i862— i88i] mi: la stoma: 301 



were aborting, and that the pollen was not good. The mature Letter 629 

 pollen is incoherent, and mu>t be [word illegible] ;t :he 



visiting insect's body. I remembered this, and I find it said 

 in my earl) notes that bees would never visit the flowers for 



pollen. This made me afterwards write to the late Dr. Criiger 

 in the West Indies, and he observed for me the flowers, and 

 saw bees pressing the anthers with their mandibles from the 

 base upwards, and this forced a worm-like thread of pollen 

 from the terminal pore, and this pollen the bees collected with 

 their hind legs. So that the Melastomads are not opposed to 

 your views. 



I am now working on the habits of worms, and it tires me 

 much to change my subject ; so I will lay on one side your 

 letter and my notes, until I have a week's leisure, and will 

 then see whether my facts bear on your views. I will then 

 send a letter to Nature or to the Linn. Soc, with the extract 

 of your letter (and this ought to appear in any case), with 

 my own observations, if they appear worth publishing. The 

 subject had gone out of my mind, but I now remember 

 thinking that the imperfect action of the crimson stamens 

 might throw light on hybridism. If this pollen is developed, 

 according to your view, for the sake of attracting insects, it 

 might act imperfectly, as well as if the stamens were becoming 

 rudimentary. 1 I do not know whether I have made myself 

 intelligible. 



To W. Thiselton-Dyer. Letter 630 



Down, March 21st [1SS1J. 



I have had a letter from Fritz Muller suggesting a novel 

 and very curious explanation of certain plants producing two 

 sets of anthers of different colour. This has set me on fire to 

 renew the laborious experiments which I made on this subject, 

 now 20 years ago. Now, will you be so kind as to turn in 

 your much worked and much holding head, whether you can 

 think of any plants cially annuals, producing 2 such sets 



of anthers. I believe that this is the case with ia 



eUgans, and I have just written to Thompson for seeds. The 

 Lythraceae must be excluded, as these are heterostvled. 



1 As far as it is possible to understand the earlier letters it seems that 

 the pollen of the shorter stamens, which are adapted for attracting insects, 

 is the most effective. 



