Page:Miscellaneousbot01brow.djvu/573

 ON THE DIVISIONS OF STIGMA,^ kc. ^07

��To estimate correctly the importance of the relation be- tween the divisions of the Stigma and the pai'ictal placentae of the componnd ovarium, namely, Avhetherwhen agreeing nos in number they are placed opposite to or alternate with each other, it is necessary to take into consideration the theoretical view which appears the most probable of the orio'in or formation of a simnle ovarinm, and that of the stigma belonging to it, as well as the various kinds and degrees of confluence by which the real nature of b^tli organs, but especially the latter, is so often obscured.

It is at present, I believe, universally agreed to consider a polyspermous legumen as that state of the simple ovarium, which best exemplities the hypothetical view of the forma- tion of this organ generally adopted ; namely, that it consists of the modification of a leaf folded inwards and united by its margins, which in most cases are the only parts of the organ pi'oducing ovula ; or, at least, where this power of production is not absolutely confined to the margins, it generally commences with or includes them.

The exceptions to the structure as here stated are of two kinds : —

First. Where the whole internal surface of the carpel is equally ovuliferous, which is the case in a few famihes of very small extent, as Butomece, Nijmphceacecp, and LnrdhnhaJerp.

' [Extracted from Mr. Brown's account of CyrtandrecP, given in the second part of Dr. Horsficld's 'Plnntrc Javanice l^'iriore?,' pul)lished in IStO. Sepa- rate copies of this nolo were dislribufcd in December, I'^.'i'). — Elit. Ann. Nat. IfhL]

�� �