THE GENUS ROSA IN ‘LONDON CATALOGUE’ ED. 10 357 
iliated”’; and on p. 215 he refers to a Thirsk plant in which 
«the ae: is beautifully gland-ciliated.” Ciliate petals therefore 
occur both in R. mollis and R. pomifera, and cannot be relied on 
to Aomdiaas the one from the other. 
F. mollis var. recondita Pug. Mr. Ley’s contention is that 
this rose, which is very widely spread in Switzerland, occurs also 
in Britain, and bk continental botanists have wrongly d prea 
"gg 
it as a variety of R. pomifera, whereas it is really a var. of R. mo 
It is hardly likely that these continental botanists, — of chi 
rhodologists of the first rank, could have gone so far astray, more 
especially as they universally aeacribe it as the ater ep the 
of Herrmann. But an interchange of fresh specimens has 
enabled me to recognize edness Mr. Ley’s var. recondita really is. It 
is simply the large-fruited form of R. mollis, pene: we in this part 
of the country (Perthshire) have looked upon as the type of that . 
species. True, it mall subfoliar gata, and if it is 
to be considered on sag pany as a variety and not as the type, 
I do not object; but it certainly is the form which is nearest to 
the type of Smith. Mr. Ley’s discovery therefore amounts to 
this, that RB. pomifera Herrm. and R. mollis Sm., as regards the 
chief variety of each, cannot be separated. Surely the logical 
conclusion from this is what Crépin maintained so strongly and re- 
iterated _—_ and again, viz. that these two a species 
are in reality one, and that the characters relied on to separate 
ieee aes colour, shape and size of the leaves, the spiny fruit— 
are ‘merely, superficial and very variable, and at the most can only 
se} te 
para’ as varieties and not as species. Mr. er, too, 
seems to ancy been substantially of the same mind, for he joins to 
mollis synonyms var. recondita Pug. and var. Grenterit 
Déségl., the "iather a small-leaved and small- Fraited var. of R. pomi- 
fera, which, if it occurred in Britain, would certainly be set down 
as a var. of R. mollis. It is extremel likely, I think, that 
Mr. Ley’s R. pomifera Herrm., with ciliated petals but differing 
from the garden form, will turn out to be a discovery of the same 
kind as his discovery of var. recondita ug. 
. omissa Déségl. There can be no doubt that, in Scotland at 
least, we have forms of R. tomentosa Sm. on which the sepals per- 
sist till the fruit is fully mature, These may be con sidered as 
our forms pi memsaan§ with — foreign species st its sor i 
. suberecta Ley. As regards this form, my question was, 
How ecnie Wr te identify this = with RB. suberecta Woods, 
described by Woods and Mr. er as a form of R. mollis Sm. ? 
er. 
Mr. S species s to embrace more forms than one was 
suggestod partly by the terms of his diagnosis, and also because I 
see no real difference, so far as his descriptions go, between 
his R. suberecta and his R. Andreseiows and very little between 
his suberecta and his Sherardi 
