418 DR. OTTO STAPF AND MR. J, HUTCHINSON ON 
conceived the idea of revising the whole group, and of placing its con- 
stituents on a more equal basis. For that purpose we studied, in the first 
place, the material at Kew ; but we had also, of course, to direct our attention 
to the valuable material in the herbarium of the British Museum ; we asked 
for and had, thanks to the courtesy of. Dr. De Wildeman, the loan of the 
“ Gardenia Thunbergia ” specimens of the Congo State collections ; and 
finally we availed ourselves of Mr. Casimir De Candolle’s ever ready help for 
the elucidation of some critical species, the types of which were in the 
Candollean herbarium. We might have extended the field of our investiga- 
tion, but for want of time and the conviction that after all the time had not 
yet come for attaining anything like finality. What we ean and ought to do, 
in cases like the present, is to revise our knowledge from time to time and 
clear the basis for a further advance. Our paper is an attempt in this 
direction, The Gardenias of the “ Thunbergia " group are rather striking 
objects when in flower, and it may be hoped that the collectors and field- 
students will, in the future, pay more attention to them, and supply us with 
more complete material and. their own observations on the mode of growth, 
the range of individual variation, and any biological conditions which may 
influence their morphological characters. 
The genus Gardenia, in the sense of Bentham and Hooker’s * Genera 
Plantarum,’ consists of three sections: fu-Gardenia, Ceriscoides, and. Roth- 
тапта. Те first section ranges over the whole of the palæotropic region, 
and is represented in Africa by the Gardenia Thunbergia group—the subject 
of the present paper—and two more species which are characterised by long, 
almost cylindrical fruits. From the Indian, Malayan, and Polynesian 
species the “ Thunbergia” group differs on the whole in the number of 
the placentas, which rarely falls below six, and the more or less abundant 
development, in the mesocarp, of sclerotic fibres, which in the endocarp fuse 
into a very hard woody shell, and finally in the more xerophilous structure 
of the leaves. The species forming this group are very closely connected. 
This and the very considerable variation in the size and shape of the leaves, 
the size of the flowers, and the degree of development of the calyx-teeth or 
lobes—even in the same individual—make the discrimination of the species 
very dificult if the specimens are scrappy or badly selected. The number of 
parts in the plan of the flower, though also subject to a certain amount of 
variation, is much more constant, and the same may be said of the surface 
condition and the nervature of the leaves, and the shape of the calyx-teeth or 
lobes. The very copious material of G. Jovis-tonantis which we had before 
us was most instructive in this respect. Then there are certain species 
which stand out from the rest on account of some special peculiarity, as the 
cucullate and more or less petioled sepals of G. Thunbergia, the cuff-like 
prolonged calyx-tube of G. cornuta, with its laterally attached segments, the 
