254 M. COHREA DE MELLO O^' SOME BRAZILIAN" PLANTS. 



the ovaries of both from penetrating into the ground, either by the 

 interposition of a resisting body (a dry leaf or a slip of wood), or 



(when 



) 



the ground. Under these circumstances, in the A. Tiypogea the 

 stipes proceeding from the torus, at the end of which the ovary 

 is formed, lengthened out even to 3 or 4 inches ; but the ovary 

 never enlarged, and remained in the same state till the plant 

 perished ; and this I observed in many ovaries of many indivi- 

 duals. On this occasion I made the following observations on 

 the flower of this plant. The calyx-lobes are membranous, their 

 base is produced into a filiform somewhat fleshy tube, the alary 

 and carinal petals (wings and keel) are membranous and appear 

 articulate on the calyx, at least they readily separate without 

 laceration ; but the vexillum, which is somewhat fleshy, especially 

 at the base, is intimately consolidated at the base with the base 

 of the staminal tube, and both are continuous with the calyciual 

 tube — a circumstance which induces me to believe that the fili- 

 form part of the base of the calyx does not consist of that 

 alone, but also of the consolidated bases of the vexillum and 

 of the staminal tube. The Voandzeia siihterranea^ whose slender 

 and somewhat compressed stems spread along the ground, and 

 even penetrate under the surface when the soil is sufficiently 

 soft and porous, produces on the surface of the soil its small pale 

 yellow and slightly greenish flowers, the peduncle drying up 

 immediately if the ovary cannot penetrate the ground ; and thus 

 I can aflBrm that neither the one nor the other of these Legu- 

 minosse produces aerial pods. 



On this occasion I may observe that, notwithstanding all the 

 care and attention I could bestow, I failed to discover those 

 female apetalous flowers mentioned by Sprengel (under Crypto- 

 lohus), by DeCandolle (' Prodromus'), by Lindley and Moore 

 (' Treasury of Botany '), and by Bentham and Hooker (Gen. 

 PI. i. 539) *. The flowers in twelve individuals examined in the 

 plantation I made were all uniform, and provided with petals 

 (papilionaceous). The peduncle is slender, of variable length, 

 terminated by a nodosity, on which are inserted one or two pedi- 

 cellate flowersj of which one constantly falls off", carrying the 

 ovary with it ; the remaining one, after the petals have fallen, 

 turns downwards enclosed in the calyx, which persists for a con- 



** Flores apctali non satis noti." Bcntli, et HooJc, I. c. 



