i843 iS6j] INSECT FERTILISATION 255 



not-rare, abnormal, early maturity of the pistil as described Letter 586 

 by Gartner. 1 have hitherto failed in meeting with detailed 

 accounts of regular and normal impregnation in the bud. 

 Podosiemon and Subularia under water (and Leguminosae) 

 seem and are strongest cases against me ; as far as I as yet 

 know. I am so sorry that you are so overwhelmed with work ; 

 it makes your very great kindness to me the more striking. 



It is really pretty to see how effectual insects are. A short 

 time ago I found a female holly sixty measured yards from 

 any other holly, and I cut off some twigs and took by chance 

 twenty stigmas, cut off their tops, and put them under the 

 microscope : there was pollen on every one, and in profusion 

 on most i weather cloudy and stormy and unfavourable, wind 

 in wrong direction to have brought any. 



To J. D. Hooker. Letter 



Down, Jan. I2th [1858], 



I want to ask a question which will take you only few 

 words to answer. It bears on my former belief (and Asa 

 Gray strongly expressed opinion) that Papilionaceous flowers 

 were fatal to my notion of there being no eternal herma- 

 phrodites. First let me say how evidence goes. You will 

 remember my facts going to show that kidney-beans require 

 visits of bees to be fertilised. This has been positively stated 

 to be the case with Lathy rus grandiflorus, and has been very 

 partially verified by me. Sir W. Macarthur tells me that 

 Erythrina will hardly seed in Australia without the petals 

 are moved as if by bee. I have just met the statement that, 

 with common bean, when the humble-bees bite holes at the 

 base of the flower, and therefore cease visiting the mouth of 

 the corolla, " hardly a bean will set." But now comes a much 

 more curious statement, that [in] 1842-43, " since bees were 

 established at Wellington (N. Zealand), clover seeds all 

 over the settlement, which it did not before" J The writer 

 evidently has no idea what the connection can be. Now I 

 cannot help at once connecting this statement (and all the 

 foregoing statements in some degree support each other, as 

 all have been advanced without any sort of theory) with the 

 remarkable absence of Papilionaceous plants in N. Zealand. 

 I see in your list CliantJius, CarmicJiaelia (four species), a new 



1 See Letter 362, Vol. I. 



