SELF-FERTILIZATION OF PLANTS. . 337 
increases by work, or as the mamme may be made to secrete for prolonged periods ; so 
that, in my view, it is not that insects have gone to the flowers, because they were first 
conspicuous, but have actually themselves determined their conspicuousness. The final 
result has been that nourishment has been delayed from the pistil, and the flowers have 
become dichogamous and proterandrous. | 
That there should be no à priori objection to the idea of an external mechanical 
structure giving rise to organie changes in the tissue of a plant is obvious from such well- 
known cases as tendrils and carnivorous plants, as also from Mr. H. Spencer's experi- 
ments with Cacti, which showed an increase of vascular tissue when subjected to unusual 
mechanical strains. Perhaps these and other * responsive" activities may hereafter 
prove to be due to a widely extended principle of reflex action, to which Mr. Darwin 
attributes the curvature of the tentacles of Drosera. 
On the other hand, in the absence of insects, there is no such increase of energy in the 
corolla, &c., and the balance is restored with the result of self-fertilization. — 
A restoration to a homogamic equilibrium, then, I take to be the true explanation 
of pale varieties of Pelargonium being great seeders, and of Dianthus and other dicho- 
gamous flowers becoming highly self-fertile. Mr. Darwin records of Dianthus caryg- 
phyllus that it became highly self-fertile in three generations, and that the value of its 
fertility, and that of intercrossed plants, was as 125: 100. It was similarly with varieties 
of Primula veris and P, Sinensis (to be alluded to again). And I have myself found a 
wild Primrose with the style (of a “ short-styled form”) elongated so that the stigma 
was surrounded by the anthers at the orifice of the tube, and which was “setting” seed 
abundantly. 
Ido not, therefore, see how the conclusion can be avoided that self-fertilization is 
per se a decided advantage, and that intercrossing, as far as the production of seed is 
only regarded, a compensatory process for the loss of self-fertility. 
A similar explanation will, I think, apply to proterogynous flowers. If we regard 
homogamie equilibrium as the normal condition of hermaphroditism, then the pro 
terandrous state, as shown, results from the energy preponderating in the whorls 
external to the pistil; but if these be more or less suppressed, as seen in what is called 
contabescence of the anthers (Anim. and Pl. under Domest. ii. p. 165), which will be spoken 
of again below, there is a corresponding gain to the pistil, so that it may emerge con- 
jointly with the stamens, as with Nasturtium officinale, Lepidium campestre, or it may 
precede them in emergence, but mature together, as in Cerastium glomeratum, Arenaria 
trinervis and serpyllifolia. If, however, the pistil not only starts first in the race but 
retains the lead and ultimately matures its stigma first, then proterogyny is the result. 
Now as far as it has been observed as yet, proterandry is recognized as being by far 
the commonest condition amongst conspicuously flowering plants; white proterogyny is 
the exception with large-flowering dicotyledonous genera. Thus Helleborus viridis 
is, according to Hildebrand, proterogynous,. and here the corolla is replaced by small 
nectariferous tubes. Geraniwm pratense and other large-flowered species are pro- 
terandrous. G. pyrenaicum matures its pistil conjointly with the inner whorl of stamens; 
