372 NATURAL HISTORY OF PLANTS. 



It cannot be denied that the organization of the Ochnacece 

 give them the greatest analogy with Itutacece, of which they 

 might possibly be considered a tribe. It is true they have not 

 glandular punctuations, they do not contain odoriferous essen- 

 tial oil, they have stipules, the flowers are destitute of glandular 

 disk, and in their fruit the endocarp, when it is capsular, does not 

 separate from the exocarp ; but these are characters which may 

 possibly be wanting in the Rutacece, and which have not much 

 value in themselves. The functional organization of the gynseceum 

 in the Ochnacece is, moreover, much more similar to that of the 

 Rutacea than is generally supposed. Certain Hypericacece have ap- 

 parent analogies with the Ochnacece; but their leaves are opposite 

 and odoriferous. Perhaps the Dilleniacece of those genera with yellow 

 imbricated petals, independent carpels, and fleshy fruit, are more 

 closely allied to the Ochnacece than they are said to be. Crossosoma 

 seems also to nearly approach the Ochnacece with independent car- 

 pels, and certain SimarubecB. But where this great analogy presents 

 itself, the Dilleniacece have different leaves from those of the Ochnacece, 

 arillate seeds, and a small embryo situated towards the summit of an 

 abundant fleshy albumen. 



We have seen how 1 Ternstrcemiaccce, to which the genus Pceci- 

 landra has been attributed, is distinguished from Ochnacece, only by 

 unimportant characters it is true. The same may be said of those 

 separating Ochnacece from certain Violaccce, such as the Sauvagesice. 

 It is not without strong reasons that these latter have been classed 

 by some authors 2 in the same group with Luccembiirgice, and perhaps 

 they are as well placed there as among the Ochnacece. The Luxem- 

 hurgice are always distinguished from the Sauvagesice by " their simple 

 anthers dehiscing by two short pores, 3 not connivent round the gynae- 

 ceum, and by the very different character of the stamina! filaments." 

 It may be admitted to a certain point, in consequence of what has 

 been said, and by the comparing of the vegetative organs, the 

 leaves, &c, of certain species of Rinorca with Ochnce, that these latter 

 represent a regular form, with independent carpels, of the arbo- 

 rescent Violacece. The woody Cisfacece, so nearly allied to Violacece, 

 have, for the same reason, some affinity with the Ochnacece. 



1 Sou p. 25G. 



; See 1. iMii,., Veg. Kingd., 343. — Tb. A: Pi., in Ann. Sc. Nat., ser. I. xviii. 27o. 



3 Except Wallaoea (1!. II., Gen., 316). 



