S. Curtisti, the subject of the present plate. I cannot 
recall any Liliaceous genus in which the stem is erect in | 
one species and scandent in others, the leaves opposite in | 
some and alternate in others, and with the flowers appear- 
ing with the leaves in their axils in most of the species, 
but in one appearing on leafless stems; nevertheless, the 
above named plants are undoubtedly congenerics. When | 
describing S. Curtisii for the Flora of British India, Tfound | 
the only three flowers available for analysis to be unisexual,) 
but as both ovary and stamens are formed in the flower of} 
uh 
the Kew individual, the plant is probably polygamous. 
The only other genera of the group Roxrburghiacex are 
Croomia, Torr. and Gr. of Florida and Japan, and my 
Stichoneuron, a plant of the Khasia Hills; both resemble 
Stemona in the consistence and nervation of the leaves, 
but differ wholly in their minute flowers with simple oblong 
or didymous anthers, without appendages, and in the more 
important character of the ovules being pendulous from 
the top of the cell of the ovary. These differences suggest 
that Roxburghiacee should be broken up into two tribes of 
one Order, or perhaps better form two tribes of Liliacee, 
probably to be placed near Uvulariee or Medeolex, but 
differing from both in the introrse dehiscence of the anthers. 
S. Curtisii was discovered by Mr. Curtis, Superinten- 
dent of the Penang Garden, where it grows near the 
Waterfall. He sent Herbarium specimens to Kew in 1888, 
but the plant figured was received from the Botanical — 
Garden of Singapore in July, 1891. Iam not informed as — 
to whether it is a native of the Malayan Peninsula. It — 
was received at Kew in July, 1890, and flowered in a stove — 
in April of the present year.—J. D. H. 
Fig. 1, Flower with the perianth removed; 2, stamens seen in front; a 
3, ovary; 4, transverse; and 5, longitudinal section of the same :—@ 
enlarged. : : 
