247 
THE COLLECTION AND IDENTIFICATION OF ROSES. 
By rue Rev. A. Ley, M.A., anp Masor A. H. Wotury-Dop. 
heading may facilitate the study of our British roses, in which, it 
may be feared, we have fallen considerably behind Continental 
authors. There has no doubt been a tendency on the part of the 
latter to establish species on inadequate grounds; British authors, 
on the contrary, have n too conservative, in some cases only 
recognizing considerable aggregates, and in others continuing the 
use of totally incorrect names. We venture to suggest that the 
latter fault has been due to the too confident acceptance of names 
which have been given, doubtless in error, by competent authority, 
without any attempt at comparison with authors’ descriptions or 
with Déséglise’s magnificent collection of roses at South Ken- 
sington, which was purchased by the Natural History Museum 
in 1884 
The siz 
collection, and though these have their importance, they are far 
less indispensable than fully-developed fruit, without which it is 
often not easy to determine even the group to which an example 
should be referred. Specimens are also very commonly too small. 
The end of a flowering-shoot rarely shows characteristic prickles, 
which are to be found on the barren shoots, or on the old stems 
from which the flowering-shoots grow. Hither of these are 
admissible, but the very strong shoots of the year, arising from 
the rootstock, should be avoided, because their strength gives 
deceptive characters to the prickles and leaves which they bear. 
A few notes on the characters afforded by the various organs 
may be helpful. : . 
Habit, colour, &c.—As with Rubus, so with Rosa, there is 
much to be deduced from the habit and general appearance. The 
direction of the stem, and its colouring, as well as that of the foliage 
and prickles, though not easy to express in words, may often help 
towards identification. Notes made on these points at the time 
