370 REV. GEORGE HENSLOW ON THE 
potent at first, if continued for a long while completely lose their effect by the system 
becoming attuned to them. 
Sotanacea.—Solanum dulcamara is said to be not much visited by insects, yet it 
sets its berries very freely and is more or less proterandrous. S._nigrwm, however, 
is decidedly self-fertilizing. The flowers are tolerably inconspicuous and pendulous; the 
anthers are close round the style, but are not * sub-syngenesious," as in S. dulcamara. 
They dehisce by slits all the way down, and so shower their pollen over the stigma, 
which sets its berries in great profusion. (Tab. XLIV. fig. 22.) The anthers do not | 
appear to have the spirals in the cells of the walls, which indieates, therefore, that this 
method of dehiscence is a subsequent acquirement from the condition obtaining in 
S. dulcamara, the anthers of which dehisce by pores. A. nigrum, therefore, affords 
another proof that existing self-fertilizing plants are all de&radations from more con- 
spicuously flowering ancestors. S. nigrum is also white, the commonest colour of self- 
fertilizing plants. > 
Nicotiana tabacum proved to be fully self-fertile with Mr. Darwin. A plant growing | 
in Kew Gardens was clearly proterandrous; but W. rusticum, which has smaller dingy 
green flowers, was as obviously self-fertilizing. The pollen-tubes were abundant. 
SCROPHULARIACEX.— The many genera with conspicuous flowers, and all being more ` 
or less irregular, are obviously adapted for insect-fertilization ; but there are several 
inconspieuous ones which are habitually seif-fertilizing. "Thus of the genus Veronica, 
V. spicata, blue garden-variety, is proterandrous, and so much so, that in some instances 
it is not until the corolla and stamens have actually fallen off that the style elongates * 
(Tab. XLIV. figs. 26 a, b, c). "This and V. Chamedrys, with its brilliant blue corolla, are 
intercrossed. The order of development of This species is—calyx, stamens, pistil, corolla ; 
and though the anthers and stigmas mature apparently quite or nearly simultaneously, 
yet the stamens spread away, while the style projects forwards and is not, therefore, 
specially adapted for self-fertilization. On the other hand, V. hederifolia, V. serpylli- 
folia, and V. agrestis, which is, perhaps, the self-fertilizing form of V. Buxbaumii, 
as well as V. Anagallis, are self-fertilizing (Tab. X LIV. fig. 27). In these the order of 
development is the same, viz. calyx, stamens, pistil, corolla. In V. serpyllifolia the ` 
pistil rapidly elongates, even beyond the stamens, just before expansion; the stamens - 
* Such, at least, applied to some specimens I found growing in a much-shaded garden, and have recorded in my 
notes that the corolla had the tube densely clothed with upturned hairs. The style, only just or not at all protruded» 
is green and the stigma immature, while the anthers, on elongated filaments, are turned vertically downwards. As 
soon as they have shed their pollen the corolla is ejected. The style then elongates and curves slightly downwards, 
becomes bright blue, and the stigma enlarges. The relative length of the style is now the same as that of the 
stamens, or one fourth of an inch. The development of the stigma subsequent to the ejection finds its parallel in 
Sambucus Ebulus, antea, p. 366. Much the same occurred in a white variety, only the style was of the same length ` 
as the stamens when the anthers dehisced, but the stigma still immature after the ejection of the corolla; the style a 
elongated to about half of its former length, then shrivelled. No seeds were detected in any capsules. The ` 
order of development is—calyx, stamens, corolla, pistil. Since this note was penned, I find that Müller has also dis- - 
covered this plant to be very peculiar; for he has met with plants which present exactly the reverse conditions, in ` 
that they are are while = like those seen by me, were Leg Zeie ad) He Bat both. kinds i c. 
p. 287).. ae 
