ON THE NECESSITY FOR A NEW MONOGRAPH OF ROSES 45 



southern counties. Its varieties are very numerous, and 

 several among them have been raised to the rank of species : 

 R. subglobosa, Sm., R. scabrinscula, Sm., etc. Some have the 

 sepals remaining reflexed after flowering, and more or less 

 early caducous ; while others have them erect and crowning 

 the receptacle till complete maturity, then caducous. It is 

 probable that ultimately there will be distinguished among 

 these numerous varieties two or more natural groups, as has 

 occurred among the old varieties of R. canina. Up to the 

 present time the classification of the varieties of R. tomentosa 

 has been artificial. If the group of varieties or forms with 

 sepals reflexed is always easily distinguished from R. mollis, 

 by the single character of the direction of the sepals, it is not 

 the same with the group that have the sepals erect on the 

 receptacle during ripening. Although the erection of the 

 sepals may be less early and less marked in R. tomentosa 

 than in R. mollis, one must, to distinguish these two types 

 from one another, have recourse to other distinctive characters. 

 These are not wanting ; but it is necessary to be able to 

 grasp them well, not to be duped by mere appearances. 

 These appearances, incorrectly appreciated on herbarium 

 specimens, have led many botanists to confound these two 

 types, which are essentially distinct, and thus to render the 

 geographical distribution very uncertain. This confusion has 

 even led, as a consequence of it, to some botanists uniting 

 R. tomentosa and R. mollis specifically, and to not seeing in 

 them more than one and the same species. At certain 

 places on the mountains of Switzerland and of France one 

 observes a species very near R. tomentosa, and which seems 

 to be a mountain race. Its characters have appeared 

 sufficiently remarkable and constant to warrant its separation 

 from Smith's plant as a subordinate species. We speak or 

 R. omissa, Desegl., which for a long time Swiss and French 

 botanists have confounded with R. mollis. Up to the present 

 time this species does not appear to have been determined 

 from the British Islands, where, indeed, it appears not to 



exist 



Rosa mollis, Sm. 



is rather abundantly distributed in the north-west of Europe, 

 in Scandinavia, in Finland, in the Baltic districts, and in the 



