1 THE ORIGIN OF SPECIES' 103 



species. The more brightly coloured <among them, 

 again, would be more readily discriminated than the 

 less brightly coloured ; and this would give them such 

 an advantage that in the long run, as we actually see, 

 almost all habitually insect-fertilised flowers would come 

 to have brilliant petals. The germ of this luminous 

 idea, once more, is to be found in Sprengel's remarkable 

 work on the fertilisation of flowers a work far in 

 advance of its time in many ways, and to which Darwin 

 always expressed his deep obligations; but, as in so 

 many other instances, while Sprengel looked upon all 

 the little modifications and adaptations of flower and 

 insect to one another as the result of distinct creative 

 design, Darwin looked upon them as the result of 

 natural selection, working upon the basis of indeter- 

 minate spontaneous variations. 



How do these variations arise ? Not by chance, of 

 course (for in the strict scientific sense nothing on earth 

 can be considered as really fortuitous), but as the out- 

 come for the most part of very minute organic causes, 

 whose particular action it is impossible for us to predict 

 with our present knowledge. Some physical cause in 

 each case there must necessarily be ; and indeed it is 

 often possible to show that certain changes of condition 

 in the parent do result in variations in the offspring, 

 though what special direction the variation will take 

 can never be foretold with any accuracy. In short, our 

 ignorance of the laws of variation is profound, but our 

 knowledge of the fact is clear and certain. The fact 

 alone is essential to the principle of natural selection ; 

 the cause, though in itself an interesting subject of 

 inquiry, may be safely laid aside for the present as com- 



