348 THE BIOLOGY OF FLOWERING PLANTS 



is avoided. It may be doubted whether the more extreme 

 cases of speciaHsation bring any further advantage, for it is 

 unHkely that the pollen grain often gets a second chance if 

 it has once been deposited on the wrong stigma. 



We may mention here a subject to which we shall return 

 in another connection — those cases of self-sterility where 

 failure is due to the fact that pollen will not germinate on 

 the stigma of the flower which produced it. 



Protection of Pollen. — We have seen that barley pollen 

 is readily damaged by water, and the same is true for many 

 plants. Lidfors (1896, 1899) found that in 38 families out 

 of 80 which he investigated, species were found with 

 pollen damaged by water. All degrees were found, from the 

 extreme sensitiveness of the barley, where instant bursting 

 occurs, to pollen which germinates in pure water. 



Such sensitive pollen must be affected by rain, and it is 

 natural that certain features in the floral structure and 

 behaviour should have been interpreted as offering protection 

 to the pollen against this danger and against the grosser one 

 of being washed away. 



Many pollen grains are covered by a thin film of oil, 

 which makes wetting difficult. Especially in zygomorphic 

 flowers the structure of corolla and calyx is frequently such 

 as to protect the stamens. It is only necessary to think of 

 the arched hood of the sage, or monkshood, of the pursed- 

 up corolla of the toad-flax or of the snapdragon, of the 

 enclosed stamens of the broom or whin, of the sheltering 

 scales which close the corolla tube of the forget-me-not, of 

 the narrow tube entrance of the centaury, to see how fre- 

 quently and well pollen may thus be protected from wet. 

 In hanging flowers the pollen is again shielded, as in the 

 wood hyacinth and harebell. It is not necessary to regard 

 these structures as primarily related to pollen protection, 

 probably they are not ; yet their protective action may be 

 none the less effective. 



Floral Movements.— It is rather more difficult to assess 

 the usefulness in this respect of flower movements. In 

 Anemone sylvatica, A. japonica, and other species, the 



