THE CARPEL AND OVULE 291 



seen in the Lily (Fig. 209) . Such fusion of carpels, with or without their sinking 

 into the receptacle, has developed progressively, and has appeared repeatedly 

 in distinct evolutionary lines. The result is a solid and massive gynoecium, 

 whether with superior or inferior ovary. A transverse section of the inferior 

 ovary may still show evidence of its carpellary origin almost as clearly as in 

 the superior ovary. This is seen in Iris, where notwithstanding that the 

 carpels are sunk in the receptacle, their struc- 

 ture, and even the arrangement of the chief 

 vascular strands resembles in some degree that 

 seen in the superior ovary of the Lily (Fig. 

 211). The conclusion follows that in such cases 

 the gynoecium is still to be referred in origin 

 to foliar structures, more or less completely 

 fused with or sunk into the tissue of the 

 receptacle. 



The structure of the carpel, where it 

 is distinctly leaf-like, as it is in the pod FlG - 2I 3- 



. . Transverse section of the inferior 



Of the Pea Corresponds in essentials tO ovary of Iris. Compare the superior 

 ,\. r 1 • i-r j a ovary of Lily, Fig. 211. F. O. B. 



that of a foliage leaf, but simplified. A 



vascular strand usually traverses each margin, as in Caltha (Fig. 210). 

 This is related to the fact that the ovules are seated there ; or, as 

 it is described, the placentation is marginal It is probable that 

 this was the regular primitive position for ovules. But sometimes 

 they appear scattered over the inner surface of the carpellary wall, 

 as in the Flowering Rush (Butomus), the Poppy, or the Water-Lily. 

 This is described as superficial placentation, and it probably originated 

 by the spread of the ovules to the surface. Sometimes they appear as 

 though seated on a prolongation of the axis into the ovarian cavity, 

 as in the Pinks and Primroses (Appendix A, Fig. 47 1, of Lychnis). This 

 is called free-central placentation, and it also is probably derived from 

 the marginal type, by breaking away the partitions, or septa, of the 

 syncarpous ovary, leaving the margins with their ovules at the centre. 

 Thus the marginal was probably the primitive, as it is certainly the 

 prevalent position for the ovules on the carpellary leaf ; and this holds 

 equally for syncarpous ovaries (Figs. 211, 213). 



The Stigma, or receptive surface for the pollen-grains, is at the distal 

 end of the carpel. Where the carpels are separate each has its own 

 stigma, a condition which is probably the primitive state. It is seen 

 in Caltha and the Buttercup, the stigmatic area being recognised by 

 its roughened surface. Even where the carpels are united below into 

 a syncarpous ovary, each may separate upwards to form a distinct 

 stigma, as in the Apple or Quince (Fig. 212, A, B). But in cases which 

 are regarded as more advanced the fusion of the carpels may extend 



