254 THE FLOWER. 



ilar leaflets of a compound leaf, in what is called transverse chori- 

 sis there is seldom if ever such a division or ramification into 

 homogeneous parts ; but the original organ remains, as it were, in- 

 tact and unmodified, while it bears an appendage of some different 

 appearance or function on its inner face, or at its base on that side. 

 Thus the stamens of Larrea, &c., bear a scale-like appendage ; the 

 petals of Sapindus, Cardiosperrnum, &c., a petaloid scale quite 

 unlike the original petal ; the petals of Parnassia, a cluster of bod- 

 ies resembling sterile filaments united below. In a still greater 

 number of instances, the accession to the petal consists of a real 

 stamen placed before it, and often more or 

 &_ less united with its base, as in the whole 



J Buckthorn Family (Fig. 315), and in the 

 Byttneriaceae ; or of a cluster of stamens, 

 as in the Mallow Family, and indistinctly in 

 most European Lindens, or of such a cluster 

 with a petal-like scale in the midst, as in the 

 American Lindens (Ord. Tiliacea?, Fig. 306). 

 In the first-named cases, the accessory organ developes entire and 

 simple ; in the latter, it is multiplied by collateral chorisis.* 



460. A most able writer in a recent number of the Journal of 

 Botany, (with whom we entirely accord as to the nature of collat- 

 eral chorisis,) " being totally at a loss to find any thing analogous in 

 the ordinary stem-leaves " to this transverse or vertical multiplica- 

 tion of parts, inclines to consider such appendages as those of the 

 petals of Silene, Sapindus, Ranunculus, &c., as deformed glands, 

 and the stamens thus situated, whether singly or in clusters, as de- 

 velopments of new parts in the axil of the petals, &c.t It appears 

 to us, however, that the leaves do furnish the proper analogue of 

 these appendages (especially those of Fig. 302, 303, 305, and the 



* For illustrations, and more detailed explanation of these points, the stu- 

 dent is referred to the figures and text of The Genera of the United States Flo- 

 ra Illustrated, especially to Vol. 2. The opposition of the exterior circle of 

 stamens to the petals in Geranium, &c., we explain in a different way. 



t Namely, in Hooker's Journal of Botany and Kew Garden Miscellany, Dec., 

 1849, p. 360. The morphology of true glands is still obscure, notwithstand- 

 ing the interesting light that is thrown upon them in the article here referred 

 to ; and stipules often tend to assume the glandular character. 



FIG. 306. Diagram (cross-section) of the unopened flower of the American Linden, to show 

 the scale and the cluster of stamens before each petal. 



