136 TRANSACTIONS OF THE [Sess. LXxv. 
has already been done in “Journ. Bot.,’ 1910. I propose 
merely to indicate the three groups which are represented 
in this country, with occasional reference to the forms which 
are found in the neighbourhood of Edinburgh, and to give 
some idea of the work which has already been done on 
this section of the genus. 
On the whole the hybrids are easily recognisable. Their 
foliage generally, but not always, recalls that of R. pim- 
pinellifolia in shape and colour. There is always and often 
a decidedly marked admixture of acicles on the stem and 
branches, and the fruit is in most cases very imperfectly 
developed. 
The largest of the three groups is Pimpinellifoliz x 
Villosze, collectively called R. involuta, Sm. Several var- 
ieties have been described, but these seem to run so closely 
into each other that it is very difficult to distinguish them, 
however well defined on paper. 
R. «nvoluta was first described in 1804 by Smith in 
“Flora Britannica,” but at that time was not recognised as 
a hybrid. Baker, in his “Monograph of British Roses,” 
1869, includes R. involuta in the group Spinosissime, with 
which of course it must have some affinity since R. pim- 
pinellifolia is one of the parents. But this is sufficient to 
show, I think, that Baker was not aware of the hybrid 
origin of the rose. It was not, in fact, for a number of 
years after the appearance of the monograph that Dr. 
Christ’ promulgated the idea that R. involuta is a hybrid 
of R. pumpinellifolia and R. tomentosa, a member of the 
subsection Villosz. 
Several years after Smith’s earliest description there 
appeared in “ Trans. Linn. Soc.,”’ 1818, an account by Woods 
of a variety of R. involuta which its author named R. sabini. 
This differs from the former in having its leaflets fully 
compound-glandular-serrate, its sepals pinnate, and _ its 
prickles more scattered. But since Woods’ description 
appeared, a number of intermediate forms have been 
figured, and under R. involuta, Sm. in the “London 
Catalogue Ed. 10,” we find a list of twelve varieties. 
R. sabini, Woods, is considered by Baker to be much more 
common than A. imvoluta, reaching its maximum frequency 
1 “ Botanisches Centralblatt,” 1884. 
