112 
LOCK: ECOLOGICAL NOTES 
attention was naturally attracted by the behaviour of bees 
and ants visiting the plants, and the results of my watching 
are embodied in the following notes. It appeared worth 
while also to record the facts of heterostylism and self¬ 
sterility. As might have been expected, my notes fully bear 
out Darwin’s comprehensive work on other plants. 
Heterostylism .—In the long-styled form the average length 
of the styles of twenty flowers was 12 mm. including the 
stigmas. The total length of the stigmas themselves was on 
the average 2*5 mm. The average length of the stamens 
from their point of first fusion with the receptacle to the 
extremity of the extrorse anther was 7*5 mm. In the short- 
styled form the average length of the styles from the same 
number of flowers was a little over 6 mm., the stigmas were 
1*5 mm. in length, and their branches were finer and more 
numerous than in the long-styled form. The average length 
of the stamens was 108 mm. The average diameter of 
pollen grains from the long-styled flowers was 36 from 
the short-styled 44 p ; they are in the ratio of 9 to 11. 
The size of the unfertilized ovules was approximately the 
same in the two forms ; their diameter was about *4 mm. 
All these measurements were made upon good-sized 
flowers from vigorous parts of the plant, and in these the 
variation from the sizes given was not great. Much smaller 
flowers occurred at the extremities of old flowering branches 
in both forms. 
Contrary to the state of things described by Darwin* in 
the case of the cowslip (Primula veris), the long-styled 
form regularly produces a much greater quantity of seed 
than does the short-styled. Three well-grown bushes of 
each kind were examined in May. With free pollination 
practically every flower of the long-styled form produced a 
capsule. Two of the short-styled plants produced some 90 
per cent, of capsules; the third, not 50 per cent. Moreover, 
the capsules produced by the long-styled plants were 
Different Forms of Flowers,” ed. II., 1880, p. 20. 
