SOUTH AFRICAN ORCHIDS: 1863 219 



ments ; and indeed in those you describe such move- 

 ments would be superfluous. If you have time to wander 

 about do watch some kinds and see insects do the work. 1 

 Those with long nectaries would be probably hopeless 

 to watch as probably fertilized by Moths. But since 

 my publication I have ascertained that with Orchis, 

 Diptera are chief workmen. They certainly do puncture 

 the walls of nectary, and so get juice. Disperis would 

 be grand to watch, and discover what attracts insects. 

 You draw so well, and have so seized on the subject, 

 that you ought really to take up 2 or 3 of the most 

 distinct genera, and watch them, experiment on them 

 by mutilation of parts, and describe them and send 

 over an excellent paper to Linnean Soc y or some other 

 Soc y . I have so much other work, that I hardly know 

 whether I shall ever publish again, not but what I have 

 already collected some curious new matter; for the 

 subject delights me, and I cannot resist observing. 



I am very glad to hear that you do not now think me 

 so dangerous a person ! 2 You will gradually, I can see, 

 become as depraved, as I am. I believe, or am inclined 

 to believe, in one or very few primordial forms, from 

 community of structure and early embryonic resem- 

 blances in each great class. 



With most cordial thanks I remain my dear Sir 

 Yours sincerely 



CH. DARWIN 



P.S. Would it be asking too great a favour to beg you 



1 Mr. Trimen writes as follows of his attempts to carry out 

 Darwin's advice : ' I had no success with this, though I watched 

 a variety of orchids as opportunity offered. A good many visitors 

 of various orders came, but they were evidently not regular 

 customers (" unbidden guests," as Kerner says), and I never saw 

 a pollinium actually removed by any one of them.' 



Trimen found, however, that one or both pollinia had been 

 removed from 12 out of 78 flowers of Disa grandiflora. Fertilisation 

 of Orchids (1877), 78. 2 See pp. 214-15. 



