AVIAN SYSTEMATICS AND EVOLUTION IN THE GULF OF GUINEA AVIAN SYSTEMATICS AND EVOLUTION IN THE GULF OF GUINEA 0 THE J. G. CORR(EIA COLLECTION DEAN AMADON Associate Curator Department of Birds BULLETIN OF THE AMERICAN MUSEUM OF NATURAL HISTORY VOLUME 100: ARTICLE 3 NEW YORK : sQSw I BULLETIN OF THE AMERICAN MUSEUM OF NATURAL HISTORY Volume 100, article 3, pages 393-452, text figure 1, plates 1-4, tables 1-2 Issued January 20, 1953 Price: $1.00 a copy CONTENTS INTRODUCTION . * . 399 THE GULF OF GUINEA ISLANDS 401 SYSTEMATIC NOTES . 404 LISTS OF THE BIRDS OF THE FOUR ISLANDS. 437 ZOOGEOGRAPHY . .* . 440 SOME EVOLUTIONARY FACTORS . * * . 442 Endemism . 442 Evolution of Color and Plumage. 443 * Variation in General Size. 445 * Increase in Size of Bill. 447 SUMMARY. * * * . 449 LITERATURE CITED . .. .. .. 449 397 INTRODUCTION JosE G. CORREIA, a cooper on the whaling Africa." I sent a list of the species obtained brig "Daisy" during her now famous cruise to by Correia on Principe and Sao Tome to South Georgia in 1912, was instructed in the David Snow, who incorporated some of this methods of preparing museum specimens by information in his recent paper (1950) on Dr. Robert Cushman Murphy. He has since the birds of these islands. Finally, one must collected, chiefly for the American Museum, mention the revision of Chrysococcyx cupreus in many parts of the globe. The present paper by Moreau and Chapin (1951) in which a is based on a collection made by Correia and new race of the Emerald Cuckoo is described his wife in 1928 and 1929 on the islands of from Correia's collection. Fernando Po, Principe, and Sao Tome, in Among those who have published general the Gulf of Guinea off the west coast of accounts of the ornithology of the Gulf of Africa. Their field work was supported by the Guinea islands are Bocage, Salvadori, Alex- late S. Brinckerhoff Thorne, then a trustee of ander (Fernando Po), Bannerman (Annobon, the American Museum. Representative col- Sao Tom6, Principe), and Snow (Sao Tom6, lections were obtained on Principe and Principe). Fernando Po has been relatively Sao Tome. On Fernando Po, Correia was in- neglected, for there has been no general capacitated by sleeping sickness, and his account of which I am aware since Alex- plans to explore the higher levels of Clarence ander's article of 1903. Bannerman's papers Peak (Pico Santa Isabel) had to be aban- (1914, 1915) on Principe and Sao Tome are doned. He thus missed many of the montane much more up to date, particularly as sup- species but did add a number of birds to the plemented in a few details by Snow. Never- list of the avifauna of Fernando Po. More theless, it seems worth while to present here important perhaps is the fact that he ob- (see below) a list of the resident, non-marine tained fine series of many of the West African birds of these two islands, as well as of species described from Fernando Po by Fra- Fernando Po, both to incorporate certain ser, Strickland, Jardine, Alexander, and other changes in specific or subspecific status sug- early ornithologists. Critical comparisons gested by the present studies and also to with mainland material were thus made pos- provide a reference list for the discussion of sible. evolution. The Correias did not visit Annobon, the Both Drs. Murphy and Chapin encouraged only other island of any consequence in the me to write the present report on the Correia Gulf of Guinea. Annobon is much smaller collection. I am indebted to them and to -than the other islands but has a few interest- other members of the staff of the American ing endemic land birds, all unrepresented in Museum for many suggestions during the the American Museum. course of the work. I have, of course, leaned Correia's collection was identified for most heavily upon Dr. Chapin, whose en- cataloguing by Dr. James P. Chapin, but cyclopedic knowledge of African birds and no report on it has ever been written. A few voluminous notes were both placed at my records were incorporated in the second part service with characteristic generosity. Dr. of Chapin's "The birds of the Belgian Con- David Bannerman of the British Museum go," and others are included in the third and was kind enough to send me his personal fourth parts of that work, which are still bound copies of his own and Salvadori's unpublished. Such a record is Chapin's papers on the Gulf of Guinea avifauna. The statement that Columba arquatrix sjistedti eight volumes of Bannerman's "The birds of occurs on Fernando Po as well as on Cam- tropical West Africa" were used constantly, eroon Mountain. Dr. Chapin also advised for they give diagnoses of all the West Dr. David A. Bannerman of the presence in African species and subspecies. I am also Correia's collection of such rarities as Amasu- indebted to Mr. David Snow of Oxford rocichla bocagei, and this is duly noted in University and Mr. Derek Goodwin of the Bannerman's "The birds of tropical West British Museum for various courtesies. In 399 400 BULLETIN AMERICAN MUSEUM OF NATURAL HISTORY VOL. 100 America I was able to borrow comparative Natural History Museum). The map was material through the courtesy of Mr. W. E. prepared for publication by Dr. Charles Clyde Todd (Carnegie Museum), the late Vaurie. Dr. Chapin, Dr. Ernst Mayr, and James Lee Peters (Museum of Comparative Dr. R. E. Moreau read sections of the Zoology), and Dr. Austin L. Rand (Chicago manuscript. THE GULF OF GUINEA ISLANDS FERNANDO PO, PRINCIPE, and Sao Tome lie KILOMETERS in the Gulf of Guinea on an arc of volcanic Fernando Po to coast of the activity which includes Mt. Cameroon on Cameroons 32 the mainland and which extends beyond Fernando Po to Principe 220 Principe to include Annobon Island and pos- Principe to Sao Tome 146 sibly even the distant St. Helena. Geologically Principe to mainland 220 Sao Tom6 to mainland 280 these islands are of volcanic origin. Some Sao Tome to Annobon 180 biologists believe that they represent vol- Annobon to mainland 340 canic elevations on a peninsula which was once connected with the mainland. So far as Fernando Po is a rectangular island with Fernando Po is concerned, there can be no dimensions of 69 by 32 kilometers. There are reasonable doubt that it was connected with extensive mountains at each end, the highest the mainland in the not too distant past. It mountain, called Clarence Peak or Mt. Santa lies on the continental shelf, only 32 kilo- Isabel, being 2850 meters high. Fernando Po meters from the coast. The greatest depth is almost covered by rain forest, but there separating it from the mainland is about are some natural grasslands at moderate 60 meters (Exell, 1944, p. 5). elevations near Moka, while approximately As regards Principe, Sao Tome, and Anno- the highest 1200 meters of Clarence Peak is bon, however, the present view, again ex- covered by grassland and heather. Never- pressed by Exell, is that "All the evidence theless, these grasslands are evidently too tends to show that the three islands never small or too recent to have made much im- had any land connections either with each print on the avifauna. The few open-country other or with the continent. They are isolated, birds that have been attracted by them, such moreover, by ocean depths of over 1800 as Linurgus olivaceus olivaceus and Saxicola metres." Since the volcanic activity that gave torguata pallidigula, usually do not differ rise to these islands (and to Mt. Cameroon) even subspecifically from the forms occupying is believed to have occurred 'in the middle or the more extensive grasslands of Mt. Cam- late Tertiary (there is still some slight active eroon and the other Camer6on highlands. vulcanism on Fernando Po and Sio Tome), Even among the forest birds there is little there is no reason to think that changes of taxonomic zonation into tropical and sub- sea level of a magnitude necessary to unite tropical forms. Among the latter are such these outer islands with the mainland have species as Columba arquatrix sjistedti, Phyl- occurred since their origin. lestrephus poensis, Cyanomitra oritis poensis, The biogeographical evidence fully sup- and others, most of which do not differ ports the above views. Fernando Po has an subspecifically from Mt. Cameroon repre- essentially continental biota, with endemism sentatives of the same species. Since Correia (at least in the birds) almost entirely con- barely entered the highlands, it is precisely fined to the subspecific level. The island has these highland forms that are usually lack- some endemic mammals, including monkeys ing in his collection. Some of them are re- and a small antelope (Krumbiegel, 1943). The presented in the American Museum by a outer islands, on the other hand, are said by specimen or two taken by Alexander, but in Exell to have lacked endemic mammals other general the systematic and other remarks than bats. This is excellent evidence that on Fernando Po birds in this paper are re- they were never connected with the main- stricted to the lowland species. land. The floras, too, have an "insular char- Perhaps the best general account of Fer- acter" (Exell, op. cit., p. 5), and the avifauna nando Po and its history is the chapter by has the unbalanced and impoverished yet the botanist Mildbraed in volume 2 of the highly endemic aspect usual in islands that popular account of the Duke of Mecklen- have always been well isolated. burg's expedition called in the English edi- The distances between these various land tion, "From the Congo to the Niger and the units are as follows: Nile" (1913).