182 BRITISH FLOWERING PLANTS chap. 



the anthers are not. In the second stage the anthers 

 are open, but turned away from the stigmas ; finally, 

 in the third they bend over towards and fertilise 

 them, if they have not been already fertilised from 

 another flower. 



Rosa 



In this variable genus there are great differences of 

 opinion as to the limits of species. Bentham admits 

 five British species, while other botanists extend the 

 number to fifteen or twenty. I often wonder that the 

 older school of naturalists who regarded species as fixed, 

 aboriginal, and invariable, were not staggered by such 

 genera as Rosa, Rubus, or Hieracium, where the forms 

 are so various and so variable. 



The Rose is much visited by insects for the sake of 

 the pollen, but does not produce honey. The numerous 

 stamens ripen at the same time as the pistil ; but from 

 the convenient position of the latter, insects very fre- 

 quently alight upon it, and thus fertilise it with pollen 

 from other flowers, though self- fertilisation probably 

 often occurs. The carpels are numerous, one-seeded, 

 hairy, and embedded in the receptacle, which thus 

 forms a more or less bottle-shaped tube, open at the free 

 end, the walls of which become succulent when ripe, 

 forming the red "hips" which are so great an orna- 

 ment of our hedge-rows in autumn and winter. Roses 

 are well protected by numerous sharp curved prickles, 

 which in the climbing species are also useful as grapples. 



In most cases of upright flowers with separate sepals, 

 the sepals are more or less similar. In the Rose the 

 differences are considerable, and have long been observed. 

 Mr. Landon has kindly called my attention to the 

 following Latin lines, which well describe them : — 



Quinque sumus fratres, unus barbatus et alter, 

 Imberbesque duo, sum semiberbis ego. 



The reason, I believe, is to be found in the arrangement of 

 the leaves in the bud. They overlap one another to some 



