PORTUGUESE EAST AFRICA (SOUTH) 95 
which feed on the flowers of the high trees, and do not descend 
near the ground. These birds do not normally fly through the 
branches of trees, as their food is on the outside, so they are ex- 
tremely difficult to catch in the ordinary way. It is perhaps as well 
for my reputation that I was in the wilds of Africa while carrying 
out this experiment, for I feel sure if anyone had seen me pulling 
a thread and causing a stuffed owl to perform strange antics on a 
stick, he would have had serious doubts as to my mental condition, 
especially as I could not help grinning when a few birds arrived 
and gazed on the restless yellow-eyed demon below with great 
alarm. Strange as it may seem, this “Heath Robinson” device was 
very successful and enticed quite a few rare sunbirds. However, 
they were not inclined to fly through the gaps where the nets 
were set, but merely hopped from one branch to another. 
Unless a bird is going at speed, hanging nets will not function, 
as some force is required to form a pocket, also a net is easily 
detected if it is approached gradually. To remedy this state of 
affairs, I crept from my hiding-place and threw a fair-sized stone 
high into the air, so that in falling it crashed through the centre 
of the tree, causing all the birds to fly outwards through the gaps 
at full speed. I, of course, rushed to the tree and climbed it rapidly 
to find that quite an assortment of birds was caught, including 
three of those I specially wanted. 
After that, this same stuffed owl performed signal service in 
many lands, but finally perished on the field of battle on another 
expedition. I was trying him out one morning, but was unable to 
get a start. This is always the most difficult part, for once a few 
birds arrive and begin chattering excitedly, the rest follow as a 
matter of course. As it was a very hot day, I went home for a 
drink and left the owl on a tree stump, but when I returned it 
had vanished. After a brief search I found a few tell-tale feathers, 
and following the trail, found my poor owl under a thick bush 
with his cottonwool innards hanging out. Evidently during my 
absence it had gathered an audience, and the pandemonium that 
ensued must have attracted a carnivorous animal, such as a mon- 
goose, on the look-out for his dinner. The owl, being nearest to 
the ground and not being able to look after himself, was attacked 
and “killed” for the second time in his long and varied career. | 
