ROLLER, a very beautiful bird, so called from its way of occasionally rolling or turning over in its flight, somewhat after the fashion of a tumbler-pigeon. It is the Coracias garrulus of ornithology, and is widely though not very numerously spread over Europe and Western Asia in summer, breeding so far to the northward as the middle of Sweden, but retiring to winter in Africa. It occurs almost every year in some part or other of the British Islands, from Cornwall to the Shetlands, while it has visited Ireland several times, and is even recorded from St Kilda. But it is only as a wanderer that it comes, since there is no evidence of its having ever attempted to breed in Great Britain; and indeed its conspicuous appearance—for it is nearly as big as a daw and very brightly coloured—would forbid its being ever allowed to escape a gun. Except the back, scapulars and tertials, which are bright reddish-brown, the plumage of both sexes is almost entirely blue—of various shades, from pale turquoise to dark ultramarine—tinted in parts with green. The bird seems to be purely insectivorous. The genus Coracias, for a long while placed by systematists among the crows, has really no affinity whatever to them, and is now properly considered to belong to the heterogeneous group of birds now associated as Coraciiformes, in which it forms the type of the family Coraciidae; and its alliance to the bee-eaters (Meropidae) and king-fishers (Alcedinidae) (q.v.) is very evident. Some eight other species of the genus have been recognized, one of which, C. leucocephalus or C. abyssinicus, is said to have occurred in Scotland. India has two species, C. indicus and C. affinis, of which thousands upon thousands used to be annually destroyed to supply the demand for gaudy feathers to bedizen ladies' dresses. One species, C. temmincki, seems to be peculiar to Celebes and the neighbouring islands, but otherwise the rest are natives of the Ethiopian or Indian regions. Allied to Coracias is the genus Eurystomus with some half-dozen species, of similar distribution, but one of them, E. pacificus, has a wider range, for it inhabits Australia and reaches Tasmania.
Madagascar has four or five very remarkable forms which have often been considered to belong to the family Coraciidae; and, according to A. Milne-Edwards, no doubt should exist on that point. Yet if any may be entertained it is in regard to one of them, Leptosomus discolor, which, on account of its zygodactylous feet, some authorities place among the Cuculidae, while others have considered it the type of a distinct family Leptosomatidae. The genera Brachypteracias and Atelornis present fewer structural differences from the rollers, and perhaps may be rightly placed with them; but the species of the latter have long tarsi, and are believed to be of terrestrial habit, which rollers generally certainly are not. These very curious and in some respects very interesting forms, which are peculiar to Madagascar, are admirably described and illustrated by a series of twenty plates in the great work of A. Grandidier and A. Milne-Edwards on that island (Oiseaux, pp. 223–250), while the whole family Coraciidae is the subject of a monograph by H. E. Dresser, as a companion volume to his monograph on the Meropidae. (A. N.)
- Gesner in 1555 said that the bird was thus called, and for this reason, near Strassburg, but the name seems not to be generally used in Germany, where the bird is commonly called Rake, apparently from its harsh note. The French have kept the name Rollier. It is a curious fact that the roller, notwithstanding its occurrence in the Levant, cannot be identified with any species mentioned by Aristotle.