S^ soar/I AFRICAN FLOWERING PLANTS. 



why flowers are so different in their shapes and colours, 

 why they vary in the numbers of ' their parts, or why 

 some one or more may differ in size or shape from the 

 rest of the parts of one and the same whorl, one asks, 

 what are' the causes which have given rise to the 

 infinite diversity which exists among flowers ? 



Again, we see some flowers with all their parts 

 perfect, their corollas very large, brightly coloured or 

 white, often strongly scented, and producing much 

 honey. On the other hand, many flowers are exceed- 

 ingly minute, with a very small corolla, or even none 

 at all, totally wanting in scent, and producing no honey. 



Collectively, in many instances, one describes these 

 latter features as " degradations," since in many cases 

 they have " rudimentary organs," as they are called, 

 being of no use now, but representing ancestral states, 

 when they doubtless were possessed of perfect functions. 

 Thus a remnant of a pistil often exists in male flowers. 



To try and solve this problem we must observe 

 what goes on in JSTature. The first thing to note is 

 that insects are habitual visitors of many flowers, 

 either for pollen or honey, or both as food. As they 

 generally visit the more conspicuous flowers, which 

 are white or coloured, or else discoverable by their 

 scent, as Mignonette, one concludes that there is some 

 " correlation " between such flowers and insects. 



On the other hand, innumerable small flowers may 

 be, and indeed are, never visited at all ; in these one 



