This order contains seventeen British genera, including Prunus (the Cherry, &c), Spiraea, Geum, Rubus (Blackberry, &c), Fragaria (Strawberry), Potentilla, Alchemilla, Sanguisorba, Poterium, Agrimonia, Rosa, Pyrus, Crataegus, etc.

Prunus. Our three species of this genus differ somewhat in the relations of the anthers to the stigma. In P. cerasus (the Cherry) both mature at the same time, while in P. spinosa (the Black-thorn) and P. padus (the Bird Cherry) the stigma reaches maturity before the anthers: though as it retains the capability of fertilisation after the anthers have opened, the flowers are doubtless often self-fertilised; which from the position of the anthers probably happens more frequently in the Bird Cherry than in the Blackthorn. The flowers are melliferous. The British species of Spiraea, on the contrary, contain no honey, but are rich in pollen and are consequently visited by insects; which, from the weakness of the petals, generally alight on the stigma, and thus effect cross-fertilisation; though the flowers, if not visited by insects, fertilise themselves. Among the foreign species of this genus, several are melliferous.

Both our English species of Geum (G. rivale and G. urbanum) are melliferous: but the flowers of G. rivale are much larger than those of G. urbanum, and more frequently visited by insects. Muller mentions that Primula elatior is deserted by bees as soon as Geum rivale comes into flower. Yan Tieghem states that while G. urbanum produces honey in the north, this is not the case in France, at least, near Paris.

The genus Rubus is very variable, and there are great differences of opinion among botanists as to the specific limits, and the number of species. Bentham admits five, though even these, he adds, "will very frequently be found to pass imperceptibly one into the other." The Raspberry (Rubus idaeus) is so called because it is said to be very frequent on Mount Ida, where in 1872 Mr. (now Sir M. E.) Grant Duff and I found in abundance a species, which if not identical with, was very near, our R. idaeus. This species, though it secretes honey, is not apparently a great favourite with insects, and frequently fertilises itself. The flowers of the Blackberry (R.fruticosus), on the contrary, are much more conspicuous, and the stamens are turned more outwards, so as to leave more room between themselves and the pistil. They are very much frequented by insects, and as the stamens ripen gradually, and from the outside inwards, there is a considerable interval during which, though the pistil is mature, and some of the anthens are ripe, self-fertilisation is difficult; while from the great frequency of insect visits, fertilisation is generally effected before the inner anthers are mature.

In the Strawberry (Fragaria vesca) also, the stigma arrives at maturity some time before the anthers, so that cross-fertilisation generally takes place. The species of Potentilla agree with Fragaria in habit, foliage, and flowers, but the fruit is not succulent. The honey, however, is not secreted in drops, but forms a thin layer. According to Van Tieghem, P. tormentilla produces honey abundantly in the north, but scarcely any in the neighbourhood of Paris. Agri-monia Eupatoria appears to secrete no honey, and is but seldom visited by insects. Alchemilla vulgaris is remarkable for variability. The honey is scanty, so that it is little visited by long-lipped insects; while, from its greenish colour, it is not attractive to beetles, or other colour-loving species. Self-fertilisation is, however, comparatively rare, since the flowers seldom possess both anthers and stigmas; one or the other being generally more or less rudimentary. This plant, therefore, may be considered to be becoming dioecious.

The next two genera of Rosaceae, Sanguisorba and Poterium, each of which contains a single British species, have been already alluded to in the opening chapter (ante, p. 10). Sanguisorba (Fig. 10) officinalis is monoecious and fertilised by insects. In Poterium sanguisorba (Fig. 9) some flowers are male, some female, and some hermaphrodite, and the pollen is said to be wind-borne. In other respects these two plants are curiously similar.

There is almost as much difference of opinion with reference to the specific limits in the genus Rosa as is the case in Rubus. Bentham admits five British species, while others, as, for instance, Babington, extend the number to fifteen or twenty. The flowers do not appear to secrete honey, but are much visited by insects for the sake of the pollen. The numerous stamens ripen at the same time as the pistil, but from the convenient position of the latter, insects very frequently alight upon it, and thus fertilise it with pollen from other flowers, though self-fertilisation probably often occurs.

Pyrus mains (the Apple), on the contrary, and Crataegus oxyacantha (the Hawthorn) are melliferous, and the stigma comes to maturity before the anthers.