburg by Rau. 
The union of styles was long ago pointed out in 
R. arvensis by Lachenal and adopted by Haller and 
Villars. Afterwards it was strangely neglected, an 
has only been reconsidered within a few years. M. De 
Candolle was the first to employ it as a means of form- 
ing a natural assemblage among Roses, in his Hortus 
Monspeliensis, where he defines six species from which 
the last is to be excluded. I have four to add; and R&R. 
setigera of N. America has the same structure; but, 
on account of its habit and subulate stipulz, belongs to 
my division Banksiane. 
63. ROSA abyssinica. Tab. 13. 
R. surculis -scandentibus, aculeis confertissimis fal- 
catis, foliolis ovatis sempervirentibus, calycibus pe- 
dunculisque tomentosis. 
R. abyssinica Brown! in Salt’s Abyssin. app. lxiv. 
Hab. in Abyssinia Salt (v. s. sp. herb. Banks et Lam- 
bert.) | , 
This is one of the very few Roses indigenous to 
Africa. It was first noticed as a distinct species by 
Mr. Brown, in his appendix to the travels in Abyssinia 
of Mr. Salt, who discovered it. It can be confounded 
with nothing except R. sempervirens, from which it 
differs in the following particulars: its leaflets are 
shorter with a little stalk, broader towards the point 
than at the base; the petioles are exceedingly rough 
with unequal glands and setz; the peduncles and calyx 
are covered over with a thick down; and the prickles 
are exceedingly numerous and strong. 
