498 DR. KIRK ON A NEW GENUS OF LILIACEAE. 
Africa, by Mr. H. Waller, at an elevation of 4000 feet above the sea-level, not far from 
the mission-station of the late Bishop Mackenzie. 
Neither root nor ripe fruit have been sent: the latter, judging from young specimens, 
promises to be a many-seeded capsule. 
In the absence of these important parts, some doubt remains as to their position in the: 
natural system, from the complicated affinities existing in the Liliaceæ. That they form 
part of the Conantherous group seems, however, almost certain, from the remarkable 
similarity of the perianth and stamens. The ovary, however, in being quite or almost 
free, presents a marked point of difference. 
Having been kindly permitted to examine the very complete series of allied plants 
in the Hookerian Herbarium, I shall mention briefly the leading peculiarities of each 
genus. 
This group differs from other Lilies in its semiadherent ovary and anthers, which are 
not versatile, opening by terminal pores. The amount of adhesion, however, varies, 
while in one of the genera we have anthers opening by longitudinal valves. The present 
genus, which it is proposed to add, has the ovary in one quite free, in the other slightly 
immersed, while the stamens are identical with those of Conanthera in attachment, shape, 
cohesion, and dehiscence. 
At first it may seem anomalous that there are in Africa members of a group other- 
wise peculiar to South America; but it will be found that the Cape genus Cyanella must 
also be referred here, and take a position nearer to the typical genus than some others 
commonly admitted. 
The characters of the group are most fully represented in Conanthera and Cummigia, 
which differ from each other in the perianth of the one being 6-parted, of the other 
campanulate ; in these the stamens are united near the apex in the manner of Solanum. 
Zephyra has two stamens abortive, represented by curved filaments, while the fertile 
anthers are spurred at the base. Cyanella, a genus of Cape plants, has not only the habit 
of the group, but also the structure : its ovary is semiimmersed in the stalk; the stamens 
open at the apices, while the amount of irregularity is less than in Zephyra—one of the 
stamens being larger than the others, but all fertile. 
In Pasithea the ovary is semiadherent; but the stamens open by longitudinal valves, 
offering a transition in this respect to the ordinary structure of the Anthericez. 
The new genus connects these two groups in like manner: retaining the staminal 
arrangement of the Conantherez, it has the ovary of the Anthericez, which in one of the 
species shows a tendency to adhesion. 
It is proposed to name the genus after its discoverer, Mr. Horace Waller, one of the 
few survivors of the ill-fated Central-A frican mission, to whom we are indebted for many 
additions to the flora of that region. 
