1.]

EXPLANATION OF THE DEADNETTLE.

19

them several rows of anthers (a). It might be supposed therefore that the pollen from the anthers would fall on and fertilise the stigmas. This, however, is not what occurs. In fact the stigmas come to maturity first, and have lost the possibility of fertilisation before the pollen is ripe. The pollen must therefore be brought by insects, and this is effected by small flies, which enter the leaf, either for the sake of honey or of shelter, and which, moreover, when they have once entered the tube, are imprisoned by the fringe of hairs (). When the anthers ripen, the pollen falls on to the flies, which in their efforts to escape get thoroughly dusted with it. Then the fringe of hairs withers, and the flies, thus set free, soon come out, and ere long carry the pollen to another plant.

Now let us return to our White Deadnettle and see how far we can answer the questions which I began by asking.

In the first place, the honey attracts insects. If there were no honey, they would have no object in visiting the flower. The bright colour is useful in rendering the flower conspicuous. The platform serves as an alighting stage for bees. The length of the tube has reference to that of their proboscis, and prevents the smaller species from obtaining access to the honey which would be injurious to the flower, as it would remove the source of attraction for the bees, without effecting the object in view. The upper arch of the flower protects the stamens and pistil, and also presses them firmly against the back of the bee; so that, when the bee alights on the stage and pushes its proboscis down to the honey, its back comes into contact

20

SALVIA.

[CHAP.

with them. The row of small hairs at the bottom of the tube prevents small insects from creeping down the tube and stealing the honey. Lastly, the small processes on each side of the lower lip are the rudimentary representatives of parts, formerly more largely developed, but which, having become useless, have almost disappeared.

In the Deadnettle, it would appear that the pistil matures as early as the stamens, and that cross-fertilisation is attained by the relative position of the stigma, which, as will be seen in the figure, hangs down below the stamens; so that a bee, bearing pollen on its back from a previous visit to another flower, would touch the pistil and transfer to it some of the pollen, before coming in contact with the stamens. In other species belonging to the same great group (Labiata) as Lamium, the same object is secured by the fact that the stamens come to maturity before the pistil; they shed their pollen, and shrivel up before the stigma is mature.

Fig. 14 represents a young flower of Salvia officinalis in which the stamens (a a) are mature, but not the pistil (p), which, moreover, from its position, is untouched by bees visiting the flower; as shown in Fig. 15. The anthers, as they shed their pollen, gradually shrivel up; while, on the other hand, the pistil increases in length and curves downwards, until it assumes the position shown in Fig. 16, st, where, as is evident, it must come in contact with any bee visiting the flower, and would touch just that part of the back on which pollen would be deposited by a younger flower. In this manner cross-fertilisation is effectually secured.

1.]

SALVIA.

21

There are, however, several other curious points in which S. officinalis differs greatly from the species last described.

FIG. 14.-Salvia officinalis. Section of a young flower.

FIG. 15.-Ditto, visited by a Bee.

FIG. 16.-Ditto, older flower.

st

The general form of the flower, indeed, is very similar. We find again that, as generally in the Labiates, the corolla has the lower lip adapted as an alighting board for insects, while the arched upper lip covers and protects the stamens and pistils.

22

SALVIA.

[CHAP. I.

The arrangement and structure of the stamens is, however, very peculiar and interesting. As in Lamium, they are four in number, but one pair is quite rudimentary (Fig. 14). In the other (a a) the two anthers, instead of being attached close together at the summit of the filament, are separated by a long movable rod, or connective (Figs. 17, 18, m), so that they can play freely on the stalk of the stamen. In a natural position, this connective is upright, so that the one

the other under the arched hood. The lower anther, moreover, is more or less rudimentary. Now when a bee comes to suck the honey, it pushes the lower anther out of the way with its head; the result of which is that the connective swings round, and the upper fertile anther comes down on to the back of the bee (Figs. 15 and 18), and dusts it with pollen, just at the place where, in an older flower (Fig. 16) it would be touched by the stigma, st.

AT first sight, it may seem an objection to the view set forward in the preceding chapter, that some flowers-as, for instance, those of the common Antirrhinum-which, according to the above-given tests, ought to be fertilised by insects, are entirely closed. A little consideration, however, will suggest the reply. The Antirrhinum is especially adapted for fertilisation by humble bees. The stamens and pistil are so arranged that smaller species would not effect the object. It is therefore an advantage that they should be excluded, and in fact they are not strong enough to move the spring. The Antirrhinum is, so to say, a closed box, of which the humble bees alone possess the key.

The common Heath (Erica tetralix) offers us a very ingenious arrangement. The flower is in the form of