THE EVOLUTION AND SYSTEMATICS OF BEE-EATERS (MEROPIDAE)

Behavioural and ecological characters are used in addition to structural ones in considering the systematics of the Meropidae. The two species of Nyctyornis are the most primitive extant forms. Meropogon is retained as a monotypic genus for forsteni, and all other bee-eaters are placed in the single genus Merops, being ecologically and morphologically rather uniform except in details of wing and tail structure, which should be considered only of specific importance. M. breweri and M. oreobates are thought to be secondarily rather than primarily forest species. With the submergence of Aerops and Melittophagus, and the monotypic genera Bombylonax and Dicrocercus, which are considered to be closely related to the pusillus species-group, Merops is enlarged to 21 species which are uniform except in tail-shape, wing formula and throat feather structure. These characters are of specific importance only, and have a mosaic distribution within the genus. Their use in the definition of the formerly recognized genera results in an artificial classification. The proposed delimitation of species-groups within Merops differs somewhat from previous arrangements, and affinities argued in the text are summarized in Fig. 15, which shows superspecies and species-groups. The species recognized are formally tabulated below. The Meropidae probably originated in southeast Asian forest and spread through former forest to Africa. Only in Africa was the open-country environment invaded, and speciation in the savanna was in two main directions, producing small sedentary and large migratory species. Representatives of both types returned to Asia in open country. From the present distribution of species it is inferred that speciation has proceeded under the main influences of (1) isolation of a population and its habitat in Pleistocene Africa and (2) isolation of migrants away from their breeding range. The distributions of species with wide or rather limited ranges are discussed in terms of physiological adaptation and ecological competition. Opinion has not been expressed on the validity of subspecies, which have been discussed only in delimiting controversial species. In the following summary, the only subspecies named are those affected by changes from Peters' scheme (cf. Table 1). Superspecies are bracketed. Nyctyornis amicta N. athertoni Meropogon forsteni Merops guloris M. mülleri M. bulocki M. bullockoides M. pusillus M. variegatus (?loringi, oariegatus, lafresnayii, bangweoloensis) M. oreobates M. hirundineus M. breweri M. revoilii M. albicollis M. orientalis M. boehmi M. viridis M. superciliosus (persicus, chrysocercus, superciliosus) M. philippinus (philippinus, salvadorii) M. ornatus M. apiaster M. leschenaulti M. malimbicus M. nubicus (nubicus, nubicoides)