SELF-FERTILIZATION OF PLANTS. 323 
16. They may be much more productive than flowers dependent upon insects. 
17. The most widely dispersed of British plants are almost, if not quite, always self- 
fertilizing. 
18. Naturalized abroad they often gain great vigour. 
19. They are **the fittest to survive in the struggle for life." 
I will now proceed to treat of each of these facts in succession. 
l. The majority of flowers are self-fertile. 
It is to Mr. Darwin's and H. Müller's investigations that I am partly indebted for 
this fact. Until their works were published I had, perhaps in common with others, a 
general belief that conspicuous flowers were much more self-sterile from dichogamy or 
other cause than is really the case. The former gives two lists of forty-nine species in 
each, one of self-sterile and the other of self-fertile plants; but he adds (J. c. p. 370):— 
* T do not, however, believe that if all known plants were tried in the same manner, half would be 
found to be sterile within the specified limits; for many flowers were selected for experiment which 
presented some remarkable structure; and such flowers often require insect aid." 
The limit was the production, when covered, of less than about half the number of 
seeds produced by unprotected plants. So that while 49, in Mr. Darwin's experiments, 
were generally highly self-fertile, the other 49 were not absolutely self-sterile when 
protected. Moreover, a large number of inconspicuously flowering plants from a great 
variety of Orders may be mentioned which are habitually self-fertile. Hence the above 
statement may be considered as established notwithstanding Asclepiadacez and Orchid- 
acee being large Orders and exceptional in requiring almost universally insect aid 
for the cross-fertilization of their species. 
2. Comparatively few flowers are known to be physiologically quite self-sterile. 
When the pollen of a self-sterile flower is placed upon the stigma of the same flower it 
is impotent. Mr. Darwin has dealt with this subject in his work on ‘ Animals and 
Plants under Domestication,’ vol. ii. p. 132, seqq., to which the reader must be referred 
for details. It appearsthat such plants are chiefly Orchids and of the genus Onicidium ; 
‘but examples are to be met with from widely different Orders: thus Corydalis cava, 
Lobelia fulgens, Verbascum nigrum, species of Passiflora, Ee, are in this condition; and 
I repeat that I cannot but think Mr. Darwin’s interpretation given in the above work is 
correct, and not that which he has lately advanced in his ‘ Cross and Self-fertilisation of 
Plants.’ In the former work he says, ** The sexual elements of the same flowers have 
become differentiated in relation to each other almost like those of two distinct species ;” 
but in the latter book he says, such sterility is ** due to the sexual elements not having 
been sufficiently differentiated " (p. 456). All the flowers here mentioned have their 
perianths in a highly differentiated state, as witnessed by their great irregularities, and it 
is at least reasonable to infer that the pollen is correlatively changed as well. This idea 
is certainly corroborated by the fact that the pollen of some species of Oncidium actually 
affects the stigma as a poison. 
But what is particularly noticeable is that, in the first place, allied species of such 
