OIST TURNERA ULMIFOLIA. 
117 
Both honey and pollen are produced in considerable 
abundance and appeared about equally attractive. The only 
other plants at Peradeniya which I have seen visited by bees 
with equal eagerness are Zea Mays—the male flowers— 
and Oreodoxa regia. The latter case was pointed out to me 
by Mr. Willis. It is noteworthy that all three are introduced 
plants. 
Seed Distribution by Ants .*■—The seeds of Turnera elegans 
are 2 or 3 mm. in length and somewhat pear-shaped. The 
surface, although covered with an indented pattern, is very 
smooth and hard. Attached near the small end of the 
seed is a large shield-shaped aril which covers nearly half 
the surface. When the seeds are freshly ripe the aril is 
soft and fleshy, and under the microscope its cells were 
observed to contain much starch and oil. The fruit is a 
capsule, which when ripe opens by three apical teeth, which 
gradually widen into valves and allow the seeds to fall 
quietly out. 
Certain species of ants, among others Phidole spathifera 
Forel , on coming across the seeds appeared almost invari¬ 
ably to pick them up and carry them away. I watched 
a member of this species carry a seed straight to the mouth 
of its nest, a distance of five yards. Watching the nest 
on another occasion, I saw more than thirty seeds carried 
in at a rate of quite one per minute. This was shortly after 
I had assisted by scattering numerous seeds under the plants. 
The seeds were always carried by the aril. But I also saw 
occasionally ripe seeds without the aril brought up out 
of the nest, carried away, and dropped at a little distance. 
These seeds were quite undamaged, but appeared dull on 
the surface as if they had been underground some time. 
I placed a number of seeds close to the opening of the nest 
and these were usually carried into it at once by the finders. 
But there appeared to be some conflict of instincts, for 
* Cf. Sernander, Den skandinaviska Vegetationens spridningsbiologi, 
