txii INTRODUCTION. 



carpial leaves right and left with respect to the common axis of inflores- 

 cence, it is never truly present ; while in Scrophularineae and their allies, 

 the pericarpial leaves of which are anterior and posterior, it is as uniformly 

 present in one shape or other. 



The last modification of leaves in the fructification consists in their con- 

 version into what is called the pistillum, or Ovarium ; that is to say, 

 into the case which contains the young seeds or ovules. Now that the 

 structure of this part is well understood, we know that an ovarium either 

 consists of one or several connected pericarpial leaves, called carpella, ar- 

 ranged around a common axis, or of several combined into a single body. 

 Upon this difference the distinction depends of what I call apocarpous 

 ovaria, or those of which the carpella are distinct ; and syncarpous are those 

 of which the carpella are compactly combined. These differences appear 

 to me of much importance, and subject to as few exceptions as any modi- 

 fications that botanists make use of. Thus Berberideae are distinguished 

 from Papaveraceae, Nelumboneae from Nymphaeaceae, Amyrideae from 

 Burseraceae, Boragineae from Ehretiaceae, and the like. But, at the same 

 time, it will be seen that cases exist of both forms being found in the same 

 natural order, as Zanthoxyleae. This, however, is rare. — The cohesion 

 of the ovarium with the calyx, or its separation from it, has been already 

 treated of in speaking of the calyx. — An ovarium may be either one-celled, 

 in consequence of its consisting of a single carpellum, in which case it will 

 belong to the apocarpous division ; or it may consist of several carpella 

 strictly cohering, and therefore syncarpous, but nevertheless one-celled, in 

 consequence of the obliteration of the dissepiments. Peculiarities of this 

 latter nature are almost always of ordinal importance, at least if the pla- 

 centa are parietal ; for instance, the latter is the structure of Papaveraceae, 

 Homalineae, Flacourtiaceae, Cucurbitaceae, Papayacae, and Violaceae, to 

 which there is no exception ;. but Caryophylleae and Bruniaceae, the usual 

 structure of which is to be one-celled, have the placentae in the centre ; 

 and in both these orders there are genera, the ovarium of which contains 

 several cells. — Another point that deserves particular attention is the rela- 

 tion borne to the axis of inflorescence by the pericarpial leaves, of which an 

 ovarium is formed. What the exact value of this character may be, is 

 not yet known ; but it is certain that Gentianeae and their allies have their 

 principal leaves right and left of the axis, while Scrophularineae and their 

 allies, which are sometimes to be distinguished with difficult}', have the 

 pericarpial leaves anterior and posterior with respect to the axis. Rosaceae 

 and Leguminosae differ in a nearly similar way. — Connected with the 

 apocarpous or syncarpous state of the ovarium is the union or separation 

 of the styles, which, therefore, scarcely require distinct mention. It is as 

 well, however, to remark, that the separation of styles is commonly a sign 

 of the apocarpous state of the ovarium, provided the latter is not very 

 apparent otherwise ; and the cohesion of the styles is constantly an evidence 

 to the contrary ; and in this view the Elder and Hydrangea tribes may be 

 justifiably separated from Caprifoliaceae. 



The Stigma seldom offers any good characters. In some cases, how- 

 ever, advantage is taken of it, as in Lineae, the capitate stigmas of which 

 distinguish them from Caryophylleae, in which they occupy the whole 

 inner face of the styles ; and in Goodenoviae, Scaevoleae, and Brunoniaceae, 

 there is a peculiar membranous appendage enveloping the stigma, and 

 called an indusium, which distinguishes those orders from all others. 



