| 242 MR. W. FAWCETT ON A NEW SPECIES OF THONNINGIA. 
There are no stomata present on the scales, or, in fact, on any part of the plant. 
The flattened receptacle of the male capitulum contains few flowers as compared with 
the female capitulum. Round the circumference there are a few scales shorter than the 
flowers. Surrounding the base of the male flowers there are scale-like bodies of fleshy 
character, with the surface of the upper part papillose, which possibly correspond to the 
bodies on the male capitulum of Langsdorffia, considered by Hooker * and Eichler to 
be abortive ovaries. 
In the female capitulum the flowers are exceedingly numerous, probably as many as 
4000. As they are slightly connected together, they form a continuous layer, completely 
covering the receptacle, which is convex. The inflorescence is centripetal. At the 
circumference there are scales, sometimes as long as the flowers, but generally shorter ; 
they are probably true bracts, but none occur amongst the flowers themselves. 
The tissue of the male flower and its pedicel is very like that of Langsdorffia, the 
cells being very thin-walled, and longer than broad. The pedicel contains a single 
fibro-vascular bundle, and therefore cannot be regarded as a staminal column with the 
tube of the perianth adnate, as Eichler has remarked for Langsdorffia i. The lobes of 
the perianth are reflexed in flowering. There is a distinct layer of epidermal cells with 
rugose markings on their outer walls; there is no fibro-vascular bundle nor any 
sclerenchymatous cells. The staminal column is solid, with three fibro-vascular bundles, 
which indicate three anthers. I have not been able to determine exactly the number of 
cells in each anther, as the flowers are too far advanced, but there appear to be 4, and it 
is evident that they dehisce by longitudinal chinks, as in 7. sanguinea, Vahl. The pollen 
is subglobose. 
The female flowers agree in appearance with those of T. sanguinea, Vahl, as described 
by Hooker $. І have not been able to trace the development of the flower, but have 
little doubt that the so-called “perianth” is really a prolongation of the ovary, as 
Kichler has shown to be the case in Lophophytum and Нео |. Even after the 
flower has become fully developed, the perianth-like limb increases in width, especially 
at the base, the mouth becomes contracted and covers up the persistent lower portion of 
the style. This does not appear to be the result of fertilization, for in flowers which 
never produce fruit the same thickening occurs, so that finally they have a very different 
appearance from those in the flowering condition ; for, instead of being cylindrical with 
the styles protruding, they become clavate. The flowers of the three or four rows 
nearest the circumference are fleshy and thicker than the rest. In proportion as the 
fruit ripens, the receptacle grows more convex, thus providing a greater surface for the 
enlargement of the fruit, while the hardened limb forms a protecting cap. The ripe 
fruit is similar in most respects to that of Langsdorffia, as described by Eichler Т. Тһе 
* Trans. Linn. Soc. xxii. р, 41, tab, 2. fig. 4, 
+ Mart. Fl. Bras. fasc. 47. Balanophores, p. 18, tab. 3. fig. 14. 
+ Ibid. p. 17 (note). 
$ Trans. Linn. Soc. xxii. p. 43, tab. 3. figs. 10-15. ; 
| A. W. Eichler, “Sur la structure de la fleur femelle de quelqu 5 
es Bal h "d оте j 
tional de Botanique, Paris, 1867. que anophorées," Actes du Congrés Interna 
4 Mart. Fl. Bras. fasc. 47. Ва!апорһогеге, p- 20, tab. 3. figs. 15-21. 
