THE AMERICAN NATURALIST [Vol. XLIII 



chapter of his book on the effects of cross and self-fertil- 

 ization in the vegetable kingdom indicates that this con- 

 clusion," based on his personal observations on plants, 

 was guided to a certain extent by the experience of 

 breeders of animals (which is detailed at length in his 

 book on the variation of animals and plants under 

 domestication). 



It was, therefore, not a novice who entered the botanical 

 field when, in October, 1857, Darwin contributed a one- 

 column note to the Gardiner s' Chronicle on bees and fertil- 

 ization of kidney beans, which he was attracted to through 

 "believing that the brush on the pistil, its backward and 

 forward curling movement, its protrusion on the left 

 side, and the constant alighting of the bees on the same 

 side, were not accidental coincidences, but were connected 

 with, perhaps necessary to, the fertilisation of the flower." 

 His subsequent papers and books, vivifying floral ecology, 

 were therefore as inevitable from one who was moved 

 by a new teleology in natural science as was the superb 

 publication of a half century earlier by Sprengel, who, 

 seeing the petal-hairs of a Geranium, sought after their 

 meaning in the firm conviction that the wise Author of 

 nature had not created even a hair in vain. 



Darwin's publications in this field are not numerous; 

 omitting abstracts and reprints of papers published in 

 Cull, and later editions and translations, they number only 

 twenty-two. 1 They are remarkably direct in purpose, 

 simple in treatment, clear in reasoning, and free from a 

 controversial or dictatorial spirit. His own observations 

 and experiments were rarely accepted at their face value, 

 but were checked up with rather unusual care; and, as 

 we now know from his published letters, his conclusions 

 w ore commonly subjected to the criticism of a very select 

 private audience before going to the world at large. 



The comprehensiveness of the studies underlying these 

 publications may be indicated by the statement that 



