456 Mr. B. Clarke on Relative Position ; 



shallow furrow extending from its summit to its base, and this 

 furrow is always posterior (PI. XV. fig. \2). This may be taken 

 as an indication of a single carpel uniform in its position, and 

 (considering the near affinity of Circaa) in all probability ante- 

 rior, the stamen also being anterior. 



Composite. That a tendency exists in Compositae to sup- 

 pression of the posterior carpel is shown by the achenia of the 

 Sunflower, and especially by those of Centaurea, where the ante- 

 rior portion of the ovary is both larger and thicker than the pos- 

 terior ; and as in both these plants the stigmas are anterior and 

 posterior, it would be difficult to account for the inequality of 

 the ovary on any other hypothesis. In Centaurea nigra there is, 

 besides the shrinking of the posterior portion of the ovary, a line 

 on each side like a suture ; there are also two others, one anterior 

 and the other posterior ; and as the flower is quinary, these four 

 lines may be relied on as marking the presence of two carpels, 

 and the two lateral of these the juncture between them : it fre- 

 quently also occurs that two-thirds of the ovaiy are on the ante- 

 rior side of the two lateral lines and sometimes more, so that the 

 posterior half of the ovary becomes almost rudimentary, being 

 also much shorter than the anterior from contraction at its base 

 (PI. XV. fig. 13). This latter character is also well represented 

 in Dr. Lindley's ' Vegetable Kingdom,' the figure being that of 

 C. Cyanus. 



The ovary of Aster sibiricum has nearly the same structure as 

 that of Centaurea, and the anterior portion of the stigma is in 

 the marginal florets frequently elongated like that of Ruellia or 

 Stemocanthus* (vide fig. 9 in Plate II.). But it should not here 

 pass unnoticed that the position of the stigmas, except where the 

 style is thickened or very short, as in Euxenia, cannot be relied 

 on as an index of the position of the carpels, as the style very 

 commonly becomes twisted half round, of which Lasthenia is an 

 example. 



ChamjElauciace^. In Calytrix virgata the placenta is cen- 

 tral, is a pi'olongation of the pith, and is continued uninter- 

 ruptedly into the style, and as neither style nor stigma exhibit 

 any evidences of division the carpel must be wanting, and the 

 style only a prolongation of the axis, as suggested by Dr. Schleiden 

 in certain cases where the flower is superior. The cavity more- 

 over is imperfect, being occupied by cross bands of cellular tis- 

 sue. The ovules are attached to the anterior side of the central 

 column, or very rarely more or less laterally, which can scarcely 

 be regarded as a character less than equivalent to a single carpel 



* To this it may be added, that the attachment of the ovule incliaes 

 more or less to the posterior side of the ovary, and the position of the 

 raphe being uniform tends to show that that of the fertile carpel is also. 



