SELF-FERTILIZATION OF PLANTS. : 353 
the same species. Other cases seem to bear out this somewhat important fact, such as 
the case of the Viola tricolor with the protruding placentiferous process described above; 
as also that of the peculiar form of the same plant with the prolonged * tongue," 
enabling it to “lick” up the pollen from the spurred petal. 
Again, the arching of the styles outwards, and the filaments inwards, in the self- 
fertilizing condition of Alisma plantago, as also the subsequently inflexed condition of 
the stamens of Agrimonia, seem to point to what may be called individual adaptabilities 
for securing self-fertilization; and such structures as these may not precisely or ever 
occur again, while other peculiarities may be found adopted to secure the same end. — I 
would invite the attention of observers to this point, which appears important, but 
requires further corroboration. 
CARYOPHYLLACEJX.—The larger-flowered species of this Order are probably always pro- 
terandrous, and adapted for insect-fertilization ; yet, as Mr. Darwin has shown in the 
case of Dianthus Caryophyllus, which is very strongly proterandrous, there appears to be 
no difficulty in their becoming self-fertilizing by losing their dichogamy and so becoming 
homogamous; that is, the anthers and stigmas mature together. Mr. Darwin found, in 
his first experiment with Carnations, that there was little difference in the number of 
seeds produced by cross-fertilization and self-fertilization (artificially produced) in this 
plant. But in the second generation the crossed was more fertile than the self-fertilized 
in the ratio of 100: 65, both being grown much crowded. But when plants of the third 
generation were placed under a net, and both the crossed and self-fertilized were allowed 
to set seed spontaneously, the self-fertilized plants now produced more seeds than the 
self-fertilized ** crossed " plants, in the ratio of 125:100; and he adds :—“ This anomalous 
result is probably due to some of the self-fertilised plants having varied so as to mature 
their pollen and stigmas more nearly at the same time than is proper to the species,” 
. It must be borne in mind that proterandry in flowers admits of degrees; that is to 
say, there is no absolute length of time between the maturation of the pollen and that 
of the stigma. When the period is relatively long, as in Dianthus, some species or 
varieties of Pelargonium, Geranium pratense, and Malva sylvestris, every grain of pollen 
will have been shed, and sometimes the anthers fallen off as well, before the stigmas are 
nearly ready. On the other hand, Stellaria Holostea and Geranium pyrenaicum develope 
their two whorls of stamens in succession; and the stigmas maturing about the same 
time as the second whorl, these flowers are only partly purum upon insects. This 
condition of things accounts for many flowers retaining, or else regaining, their self- 
fertilizing properties, though still being adapted for intercrossing. 
` The point, then, I wish to bring forward as partieularly well shown by members of 
this Order, is that complete self-fertility may be regained with great ease, namely, by 
shortening the time between the periods of maturity of the anthers and stigmas, The 
causes to bring this about are ënn various, though somewhat difficult to isolate and 
specify exactly, 
With Dianthus Mr. Daititin records that it was of a pale’ pink or rose-colour, which 
mek indicate the approximate cause, assisted by continual and artificial homogamy ; so 
~ 
