98 ALFALFA FARMING IN AMERICA. 
close the pistil. A set of interlocking processes for the keel 
further assist in forming this.spring-trap arrangement. When 
an insect of sufficient weight alights upon the keel, it depresses 
the latter, together with the enclosing wing petals; the trigger- 
like processes are pushed down past the upcurved column of the 
pistil and stamens, releasing them, and allowing the whole col- 
umn to spring up with considerable explosive force against the 
erect standard. At the time of pollination the style with the 
stigma has grown up above the stamens, and when released the 
stigma precedes the stamens, striking the insect’s body first, in 
case the latter rests upon the keel, bearing its deposit of pollen 
brought from another flower. The burst anthers in turn dust off 
a new deposit of pollen as they are driven past the insect, which 
is thus equipped with a fresh supply of pollen to become avail- 
able for the next flower. Sometimes the shaking of the flower 
stems by the wind, or by the pelting of rain, many accomplish 
the same result. Self-fertilization may be secured also by visits 
of insects not yet loaded with pollen, which may, by setting off 
the explosive mechanism, bring about self-pollination. Since the 
pollen is shed before the stamen-pistil column is released, it 
happens that the stigma is already partly covered with pollen. 
Nevertheless, self-fertilization seems to occur but seldom in en- 
closed plants protected from insect visits. 
The explosive mechanism of the alfalfa flower has long 
been known, having been discovered as early as 1832 by A. P. 
De Candolle. 
In 1894, Burkill found it impossible to make seeds set in the 
unexploded flower, even though pollen were in contact with the 
stigma. He considers this fact to be due to the circumstance 
that the stigma does not become receptive to the pollen until its 
cells are injured by violent contact with some object. In proof 
he adduces the fact that he had caused unexploded flowers to 
set seed by pinching the stigma, by cutting off the tip of the keel 
or by rubbing the stigma with a stiff brush. It appears, there- 
fore, probable that insects secure the fertilization of alfalta flow- 
ers largely by accidental injury to the stigma while endeavoring 
to cause the proboscis to enter; or else by exploding the flowers 
and causing the stigma to be dashed against the standard, the 
necessary amount of injury may be accomplished to enable tk<¢ 
pollen to become effective, in which case it may either be the 
already present pollen of the same flower, or foreign pollen 
brought by the insect that is utilized. 
Thus it is plain that insects play a large part in 
