SUMMARY 155 
the Herb Robert itself being a case in point. The real 
interest is not in the disparity but in the uniformity, 
both in the shortness of the minimum resting period 
and in the power, under suitable conditions, of germin- 
ating promptly ; moreover, my observations on this 
point extend as yet only to six of our twelve species 
of Geranium and one of our three Erodiums. 
I may commend to the reader’s notice the leafy 
winter rosettes of our annual Cranesbills and of the 
Storksbill; they are beautiful in themselves and in 
the arrangement of their foliage. The large-flowered 
species of Geranium that I have grown generally die 
down to the ground-level in the autumn after the usual 
manner of herbaceous perennials, save for a few 
radical leaves. 
The reproductive arrangements agree in the absence 
of spontaneous vegetative propagation, and there is 
not a great deal of difference in the pollination plans ; 
it is rather a case of the same tune with some decidedly 
pretty variations. 
As regards the separation of the sexes in time, 
instability characterizes the Storksbill especially, 
and our annual, as opposed to our perennial Cranes- 
bills, are not altogether innocent of the charge. Some- 
times the anthers have the first innings, but it is more 
usually the stigmas: simultaneous maturity is by no 
means uncommon, and there is always a decided over- 
lap. 
The perennials, on the other hand, are considerably 
more steady, and in all of them the rule is for the 
anthers to mature before the stigmas. I know no 
instance of the order being reversed, but in the Wood 
Cranesbill on rare occasions they both mature at the 
same time. 
