273 
in South Africa. 
fusion of reddish yellow flowers. There are some very peculiar 
features about this flower. The lower lip of the corolla is 
withered and bent back just as in Leonotis: The two side 
halves of the upper lip, moreover, are bent downwards so that 
they touch one another, thus completely enclosing the anthers. 
The lever arrangement is very perfect : the short stout fila- 
ments of the stamens are about three lines apart, and this 
interval is completely blocked by the broad spoon-shaped 
barren lobes of the connective 1 . The style projects about one 
and a-half lines out of the upper lip, and is curved forwards. 
The nectary forms a sort of cushion two lines in diameter and 
one line high, on the top of which are the nullets. 
I found this flower was visited by Zosterops capensis in the 
Cape Town gardens. I could see the process distinctly as it 
was within two yards, and cross-fertilisation is certain. I 
have also seen wild plants near Seapoint visited by birds (not 
Nectariniae) taking sips at the honey. I have never seen in- 
sects on it though I have watched it several days at different 
•times for an hour or so, and it is probably truly ornithophilous 
like some other Salviae 2 . 
SARCOCOLLA SQUAMOSA, Bth. 
The flowers are bright scarlet and closely packed together at 
the ends of the branches. Both the corolla-tube and involucral 
bracts are excessively viscid, so that insects are largely kept out. 
The corolla-tube is about one inch long, and the free ends 
are scarlet and bent outwards. The stamens converge together 
and are united to one another and the style just below the 
stigma. When pushed apart by a needle a cloud of pollen is 
scattered from the anthers. The anthers in older flowers drop 
off outside the corolla as the stamens bend outwards till the 
anthers are beyond the corolla-tube. Honey is abundantly 
secreted by the base of the ovary. 
1 Each of these is one and a-half lines broad ; there is an oblique depression on 
the inner side at the end of the filament. In the centre of this depression is a 
small peg attached to the connective. 
2 Cf. F. Muller, Bot. Zeit. xxviii. p. 274. 
T 
