INTRODUCTION. XXVll 



all the gradations from a corolla of the most irregular form to one 

 of the most perfect symmetry. In Compositae both are found con- 

 tinually in the same head ; and Lobeliaceae, which may be almost 

 always distinguished from Campanulaceai by their irregularity, be- 

 come nearly regular in Isotoma. — The venation of the petals is 

 scarcely ever employed for distinction, little being at present known 

 of it. Composita3 are distinguished by the peculiar arrangement 

 of the veins of their corolla ; and they are always oblique in 

 Hypericinese. 



From within the corolla arise certain metamorphosed leaves, 

 which are called the Sexes of plants. From the manner in whicli 

 they are combined, good characters may sometimes be derived, but 

 frequently no characters at all. Thus, Xanthoxylese are known from 

 Diosmese and Terebintaceoe by their unisexual flowers ; all Euphor- 

 biaceae, Begoniacese, Amentacese, Coniferae, Myriceae, are unisexual. 

 But Vites, Gramineae, Cyperaceae, Chenopodeae, Umbelliferae, and 

 even Ranunculaceae, contain hermaphrodite and diclinous genera; 

 and it is familiar to every one, that flowers of all these kinds (that is, 

 male, female, and hermaphrodite), stand side by side in Compositae. 



Of these sexes the Stamens are what are called the male 

 organs, and are undoubtedly the apparatus by means of which vivi- 

 fication is communicated to the ovula or eggs. They either arise 

 immediately from below the ovarium, having no adhesion to the 

 calyx, when they are said to be hypogynous, or they contract an 

 adhesion of greater or smaller extent with either the calyx or corolla, 

 when they become perigynous, or, finally, they appear to proceed 

 from the apex of an inferior ovarium, in which case they are named 

 epigynous ; but it is usually now understood that all stamens take 

 their origin from below the ovarium; and if this opinion be well 

 founded, there will be no material difference between those which 

 are perigynous and those which are epigynous; and these two 

 modifications are accordingly confounded together by most modern 

 botanists. M. Ad. Brongniart, however, conceives epigynous stamens 

 to be essentially distinct from perigynous, founding his opinion upon 

 the genus Raspailia, which has a superior ovarium, from the top of 

 which arise the stamens; but it is possible perhaps to explain this 

 apparent anomaly. To the difference between perigynous and hypo- 

 gynous stamens the French school attaches the greatest value, not 

 being willing to admit any genus with hypogynous stamens into an 

 order with perigynous ones, and vice versa ; and there is somewhere 

 an observation, that of such primary importance is this distinction, 

 that while poisonous orders are to be known by their stamens being 

 hypogynous, all in which they are perigynous are wholesome. Setting 

 aside, however, this hypothesis, which has not the general applica- 



