Page:Miscellaneousbot01brow.djvu/567

 ox ORnilDK.I^. AND asc]<kpiadk.t:. 549

of the innei' surface of the Corolla of sonie species of Ceropegia, in the wool enveloping the spines in several species of Manimillaria and Melocactus, and in the Coma of the seed of an Apocy neons plant from Brazil: for the spiral vessels in the seeds of CoUomia?, first observed by Mr. Lindley, and described by him as external, are seatecl between the two membranes of the testa, as I have lon^r since described those of Casuarina. They differ, however, in direction; being in Collomia transverse or perpendi- cular, while in Casuarina they are longitudinal, or parallel to the membranes.

��ASCLEPIADEiE.

With regard to this Pamily, it was remarked, both in the Pamphlet which was distributed in the beginning of November, 1831, and in my Paper in the 'Transactions of the Linnean Society,' published in lSo3, that I had '3 never been able to find the Pollen tubes descending lower than the commencement of the ovuhferous portion of the Placenta. ]hit as this was far from satisfactory, especially after the further course of the analogous Tidjcs in Orchideae had been ascertained, I determined to re-examine the sub- ject.

For this purpose yhclcjjias phiitolaccoides was selected; and on the 12th of the present month I succeeded in tracing the Pollen tubes in that species, not only over the whole ovuliferous suiface of the Placenta, but also going off to the Ovula, to a definite point of each of which a single Tube was found in many cases attached.

These observations I have now so frequently repeated, and always with results so exactly similar, that I have great confidence in the correctness of the following statement:

In the newly expanded flower, the Ovulum in Asclcpias jjJii/toIaccoides is nearly obovate, and is compressed in the same direction as the ripe seed, but in a much less degree: its umbilical cord is inserted on the axis of the inner or ventral side, about one fourth from the apex, and a process

�� �