In Chrysanthemum leucanthemum according to Muller, the pistil of the ray florets possesses a terminal brush, which, however, is much less developed than in the disk florets. Matricaria camomilla agrees in most respects with Chrysanthemum. The strong smell of this flower, however, seems to be distasteful to bees, though Muller has observed it to be visited by Prosopis signata and Sphecodes gibbus. It is said to be generally fertilised by flies. Anthemis resembles the two preceding genera in many respects, but differs in possessing scales between all, or at least the central, florets of the receptacle.

The Common Daisy (Bel/isperennis) has ray florets 1 - 2 mm. in length, united into a yellow disk 6 mm. in diameter, and surrounded by a row of florets, each terminating in a white "flag" 5 mm. in length. These ray florets are exclusively female, and the pistil has lost the terminal brush of hairs. The two branches are long and clothed on their whole upper surfaces with rows of stigmatic papillae. The pistil of the ray flowers, on the contrary, has short branches, terminating in a tuft of hairs, and only provided with a small number of stigmatic papillae. When fertilised, the pistil retires again into the tube of the floret.

In Inula dysenterica (the Fleabane) the disk florets contain both stamens and pistil; the ray florets a pistil only, which, however, agrees exactly with that of the disk florets, even in the position of the terminal hairs, which in the absence of pollen, must apparently be useless.

In Tussilago farfara the disk florets are male, the ray florets female. In the disk florets the ovary is rudimentary; they contain honey at the base of the tube, which has a length of 4 mm. The pistil terminates in the usual tuft of hairs. The ray florets, on the contrary, produce no pollen; they open, and as the stigmas are mature, before the anther tubes of the disk flowers have opened, they are in fine weather almost always fertilised by the pollen from other flowers.

In the Common Groundsel (Senecio vulgaris), 60 to 80 florets are united on one receptacle. The lower, tubular, portion of the floret has a length of 3 1/2 to 4 mm.; the bellshaped portion, only of 1 - 1 1/2 mm. The flower heads have no ray flowers, and being therefore much less conspicuous than the allied species, are rarely visited by insects.

Carduus arvensis (Cirsium of some authors) is the commonest of our thistles. Each head contains about 100 florets. The tube of the florets is 8 - 12 mm. in length, the upper part forming a bell-shaped reservoir 1 - 1 1/2 mm. in depth, with five diverging linear lobes. As the lateral florets turn outwards, the whole form a flower head, as much as 20 mm. in diameter. Being therefore very conspicuous, and as the honey in this species and most of its allies rises into the cup of the flower, so as to be accessible even to insects with very short tongues, it is visited by a large number of species. Muller records no less than 88. In C. lanceolatiim, on the contrary, though it is also a very common species, still in consequence of the cup being somewhat deeper (4 - 6 mm. against 1 - 1 /2 in C. arvensis), and the honey therefore rather less accessible, he only records twelve. In C. palustris the depth of the cup is intermediate between those of the two preceding species, and also the number of insect visitors, namely 22.

Onopordon differs from Carduus only in the character of the receptacle, which does not bear chaffy bristles, as in that genus.

The genus Centaurea offers several interesting points. In C.jacea, which is sometimes, for instance by Bentham, regarded as a variety of C. nigra (the Knapweed), 60 - 100 florets are united into a head; the tubes of the florets are 7 - 10 mm., the cups 3 - 4 1/2 in length, each with five long, linear, lobes. The divergence of the outer florets gives the whole head a diameter of 20 - 30 mm. The hairs constituting the pollen brush are not situated at the extremity of the stigmas as in the preceding species, but form a ring round the pistil at the spot where it bifurcates. When the flower opens the pollen has been already shed into the anther tube in the upper end of which it lies, occupying the space between the anthers and the pistil, and supported by the ring of hairs. If now the flower remains untouched, after a while the stig-matic lobes separate, and some of the pollen falls on them. But if, as generally happens, an insect alights on the flower, or if in any other way the tip of the anthers is touched, immediately the stamens contract, exposing the pollen, which is supported by the stigmatic lobes. Gradually the pistil elongates, and the stigmatic lobes separate; by which time the pollen has generally been all removed, as the flowers, in consequence of their richness in honey, are much frequented by insects.

In C. nigra the outer florets are sometimes of the same size as the rest, sometimes larger, and without either stamens or pistils. In C. scabiosa this is always the case. The tubes of the florets also are longer, the cups deeper, and the honey less accessible, in consequence of which it has fewer insect visitors. Muller records only 21 against 48 in C. nigra.. In C. Cyanus also the ray florets are neuter. The contractility of the stamens is very marked. In flowers kept in a room, Muller observed that when touched, they rapidly withdrew themselves 2 - 3 mm., and then more slowly, 4 - 6 mm.

Taraxacum (the Dandelion). In T. officinale the heads consist of 100 - 200 florets. In fine weather they stand open, but at night and during rain they close completely. The two lobes of the stigma gradually curl over, so that if the visits of insects are delayed the flower always fertilises itself. The honey, however, is so abundant, and rises therefore so high in the floret, that it is very accessible to insects, no less than 93 species of which have been observed by Muller to visit this plant. The brightness of its colour, the quantity of its honey, the habit of closing in unfavourable weather, and the power of self-fertilization, go far to explain the great abundance of the Dandelion.

The genus Artemisia has minute greenish florets, and is said to be wind-fertilised.

This order is the subject of an admirable memoir by Hildebrand (Ueber die Geschlechtsverhaltnisse bei den Compositen).