OBOLARIA VIRGINICA. 27 



detected at those sinuses which correspond with the axis of each car- 

 l)el ; and besides, this same endocarpic portion certainly makes up a 

 part of the thickness of the style, as well as of the walls of the ovary. 

 Some points concerning the position of the flowers and their 

 parts deserve notice. The axillary flowers are often solitary, when 

 the short peduncle is bractless ; otherwise they are three in a clus- 

 ter or cymule ; the two lateral being sessile, or nearly so, close at 

 the base of the terminal, each arising from the axil of a bract re- 

 sembling the sepals, or, indeed, the leaves. When the flower is 

 solitary, the sepals (as we are obliged to term them) are uniformly 

 lateral, as in the diagram, Fig. 6. Where the lateral axis bears 

 three flowers, these are commonly disposed as is represented in the 

 diagram. Fig. 1 ; that is, the two additional flowers are placed 

 right and left, having, of course, the same relation to the central 

 flower which that, when solitary, has to the main axis. The 

 sepals of the central flower in this case are not lateral, but anterior 

 and posterior, namely, one next the axis, the other next the bracteal 

 leaf. It is obvious, therefore, that the same organs which stand for 

 the calyx of the solitary flower, Fig. 6, form the bracts of the 

 three-flowered cluster in Fig. 1 ; the calyx of the central flower in 

 this latter case being just the next pair of leaves of the branch, 

 decussating with the first pair, and therefore necessarily standing 

 fore and aft, as respects the primary axis and cauline leaf. This 

 gives some apparent confirmation to the Linnsean view, that what 

 is called the calyx of Obolaria is no part of the flower, but rather a 

 pair of bracts. The three flowers of the axillary clusters are not 

 always thus disposed in a line at right angles (or nearly so) with 

 that passing through the stem and cauline leaf. This is the preva- 

 lent, but not the uniform, mode. Dr. Torrey called my attention 

 to the fact, that not unfrequently one or more of the clusters stand 



