898 МВ. С. Е, М. SWYNNERTON ON 
to the left had their right sides punctured, those to the right their left sides, 
those leaning pointwards their near side. One flower had as many as five 
punctures, some others three or two. Drops of nectar were oozing from the 
more basal punctures, and the damage was in all cases such as would obviate 
contact with the anthers and stigma. In addition to these punctures and 
bill-marks were a few long slits of the type already described, continuous 
with the margin of the perianth. These were not entirely confined to the 
exposed surfaces *. The six undamaged exposed flowers were those nearest 
the tip, and had possibly only opened since the bird’s visit. 
“Of the remaining 114 flowers that were normally placed and mutually 
protective only six were damaged, and these with only the usual slits I have 
just referred to. 
* No. 2. (No. 1’s immediate companion) : 5 slits, 4 small bill-point marks 
scattered at random and on non-exposed surfaces, probably accidental. 
Flowers every where densely mutually protective. 
“No. 3. A very perfect specimen with a very large number of flowers 
(approximately 306 out at the moment), densely packed. There were 48 of 
the usual slits, most of them very slight and all of them most certainly 
accidental, as none could have assisted in the procuring of the honey or the 
evading of the pollen. 
* No. 4, A small head with few flowers (about 100 out) had a bare tract 
half an inch wide extending spirally from the tip of the raceme to nearly 
the base of the flowers actually out. The flowers immediately below it had 
their bases exposed. The highest (latest out), to the number of 10 or 11, 
showed no marks on their exposed side and only one of the usual underside 
slits. Of the 6 flowers exposed lower down the spiral, 4 were slit on the 
side exposed, none elsewhere. Of the remaining approximately 84 flowers, 
mutually protective but less utterly densely packed than in No. 3, 7 had the 
usual slight slits. 
“No. 5. A still smaller head. The latest flowers to commence to open were 
already those at the tip of the raceme, and a bird perching there might well 
have been expected to slit their upper surface as is done т Leonotis. There 
was no such damage, but the flowers were in any case barely ready. 
Slightly lower, however, there was a fault—a number of flowers, a closely- 
packed bunch of them, pointing upwards at as strong an angle as they should 
have taken downwards. The flowers immediately below them pointed in the 
normal direction. "The result, of course, was that both lots were exposed, 
8 аз to their lower surface, 5 as to their upper. Four of the latter were punc- 
tured (3) and slit (1) on their upper exposed surfaces only (none elsewhere), 
* An observation yet to be described subsequently threw light on the causation of these 
slits, It showed that they were purely accidental and need not be found more on exposed 
suifaces than on others. 
