BY E. HAVILAND, P.L.S. 461 



however not being in great abundance. As this careful protection 

 of the pollen and stigma exists until the flower has withered, I 

 can only arrive at the conclusion that the plant is closely self- 

 fertilised. While however I say this, I wish to repeat, that I 

 have only had the opportunity of examining the flowers of a single 

 plant. There is therefore just a possibility of plants existing which 

 at some stage or other 0[)en their flowers and thus expose them to 

 cross-fertilisation, Bentham however mentions the exserted coue 

 of anthers as common to the whole genus. 



It is impossible to examine the flowers of different plants , 

 without being struck with the difference in their various 

 pollens. In some cases the grains are powdery, loose and diy, 

 and liable, one would think, to absorb moisture readily. In 

 others waxy and more likely to repel it. I am strongly impressed 

 with the idea, that in cases where the anthers and pollen are so 

 carefully shielded, as in Lyonsia, there is another purpose to be 

 served than the self -fertilisation of the plant. There are so many 

 plants known to be cross-fertilised, and yet which have their 

 anthers very carefully shielded ; that the conviction is forced upon 

 one, that the primary intention is to preserve the pollen from 

 injury by rain or dew, in cases where it is absorbent and likely to 

 be injured. In the case before us, Lyonsia reticulata, as I have 

 already said, the backs of the anthers form so complete and 

 compact a roof to the pollen-cells, that it would seem impossible even 

 for a heavy shower to injure the pollen. In the genus Cryptandra, 

 and especially in G. amara, the very small petals are hood-shaped, 

 completely covering the anthei-s, until after the pollen is 

 ripe ; while it is quite possible for small insects to obtain access to 

 it from below. There are too, hundreds of cases in which at 

 might, when dew is likely to fall, flowers not only close, 

 but droop ; thus inverting the corolla, and making a roof of it to 

 protect the anthers and pollen. 



Acting on the supposition that the intention is to protect the 

 pollen of those plants which would be injured by moisture ; I 

 selected that of three plants, Lyoiisia reticulata, Cryptandra a.mara 



