SHORT CUTS BY BIRDS TO NECTARIES. 403 
I stripped only one side. I expect no results, however, as I examined a 
large number of Leonotis heads in various spots thereabouts and could only 
find two individual flowers that had even the corolla split. Also I examined 
eight Kniphofia plants the flowers of which had reached the apex and so had 
their bases exposed, and racemes with faults resulting in the same thing, and 
only found one flower in all that had been punctured. Evidently there are 
few or no birds of destructive tendency haunting that slope at present. For 
that matter sunbirds appear to be visiting the flowers there very little in any 
ase : I saw nothing at the numerous Leonotis plants and only one bird at 
a Kniphofia. This was а shrike, Chlorophoneus olivaceus (Shaw) Cab. He 
was perched on the ‘mat’ of a raceme and was thence probing quite slowly 
and carefully one after the other the flowers above him. It seemed utterly 
unlikely that his bill could enter the average flower without splitting it, and 
it struck me as a good opportunity to test this view as to the cause of the 
split perianths. On the shrike’s leaving, I took the flower: also, for com- 
parison, a neighbour 8 yards away and one 80 yards away. It was possible 
that the former and probable that the latter had not been visited by him. 
I noticed at once that many of the flowers in the shrike’s raceme were split, 
and absolutely freshly split, the edges glistening and wet; but the shrike 
had used no force and had done no tearing. I felt sare of this, having been 
close to the bird and having used my glasses as well. There were no wet 
edges in the 8-yard raceme.” 
I have now examined all three heads with the following result :— 
- | j i Split | More th ,ess the 
їз. Ый, кы р hal way. Dalt way, Halt way. 
1. Shrike's raceme.. 3l 24 22 6 22 10 18 
2. 8-yard 5 +. 78 4 9 10 | 1 18 4 
3. 80-yard ,, .. 46 0 0 0 | 0 0 0 
No. 1 has the largest flowers, No. 2 the next largest, Мо. 3 the smallest 
through a somewhat severely constricted neck which the others lack. 
In No. 1 at least 31 slits, from their fresh condition, seem to have been the 
work of the shrike. The remaining 18 are more doubtful, but even here 
some seem thoroughly fresh in part, as though the shrike in pushing well 
in had extended a previous slit. Comparing the fresh with the doubtful 
Т found: 
More than About 
half way. half way. Less. 
Probably shrike's flowers ........ 17 11 4* 
Possibly not his ............ .... 5 8 6 
* There is a discrepancy between the total number of slit flowers given in these three 
tables. In the first the total is 52, in the second 50, and in this 51. I can only suppose 
that I did not compare the totals at the time of counting, and that once one, once two 
flowers were omitted, probably amongst older flowers that were beginning to join the 
“ mat.” 
