160 
spread over our county ; ranking, I think, next to urbica to form the fourth in 
commonness among the varieties of cantina. As far as I have observed, it is not 
confined to any one soil or situation. This and the last agree in having the styles 
commonly agglutinated, so as to form a column more or less projecting, similar to 
that in R. arvensis, though not so conspicuous. 
(i) Arvatica. This, as one of the glandular forms, is quite rare in our 
county, though it is frequent in the north of England. From Herefordshire I 
have a specimen gathered at Hoarwithy by Mr. B. M. Watkins, but I never 
gathered it myself in the County, till last year, when I met with it at a second 
station, on the Maes-coeds above Peterchurch, and at a third station in Orcop. 
(j) Dumetorum. Fairly common in Herefordshire ; but not common in a 
typical state. In our county it often runs so near to urbica as to leave one in 
perplexity to which variety a bush is to be assigned. Again, other bushes will 
have leaves of “‘ good” dumetorum, but fruit which would stand for dwmalis or 
lutetiana just as well. When both leaves and fruit possess the large fine character 
which marks dumetorum, it becomes a very fine and distinct Rose. 
(m) Tomentella. This neat and distinct variety is either wanting, or at 
least quite rare, on the light soils of Herefordshire; but appears to become much 
more common upon the heavy clay lands. I have gathered it at Marcle, and near 
Allensmore ; and recently in abundance at Weston Beggard. Ihave a specimen 
gathered by Mr. Burton Watkins at Hoarwithy ; but I never, myself, met with 
it in the Ross neighbourhood, nor in any part of the south of the county. 
(n) Andegavensis. We here enter upon a set of varieties of R. canina 
bearing aciculate peduncles. This, the first of them, can searcely be said to exist 
apart from lutetiana ; the only character separating them being the peduncular 
aciculi ; and these are often very weak and slightly developed. It is scattered 
and not uncommon, so far as T have noticed, all over the county. 
(o) Verticillacantha. This, as appears to me, is far more distinctly 
marked from dwmalis than andegavensis is from lutetiana. It seems to be common 
in many parts; but in Herefordshire (like the other glandular varieties) is of 
extreme rarity. Indeed we have no positive evidence yet of its occurring in 
Herefordshire proper ; but I have gathered it in the Llanthony valley, within the 
limits of our Flora ; and it occurs again abundantly round Malvern, on the 
Worcestershire side of the hills. Thus it occupies the extreme west and east 
flanks of the county. Round Malvern it often puts on the extreme clothing of 
aciculi, by virtue of which it becomes the R. aspernata of Deséglise. It is, in this 
state, a very marked Rose. I have gathered it in this state myself at Malvern, 
and have received beautiful specimens gathered in the same neighbourhood by 
Mr. R. F. Towndrow. T trust that it will be found to occur on the Herefordshire 
as well as the Worcestershire side of the hills. 
(p) Collina, Jacq. (Kosciniana, Besser ). This is not an abundant form 
in Herefordshire; and I must confess to knowing little about it. Specimens 
(which appear poorly marked) from several stations in Districts 1 and 2, have 
been communicated to me by Mr. Burton Watkins. Others, much better marked, 
