The name of Ericaceæ, which most people are aware signifies the Heath family, conjures up immediately the image of a number of narrow-leaved plants, with globular, ventricose, or bell-shaped flowers; and we are apt at first to think that the family is so natural a one, as to require very little explanation. Did the order include only the Heaths, this would be the case, for all the Heaths, different as they are in some particulars, may be recognised at a glance: but as the order includes the Rhododendrons, Azaleas, and Kalmias, besides several other plants which have not so strong a family likeness to each other as the Heaths, it becomes necessary to say a few words on the botanical resemblances which connect them together. The first, and most striking, of these is the shape of the anthers, each of which appears like two anthers stuck together, and the manner of their opening, which is always by a pore or round hole, in the upper extremity of each cell. The filaments, also, in all the genera, except Vaccinum and Oxycoccus, grow from beneath the seed-vessel, being generally slightly attached to the base of the corolla. There is always a single style with an undivided stigma, though the capsule has generally four cells, each containing several of the seeds, which are small and numerous. The calyx is four or five cleft, and the corolla is tubular, with a larger or smaller limb, which is also four or five cleft. The order has been divided into four tribes, which I shall describe in this chapter, though some of these are considered as separate orders by Dr. Lindley and other botanists. TRIBE I.—ERICEÆ. [Illustration: FIG. 47.—THE BESOM HEATH (_Erica Tetralix_).] This tribe, which comprehends all the heath-like plants, has been re-divided into two sub-tribes, one containing the genera most nearly allied to the heaths, and the other those belonging to the Andromeda. In both there is a honey-bearing disk under the ovary, and the leaves are generally rolled in at the margin, as shown at a, in _fig._ 47. SUB-TRIBE I.—ERICEÆ NORMALES. All the genera in this sub-tribe, twenty-two in number, were formerly included in the genus Erica; and some botanists still consider all the species to belong to that genus, with the exception of those included in Calluna, while others adopt about half the new genera. In this uncertainty, I shall only describe two of the doubtful genera, partly because the distinctions between them and the true heaths are strongly marked, and partly because the species they contain are frequently met with in British gardens and greenhouses, where they are sometimes labelled with their old names and sometimes with their new ones.