434 BOTANY. 



latter unite, forming a common ovary cavity ; here the placentae gener- 

 ally occur along the sutures, and are said to be parietal i.e., on the 

 walls. Between such unilocular pistils and the uiultilocular ones 

 described above there are all intermediate gradations. In one series of 

 gradations the placentae project farther and i'arther into the ovary cav- 

 ity, at last meeting in the centre, when the pistil becomes multilocular 

 with axile placentae. On the other hand, a multilocular pistil sometimes 

 bt comes unilocular by the breaking away of the partitions during 

 growth. In such a case the placentae form a free central column, 

 commonly called nfree central placenta. 



In other cases a free placental column of an entirely different origin 

 occupies tlie axis of a unilocular, but evidently polycarpellary pistil. 

 In Anagallis, ior example, the placental column grows from the base 

 of the ovary cavity, and there is at no time a trace of partitions (see 

 illustrations of the Order Primulacese, p. 507). 



The Qyno3cium may be free from all the other organs of the flower, 

 which are then said to be liypoyynous* and the gynoecium itself su- 

 perior. Sometimes the growth of the broad flower-axis stops at its 

 apex long before it does so in its marginal portions ; a tubular ring is 

 thus formed, carrying up calyx, corolla, and stamens, which are then 

 said to be perigynous,^ and the gyncecium half inferior. These terms 

 are used also in the cases where the gyncecium is similarly surrounded 

 by the tubular sheath composed of adnate calyx, corolla, and androe- 

 cium. In some nearly related cases, in addition to the structures de- 

 scribed above as perigynous, there is a complete fusion of the calyx, 

 corolla, and stamen -bearing tube with the gynoecium, so that the ovule- 

 bearing portion of the latter is below the rest of the flower, e.g., Com- 

 positae. The perianth and the stamens are said to be epiyynom\ in such 

 ilowers, and the ovary is inferior. Some cases of epigyny are doubtless 

 to be regarded as due to the adnation of the calyx, eorolla, stamens, 

 and ovaries ; in others, the ovaries are adnate to the hollow axis which 

 bears the perianth and stamens ; in still others, it seems probable that 

 the hollow axis is itself ovule-bearing, and that the true carpels are 

 borne on its summit. 



Certain terms descriptive of relations between the stamens and pis- 

 tils which have recently come into use require explanation here. 



In many flowers the stamens and pistils do not mature at the same 

 time, such are said to be dichogamous ; when the stamens mature be- 

 fore the pistils the flower is proterandrous ; and when the pistils ma- 

 ture before the stamens they are proterogynous. 



In some species of plants there are two or three kinds of flowers, 



* From Greek into, under, and yvvrj, female i.e., the pistil, 

 f From the Greek nepi, about, etc. 

 | From the Greek i-xl, upon, etc. 



