2 THE SUBSECTION EU-CANINZE) OF THE GENUS ROSA 
paratively — Meantime, continental botanists have not 
been i 
e, and the number of recognized species has vastly in- 
creased, as is sii tined by the works of Déséglise, Crépin, Keller, 
Rouy & Foucaud, &c.; and during the past two or three sI 
year: 
have been endeavouring, by means of books and herbaria, to pick 
up the threads of their labour 
In studying the subsection sae canine I have had the immense 
advantage of being able to consult Déséglise’s: fing collection, 
sell was acquired for the National Herbarium 84. I have 
also ined Smith’s & Woods’ herbaria at the Linssen Society, 
the British collection of the National Herbarium, and the general 
collection at Kew, as well as the writings of the authors above 
and in many cases I am ‘ to D disungiah al een specimens 
of different species named by their own authors. In going through 
aerate s herbarium, some of the covers in which contain nigrce 
or t sionally more s 0 on 
can fail to be struck with the great rep tig of the individual 
specimens under one name; in fact it wo e no great exaggera- 
tion to say that no two quite = it that if one were to wor 
coir oe one species, be ein uided by dese ript 0 alone, there 
o di 
o happens in many cases that Déséglise’s British examples 
differ more widely from the average of the continental specimens 
and from their descriptions than do any _— ae in the 
covers. Finally, there is the impossibili 
: 5 s ; Ps 
troubled to quote types, and when they do it is iB obvious 
that the species have been founded on single s S. 
Before enc narees g to give in detail Elia and notes on 
the species and varieties which have been identified as British, I 
will call Soin to the principal rae ere which have been 
relied ae for their differentiation 
halit is generally mentioned in descriptions, but in 
= sage wt many 0 es to ur Ro ow in hedges, the habit 
ickles may present more rr characters 
aaa by foreign botanists, but I have found din ae Tittle 
8 ¢ value. 
