132 THE BRITISH ROSES 
least in the South and Midlands, forms, with that of R. dumetorum, 
the largest portion of our Rose Flora. In the north it is replaced, 
in part at least, by the groups of R. i and &. corvifolia, and 
by t Villose subsection. Its mbers are sometimes with 
difficulty distinguishable from si of the RB. glauca group, and, 
as has already been mentioned under the R. stylosa group, they 
sometimes bear the leading features of that section. The entire 
absence of hair, even from the midribs beneath the leaflets, seems 
to be the only certain distinetiod: from the groups of &. dwmetorum, 
= and even Borrert. 
Its subgroups are quite artificial, although they contain some 
of the oldest lastablichol species in the genus. The Lutetiane 
have their leaflets quite uniserrate, without subfoliar glands, and 
-with smooth peduncles. If the leaflets are eglandular, with their 
teeth all or mostly biserrate, and smooth peduncles, they belong 
to the Dumales, and the Transitorie come between them, having 
the lower leaflets of the flowering shoots mostly biserrate, and 
the upper uniserrate. The last-named subgr roup is one that is 
: b 
frequent British member of this subgroup. It has large, often 
glaucous leaflets, longly acuminate, and frequently more or less 
narrowed below, rather large ovoid or subellipsoid fruit, hispid styles 
and often rising sepals. &. globularis Franch., R. syntrichostyla 
ip., var. ramosissima Rau, and R. viridicata Pug. are also mem- 
bers of this subgroup 
Glandular-hispid ‘peduncles, sometimes very sbgnty so, and 
leaflet g feat of the Ande- 
alas subgroup, while : strongly biserrate i ra ith subfoliar 
glands on the secondary nerves distinguish the Scabrate. All five 
subgroups contain members very difficult to separate from one 
. : 
ern me features, do not always occur in the combinations 
For this reason, as well as on account of the striking 
sai between species which are now assigned to quite 
different ‘Groups or $s paar because. of their technical paneer 
absence oO ities or poms either on the leaflets or the comes 
and sim: care ae as varietal ones ouly. Our species o 
= Eu-canine might thus be reduced to quite a few, but each 
have several, and perhaps some a conside number of, 
on This would of course Id step, and the determi- 
distinguishing species, which are are practicall identical, simpl 
pewrer a one sta a few hairs on a midrib which on : 
ee tareren meta tia or ping fr > 
