396 MR. С, Е. М. SWYNNERTON ON 
Occasionally “faults” are present in а raceme—flowers twisted or un- 
symmetrically placed or over-scanty in some particular part. In these cases 
they themselves or their neighbours are usually more or less exposed, 
and slits or punctures in their exposed surfaces would often enable birds 
to extract the honey without coming into contact with the anthers and 
stigma. 
There are two main varieties of the plant at Chirinda—one with the 
perianth-segments well marked and commonly slightly splayed back, the 
other with these parts quite small and inconspicuous and т practically 
the same plane as the rest of the perianth. The former flowers are usually 
(not always) the larger and are very commonly rather less densely 
packed. In each variety there is a little further variation in the matter of 
the diameter of the perianths. Sometimes the flowers of a raceme are charac- 
terized by a much-constricted neck as in some heaths, 
“June Ist, 1913. Examined a few Kniphofia heads on the western ont- 
skirts of the Chirinda Forest. A very few split flowers were present, the 
slits probably not prejudicial to pollination. Abundant honey was welling 
forward in the tubes. 
“June 2nd. Examined two heads standing close together on the northern 
outskirts of the Chirinda Forest. Some of the lowest flowers showed damage, 
and all the open flowers on one side of the head were broken off to at least 
half their length. The nature of the laceration and the fragments lying 
below indicated that it was the work of a bird, as did claw-marks on the 
stem and very definite bill-marks on some of the flowers. Impatience with 
the dense array of downward-pointing stamens struck me as a possible 
reason for this wholesale damage. 
"Later I examined several heads above the dam. Very few were slit 
above, and these solely amongst the Howers that were already beginning 
to lie down fairly low, just above the mat—the work perhaps of claws. А 
few were slit on their under surface." 
“June 3rd. Watched in turn, for quite a long time, two bulbuls (Руспо- 
notus layardi, Gurney) at the flowers above the dam in the early morning. 
They always perched below the flowers, one foot well above the other and 
on the mat. If the mat was already extensive both feet were on it, other- 
wise the lower foot was amongst the triangular bracts that sparsely cover 
the upper end of the bare stem. The birds probed the flowers in a fairly 
leisurely way one after the other until the round was completed, when they 
would fly on to the next head. No attempt at breakage was seen, and 
whenever the bird's face was not in shadow the light appearance of the 
parts immediately round the bill showed that these must have been well 
dusted with pollen. 
“Г saw two sunbirds— $ and $ Cinnyris chalybeus—visit them too, and 
watched the female, which was nearer. She probed the lower flowers of each 
