132 BOTANY. [chap. iv. 



Cross-fertilization was undoubtedly the first and main 

 incentive in promoting the various changes that flowers 

 have undergone for the purpose of facilitating the visits 

 of insects ; but even if this were not the case, the idea 

 would be a success from the economical point of view, as 

 insect - fertilized flowers by their greater exactness of 

 arrangements insure fertilization with a much less expen- 

 diture of material than is the case with either wind- 

 fertilized or self-fertilized plants. In the latter class 

 the flowers are regular in form, that is, the sepals are of 

 equal size, as are also the petals, as illustrated by the 

 buttercup, wild rose, and tiger-lily, and in addition are 

 perfectly free from each other; in fact, the stamp of a 

 primitive flower is the perfect freedom from each other 

 of all its components, a character illustrated by our 

 buttercups ; the stamens in such plants are usually 

 numerous, the haphazard method of fertilization re- 

 quiring a very large amount of pollen to make certain 

 this indispensable process. In some flowers colour has 

 not yet been evolved, but this is usually the case, and 

 numerous flowers have not yet got beyond this stage, 

 which is the first step in the sequence leading up to the 

 fully evolved insect-fertilized flower. Along with the 

 advent of colour, we find as a rule arrangements for the 

 secretion of nectar, the production of which is confined 

 to specially modified portions of the flower called nec- 

 taries. Nectaries occur on difierent parts of the flower 

 in difierent plants. In the buttercup a nectary or honey 

 gland is present at the base or narrow point of attach- 

 ment of each petal on its upper surface ; in the colum- 

 bine, a plant belonging to the same order as the butter- 

 cup [Ranuneulaeece) , we find each of the five petals 



