192 Chapters in Modern Botany chap. 



this time willing to take the writer for examiner, he will be 

 asked, as always, not to answer questions — with second- 

 hand information — at the end of his day's work ; but to ask 

 questions — about such actual flowers and fruits and seeds as 

 he can see — at the very beginning ; this done, it is the books 

 and the teacher whom he has to examine, the limits of their 

 knowledge which it is his business to work up to and to 

 define. He must look how flowers are arranged before 

 "reading up Inflorescence''''', and take them to pieces before 

 reading of sepals and petals, of stamens and ovules ; he 

 must puzzle about what pollen and ovules are for, he must 

 watch the bees and butterflies among the flowers, and find 

 out which flowers no insects go to, before reading of insect- 

 and wind-fertilisation ; then the books of Grant Allen and 

 Sir John Lubbock may be read one summer as pleasant 

 introductions to the larger volume of Hermann Miiller (the 

 standard work of reference upon the subject) for the next. 

 Finally its bibliography, supplemented by that of MacLeod,^ 

 is exhaustive. Kerner's Flowers and their Unbidde7i Guests^ 

 and his Pflaiisenleben^ now being translated, will also be 

 of special interest. 



This widening knowledge of flower -function will of 

 course involve an increasing minuteness of observation in 

 detail, and the Darwinian interpretation of the utility of even 

 the smallest of these — the shape and relative position of 

 parts, the colour, the markings, the perfume — will give them 

 interest. Thus arises a knowledge of flower form not only 

 far more interesting and more genuine, but more per- 

 manent and more intimate and thorough than that of the 

 conventional "anatomy before physiology" method, against 

 which this little volume is a continuous protest. For it is 

 only when we have first seized the essential parts of the 

 flower (stamens and carpels), and seen how they are adapted 

 ^ Botanisches Jaarboek, vol. ii. ' 



