116 
compensates in a measure for this by bringing a good part of the 
stigma to the upper side of the style, so as to rub against the 
body of an insect trying to reach the nectar in the most advan- 
tageous way. For, in looking at a flower from the outside, the 
twisting is seen to be towards the left hand. The style makes a 
turn, bringing that part of the fork axially on the right over to 
the left. Since the cleft extends about half way down, this part 
of the fork, in the effort to withdraw from the other and turn at 
the same time, rolls its inner face upward for a good portion of its 
length, being still held to the opposite part at the upper end; 
and turning it wholly upward, except at the tip (which may have 
turned too far for this), in case it is released. By the same pro- 
cess, the part of the fork axially on the left passes over to the 
right, much of the inner surface being turned downward. Some- 
times the tips of the parts in either case may turn so far as to 
bring the ends of the stigmas upward or downward, and opposite 
to the position of their main parts respectively. This is a great 
advantage, by making the larger part of the stigmatic surfaces 
easily accessible, since the parts of the style diverge but little © 
when they succeed in separating, for the interlocking is often 
quite strong. The erection of the style, a characteristic of species 
of Sadbaiza in later anthesis, is not a prominent feature of S. angu- 
laris. In fact, when it occurs, it is of no special advantage for 
cross-fertilization in this plant, the stigmas being well exposed 
while the style is turned to one side. Fertilization may thus 
begin at an earlier stage, with fewer chances of failure, should the 
parts remain locked together. Besides, the style, being quite 
stiff, when turned to one side offers a better landing-place to a 
large insect, by affording more room on one side of the corolla of 
a flower not very broad. 
_ The object of these visits is to get the nectar stored in recep- 
tacles attached to the base of the ovary. These project outward 
and downward as little swellings, each nectary opposite a petal 
and to a point of the star forming the eye at its base. Hence - 
they alternate with the stamens, which are of the same number 
_ as the lobes of the corolla and opposite its clefts. It is between 
the filaments that an insect would run its proboscis to reach ‘the 
-nectar-glands, the bright line raised up into blunt points exactly 
