234 Flowers and their Pedigrees. 



its calyx instead. It is much more natural, therefore, 

 to suppose that the stanch-wound, with its few sta- 

 mens and its clumsy device of a coloured calyx 

 instead of petals, is descended from the salad-burnet, 

 than that the pedigree should run the other way ; and 

 there are many minor considerations which tend in the 

 same direction. Most correctly of all, we ought perhaps 

 to say that the one form is probably a descendant of 

 ancestors more or less like the other, but that it has 

 lost its ancestors' acquired habits of wind-fertilisation, 

 and reverted to the older methods -of the whole tribe. 

 Still, it has not been able to replace the lost petals. 



I ought likewise to add that there are yet other 

 roses even more degenerate than the burnets, such as 

 the little creeping parsley-piert, a mere low moss-like 

 plant, clinging in the crannies of limestone rocks or 

 growing on the top of earthy walls, with tiny green 

 petal-less flowers, so small that they can hardly be 

 ^jstinguisMted with the naked eye. These, however, I 

 cannot now iind space to describe at length ; and, 

 indeed, they are of little interest to anybody save the 

 professional botanist. But I must just take room to 

 mention that if I had employed exotic examples as 

 well as the familiar English ones, I might have 

 ,tfaced the lines of descent in some cases far more fully. 

 It is perhaps better, however, to confine our attention 



