EXAMINATION OF A BUTTEROtTP. '9 



looks less like a, leaf than any other part of the flower. 

 Fig. 12 will, however, serve to show us the plan upon 

 which the botanist considers a stamen to be formed. The 

 anther corresponds to the leaf-blade, and the filament to 

 the petiole. The two cells of the anther correspond to 

 the two halves of the leaf, and the cells burst open 

 i^long what answers to the margin of the lea£ 



10. In the case of apocarpous pistils, as that of the 

 Buttercup, the botanist considers each carpel to be 

 formed by a leaf-blade doubled lengthwise until the edges - 

 meet and^unite, thus forming tlie ovary. Fig. 13 will 

 make this plear. 



11. There are many facts which support this theory 

 as to the nature of the different parts of the flower." 

 Sufiice it to mention here, that in the white Water-Lily, 



in which there are several circles of sepals and 

 petals, it is difficult to say where the sepals end 

 and the petals begin, on account of the gradual 

 change from one set to the oth( r. And not only 

 Fig. 13. is there a gradual transition from sepals to 

 petals, but there is likewise a similar transition from 

 petals to stamens, some parts occurring which are neither 

 altogether petals, nor altogether stamens, but a mixture 

 of both, being imperfect petals with imperfect anthers at 

 . their summits. We can thus trace ordinary leaf-forms, 

 by gradual changes, to stamens. 



We shall then distinguish the leaves of plants as 

 foliage-leaves' and flower-leaves, giving the latter name 

 exclusively to the parts which make up the flower, and 

 the formei: to the ordinary leaves which grow upon the 

 stem and its branches. 



Fig. 13. — ^Diagram to illustrate the leaf -structure of the carpei. 



