Chap. III. CEPHALANTHEEA GRANDIFLORA. 85 



uncovered plants weighed 1 • 5 grain ; wliilst those 

 from an equal number of capsules on the covered plant 

 vveighed under 1 grain ; but this does not give a fair 

 idea of the relative difference of their fertility, for I 

 observed that a great number of the seeds from the 

 covered plant consisted of minute and shrivelled husks. 

 Accordingly I mixed the seeds well together, and took 

 four little lots from one heap and four little lots from 

 the other heap, and, having soaked them in water, com- 

 pared them under the microscope : out of forty seeds 

 from the uncovered plants there were only four bad 

 ones, whereas out of forty seeds from the covered-up 

 plants there were at least twenty-seven bad ; so that 

 there were nearly seven times as many bad seeds from 

 the covered plants, as from those left free to the access 

 of insects. 



We may therefore conclude that this orchid is 

 constantly self-fertilised, although in a very imperfect 

 manner ; but this would be highly useful to the plant, 

 if insects failed to visit the flowers. The j^enetra- 

 tion of the pollen-tubes, however, is apparently even 

 more serviceable by retaining the pillars of pollen in 

 their proper places, so that insects, in crawling into the 

 flowers, may get dusted with pollen. Self-fertilisation 

 also may, perhaps, be aided by insects, carrying pollen 

 from the same flower on to the stigma ; but an insect 

 thus smeared with pollen could hardly fail likewise to 

 cross the flowers on other plants. From the relative 

 position of the parts, it seems indeed probable (but I 

 omitted to j)rove this by the early removal of the 

 anthers, so as to observe whether pollen was brought 

 to the stigma from other flowers) that an insect \\»ould 

 more frequently get dusted by crawling out of a flower 

 than by crawling into one ; and this would of course 

 facilitate a cross between distinct individuals. Hence 



