156 THE FERTILISATION OF FLOWERS. | PART II. 
cross-fertilisation is thoroughly insured through the separation of 
the sexual products both in time and space; (2) in G. pyrenaicum, 
self-fertilisation never takes place in the earlier period, but it takes 
place regularly though not abundantly tow ards the end of the 
flowering period if insect-visits do not occur,—cross-fertilisation is 
insured, as in the two preceding species, if insect-visits occur early, 
but if they are delayed till the second stage they result more easily 
than in the other two species in self-fertilisation also; (3) in 
G. molle, even if insect-visits occur early and abundantly, self- 
fertilisation is not excluded; and it takes place to a greater extent 
than in the former species long before the flowering period ends, if 
insect-visits do not occur,—though self-fertilisation is at no time 
excluded, cross-fertilisation is favoured in case of insect-visits, by 
the position of the reproductive organs in both stages of the 
flower; (4) in @. pusillwm, self-fertilisation takes place abundantly 
in every case before the end of flowering, but if insect-visits 
take place in time, the stigma, which is developed before the 
anthers, receives pollen from outside before self-fertilisation can 
take place, and when pollen of the same flower is afterwards 
applied to the stigma it is probably rendered useless by the 
“prepotency ” of the other: if insect-visits are delayed till a later 
period, so that pollen from outside is brought only to a stigma: 
already laden with pollen of its own flower, we may still suppose 
that the former outstrips the latter. 
Thus in these Geraniums we see clearly how, as the number of 
insect-visits diminishes with the censpicuousness of the flower, the 
plant makes amends by an increasing likelihood of self-fertilisation 
for the diminished chance of cross-fertilisation. 
Geranium dissectum, L.—The manner of fertilisation is similar 
to that of G. pusillum, but the flower has ten anthers. When 
it opens, the stigmas already stand widely outspread, and 
the anthers which stand close around them are still closed and 
dehisce gradually one after another, dusting the stigmas with 
pollen. I have seen it visited by flies and by Andrena (vide 
No. 590, II. p. 217). 
82. GERANIUM ROBERTIANUM, L.—When the flower opens, 
the five stigmas are folded close together; the five outer stamens 
stand close round them, in the middle of the flower, and their 
anthers, which project a little above the stigmas, dehisce, covering 
oT 
ee ee ee ee ee eee ee ee 
