PREFACE TO THE SECOND EDITION. 



The publishers having called for a second edition 

 of this book, I take occasion, in addition to making 

 a few corrections and minor changes in the plates, 

 to more fully express my convictions upon two or 

 three points to which reviewers have taken exception. 

 I do this not as a defense, but for the purpose of 

 making my contention more clear. I should say, 

 also, that the book, from its very nature, is not a 

 treatise, and that it does not attempt to construct 

 any philosophy of the progress of the evolution of 

 plants, — a subject which the author may hope to con- 

 sider at another time. 



It has been objected that the statement that un- 

 like produces unlike cannot be strictly true, since it 

 might follow that the offspring should be so different 

 from the parents as not to carry over any of the 

 parental characteristics. In other words, its logical 

 termination might be a general scattering of forms 

 and the absence of resemblances. All this is perfectly 

 true, but the reader is expected to bear in mind the 

 fact that hypotheses are methodological: their great- 

 est purpose is to suggest a method rather than to 



(5) 



