^' *. if Ji HOLOGY LIBRARY IN MEMOBrAM. THE COLLECTED SCIENTIFIC PAPERS OF THE LATE WILLIAM ALEXANDER FORBES, M.A., FELLOW OF ST. JOHN'S COLLEGE, CAMBRIDGE ; LECTURER ON COMPARATIVE ANATOMY AT CHARING CROSS HOSPITAL ; PROSECTOR TO THE ZOOLOGICAL SOCIETY OP LONDON. EDITED BY F. E. BEDDARD, M.A., PROSECTOR TO THE ZOOLOGICAL SOCIETY OF LONDQN. WITH A BY P. L. SC LATER, M.A,, PH.D., F.R.S., SECRETARY TO THE ZOOLOGICAL SOCIETY OF LON1>ON. LONDON: E. H. PORTER : 6 TENTERDEX STREET, W. 1885. UfiRXKY PRINTED BY TAYLOR AND FRANCIS, RED LION COURT, FLEET STREET. F57 I1BRARY PAGE PREFACE v LI8T OF SUBSCRIBERS ix TABLE OF CONTENTS ri LIST OF PLATES XT SCIENTIFIC PAPERS .... 1 PREFACE. WILLIAM ALEXANDER FORBES, Fellow of St. John's College, Cambridge, Prosector to the Zoological Society of London, and Lecturer on Comparative Anatomy at Charing Cross Hospital, was born at Cheltenham on June 24, 1855, the second son of Mr. J. S. Forbes, the well-known Railway Director. He was educated at Kensington School and Winchester College, which he entered at the early age of eleven. On leaving Winchester in 1872 Forbes passed a year at Aix-la-Chapelle, studying German, and then became a student of the University of Edinburgh, where he pursued the regular medical course, paying special attention to Zoology and Botany, and commencing collections of insects and plants. In 1875 Forbes transferred his residence to London, and entered himself as a student of University College, with the idea of taking a medical degree in the metropolis. Here he quickly became intimate with other zoologists, who were very soon attracted by the astounding general knowledge of zoology and the acute intelligence of one so young. By the advice of the late Prof. Garrod and other friends Mr. Forbes was induced, in October 1876, to leave London and to enter as an undergraduate at St. John's College, Cambridge, where he was subsequently elected Scholar, and took his B.A. degree with a First Class in the Natural Sciences Tripos in 1879. The post of Prosector to the Zoological Society of London having become vacant in October 1879, by the lamented death of Prof. Garrod, Forbes was appointed (omnium consensri) to that office in the January following : indeed Garrod on his death-bed had designated him as his most obvious and proper successor, and appointed him his literary executor. VI PREFACE. Forbes entered upon the duties of his office with characteristic energy ; and during the three following sessions of the Zoological Society brought before the scientific meetings a series of most interesting and valuable communications, mainly derived from his studies of the animals that came under his examination. He had a happy knack of presenting abstruse points of anatomy in an easily comprehensible form, and especially directed himself to the investigation of the muscular structure and voice-organs of birds, in continuation of the researches of his predecessor Garrod on the same subjects. In the summer of 1880 Forbes made a short excursion to the province of Pernambuco, Brazil, of which he published an account in ' The Ibis ' for 1881 ; and in the following year passed his holiday in the United States, in order to make the acquaintance of his American brethren in science and their collections. In July 1882 he left England on what promised to -be a splendid opportunity of visiting the eastern tropics with every advantage and without risk. Detained at Shonga (a station some 400 miles up the Niger, below Rabba) by the breaking-down of his commu- nications, Forbes fell a victim to dysentery, and died on the 14th of January, 1883, thus adding another name to the long list of martyrs to science in the deservedly dreaded climate of Western Africa. Forbes was carefully nursed, during his last illness, by Mr. Greenshields, agent of the United African Company at Shonga. His remains, deposited at first within the compound of the factory at Shonga, were subsequently disinterred under the kind care of the same excellent friend and brought to England, and buried on April 1st, 1884, in the quiet churchyard of West Wickham in Kent, in the presence of his sorrowing relatives and friends. Forbes's family have specially requested me also to mention the name of Mr. M'Intosh, Manager of the United African Company, as having most kindly assisted them in the somewhat difficult arrangements necessary for this purpose. At a meeting of the Zoological Club (of which Forbes had been PREFACE. Vll a prominent member) held on the 20th of November, 1883, it was resolved that a Committee, consisting of Prof. Flower, Prof. Bell, Mr. H. H. Johnston, Prof. Mivart, and the writer of this Preface, should be appointed to consider what would be the most appropriate memorial of our much-loved friend and fellow- worker. After some consideration it was unanimously agreed that the best and most enduring memorial would be the republi- cation of Forbes's Contributions to Science in a volume similar to that containing Garrod's Scientific Papers, which Forbes had himself edited. Such was the origin of the present volume, which has been carefully edited by Mr. Frank Evers Beddard, Forbes's successor in the Prosectorship of the Zoological Society of London. Miss Garrod has prepared the Index, Mr. H. H. Johnston has kindly contributed the life-like etching which forms the Frontis- piece, while Prof. Bell has undertaken the task of collecting and expending the necessary subscriptions. P. L. S. 3 Hanover Square, May 1st, 188-5. Churchyard of West Wickham, Kent, with Forbes's grave (beneath the two trees). LIST OF SUBSCRIBERS. B. W. Adam, Esq. H. F. Amedroz, Esq. H. Amerhoff, Esq. Dr. John Anderson, F.R.S. H. Ansdell, Esq. Lt.-Col. H. Barclay. J. Wolfe Barry, Esq. F. E. Beddard, Esq. J. H. Beddington, Esq. Prof. F. Jeffrey Bell. Dr. G. Bennett. W. T. Blanford, Esq., F.R.S. Prof. W. Blasius. Prof. Barboza du Bocage. Prof. Bonney, F.R.S. G. A. Boulenger, Esq. Major-Gen. H. Clerk, F.R.S. Lord Clifton. C. M. Clode, Esq., C.B. E. H. Cooper, Esq. A. E. Craven, Esq. C. A. Craven, Esq. F. Crisp, Esq. C. Critohett, Esq. P. Crowley, Esq. C. G. Danford, Esqv G. E. Dobson, Esq., F.R.S. Prof. Dohrn. A. Dowsett, Esq. J. W. Dunning, Esq. A. H. Evans, Esq. Prof. Cossar Ewart. Sir J. Fayrer, K.C.S.I., F.R.S. Prof. Flower, F.R.S. H. O. Forbes, Esq. W. Forbes, Esq. Prof. M. Foster, Sec. R.S. Prof. Furbringer. Dr. Gadow. J. C. Galton, Esq. Dr. Garrod. Mrs. Garrod. Miss Garrod. A. E. Garrod, Esq. H. B. Garrod, Esq. Dr. Garson. J. P. Gassiot, Esq. F. D. Godman, Esq. Lt.-Col. Godwin-Austen, F.R.S. F. E. Goodhart, Esq. Lt.-Col. Grant, C.B., C.S.I. A. Grote, Esq. Dr. Giinther, F.R.S, Prof. Haddon. Miss Browning HalL Col. Harcourt, M.P. M. Harris, Esq. Hubert Herkoruer, Esq., R.A, Rev. E. Hill. W. E. Hoyle, Esq. Prof. Huxley, P.R.S. H. H. Johnston, Esq. J. B. Johnston, Esq. J. M. 0. Johnston, Esq. Prof. Ray Lankester, F.R.S. Lewis Levy, Esq. Lord Lilford. Sir J. Lubbock, Bart, M.P., F.R.S. Manchester Free Library. H. S. Marks, Esq., R.A. Lord Mayo. Dr. Meyer. Captain Moloney. J. Morgan, Esq. Prof. Moseley, F.R.S. W. F. Mulvaney, Esq. John Murray, Esq. Prof. Newton, F.R.S. LIST OF SUBSCRIBERS. R Nicholson, Esq. H. F. Osborn, Esq. Prof. T. Jeffery Parker. T. Parkin, Esq. F. P. Pascoe, Esq. Dr. yon Pelzeln. C. B. Phillip, Esq. Henry Pollock, Esq. H. N. Pym, Esq. Dr. Quain, F.R.S. Dr. B. W. Richardson, F.R.S. Rev. E. Richardson. S. O. Ridley, Esq. Briton Riviere, Esq., R.A. C. P. Sandberg, Esq. J. E. Sandys, Esq. W. A. Sanford, Esq. Howard Saunders, Esq. H.E. Count Schimmelpenninck de Nyenhuis. P. L. Sclater, Esq., F.R.S. W. L. Sclater, Esq. J. Scully, Esq. A. Sedgwick, Esq. H. Seebohm, Esq. Rev. E. W. Sergeant. R. B. Sharpe, Esq. Captain Shelley. The Bishop of Southwell. H. Stevenson, Esq. J. B. Sutton, Esq. Rev. C. Taylor. W. B. Tegetmeier, Esq. O. Thomas, Esq. Prof. Turner, F.R.S. Lord Walsingham. Captain R. G. Wardlaw-Ramsay. Prof. Morrison Watson. John Way, Esq. W. F. R. Weldon, Esq. Dr. Westermann. John White, Esq. Prof. E. P. Wright. C. A. Wright, Esq. Herr F. C. Zillesen. Zoological Society of London. CONTENTS. PAGE 1. Late Appearance of Cetonia aurata , 1 2. Arrested Development in Timarcha coriaria and Lagria hirta 1 3. Note on Chrysomela marginata 2 4. Note on Mr. Wallace's Distribution of Passerine Birds 3 5. On the Bursa Fabricii in Birds 3 6. Recent Observations on the Parrots of the Genus Eclectus 19 7. On the Nesting of the Spoonbill in Holland. (Written in conjunction with P. L. SCLATER, M.A., F.E.S., &c.) 26 8. Lepidoptera captured during an Excursion to Switzerland and the Italian Lakes 29 9. Melanism in Lepidoptera 32 10. Reports on the Collections of Birds made during the Voyage of H.M.S. ' Challenger.' No. VII. On the Birds of Cape York and the neighbouring Islands (Raine, Wednesday, and Booby Islands) 34 11. On a small Collection of Birds from the Samoan Islands and the Island of Rotumah, Central Pacific 43 12. Letter concerning the Locality of Garrulus lidthi 45 13. Staphylimis fulvipes in the New Forest 46 14. On the Anatomy of the African Elephant (Elephas africanus, Blum.) 46 15. On the Systematic Position of the Genus Lathaimis of Lesson. (Plate I.) ... 62 16. A Synopsis of the Meliphagine Genus Myzomela, with Descriptions of two new Species. (Plates II. & III.) 71 17. On the Systematic Position and Scientific Name of " Le Perroquet Mascarin" of Brisson 96 18. Notes on Butterflies observed in the Valais of Switzerland in 1878 100 19. The Glacial Period and Geographical Distribution 104 20. On the External Characters and Anatomy of the Red TJakari Monkey (Bra- chyurus rubicundus) , with Remarks on the other Species of that Genus. (Plates I V.-VL) 105 21. On the Cause of Death of a Leopard 127 22. On Antilocapra amcricana 127 Xll CONTENTS. PAGE 23. On some Points in the Structure of Nasiterna bearing on its Affinities 131 24. Contributions to the Anatomy of Passerine Birds. Part I. On the Structure of the Stomach in certain Genera of Tanagers 133 25. Contributions to the Anatomy of Passerine Birds. Part II. On the Syrinx and other Points in the Anatomy of the Eurylcemidce 138 26. Contributions to the Anatomy of Passerine Birds. Part III. On some Points in the Structure of Pkilepitta, and its Position amongst the Passeres 144 27. On the Anatomy of Leptosoma discolor 149 28. On two rare Ploceine Birds now or lately Living in the Society's Menagerie. (Plate VII.) 159 29. Note on a Specimen of Denham's Bustard (Eupodotis denhami) 162 30. Remarks on Dr. Gadow's Papers on the Digestive System of Birds 163 31. Three Weeks' Butterfly-collecting in the Alps 165 32. On a little-known Cranial Difference between the Catarrhine and Platyrrhine Monkeys 170 33. On the Male Generative Organs of the Sumatran Rhinoceros (Ceratorhinus sumatrensis) 170 34. On some Points in the Anatomy of the Koala (Phascolarctos cinereus) ......... 173 35. On the Contributions to the Anatomy and ClassiScation of Birds made by the late Prof. Garrod, F.R.S 189 36. Notes on the Unfinished Work left by the late Prof. Garrod on the Anatomy of Birds 213 37. Note on Mr. Bar tlett's Communication on the Habits of the Darter 216 38. Contributions to the Anatomy of Passerine Birds. Part IV. On some Points in the Anatomy of the Genus Conopophaga, and its Systematic Position ... 217 39. Notes on the Anatomy and Systematic Position of the Jaganas (Parridcs) 220 40. On the Petrel called Thalassidroma nereis by Gould, and its Affinities 229 41. On the Conformation of the Thoracic End of the Trachea in the "Ratite" Birds 232 42. Eleven Weeks in North-eastern Brazil 242 43. The Descent of Birds 280 44. Note on the Systematic Position of Eupetes macrocercus 281 45. Note on the Structure of the Palate in the Trogons ( Trogonida) 282 46. On the Anatomy and Classification of the Petrels, based upon those collected by H.M.S. 'Challenger' 284 47. Observations on the Incubation of the Indian Python (Python molurus), with special regard to the alleged Increase of Temperature during that Process 285 48. Observations on the Incubation of the Indian Python (Python molurus) 292 49. The Insectarium at the Zoological Gardens 293 50. Remarks upon the Horns of the Prongbuck 295 CONTENTS. Xlll PAGE 51 . On some Points in the Anatomy of the Great Anteater (Myrmecophaga jubata). (Plate VIII.) 296 52. Note on an Abnormal Specimen of Pithecia satanas 312 53. Supplementary Notes on the Anatomy of the Chinese Water-Deer (Hydropotes inermis) 313 54. Notes on the External Characters and Anatomy of the Californian Sea-Lion (Otaria gillespii). (Plates IX.-XI.) 316 55. Note on the Gall-bladder, and some other Points in the Anatomy of the Toucans and Barbets (Capitonida} 323 56. On some Points in the Anatomy of the Indian Darter (Plotus melanog aster), and on the Mechanism of the Neck in the Darters (Plotus), in connexion with their Habits 326 57. Description of the Pterylosis of Mesites, with Remarks on the Position of that Genus 331 58. Note on a Peculiarity in the Trachea of the Twelve-wired Bird-of-Paradise (Seleucides nigra) 335 59. On the Convoluted Trachea of two Species of Manucode (Manucodia atra and Phonygama gouldi) ; with Remarks on similar Structures in other Birds... 338 60. On some Points in the Anatomy of the Todies ( Todidai), and on the Affinities of that Group 345 61. Note on some Points in the Anatomy of an Australian Duck (Sistiwa lohata). 354 62. Contributions to the Anatomy of Passerine Birds. Part V. On the Structure of the Genus Orthonyx 357 63. On the Rudimentary Hallux of Birds 359 64. Contributions to the Anatomy of Passerine Birds. Part VI. On Xenwus and Acanthisitta as Types of a new Family (Xenicid 105 VI. Brachyurus melanocephalus J VII. Fig. 1. Vidua splendens. Fig. 2. Pytelia wieneri 159 tt VIII. Anatomy of Myrmecophaga 296 IX. Otaria gillespii \ X. Head of male Otaria gillespii (one fourth the size of nature) ... J- 316 XL Anatomy of Otaria gillespii J XII. ' | Alimentary Canal, Tongue, and Palate of Petrels XVI. XIII ' -^ xterna ^ characters of Petrels XIV. XV, ,,..,.,-' ^ Myology of Petrels. Anterior extremity , XVIII. I XIX J " " Tensor patagu muscles XX. Hinder extremity XXI. Osteology of Petrels. SkuU and limb-bones XXII. Skull, pelvis, and limb-bones..., [ -I xxiv'.J stema j 363 xxin. XXIV. XXV. lurnix saturata .. 444 SCIENTIFIC PAPERS OF WILLIAM ALEXANDER FORBES. 1. LATE APPEARANCE OF CETONIA AURATA* Ent.M.M.xi. p. 208 (1875). AT p. 178, vol. x. of the * Entomologist's Monthly Magazine,' Mr. Scott records the appearance of Cetonia aurata on the 15th October. I have now to chronicle an even later date for that species, as I found a specimen of it at ivy -bloom, in the daytime, on the 29th October last year, in a garden at West Wickham. This specimen was unusually small, but otherwise in an excellent condition, and seemed to have but just entered into the imago state. [These exceptional appearances are no doubt due to the fact that the Cetonia (like Lucanus cervus, and some other beetles) assumes the perfect state late in the autumn, but remains ordinarily in the cocoon till the following summer. Hence these abnormal specimens should rather be regarded as " early," not '* late," their appearance being perhaps due to sudden rise of temperature combined with individual precocity. EDS.] 2. ARRESTED DEVELOPMENT IN TIMAECHA Ent.M.M *i. CORIARIA AND LAGRIA HIRTAj THE following instances of arrested development, causing a want of symmetry in the legs of insects, are interesting, and seem to me to be worth publishing, insomuch as, so far as I know, no similar instance has been recorded. In a $ specimen of TimarcJia coriaria taken last autumn in Switzerland, this want of development occurs in the right middle leg, all the others being of normal size. The following are the dimensions of the stunted right leg, and its normal fellow on the left * Ent. Month. Mag. xi. p. 208 (1875). t Ibid. xi. p. 279 (1875). B 2 ON CHRYSO^ELA MARGWATA. side : Femur, L. 3 rnillim., E. very slightly shorter ; tibia, L. 3 millim., R. 1-75 millim. ; tarsi, L. 2 millim., E. 1 millim. : the claws being ex- ceedingly minute, and barely projecting beyond the last tarsal joint, all of which are present and equally developed, though not attaining the normal size. In a specimen of Lagria hirta, the dwarfing occurs in the posterior pair of legs, and in this case, again, the right leg is the stunted member. The measurements of this specimen are : Femur, about 2 millim. in both legs; tibia, L. 2 millim., E. 1-75 millim.; tarsi, L. T50 millim. (last joint wanting), E. 1 millim. In both cases the want of symmetry caused by the arrested growth is more conspicuous than would appear from the above figures. Ent.M.M.xii. 3. NOTE ON CHRYSOMELA MARGINATA* p. 135 (1875). THIS species, originally found, I believe, near Pegwell Bay, near Eams- gate, seems decidedly scarce on this side of the border, though not so uncommon in Scotland, where it has been found by Dr. Syme in Orkney (on Plantayo rnaritima), and by Mr. Champion at Braemar, by sweeping alongside the Dee. Near Edinburgh it is not uncommon, though very local. As far as I know, it is confined to one particular spot on Arthur's Seat, a much exposed valley between the summit of the hill and a lesser peak known as the Lion's Haunch, about 700 feet above the sea, where the grass forms a short velvety turf, and the surface of the ground is covered with scattered fragments of the neighbouring basalt rocks. Beneath these fragments Chrysomela marginata is to be found, singly, or in twos and threes. When disturbed, it persistently feigns death. It begins to appear about the middle of June, and is most common about the first week in July, when I have taken as many as thirty specimens in the course of an afternoon's work, by assiduously turning over stones &c. in its locality. I have never seen the larva or pupa, and do not know for certain what its food-plant at Edinburgh is, as no Plantago maritima grows near. The short turf of the hill is composed in great part of millefoil (Achillea millefolium), and on that the beetles may feed, as some I kept in captivity fed voraciously on this by night, returning to their shelter at the bottom of the plants by day. I have never seen it moving about in the daytime like its congeners C. mentliastri and (according to Mr. Champion) cerealis, but only found it under the stones. In Wilson's * Entomologia Edinensis,' the Calton Hill is also given as a locality ; but I have never found it there, chiefly, no doubt, owing to my not having looked there at the right time. * Ent. Month. Mag. xii. p. 135 (1875). ON THE BURSA FABRICII IN BIRDS. 4. NOTE ON MR. WALLACE'S DISTRIBUTION Nature,' xr. OF PASSERINE BIRDS.* P- 58 (1876). IN Mr. Wallace's recently published work on Geographical Distribution, in more than one place the results arrived at from an inspection of his elaborate tables of genera and families do not agree with the numbers he uses considering the general bearing of the facts adduced. Thus, in his " General Remarks on the Distribution of the Passeres," vol. ii. pp. 299- 302, he says (1. c. p. 300) : " The families that are confined to single regions are not very numerous, except in the case of the Neotropical region, which has Jiue, the Australian has only three, the Oriental one, Ethiopian one, and the other regions have no peculiar families." Adopting his tables of the families of the Passeres, I find the numbers should be really as follows : Neotropical 7 Fams. Nos. 39a, 40, 41, 42, 44, 45, 46. Australian 5 21, 22, 25, 49, 50. Oriental 3 11, 12, 43. The Nearctic region should also be mentioned as possessing one peculiar family, i. e. ChamaeidaB. The statement that none of the turdoid Passerine families are exclusively American must also be modified to meet this fact. There are three families (i. e. Paictidae, Pittidae, Eury- laemidoo) instead of two of the Formicaroid Passeres in the Old World, of which the Pittidae can hardly be said to have only a " very restricted distribution." The Australian genus Struthidea, of doubtful position, seems omitted altogether. 5. ON THE BURSA FABRICII IN BIRDS.f P.Z.S. 1877, p. 304. PROF. GAEEOD, in his paper on Plotus mihinga (P. Z. S. 1876, p. 344) says : " In the urino-genital system of Plotus anliinga, in both sexes, the ducts open in the normal manner into the cloaca, just above its lower orifice. This orifice, however, is not on the surface, but is into a cavity behind the cloaca, which opens externally .quite close to the place where the two communicate. Except for the nearly marginal orifice, the second cavity is a csDcal sac, oval in shape, and about 1| inch high, covered at its blind * ' Nature,' xv. p. 58 (1876). t Proc. Zool. Soc. 1877, pp. 304-318. Read April 17, 1877 B2 4 ON THE BURSA FABRICII IN BIRDS, end with the crypts of shallow glands, \vhich also run down its sides. That it is a modification of the bursa Fabricii cannot be doubted." The disposition of the parts described aboye seemed peculiar enough to be worthy of further investigation ; with that end my kind friend Prof. G-arrod requested me to undertake a series of observations on the bursa in other birds, in order to throw further light on the structure of this organ, and to discover what characters, if any, it afforded for classi- P. Z. S. 1877, ficational purposes. The ample materials of the Prosector's department p. 305. k ave gj ven me opportunities for examining this organ in a considerable number of species of birds of various orders ; and though I regret to say my investigations have not turned out so satisfactorily as regards taxono- mic characters as I had hoped, I venture to bring such results as I have obtained before the Society this evening. As the subject of the bursa Fabricii has hardly attracted any notice in this country since the days of Harvey, I have added to my own notes a brief resume of the most im- portant observations and opinions as regards its structure and functions that have been brought forward by foreign anatomists. The organ in question seems first to have been noticed by the naturalist whose name it bears, Fabricius of Acquapendente. In his treatise ' De formatione ovi et pulli ' *, p. 5, he says : " Tertium quod in podice est adnotandum est duplex vesicula quse in ima ejus parte ad os pubis supereminet, et conspicua exteriorque apparet, simulatque uterus jam propositus conspectui sese offert : quse cum sit pervia, ita ut ab ano ad ipsum uterum et ab utero in ipsam, ut puta superius, infra foramen pateat, ex altero autem extreme clausa sit, hunc existimavimus esse locum, in quern gallus semen immittit porrigitque ut inibi servetur." From this and other passages in his works it is clear that he considered its function that of a receptaculum seminis in the female ; its use in the male, on such a theory, he does not explain. Harvey, in his work * De Generatione Animalium ' (London, 1651), as quoted in the Sydenham Society's trans- lation of his works (1847, p. 183), refutes Fabricius's ideas on this point. " The foramen into which Fabricius believes the Cock to inject his fluid, is discovered between the orifices of the vulva and the rump. I, however, deny any such use to this foramen ; for in young chickens it is scarcely to be seen, and in adults it is present indifferently in males and females. It is obvious therefore that it is both an extremely small and obscure orifice, and can have no such important function to fulfil ; it will scarcely admit a fine bristle and needle, and it ends in a blind cavity ; neither have I ever been able to discover any spermatic fluid within it, although Fabricius asserts that this fluid is stored up there even for a whole year, and that all the eggs contained in the ovary may be thence fecundated, as it is afterwards stated." Harvey, however, fell into error in asserting * Hieronymi Fabricii ab Aquapendente opera anatomica. Patavii, 1625. ON THE BUESA FABRICII IN BIRDS. > that in " young chickens it was scarcely to be seen ;" as we shall after- wards see, it is developed more in young than adult birds. This fact was first pointed out by Tannenberg in 1789, in his disquisition ' Circa geiiitales partes mascularuin avium ' (Grottingae), and has subsequently been recognized by most authors who have written on the subject (vide Cuvier *, Milne-Edwards t, and Gegenbaur J). Barkow, in a paper ** On the Cloaca of birds " in Meckel's 'Archiv ' , describes its condition in specimens he had examined of the Fowl, Duck, Hooded Crow, Jack- P.Z.S. 1877, daw, Bittern, Eared Owl (Ohrkauze), Honey-Buzzard, and Coot, showing P ! that it is fully developed in young birds, but absent in old ones. He figures the bursa, together with the arteries supplying it (derived from the left pudeudal), in both sexes of Gallus domesticus and Falica atra (op. cit. tab. ix. fig. 13, 15, 19, 22-24, and t. x. f. 26). Berthold devotes a special memoir to it ||, in which he describes its nervous and vascular supply. Emil Huschke^[ describes its development, showing that it arises in the superior part of the cloaca, in which it is differentiated in the embryo of the Fowl from the 8th to the 9th day of incubation, acquiring by degrees a more perfect form, but that after a time it in- creases but slowly in comparison with the other parts of the embryo. M. Martin St.-Ange, in his fine paper " Etudes sur 1'appareil repro- ducteur dans les cinq classes des vertebres " **, figures and describes the bursa in the Common Pigeon. In one adult two-year-old Pigeon he found the interior of the bursa filled up with a sort of calculus, forming a complete cast of its interior. In all others of both sexes, it was reduced to about half its size, and the cavity entirely obliterated. He found that in the egg it was better developed in proportion than other organs, but that after the age of about six months in Pigeons, and eight in Fowls, it began to lose its functional activity, and to become reduced in size. Lastly, in a paper published in the * Atti della Societa Ttaliana di Scienze Naturali,' 1875, vol. xviii. pp. 133-169 (for calling my attention to which I am indebted to Mr. Salvin), Signer Vincenzo Alesi, of Naples, has published an exhaus- tive essay on the structure and development of this organ, accompanied by two plates of histological details. His observations have been made on specimens of Meleayris mexicana, A.nser cinereus, Anas boschas, Columba livia, Turtur auritus, Corvus monedula, Turdus menda, and CoturnLv communis ; and he has also examined the cloaca of a female llhea americana, preserved in the Naples Museum. To his observations * Leons d' Anatomic comparee, 2nd ed. vol. viii. p. 276. t Physiologic et Anatoraie comparee, vol. viii. p. 514, and vol. vii. p. 347. | Vergleichende Anatomic, p. 799, note. Archiv, 1829, p. 443 et scq. || Acad. Cses.-Leop. Nova Acta, 1828. xiv. pp. 913-918. ^| De Bursas Fabricii origine. Jenae, 1838. ** Mem. pres. Ac. Sc. Frari9. par savants divers, 1856, pp. 1-232, ON THE BURSA FABRICII IN BIRDS. on the histological structure and process of atrophy of the bursa I will return after having briefly described the ordinary form and relations of this organ in the birds I have examined. These are 90 in number ; and of many of them I have examined more than one specimen. P. Z. 8.1877, p. 307. PASSERES. Oriolus galbula. Garrulax chinensis. Citta thalassina. Fregilus graculus. Amblyrhamphus Tiolosericeus. Cissopis leveriana. COCCYGOMOEPHJE. Podargus cuvieri. Dacelo gigantea. Colius castanonotus. Momotus lessonii. Merops, sp. Megalcema, sp. RhampTiastos ariel. Pteroglossus wiedi. Trogon puella. Cuculus canorus. Cacomantis, sp. Chrysococcyx, sp. Geococcyx affinis. Guira piririgua. PSITTACI. Stringops habroptilus. Psittcwus erithacus. Pionus violaceus. Tanygnaihus muelleri. Pyrrhulopsis splendens. Platycercus icterotis. ACCIPITEES. Gyparchus papa. Buteo jackal. Archibuteo lag&pus. Geranoaetus aquila. Spizaetus coronatus. Haliaetus vocifer. Strix flammea. STEGANOPODES. Sula bassana. Plotus anhinga. Phalacrocorax carbo. P. brasiliensis. HEEODIONES. Ciconia alba. C. boyciana. Abdimia sphenorhyncha. Xenorliynchus australis. Ibis falcinellus. I. rubra. Ardea cocoi. Cancroma cochlearia PALAMEDE^E. CTiauna derbiana. ANSEEES. Cygnus olor. Tadorna rutila. Fulica rufigula. COLUMB^;. Pterocles arenarius. Phaps chalcoptera. Goura coronata. Craoc sclateri. Tetrao urogallus. Lag opus scotlcus. Ortyx virginianus. Odontophorus guianensis. Calllpepla gambeli. Arboricola torqueola. Caccabis cliukar. C. melanocepliala. C. saxatllis. ON THE BURSA FABRIC1I IN BIRDS. 7 Francolinus vulgaris, GAVI^B. F. Ucalcaratus. Larus ridibundus. Euploctimus cristatus. PTGOPODES. Argus yis/anteus. Meleagris meviccma. Uria troile ' Alca torda. ALECTORIDES. TINAMI. Otis tarda. Rhynchotus rufesccns. Grus carunculata. Crypturus tatupa. Serpentarius reptilivorus. Oeydromm australis. STRUTHIONES. Porphyno alien i. Struthio camelus. P. marlinieus. Rhea americana. P. poliocephalus. Dromceus novce-hollandice. Hydrophasianus chiruryus. Casuarius beccarii. C. picticollis. GEALLJ;. ^ uniappendiculatus. (Edicnemus crepitans. Apteryx mantdli. Attagis, sp. A. oweni. For an opportunity of examining a specimen of the last I am indebted to Prof. Newton's kindness. If the cloaca of an ordinary bird be taken and laid open in front (i. e. p. z. S. 1877, on the ventral surface) along the middle line, the rectum will be seen P* ^' to terminate at the top of a more or less well-marked chamber, which usually is iwore capacious than the terminal part of the alimentary tract, from which it is generally separated off by a more or less plain valvular constriction, the different nature of the two parts being also frequently shown by the differences in the character of the mucous membrane lining them. Below the entrance of the rectum, on the posterior wall of the cloaca, and disposed symmetrically as regards the median line on the sides of which they He, are the openings of the uro-genital ducts, fre- quently elevated on papillae, which vary in shape in different birds. Of these the ureters open nearest the middle line ; externally to them are the openings or opening of the vasa deferentia or oviduct, which latter is fully developed only on one side, though sometimes a trace of it may be seen on the other also. Above the uro-genital papillae in many birds (e. g. Meleagris mexicana and Tetrao urogallus) is a well-marked valvular fold of mucous membrane, which thus separates off the upper portion of this " cloaca " from that part into which the urino-genital ducts open. Below these, again, is a similar, but usually better-marked fold (the vesical sphincter of St.-Ange), which is present in all the birds I have examined, in a more or less developed state. Below this in the middle line, and also on the posterior wall, is seen (if the specimen be not too old) a circular hole of varying dimensions. This is the opening of the 8 ON THE BUESA FABRICII IN BIRDS. " bursa Fabricii." This third part of the " cloaca" is bounded below by the external sphincter muscles, which often form in birds a great thick fleshy mass ; this chamber, from lodging the penis or clitoris, is deno- minated " loge copulatrice " by M. St.-Ange. In many birds additional glands, opening by large pores, or forming irregular arborescent patches, are developed in the walls of the cloaca in this region. It results there- fore from the above-described disposition of parts that, in its most deve- loped form, a bird's " cloaca " is divided into three chambers communi- cating with each other, and into which open respectively (from above downwards), first the rectum, secondly the urino-genital ducts, thirdly the bursa Eabricii*. This disposition of parts is seen in fig. 1. The bursa Fig. 1. Cloaca of Chauna derbiana, laid open from before. a. Rectum, b. Opening of ureters, b'. Genital papillae, c. Fold of mucous membrane d. "Vesical sphincter " of M. St.-Ange. e. Opening of bursa Fabricii. /.Bursa Fabricii (supposed to be seen through the posterior wall of the cloaca). Eabricii, when well developed, consists of a sac-like pouch, usually with thick glandular walls : it is blind above and constricted below, and * Prof. Huxley describes it (' Anatomy of Vertebrate Animals,' p. 308) as opening into the " anterior and dorsal region of the cloaca." If " anterior " mean here, as I suppose it does, " anterior " as regards the whole length of the digestive tract, this is an error ; for, as I have above stated, the opening of the bursa is in the lowest chamber of the cloaca (i. e. that nearest the vent), and therefore into the most " posterior " one as regards the general direction of the alimentary canal. ON THE BURSA FABRICII IN BIRDS. 9 opens by a simple rounded orifice without any valve* into the cloaca at P ' Z '|^g 877 ' the point already indicated. The bursa lies on the posterior wall of the cloaca (to which it is usually attached by a kind of aponeurosis) in the space between the cloaca and the pelvis. This disposition of parts, as seen from behind, is shown in fig. 2, in the common Guillemot. The Fig. 2. ~d Back view of Cloaca of Uria troile. a. Bursa Fabricii. b. Oviduct. c, c. Ureters. d. Sphincter muscles. e. Caeca. bursa when laid open frequently contains faecal matter, more or less hard, or may be empty. The walls may be thin, as in the Passeres, Parrots, &c., or thick and markedly glandular, as in the Gallinae, Herodiones, &c. The bursa is usually constricted below into a narrow peduncle, with a narrow central channel ; above, the contained cavity is more considerable but of varying dimensions. Signor Alesi, in the paper already referred to, has described at great length the minute structure of the bursa Fabricii in the common Fowl and some other birds. I must here content myself with giving a brief resume of his conclusions as regards the histological structure of the organ in question. A transverse section of the wall in the neighbour- hood of one of the ridges into which, in the Fowl and allied forms, the * Barkow, I. s. c., describes the lower part of the peduncle of the bursa as projecting into the cloaca in the common Fowl, and there forming a conical sort of flap (Zipfel) over the opening. I have never myself seen any thing like this " Zipfel " in any bird I have examined ; and all other writers are equally silent on the point. Most, indeed, say expressly that there is no valve or flap of any kind. 10 ON THE BURSA FABRICII IN BIRDS. P.Z.S.1877, interior of the bursa is raised shows the following constituent parts: (1) an external layer of connective tissue, covering (2) a thick layer of elastic membrane ; inside this is (3) a thin layer of mucous membrane, which unites together and connects the "lymphatic follicles." These " lymphatic follicles " consist of masses of minute rounded cells, on an average 0-04 millim. in diam.., enclosed in capsules of connective tissue, in which ramify their nutrient vessels &c. The lymphatic follicles are bound together by processes of the connective mucous membrane into raised processes, which project on the interior of the bursa, forming ridge-like " crests," and are covered with epithelium internally, the cells of the latter being lanceolate with oval nuclei. In Rhea, however, the follicles are not closely bound together in masses forming ridges, but are attached by peduncles of elastic tissue to a central stem, the whole having somewhat the appearance of a bunch of grapes with a few berries on it. As we have already seen, Tannenberg in 1789 was the first to point out that the bursa was more developed in young than old birds, it being gradually reduced and obliterated in the latter. This process of atrophy seems to obtain in all birds, so far as I can make out, though the periods of final disappearance seem to vary much in different groups. M. Martin St.-Ange found that the bursa began to lose its functional activity in Pigeons after six months, and in Fowls after eight ; as a rule it seems to atrophy at about the period of full growth. On the other hand, in some cases it persists for long periods, and probably throughout life ; for I found it well developed and quite open in a specimen of Platycercus P. Z. S. 1877, icterotis that had lived in the Society's Gardens for eleven years. As a p. 311. rule, the central cavity of the bursa becomes diminished and the com- munication with the cloaca obliterated in the process of atrophy; and finally, in some cases at least, the whole organ completely disappears, losing its characteristic form and structure, and becoming fused with the tract of abdominal aponeurosis that covers the back wall of the ali- mentary canal, ultimately forming a flat riband-like, or round cord-like, ligamentous structure. Spangenberg observed the conversion of the bursa Fabricii into a round cord-like ligament in the Duck ; and Alesi has observed it (and described it minutely, with figures showing the histological changes that go on) in the Fowl. He sums up the process of atrophy in the last-named bird as follows : The epithelium which covers the internal surface of the bursa becomes, as it were, invaginated into the thickness of the mucous layer, becoming pressed in between the lymphatic follicles. The connective tissue between the latter increases rapidly in bulk, and becomes full of numerous small oval corpuscles; at the same time the follicles become absorbed, and the whole of the bursa becomes obliterated and fused with part of the abdominal aponeurosis. The epithelium gradually becomes confined to certain irregular tracts in the thickness of the mass of connective tissue ; but ON THE BURSA FABRICII IN BIRDS. 11 even these disappear finally. The process of atrophy in the other species which he observed is essentially the same as that in the Fowl. I now proceed to describe the general characters of the bursa Fabricii in the groups of birds I have examined. In the Passeres it is usually small, pyriform, with a small opening and glandular walls without ridge- like elevations on the interior. It seems to disappear completely in adult birds. In the RhamphastidaB it disappears in the adult, only leaving a small pore to mark its place of opening. In the Cuculidae it presents a very characteristic shape, the peduncle being long and thin, and the extremity club-shaped, giving the whole somewhat the appearance of a P. Z. S. 1877, shortened and clumsy antenna of a butterfly (see fig. 3). It disappears P* ~' Fig. 3. Cloaca and Bursa of Geococcyx affinis. The cloaca is cut short and laid open from before. The narrow peduncle of the bursa is supposed to be seen through the cloaca, a. "Yesical sphincter.' b. Bursa. c. Opening of same, d, d'. Ureter and opening, e, e' . Vas deferens and opening. completely in adult birds. In specimens of Dacelo gigantea, Merops, and Momotus lessoni it was sac-like and of considerable size. In the Parrots it is well marked and bag-like, opening by a small pore. As already mentioned, in an old specimen of Platycercus icterotis I found the bursa well marked, though its opening into the cloaca was nearly obliterated. In a specimen of Stringop* I found no bursa. In the Accipitres it forms a moderate-sized pyriform sac ; in old birds this becomes reduced to a sort of small conical pouch in the substance of the back wall of the cloaca at the place of aperture. Of the Steganopodes, in Phalacrocorax (2 species) and Sula bassana the bursa is a large sac, l|-2 inches in length, which opens by a small pore in the usual places. The walls of the bursa are very thick ; they are traversed by about seven rows of large, irregular, crypt-like pores, separated by smooth, narrowed, raised ridges. The ridges and pores disappear towards the base of the bursa ; but the upper part has a curious honeycomb-like appearance, some- what like that of the Ruminant reticulum. The internal surface is covered with numerous, small, opaque, granular-looking corpuscles, especially 12 ON THE BUESA FABRICII IN BIEDS. towards the apex. In a young specimen of Plotus arihinga ( $> ) that I dissected, the appearance and disposition of the parts agreed very well with Prof. Garrod's description. 1 noticed, however, that the opening of the bursa was very slightly constricted by a slight fold of mucous membrane on each side, so that the bursa hardly opened by its entire width into the cloacal chamber. In other respects it showed a strong approximation to the disposition of these parts that I shall presently describe in the Ostrich and its allies. In the Herodiones the bursa is large and sac-like, with a small or moderate opening ; its interior surface has no ridges or sulci, but large crypt-like pores collected in patches. In aged birds it seems to disappear, though in a specimen of Abdimia that had lived in the Gardens more than three years it was still present, | inch long, with the opening still unobliterated. In Ckawia the bursa is a glandular pyriform sac, about 1 inch in length, with a moderate- sized opening (see fig. 1, p. 8). It seems to disappear entirely in old birds. In a specimen of Cygnus olor it was a large conical sac, 1| inch long, with a wide mouth, but slightly constricted off, and with no large glands. In aged specimens of Fuliyula rufiiia, which had lived sixteen years in the Gardens, and of Tadorna rutila, which was nearly eighteen years old, the bursa had disappeared, in the former having assumed the appearance of a round cord-like ligament, in the latter having become fused with the abdominal aponeurosis. In each case a small pore marked the place of aperture. In Pterocles, Goura (in a nine-months-old bird), and PJiaps I have never found any thing more than a small blind pore in the usual place of the opening ; as we know from M. St.-Ange, the bursa disappears very early in Columba livia as well. P. Z. S. 1877, In the Galling the bursa has a tubular or pyriform shape, with numerous well-marked alternating ridges and furrows, the latter highly glandular, on its internal surface. These ridges are most marked towards the superior (blind) end, and are formed by the projection into the interior of collections of lymphatic follicles. These last do not open by distinct pores into the cavity of the bursa, but are entirely closed, as shown by Signer Alesi and Leydig. In the process of atrophy the peduncle be- comes sclid ; also the cavity of the bursa becomes shut off from the cloaca and persists in this state for some time. Finally, however, the bursa seems to become reduced to a ligament-like structure, or to become fused with the general aponeurosis. In the Turkey the bursa is a long tubular sac, 2 inches long, with very well marked ridges and sulci. In all a pore marks the opening of the atrophied bursa. In Serpentarius the bursa is a large globular sac, with the glandular area confined to the apical region and a small aperture into the cloaca. In these respects it resembles Otis and Grus. In Forphyrio and Ocydromus it assumes the form of a long, ON THE BURSA FABRICII IN BIRDS. Fig. 4. 13 Cloaca of Casuarius uniappendiculatus, viewed from behind. a, a. Cut surface of posterior wall of bursa. b. Opening of cloacal chamber into bursa. c. Pointer passing from cavity of bursa to exterior, d'. Pointer passed from cloaca into bursa through opening b ; the upper part is supposed to be seen through the wall of cloaca, e. Clitoris. narrow, cylindrical tube, the central cavity of which becomes closed up as atrophy advances. In (Edicnemus there is a similar form of bursa. In Attac/is, on the other hand, the form is pyriform, more like that of the Passeres and Gallinse. In Larus I found the bursa represented by a small pouch. In the young Uria troile it is large and sac-like, and slightly curved from side to side (see fig. 2, p. 9). The walls are very glandular, and so thick that the central cavity is but small. There are no crests. In an adult Alca tor da it was reduced to a pore-like opening. In none of the Tinami that I have dissected have I found any bursa ; on the other hand, the posterior wall of the third cloacal chamber is covered with numerous glands arranged in a tree-like manner. In all the birds hitherto mentioned (with the exception of Plotus) the bursa, as we have seen, opens by a more or less constricted aperture into the general cavity of the cloaca. In the Struthious birds, however, the very opposite is the case. The cloaca (or at least as much of it as corresponds to the first and second chambers) opens into the bursa Fabricii. This will perhaps be best explained by looking at fig. 4, represent- ing the cloaca and bursa of a not full-grown Cassowary (Casuarius P. Z. S. 1877, p. 314. 14 ON THE BURSA FABRICII IN BIRDS. uniappendiculatus) from behind. The bursa is, as one sees, a large, somewhat triangular sac, attached above by a broad riband-like muscle to the posterior wall of the alimentary canal. Most of the back wall of the bursa has been cut away (a), to show the opening into its cavity of the cloacal chamber (6), out of which a pointer (d') is seen passing up into the rectum above through the recto-cloacal valve. Prom this, I think, it Fig. 5. Vent of young Ernu (Dromceus novce-hollandiM} with the parts still in situ, viewed from the outside. A. External sphincter. B. Cavity of bursa. C. Wall of cloaca. D. Opening of cloaca into the bursa. E. Clitoris. F. Glandular pores. will at once be evident that the cloaca does not open directly to the outer surface, but indirectly through the bursa by means of its large posterior and inferior aperture b. A similar condition of things is seen in fig. 5^ in a young Emu (Dromceus novce-hollandice) where the parts are undis- P. Z. S. 1 877, turbed and in situ, and in fig. 6, in an Ostrich (Struthio camelus, immature p. 315. female) where the bursa has been nearly all removed to show the pos- terior opening of the cloaca into its cavity, and the communication of the latter with the exterior, as indicated by the direction of the pointer (D D'). The same is the case in the young Nandou (liliea americana). In all these birds the walls of the bursa are thickly glandular ; there are no regular crests and sulci, however, but the glands are arranged in patches, the whole having a honeycomb-like or dendritic appearance. This disposition of parts, however, is not permanent. As the birds grow older, the size of the bursa gradually diminishes and its walls become less glandular; its mouth is no longer equal in extent to the whole width of the outermost chamber, but becomes narrowed ; and finally the whole bursa disappears, its remains becoming lost in the muscles of the back of the cloaca. This state of alrophy of the bursa is represented ON THE BURSA FABRICII IN BIRDS. 15 Fig. 6. Cloaca and Bursa of young Ostrich (female), viewed from behind. Most of the posterior wall of the bursa has been removed. A. External sphincter muscle. B. Cut surface of bursa. C. Opening of cloaca into bursa. D, D'. Pointer passing from bursa to exterior. E, E. Ureters. F. Oviduct. G-. Clitoris. H, H'. Pores ; beneath them the smooth, non-glandular part of the bursa. in Casuarius picticollis in fig. 7 (p. 16), the only remains of its existence being seen in the few irregular circular folds on the mucous membrane at A. At what age this change supervenes is not yet quite clear. The p. z. S. 1877, specimen of C. beccarii that died in the Society's Gardens last year, appa- P- 3i6 - rently an adult female, corresponded entirely in the disposition of these parts with the specimen of C. uniappendiculatus above described, which died shortly before, and was quite a young bird when received in 1874. On the other hand, in C. picticollis the bursa was entirely atrophied. This specimen, which also arrived in 1874, was then not quite adult ; so that at the period of its death it must have been about three years old. Whether Apteryx agrees in these points also with the other StrutJiiones I am unable to say, as both the specimens I have dissected presented no trace of a bursa. Probably therefore in this, too, when adult the bursa disappears almost completely. I have mentioned above the singular differences shown by Signer Alesi to exist in the structure of the lym- phatic follicles of the bursa of RTiea as compared with other birds. 16 ON THE BURSA FABRICII IN BIRDS. Fig. 7. A. Cloaca of Casuarius picticottis, adult male ; viewed from before. A. Circular folds of mucous membrane, being the last remains of the Bursa. B. Rectum, b. Recto-cloacal valve. 0. Cut surface of external sphincter muscle. c. Vesical sphincter. D. Urino-genital papillge. E. Glandular pore. P. Penis. Although at first sight the relation of these parts in Rliea, Struihio, &c. seems so different from that which obtains in other birds, yet a little reflection will, I think, convince one that it is not so in reality. I have represented diagram matically (figs. 8 and 9) what I conceive to be the true relations of the parts in question, denoting the homologous regions in the two forms (Serpentarius as illustrating the normal type, Casuarius the abnormal one) by similar letters. If we imagine B in fig. 8 to lose the constriction at its aperture into D, and D to become P. Z. S. 1877 proportionately deeper, we should have a form corresponding to fig. 9, p. 317. in which B passes uninterruptedly and without constriction into D, the cavity B D thus formed receiving the aperture of C. In such a form as Plotus, where the opening is but very slightly constricted, we have a type connecting the two extremes ; and I have seen RJiea in a stage very similar to that mentioned above in Cygnus olor. In confirmation of this view as to the true relation of these parts, in the Ostrich &c. the lower part of the bursa, in the region corresponding to D in figs. 8 and 9, is not glandular (vide fig. 6, p. 15, where the non-glandular part of the bursa is seen beneath the pores H H'). "With regard to the function and homologies of the bursa Fabricii, great differences have prevailed amongst authors. Thus Milne-Edwards says*, " Fabrice d'Acquapendente, a qui Ton doit la decouverte de cette * Phys. et Anat. Comp. vol. viii. p. 514. ON THE BURSA FABRICII IN BIRDS. Fig. 8. Fig. 9. 17 Diagram showing two chief Types of Development of the Bursa Fabricii. R. Rectum. B. Bursa. C. Cloacal chamber. D. Lowest chamber of "cloaca." d. Openings of urino-genital ducts. bourse, la considerait comme uii reservoir seminal, tandis que d'autres naturalistes * la regardent comme une vessie urinaire. Perrault et quelques auteurs modernest y voient 1'analogue des glandes anales des Mammiferes, et Geoffr. St.-Hilaire 1'assimile aux glandes du CowperJ; enfin, M. Martin St.-Ange la compare a la prostate." Emil Huschke, in the paper mentioned above, has studied its development, and, after a comparison of the organs of similar appearance, is inclined to consider it as the primitive urinary vesicle of the Wolfian bodies, from the fact that the ducts of this gland take origin from just that part of the cloaca which afterwards assumes the form of the bursa. Harvey and others have sufficiently disproved Fabricius's ideas as to its serving as a spermotheca; nor can the bursa be regarded as a urinary bladder, first, because it is not devoted to containing the urine ; secondly, because in other Sauro- psida and also in the Mammalia the urinary bladder is ventral, not dorsal, in position. For a similar reason, as well as from the fact that they are paired organs, the " bursce anales " of the Testudinata can be in no way p. z. S. 1877, related to that under discussion. The anal glands of Mammals, again, P- 318 - open externally on the skin, and are in fact cutaneous glands. The prostate and glands of Cowper are purely male glands, and probably play some important function in the act of reproduction ; so that they can hardly well correspond to an organ that is common to both sexes, and only proportionally developed in the young. It would be premature to accept Huschke's views without further observations on the subject. On the other hand, as pointed out by Signor Alesi, a lymphatic organ, constructed on a similar principle, but in a simpler form, exists in the * E.g. Berthold, Acad. Caes.-Leop. Nova Acta, xiv. p. 917 (1828 and Geoffrey St.- Hilaire, Mem. du Museum, 1823, t. ix. p. 394. t E. g. Carus, ' Zootomia.' I Ticdcmanr, ' Anat. der Vogel,' 1810. C 18 ON THE BURSA FABRICII IN BIRDS. patches of lymphatic follicles (which do not, however, in this instance project outside the mucous membrane of the intestine) in the appendix to the caecum of the rabbit (described by Frey, ' Untersuchungen iiber die Lymphgefasse des Darmkanales,' Leipzig, 1863). An organ still more closely corresponding in its general shape and position with the bursa Fabricii is the sac-like pouch which opens into the dorsal wall of the cloaca in many Elasmobranchs *. The glands of this, however, differ in structure from those of the bursa Fabricii ; so that at present it seems to me that we can assign no very definite analogue or homologue for the latter, but that it is a glandular outgrowth of the cloaca peculiar to birds. In conclusion, I may briefly recapitulate the chief conclusions arrived at in this paper : (1) That the bursa Fabricii exists in both sexes, and probably in all species, of birds. (2) That it is most developed in young birds, but becomes atrophied and more or less obliterated in adults, the period, however, of the com- mencement and conclusion of this process differing greatly in various birds. In some it probably persists, though in a state of functional inactivity, throughout life. (3) That in the majority of birds the bursa is a moderate-sized or small sac, that opens by a narrow aperture on the dorsal wall of the cloaca into the lowest " chamber " of that organ. (4) That in the Struthious birds, on the contrary (the state of Apteryx as regards these points being doubtful), the cloaca opens into the bursa by a posterior aperture, owing to the fact that the bursa is not constricted off at the neck, but is commensurate in extent with the third or outer chamber of the cloaca, the two being united into one. This condition, however, is only to be found in young birds. (5) That the bursa is a glandular organ, of which lymphatic follicles are the essential constituents, but has no exact homologue in other classes of Yertebrata. * Signer Alesi, in his paper, s. OX THE PARROTS OF THE GENUS ECLECTUS. 21 intermedia (green) 1 ^^ Amb Buru Ibia , 1877 , cardinalis (red) J p. 277. From this it is clear that the range of one green form (E. polychlorus} corresponds with that of two red {E. linncei and E. grandis). "As I cannot hesitate a moment," says Dr. Meyer, " in ascribing the conditions found in E. polychlorus and E. linncei from New Guinea, Mafoor, and Jobi to the other allied form (namely, that the green are the males and the red the females of one and the same species), the interesting fact comes out (unparalleled, so far as I know, in the ornis of the whole world), that differently coloured females correspond to one and the same male in different localities ; for E. linncei and E. yrandis show at first sight such differences, that, so long as we did not know their true relations to E. polychlorus, they were universally considered different species. Thus, therefore, the male remains constant, whilst the female varies." Dr. Meyer then proceeds to show that no theories of " sexual " or " natural selection " can account for these facts, of the causes of which we are completely ignorant. Schlegel (Ned. Tijd. v. d. Dierk. iii. p. 332, 1866), he observes, has already united E. intermedium and E. polychlorus into one species, the examples from Grebe and Waigiou being intermediate in their characters between these two forms. Moreover an authentic specimen of E. intermedius from Ceram, received from the Leyden Museum, and now in the Imperial Cabinet at Vienna, quite agrees with Dr. Meyer's, series from New Guinea, Mafoor, and Jobi. Hence E. polychlorus (including under this term E. intermedius) possesses in different islands three females, differently coloured according to the locality, viz. : (1) linncei, in New Guinea, My sol, Waigiou, and Gebe ; (2) grandis, in Gilolo, Batjan, and Morotai ; (3) cardinalis, in Ceram, Buru, and Aniboyna. Dr. Meyer then goes on to argue that E. wesfermanni and E. cornelice, both remarkable for being nearly uniform in colour, must also be regarded as forms of E. polychlorus. He urges that E. cornelice may well be a fourth female of E. polychlorus, as we already know that the females of this species are variable, whilst E. westermanni, he considers, is probably ^ s > ^^ an individual that has retained its juvenile plumage and has been unable to assume its adult colouring owing to captivity. Here I must join issue with Dr. Meyer on several grounds. First of all, several examples of each of these condemned species have lived at various times in the Zoological Gardens of London and Amsterdam, and no noteworthy difference has been detected in these specimens. Again, specimens of both species have lived for considerable periods at Amsterdam without undergoing any change in coloration (vide Finsch, I. s. c.). More- over Parrots, as a rule, including those of the present genus, do remarkably well in captivity, and show no tendency to lose or to fail to acquire their brilliant colours or to retain their immature dress. Eclectus cornelice and 22 ON THE PARROTS OF THE GENUS ECLECTUS. E. westermanni can hardly be man and wife, owing to their disparity in size (the wing of the former being given by Pinsch as 9" 5'", of the latter 7'' 8"' to 8" 5'", and other measurements in proportion). Hence we may conclude that in the former case the male, in the latter the female, remains to be discovered, as well as the exact habitat of each. When we reflect on the little knowledge we still have of the great mass of New Guinea, as well as of some of the neighbouring islands, it is evident that ample area for such a discovery is still left. This conclusion is strengthened by the fact that certain other Parrots belonging to the same region, likewise first de- scribed from captive specimens, and undoubtedly distinct (e. g. Lorius *i&iafta,Scl. P.Z.S. 1871, p. 449, and TricJwglossus mitcMli, G. E. Gray), have their exact habitat still unascertained. The recent discovery of Loriinse (a group of which the geographical range coincides remarkably with that of Edectus, as has been pointed out by Mr. Wallace) in such unexpected localities as Ponape (in the Caroline group), where Chalcopsitta rubiginosa occurs*, and Fanning Island, in the mid Pacific t, renders it even possible that an Edectus may turn up in some equally "unlikely" locality J. J^j 7 ' Dr. Meyer then goes on to show that Bernstein's determinations of the sexes of the specimens he forwarded to the Leyden Museum are probably erroneous, as in his three years' experience he found the sexes about equally numerous, whereas Bernstein's determinations would show great disparity in their relative abundance (in one case six males to one female, in the other twelve females to two males). The juvenile plumage of Edectus is unfortunately still unknown; but Dr. Meyer concludes that it is probably green, from the fact that twelve out of fourteen of his red specimens still preserve evident traces of green feathers. In reply to these arguments Prof. Schlegel not unnaturally hesitates to accept Dr. Meyer's conclusions, because, of 72 specimens of red Eclecti in the Leyden Museum, 20 have been determined by the collectors as males, and the remainder (52) as females, and, on the other hand, of 77 green specimens in the same museum, 56 are marked as males and 21 as females. Hence, if Dr. Meyer be right, a considerable proportion of these specimens must have been wrongly sexed by the four travellers by whom they were collected, viz. Salomon Miiller, Bernstein, Hoedt, and Von Rosenberg. Dr. Meyer returns to the charge in a paper in the ' Mittheilungen aus * Vide Finsch, ' Journal des Museum Godeffroy,' Heft xii. 1876. t Coriphilus Jcuhli, P.Z.S. 1876, p. 421. | Prof. Bietmann's "shining-red Parrots" in Guadalcanal- (P.Z.S. 1869, p. 127) might well be E. cornelia. Mus. Pays-Bas, Psittacidir, 1874, p. 17. ON THE PARROTS OF THE GENUS ECLECTUS. 23 dem k.-k. zoologischen Museum zu Dresden' (I. c. pp. 11-13). He repeats his former observations, and gives some additional ones, amongst which are some remarks on a living pair of Eclectus in his possession, green and red, the green bird on being introduced to the red at once having become friendly with the latter. A green Eclectus that died soon after it came into his possession was dissected and turned out to be a male. As regards the specimens in the Leyden Museum, Dr. Meyer disposes of them by saying that those collected by S. Miiller have been long in the Museum, and may very probably have had their labels transposed that Bernstein, during the latter part of his residence in the Malay archi- pelago (as he himself learned from one of his hunters, who had also collected for Bernstein, and knew the latter well), suffered severely from illness, and therefore may well have made mistakes in the deter- Ibl8> mination of the sexes of his specimens that Hoedt had no pretensions to any scientific knowledge and that Rosenberg has in other instances made blunders of a similar kind so that their evidence counts for little. Dr. Meyer adds some mathematical calculations showing that the chances are 32,700 to 1 against his having killed six all males of the green Eclectus, and nine all females of the red one in the same island, if they really were distinct species. So far Dr. Meyer. Important evidence in corroboration of part of his theory is given by the Italian naturalists who have lately visited New Guinea. Beccari, in his Ornithological Letters to Count Salvador! *, says, " Though it seems strange, it is nevertheless true that the green Eclecti are males of the red ones. I learnt this at Aru from my hunters; and the young have the same differences." Salvadori says again (I. c. pp. 756, 757), speaking of the sexual differences in E. grandis, that there is "no longer any doubt on this subject. D'Albertis has assured me that it is a well-known fact amongst the natives of the Moluccas and New Guinea." In his various papers on Papuan ornithology in the same journal, the green specimens of Eclectus are always determined as males> the red as females. Prof. Garrod also tells me that during his prosectorship the only two Eclecti that have died in the Zoological Society's Gardens were one E. polychlorus and one E. grandis, respectively male and female. On the other hand, the Rev. George Brown, C.M.Z.S., who has lately sent over to this country such interesting collections from New Britain and the adjacent islands, says, in a letter to Mr. Sclater, dated Sydney, October 22, 1876, " This " (i. e. the green and red Eclecti being specifically iden- tical) " is a gross error. Our attention was directed to this subject ; and I am quite sure they are two different birds. We shot the green * Ann. Mus. Civ. Storia Natur. Geneva, vol. vii. p. 701 (1875), and Ibis, 1876, p. 253. 24 ON THE PAKROTS OF THE GENUS ECLECTUS. ones, both male and female/' Two skins in the collection are referable to E. polychlorus and linncei ; the latter is marked female. It is to be hoped Mr. Brown will renew his investigations into this subject, as the *81 7 ' determination of the sexes is not always very easy without careful dissection, the suprarenal bodies in birds being particularly liable to be mistaken for the testes when the latter are not developed to the extent that they are during the breeding-season. On the whole, I think, we must conclude, in company with Dr. Meyer and Count Salvadori, that; the green Eclecti are really males, the red females. With regard to Dr. Meyer's conclusion that all the species hitherto described must be regarded simply as forms of one species (E. poly- chlorus), I have already adduced reasons for believing that E. ivestermanni and E. cornelice are good species. As regards the other five, a careful examination of a large series of skins from different localities (we now know that Eclectus extends east as far as Tule Island and Duke-of-York Island) will be necessary before coming to any definite conclusion on the subject. Count Salvadori, however, who has probably had as large a series of specimens from different Papuan islands of this genus as any- body, recognizes three distinct species (besides the two of unknown habitat), which he says may always be recognized as distinct at any age or in either sex. He gives the following table of these species as under- stood by him (I. c. p. 756) : 1. Virides: lateribus rubro-puniceis. (Mares.) a. Majores. a'. Viridis, colore obscuriore, cauda minus caerulea .................. 1. polychlorus. b'. Viridis, colore laetiore, cauda magis caerulea ..................... 2. cardinalis*. b. Minores. Cauda \-ix cserulea ............................................. 3. ffrandis*. 2. Kubrse : fascia interscapulari et abdomine cyaneo vel violaceo. (Feminse.) a. Annulo periophthalmico cyaneo .......................................... 1. polychlorus- b. Annulo periophthalmico nullo a'. Subcaudalibus auroreis vel rubro-flayis .............................. 2. cardinalis. b'. Subcaudalibus pure flav is .......................... . .................. 3. In this table the green E. cardinalis is, I suppose, the intermedius of most authors, whilst the red E. polychlorus is clearly what is usually called E. linncei, Ibis, 1877, On the whole it seems probable that we must be content with ascribing p. 282. f. Eclectus the most marked sexual differences in colour of any Parrots hitherto known. Aprosmictus (at least in some species, e. g. A. scapulatus) also presents very well-marked sexual differences in coloration, and, as Prof. Garrod has shown (P. Z. S. 1874, p. 494), agrees very closely with * In the original paper Count Salvadori has accidentally transposed these two names, as I have ascertained from a corrected copy of his paper that he forwarded to Mr. Sclater. ON THE PARKOTS OF THE GENUS ECLECTUS. 25 Eclectus in anatomical structure. Eclectus, however, differs from all known Parrots in having the female more gaudily coloured than the male. Can it be possible that, as in the few other analogous instances where the female is the more brightly coloured (e. g. Turnix, Rhynchcea, &c.*), the duties of incubation devolve on the male ? If such be the case, we can easily understand the use of the green coloration being retained by the male. Unfortunately we are still totally ignorant of the habits, nidifica- tion, and immature plumage of these Parrots. Let us hope that Signer D'Albertis or Mr. Brown will soon throw some light on this, as well as on the other interesting points noted above, which still require further examination. In conclusion, supposing that we assume the new views as to the sexual differences of the Eclecti to be correct, the following list of the species will show concisely their sexual differences and geographical distribution. 1. ECLECTUS POLYCHLOKTJS (Scop.). Maximus : mas viridis colore obscuriore, lateribus rubro-puniceis, caudd minus cceruled : femina rubra, fascia inter scapulari, abdomine et annulo perioplitlialmico cyaneis. Hab. in insulis Papuanis et Moluccanis Ternate, Gilolo, Batchian, Morotai, Guebe, Waigiou, Mysol, Gage, Ke, Aru, Papua, Nova Hibernia, et Nova Britannia. 2. ECLECTUS GRAKDIS (Gin.). Major : mas viridis, lateribus rubro-puniceis, caudd vix cceruled : femina rubra, fascia interscap. et abdomine cyaneis, subcaudalibus purbflavis. Hab. in insulis Ternate, Gilolo, Batchian, Morotai, et Gage. 3. ECLECTUS CARDINAUS (Bodd.). Ibis, 1877, Minor : mas viridis, E. polychloro similis, at colore Icetiore cauddque P* magis cceruled distinguendus : femina rubra, fascia interscapulari et abdomine cyaneis ; subcaudalibus auroreis vel rubro-Jlavis. Hab. in insulis Moluccanis Ceram, Bouru, et Amboyna. 4. ECLECTUS WESTERMANNI, Bp. Minor : mas viridis, lateribus concoloribus. Femina adhuc ignota. Hab. ? (Viv. Nat. Art. Mag. et Zool. Soc. Lond.) 5. ECLECTUS CORISTELI^B, Bp. Mas ignotus : femina punicea, colore cyaneo neque dorsi neque lateris inferioris ullo. Hab. ? (Viv. Nat. Art. Mag. et Zool. Soc. Lond.) * Vide Darwin's ' Descent of Man/ vol. ii. p. 200 et seq. (1871). 26 ON THE NESTING OF THE SPOONBILL IN HOLLAND. Ib p!'4i2. 7 ' 7. ON THE NESTING OF THE SPOONBILL IN HOL- LAND. By P. L. SCLATER and W. A. FORBES.* THAT the Spoonbill (Platalea f leucorodia) breeds in Holland is a fact well known to every ornithologist ; and most egg-collectors are aware that specimens of its eggs obtained in that country are to be purchased at a very cheap rate in the London egg-shops. But we are not sure that Ibis 1877 p.' 413. ' an J ornithologist, at least of this country, h#s actually visited the nesting-places of this bird, or, at any rate, has published any account of them. In May 1867, as is recorded in Gould's ' Birds of Great Britain ' (vol. iv. part 30), Sclater paid a visit to a nesting-place of he Spoonbill at Nieuwerkerk, near Eotterdam ; but though he saw many Spoonbills, the nesting had not then begun ; and the lake which he visited is said to have been drained since that time. We hope therefore that it may interest readers of ' The Ibis ' to have an account of our recent expe- riences on this subject. Being in Holland in the first week of May this year, Sclater made many inquiries as to where the Spoonbills could be seen performing the duties of reproduction, and finally ascertained from Hr. A. A. Van Bem- melen, Director of the Zoological Gardens at Eotterdam, that the most likely place to witness this interesting phenomenon was the .Horster Meer, between Amsterdam and Utrecht. At Amsterdam it was ascer- tained that the first week in July would be a convenient period for the proposed excursion with this object, as about that time the birds would have commenced incubation. On the 3rd of July, therefore, we found ourselves at the Amstel Hotel, at Amsterdam ; and upon visiting Mr. Hegt, the Assistant-Director of the Zoological Society's Gardens there, found that he had kindly made every necessary arrangement for our proposed expedition next day. No railway-station being very convenient for the Horster Meer, he had ordered a carriage to take us from Amsterdam to the scene of action. Next morning we started about 8 o'clock, and had about three hours' drive, passing the villages .of Abgouda and Vreeland before arriving at Overmeer an de Vecht, the little village in which Hr. van Dyk, the lessee of the Horster Meer, resided. The Horster Meer consists of a large tract of water reed- beds and swamp, lying on the right bank of the Vecht, and immediately to the south of the Zuyder Zee. It is between the * Ibis, 1877, pp. 412-416. t Mr. Dresser (B. Eur. pt. 23-24) uses Platea as the generic name of the Spoonbill instead of Platalea. It may be hoped, however, that this is a mere oversight, and that Mr. Dresser is not prepared to dissent from the canon that Linnean names are to remain inyiolate. ON THE NESTING OF THE SPOONBILL IN HOLLAND. 27 railways going from Amsterdam to Utrecht on one side, and from Ibis, 1877, Amsterdam to Amersfoort on the other. It belongs to a rich proprietor in Amsterdam, but is farmed out at a considerable rent for the sake of the fish, reeds, and bird's eggs which it produces. The last-mentioned objects are collected from the nests in which they are laid, twice a week during the months of May and June, and sold in Amsterdam to such persons as require a large supply of fresh eggs without being particular as to the source from which they are derived. On arriving at Overmeer we were received by Hr. van Dyk and escorted to a boat, which conveyed us along a short canal into the Horster Meer. No sooner had we arrived on the lake than the air above us was filled with an enormous flight of Cormorants, who well knew what a visit to their domain portended. A few minutes afterwards about 500 Spoonbills were circling in the air over our heads, their long legs stretched behind them, and their white bodies glistening in the sun. The Meer, so far as visible, was not a very extensive piece of water, being closed in on all sides by enormous reed-beds, the homes of these and other aquatic birds. Having landed at the end of a ditch which penetrated into one of these beds of reeds, we pursued a track which led us first to a breeding- place of the Cormorants. Here was a circular space, perhaps fifty yards in diameter, cleared of reeds, in which the Cormorants' nests stood thick together on the swampy soil. They were formed of rather large sticks, piled somewhat loosely together to a height of about 18 inches above the surface. The top of the nest was only slightly hollowed out, and lined with a few broken reeds. The eggs were in no case more than two in number, the poor birds having been robbed continuously up to that time, and only within the last few days allowed to commence incubation. Having inspected the Cormorants' breeding-place, we proceeded about fifty yards further through the reed-beds, over a still more treacherous swamp, to the breeding-place of the Spoonbills. The nests of these birds were not situated so near together as those of the Cormorants, but scat- tered about two or three yards from each other, with thin patches of reeds j^is ^77 growing between then. There was, however, a clear open space in the p. 415. neighbourhood, formed of broken-down reeds, in which the birds were said to congregate. The Spoonbill's nest, in the Horster Meer at least, is a mere flattened surface of broken reed, not elevated more than two or three inches above the general level of the swamp ; and no other sub- stance but reed appears to be used in its construction. What the proper complement of eggs would be if the birds were left undisturbed we cannot say ; for, as in the case of the Cormorants, the nests are robbed systematically twice a week, until the period when it is known by expe- rience that they cannot produce any more eggs. Then at last the birds are allowed to sit undisturbed. At the time of our visit the season for collecting eggs was just past ; but we helped ourselves to eight fresh 28 ON THE NESTING OF THE SPOONBILL IN HOLLAND. eggs, from different nests, laid since the last collection had been made. During all the time that we were in the reed-beds the Cormorants and Spoonbills were floating about over our heads, fully aware that there was an enemy in the camp. We were told that there were several other nesting-places of the Spoonbill in different parts of the Horster Meer, containing altogether several thousand nests ; so that we may hope that it will be some time before this fine bird becomes extinct in this locality. The only other bird we found nesting in the Horster Meer was the Black Tern, of which we captured two young chicks. After refreshing ourselves at the hostelry of Overmeer, we returned to Amsterdam in the evening by a different route, highly satisfied with our day with the Spoonbills. We may observe, in conclusion, that on looking over Mr. Dresser's account of the Spoonbill in his ' Birds of Europe,' we find him quoting from Schlegel that this bird " is found in the neighbourhood of the large rivers, at Biesboch, Nieuwerkerk, on the Tssel at B-ozenburg, and on the Maas ; and breeds in Holland, arriving there in April and leaving in September." Again, a few pages further on, Mr. Dresser says, "It breeds in Holland ; but I do not find any record of its having of late been found nesting elsewhere in Northern Europe, though in Hungary p' 416. ' an ^ South-eastern Europe it breeds numerously." In Mr. Gould's folio, too, no more detailed account is given, with the exception of the record of Sclater's unsuccessful expedition ten years ago. Now our experiences as to the position of the Spoonbill's nest certainly agree with the details given by Messrs. Dickson and Ross, who met with it breeding near Erzeroum (P. Z. S. 1839, p. 134) ; and this seems to have been the fullest account known to Mr. Dresser at the time of writing his article. So, although there seems to be no reasonable doubt that in some cases it nests in lofty trees, we may claim to have established the fact that in Holland it breeds on the ground among the reed-beds, and to be able to assure those naturalists who happen to be in Amsterdam at the right time that there is no better way of spending a spare day than an excur- sion to the Spoonbills' nesting-place on the Horster Meer. We cannot conclude this short account of a most delightful day without thanking Mr. Hegt most heartily for his kind arrangements for our trip, without which we should probably have encountered con- siderable difficulty in reaching our destination. It is to be feared that in England we could hardly promise to show our friends an equally interesting sight in such close proximity to our metropolis ! LEPIDOPTERA FROM SWITZERLAND, ETC. 29 8. LEPIDOPTERA CAPTURED DURING AN EXCURSION TO SWITZERLAND AND THE ITALIAN LAKES.* THE following list of Lepidoptera, observed by myself and a friend (Mr. M. J. Michael, of St. John's College, Cambridge), during a short trip in Switzerland and the Italian Lakes last summer, may be of interest to entomologists who have collected on similar occasions, as' showing how much (or, rather, how little) can be done in entomology on a tour where this is not the only object. When in company with non-entomological friends (we were in all a party of four) a great deal of collecting has to be done on the sly, as it were, I mean by resorting to such expedients as walking up hills when travelling along the roads ; and by these means a number of additional species were procured, though sometimes perhaps at the expense of the time of the less interested members of the party. Unfortunately, the time for departure from England (about the middle of August) coincided with the setting in all over Western Europe of the spell of bad weather w r hich prevailed for about a month continuously ; and this, combined with the lateness of the season, no doubt much contri- buted to the lack of species observed, and the entire absence of some usually common, e. g. Daplidice, Palceno, &c. It also entirely frustrated our intention of visiting Zerrnatt, where we had intended to have spent some days collecting the insects of the high Alps. The total number of species observed in the month was 107, of which 33 are not British species. Por naming some of these I am indebted to Dr. Staudinger, whose nomenclature I have throughout followed. LIST OF SPECIES OBSERVED. Papilio podalirius seen near Colico. Papilio machaon Colico, Bel- laggio, Meuaggio. Parnassius apollo one 2 specimen at rest on thistles near Andeer. Pieris brassicce, rapce, napi everywhere in cultivated grounds. Leucophasia sinapis common in the gardens of the Villa Serbelloni and elsewhere at Bellaggio ; also at Menaggio and Pallanza. Colias Tiyale occurred nearly everywhere, and usually commoner than the next. C. edusa with the last. The var. hdice occurred at Bellaggio and Menaggio, but not commonly. .Rhodocera rhamni Villa Serbelloni and Pallanza. Polyommatus virgaurece one <3 specimen near the village of Spliigen. P. dorilis Baden ; common at Bellaggio and Pallanza. In this species the J is quite dark above, the $ having the primaries orange with dark spots, and an orange border to secondaries. P. pliloeas Baden, Bellaggio, Pallanza. The Italian specimens have the * Ent. Month. Mag. xiv. p. 243. 30 LEPIDOPTERA FROM SWITZERLAND markings less distinct, the copper colour redder, the spots smaller, the costal margin of the primaries darker, and the marginal band broader and narrower than any English examples I have. The underside, too, of the primaries, is redder, leaving the circumscriptions of the eyes and . the veins of the wings paler. Lycwna argiades This little "tailed" blue occurred, but not commonly, at Baden and Bellaggio. L. argyrotoxus (wgori) Colico and Bellaggio, in the grounds of the Villa Serbelloni. L. astrarche (cigestis) Bellaggio *. L. icarus Baden, Kandersteg, Colico, Menaggio, Bellaggio. L. lellargus (adonis) Kanderthal, Chur, Bellaggio, Menaggio, Val Yedro. The var. ceronus (which also occurs in England) at Bellaggio. L. corydon Kanderthal, Chur, Menaggio, Val Vedro. Two c? Swiss specimens have all the blank points on the underside of the wings much smaller and less distinct than in English ones ; in one En< 244 'rfs^n* a ^ so ^ e row ^ oran S e 8 Pts on the post-margin of the underside of the secondaries has almost disappeared. L. damon Kandersteg, Chur. Apatura ilia. The ab. clytie, which differs from the type in having the ground-colour of all the wings reddish ochreous instead of white, occurred commonly on the poplars lining the road between Chiavenna and Colico. This species flies strongly, but does not soar like A. iris, and sits on the leaves sunning itself till disturbed, but generally too high to be in reach of an ordinary net. The females have much less metallic gloss than the males. Limenitis Camilla in the gardens of the Hotel Grande Bretagne at Bellaggio, but not at all common : one specimen also in those of the Villa Serbelloni and Menaggio. Vanessi io near Tiefenkasten. V. ata- lanta Colico, Bellaggio, and Menaggio. V. cardui Bellaggio. V. anti- opa We only saw one of this species throughout our trip, and that was at Sargans, on the railway from Zurich to Chur. V. c-album Bellaggio and Villa Serbelloni. Melitcea athalia Menaggio and Bellaggio, in meadows. M. parthenie Baden. M. didyma This species was not uncommon on the railway banks near "Waldshut, and a timely delay of the train enabled us to get out of the carriage and procure some speci- mens before it started off again. Argynnis dm Baden and the Via Mala. A. papliia Baden, Colico, Bellaggio, Menaggio, and the Val Vedro. The var. valezina also occurred at the two last-named localities. A. adippe, var. deodoxa Bellaggio and Val Vedro. A. niobe, var. eris near Spliigen : also at Pallanza. A. latlwnia common at Baden ; also at Bellaggio and near Tiefenkasten. Melanargia galatliea a worn spe- cimen in the Val Vedro, and another on the hill behind Bellaggio. Erebia stygne Kanderthal, near Frutigen, in meadows at about 2500 feet ele- vation; also on a rocky wooded hill behind Bellaggio. E. nerine a single c? specimen of this rather rare species near Spliigen. Erebia * The Bellaggio specimens have both the red and black spots on the wings larger, and the ground-colour of the underside greyer than in English specimens. AND THE ITALIAN LAKES. 31 pronoe, var. piifio This handsome species was rather common in sloping dry meadows, between Frutigen and Kandersteg, at about 3000 feet elevation, but difficult to get in good condition *. E. cethiops (blandina) Thun ; common in the valley of the Kander, and on a hill behind Bel- laggio. E. ligea near Spliigen. Satyrus hermione Colico, Menaggio, and Bellaggio ; frequented a rocky wooded hill behind the latter, and seemed to like resting in shady places. On the wing somewhat resembles a large L. Sibylla. S. circe A specimen of this handsome species was seen at Baden. S. dryas (phcedra) This fine species was abundant at Bellaggio, frequenting, like S. hermione, a rocky, bushy hill behind the town, wheeling in its flight over the bushes, the roughness of the ground making it a matter of some difficulty to catch specimens. The female is larger and lighter in colour than the male ; has the blue eyes on the pri- maries larger, and with brighter blue pupils than in that sex, and is altogether a finer-looking insect. It also occurred at Menaggio, where it frequented the flowers of the millet (Milium effusum), which is grown in the vineyards in patches between the rows of vines, at Pallanza and near Colico. S. actcea, var. cordula with the last, at Bellaggio and Menaggio, but much less common. S. semele Menaggio and Bellaggio. Pararge hiera Villa Serbelloni, at Bellaggio, Menaggio, and Yal Vedro. P. cegeria Bellaggio, Menaggio. I think those I saw belonged to the pale northern form egerides, but not having kept specimens cannot say for certain. P.janira Baden, Colico, Menaggio, Bellaggio. P. tithonus ^ Bellaggio, Menaggio, and Pallanza. Coenonympha pamphilus Baden, Bellaggio, Menaggio, and Yal Vedro. Spilothyrus alcece Bellaggio. SyriciJms alveus Baden, Bellaggio, and Menaggio. S.proto gardens of the Villa Serbelloni, at Bellaggio. Hesperia comma everywhere : on the whole, the commonest butterfly, as far as regards our experiences. Sphinx convolvuli gardens at Schafflhausen and Thun. Macroglossa stdlatarum common in gardens at all places visited, except in the Alpine valleys. Zygoma filipendulcK Baden and near Chur. Lithosia deplana one at Thun. Callimorpha hem Via Mala ; common at Bellaggio and Menaggio, fluttering about flowers in the sunshine like a butterfly. Cossus ligniperda a full-grown larva picked up in the Via Mala, and an empty pupa-case near Liuno. Bombyx rubi larvaB common, crawling in the roads, near Chur, Spliigen, and in the Val Vedro. B. trifolii a <$ specimen picked up in the road near Menaggio. Agrotis c-nigrum Thun. Plusia triplasia Bellaggio. P. gamma Schaffhausen, Bellaggio. Catocala paranympha a single worn specimen in the inn " Belle Vue " at Frutigen. Addalia perochraria common in meadows near Baden, and in the Kanderthal, near Frutigen. A. immorata Baden. A. im- mutata Bellaggio. A. strigilaria common at Pallanza, on grassy * The var. pitho has the red bands uearly obliterated, 32 MELANISM IN LEPIDOPTERA. slopes near the lake. A. ornata common at most places we visited, but not in the Alpine valleys. Timandra ornata Schaffhausen, Liuno. Numeria capreolaria one specimen near Chur. Gnoplios ylaucinaria one specimen near Chur. G. dilucidaria Baden, Bellaggio. Ortholitha limitata (mensuraria) Chur. 0. bipunctaria Chur ; larger and darker than English (chalk) specimens. Minoa murinata (cupkorbiata) Pal- lanza. Anaitis plagiata Baden, Schaffhausen. Lygris popidata com- mon in fir-woods, at about 4500 feet elevation, near Spliigen. Some specimens marked with dark, but I saw none of the var. musauaria. Cidaria variata two on the Merkur-Berg, near Baden. C. ferrugata Schaffhausen. C. ccesiata in company with L. populata, as in Scotland. G. flavicinctata one specimen in the hotel at Spliigen. C. verberata common in all the Alpine valleys we passed through. C. bilineata Baden, Schaffhausen, Chur, Liuno. Eupithecia euphrasiata one specimen on a hill behind Bellaggio. Rivula sericealis Bellaggio. Hypena obesalis a specimen near Andeer. H. obsitalis common in passages amongst vineyards, near Bellaggio. Botys purpuralis meadows at Bellaggio. B. cespitalis Baden. B. nubilalis (lupulinalis) one specimen at Bel- laggio. B. lutealis near the waterfall on the Spliigen. Eurycreon ver- ticalis Schaffhausen. Pionea forficalis Baden. Diasemia litterata This pretty little species was common at Bellaggio, flying gently in the sunshine just above the top of the grass, in grass and clover-covered meadows, in the neighbourhood of vineyards, &c. Crambus tristettus Baden, Bellaggio. C. culmellus Baden, Bellaggio. C.perlellus Chur, Bellaggio. C. geniculeus Bellaggio. Pempdia semirubella (carnella} common at Bellaggio and Menaggio, in places like those frequented by litterata. The var. sanguinella (with pale costa) also occurred. Myelois rosella one specimen of this pretty little species at Menaggio. Sericoris conchana Baden. Depressaria Tieydeni some pupSB picked up from moss under stones whilst searching for Coleoptera, on the Spliigen, near the top of the pass (at about 6500 feet), produced this species, for naming which, as well as the last, I am indebted to Mr. E. Meyrick, of Trinity College, Cambridge. W. A. FOBBES, West Wickham, Kent : January Vlth, 1877. jiv. 9 MELANISM IN LEPIDOPTERA.* p. Io(lo77). AFTER Mr. Burchell's and Dr. Buchanan White's notes on this subject (vol. xiii. pp. 130 and 145), and the very feasible explanation the theory of natural selection gives of the prevalence (though not the cause) * Ent. Month. Mag. xiv. p. 16. MELANISM IN LEPIDOPTERA. 33 of these dark varieties, I was somewhat surprised to see what may be called the " birthmark " theory revived to account for them by Mr. Fetherstonhaugh (p. 215), and subsequently supported by Mr. Tugwell (p. 256). It is almost impossible to one having any physiological know- ledge to see how any impression on the sensorium of the parent can produce any permanent change (except, perhaps, a deficiency in some parts) in the structure of its offspring. As, however, one fact is worth a hundred theories, I may perhaps be allowed to quote here a passage from Darwin's 'Animals and Plants under Domestication* (1st edit, vol. ii. p. 263), which seems to me to be decidedly " ad rem " as regards the subject under discussion. He says, " it was formerly a common be- lief, still held by some persons, that the imagination of the mother affects the child in the womb Dr. William Hunter, in the last century, told my father that during many years every woman in a large London lying-in hospital was asked before her confinement, whether anything had specially affected her mind, and the answer was written down ; and it so happened that in no one instance could a coincidence be detected between the woman's answer and any abnormal structure ; -but when she knew the nature of the structure, she frequently suggested some fresh cause ! " Natural selection perfectly explains the facts adduced by Mr. Tugwell about Gnophus obscuraria, for of course on a dark soil the darker indivi- duals, on the light the lighter ones, will be best protected by their colours and will therefore have a better chance of escaping the notice of their enemies. That the dark colour of the soil can hardly be the true cause in producing these variations is, I think, pretty certain, from their occur- rence in many places where the soil is not conspicuously dark, e. y. the Highlands of Scotland and the Alps *. I have just been looking through Dr. Staudinger's catalogue, and was much struck by the fact that in nearly every case where a local form (whether a " var. " or " ab. ") from the Alps is noticed, it is characterized as being " obscurior " or " multo obscurior," or with some of the markings " obsoleta." The great number of normally dark or black species of Lepidoptera in the Alps, as, for instance, the Erebice, Psodos, and some Pyralides (cf. Jordon, vol. xiii. p. 60), seems to me also to be worth notice in connection with this subject. In a few cases, Alpine insects are only sexually melanic, e. g. Pieris napi $ , var. bryonice, A. paphia $ , var. valezina, Polyommatus virgaurece $ t var. zermattensls. These cases are explicable on the theory that sup- posing sexual selection to have been such an efficient agent in modifying species as Mr. Darwin believes, it may have been more important for the males in the struggle for life to preserve their good looks than to have acquired sounder constitutions at the expense of the former. That the * Conversely, too, one would expect, if this theory were true, to find more melanic vars. on tbe very dark soil of peat-mosses and fen-lands than is actually the case. D 34 ON BIEDS COLLECTED BY THE 'CHALLENGER.' prime agent in this tendency to melanism is some unfavourable meteoro- logical element, probably connected with an excess of moisture and reduced amount of sunshine, is strongly suggested by the fact that, as Ent.M.M.xiv. noticed by Messrs. Birchall and White, these dark forms are most fre- p. 17 (1877). q uen t in the north of England, Scotland, Ireland, and, as I have stated, the Alps. Nor is this tendency confined to Lepidoptera. L. de Tschudi, in his ' Monde des Alpes ' (2nd edit. 1870, p. 394), says, " Les differences d'altitude produisent encore sur les insectes des modifications d'une nature particuliere. Une des choses qui frappent le plus celui qui visite nos Alpes, c'est 1'obscurcissementdes couleursdans les coleopteres alpins comrne en general dans une grand nombre d'insectes. Plus nous nous elevens plus nous voyons les scarabees qui vivent dans les trous, comme ceux qui habitent sur les plantes, dans les fumiers ou dans 1'eau, devenir unicolores. Ceux qui sont les plusrepandus dans les Alpes sont en generalnoirs ou d'un brun fonce ; et ceux qui dans les zones inf erieures sont ornes de couleurs a reflets metalliques deviennent dans les hauteurs d'un noir uniforme. Une foule des coleopteres verts et cuivres sont sur les hautes Alpes d'un noir pur, un petit nombre seulement d'un bleu d'acier, et d'un bleu fonce : ceux qui sont bruns, olivatres, et d'un vert dore, passent egale- ment au noir pur ou au noir bleuatre : merne la Chrysomda alpina jaune devient noir sur les Alpes." He then goes on to suggest as the probable cause of this, the fact that Alpine species live and undergo their meta- morphoses for a great part of the year under a thick bed of snow, and consequently in profound darkness. A similar darkening in the coloration of some of our English Coleoptera may be seen as we go northwards, e. g. the Highland dark forms of Carabus catenulatus and the mountain Calathus nubigena. It would be interesting to hear if similar cases occur in other orders. The most probable conclusion seems that darkness of coloration is in some mysterious way correlated with a constitution better fitted to encounter unfavourable conditions of life, more especially meteorological. p.z.s.1878, 10. REPORTS ON THE COLLECTIONS OF BIRDS MADE P . 120. DURING THE VOYAGE OF H.M.S. < CHALLENGER.' No. VII. ON THE BIRDS OF CAPE YORK AND THE NEIGHBOURING ISLANDS (RAINE, WED- NESDAY, AND BOOBY ISLANDS).* THE collection of Birds made by H.M.S. ' Challenger ' at Cape York and in its neighbourhood, of which the following is an account, comprises * Proc. Zool Soc. 1878, pp. 120-128. Read Feb. 5, 1878. ON BIRDS COLLECTED BY THE 'CHALLENGER.' 85 61 skins, referable to 37 species. As might have been expected, all, or nearly all, belong to well-known Australian forms, one or two only being left uncertain for want of more materials and on account of the immature condition of the specimens. Most of the skins are in excellent condition ; and their value is much increased by the notes in Mr. Murray's journal as to the colour of the soft parts &c. Besides Cape York, Raine Island (at the end of the Barrier Reef), Wednesday Island (in Torres Straits), and Booby Island (also in Torres Straits) were visited, and collections made. I copy the following extracts from Mr. Murray's journal as regards the localities where birds were obtained : " Raine Island, Barrier Beef, Australia. Ship landed two boats for nearly three hours. The following birds were taken" (several sea-birds, Rallus pectoralis, and Strepsilas interpret). " Cape York, Somerset. Ship arrived on evening of 1st Sept., 1874 ; left Cape York on Sept. 8th." (44 skins were obtained here.) " Wednesday Island, Torres Straits. Parties landed the same day (Sept. 8). Most of the birds seen were the same as those shot about p. z. S. 1878, Cape York. The following two, however, have not been procured at P- 1-1- Cape York ; they were shot on shore (Totanus incanus and ^gialites inornatus). A great many flocks of Bee-eaters were noticed making passages between the islands. " On 9th Sept., 1874, a party landed on Booby Island, a very small island, with only a few shrubs on it. The following land birds were shot or brought on board ; and in addition a Rail was seen, a Mec/apodius, and one other land bird." (The species obtained were Todirhamphus sanctus, Merops ornatus, Zosterops luteus, Pacliycephala sp., Ptilopus superbus, and Syncecus cervinus.) In the following list I have, with a few exceptions, followed the arrangement and nomenclature of Mr. Gould's * Handbook to the Birds of Australia.' 1. NINOX BOOBOOK (Lath.). Ninox boobook, Sharpe, Cat. of B. ii. p. 168, Spiloglaux marmoratus, Gld. Handb. B. Austr. i. p. 73. One male skin of this species from Cape York. No. 167. " Stomach contained insects " (Murray, MS.). The specimen sent agrees generally with two skins in Mr. Godman's collection from the same locality, and with Mr. Gould's description of Spiloylaux marmoratus, which Mr. Sharpe, in the second volume of his catalogue, treats as " the adult of the large form of JV. boobcok " (I. c. p. 170). 2. PODA.RGTTS PAPUENSIS, Quoy & Gaim. Podaryus papuensis, Quoy & Gaim. Voy. de 1'Astrol. Ois. t. 13 ; Gould, B. of Austr. Supp. pi. ; id. Handb. B. Austr. i. p. 91. 36 ON BIRDS COLLECTED BY THE * CHALLENGER.' " No. 186, ? . Eyes red, feet and bill light yellow. Stomach con- tained insects." Cape York, one specimen. 3. MEEOPS OENATTJS, Lath. Merops ornatus, Grid. B. of Austr. ii. pi. 16 ; id. Handb. i. p. 117. In all four specimens of this common Australian species. Two females from Booby Island (Nos. 199, 200), and a pair ( rf & ) from Cape York (nos. 147 and 178). Of all the eyes are stated to be " red," and the feet and legs black, except the female from Cape York, which is de- scribed as having the legs " with violet tint." The three females differ considerably from one another in the length of the produced middle tail- feathers ; in the male the produced part is thinner and at least twice as long as in the other sex. 4. TODIBHAMPHUS SANCTUS (Yig. & Horsf.). Todirhamphus sanctus, Gld. Handb. B. Austr. i. p. 128 ; Sharpe, Alced. pi. 91. One specimen from Booby Island. " No. 192, <$ . Eyes black ; mandibles black, except base of lower one, which is whitish. Stomach had remains of a crab." P ' Z 122 1878> 5 ' HALCTON MACLEATI, Jard. & Selby. Halcyon macleayi, Jard. & Selby, 111. Orn. vol. iii. pi. 101 ; Grid. B. of Austr. ii. pi. 24 ; Sharpe, Ale. pi. 78. Cyandlcyon macleayi, Gld. Handb. B. Austral, i. p. 133. One male specimen from Cape York. " No. 161. Eyes, bill, and legs black. Stomach contained insects." 6. AETAMTJS LEUCOPYGIALIS, Gld. Artamus leucopygialis, Gld. B. of Austr. ii. pi. 33 ; id. Handb. i. p. 154. One specimen from Cape York. "No. 177, rf. Bill blue tipped with black, feet black. Stomach contained insects." 7. GEATJCALUS HTPOLEUCUS, Gld. Graucalus hypoleucus, Gld. B. of Austr. ii. pi. 57 ; id. Handb. i. p. 196. One skin of this species from Cape York. " No. 151, $ . Legs, bill, and eyes black. Stomach contained insects." 8. PACHTCEPHALA, sp. inc. " No. 196, c? . Pachycephala. Eyes brown, bill and legs horn-colour. Stomach had insects." A single skin of a Pachycephala from Booby Island, immature, still retaining some of the rufous feathers characteristic of immaturity in its ON BIRDS COLLECTED BY THE 'CHALLENGER.' 37 wings. It is certainly one of the species which, when adult, are yellow beneath, the under tail-coverts being bright gamboge-yellow. The species of this group are so hard to determine, unless males and in adult plu- mage, that it seems better to leave this bird, although not exactly like any PachycepJiala I have been able to examine in the British Museum or in Mr. Godman's collection, without a name for the present, more particularly as Mr. Sharpe is, I believe, now working at this group for the next volume of his catalogue. 9. PlNAROLESTES RUFIGASTEB (Gld.), Colluricincla rujiyaster, Gld. B. of Austr. i. p. xxxvii ; id. Handb. B. Austr. i. p. 226. Pinarolestes mifigaster, Sharpe, Cat. of B. iii. p. 296. No. 170, 2 Shrike." One specimen, from Cape York, of this difficult genus. Mr. Gould, to whom I showed the specimen, was inclined to identify it with his Colluricincla parvissima* (Ann. & Mag. N. H. ser. 4, x. p. 114), and has kindly lent me the type of that species (which Mr. Sharpe, t. c. p. 297, treats as the young of P. rujlgaster) for comparison. I find, however, that the Cape-York bird has a distinctly shorter wing, a stronger and more arched bill, and is less rufous on the underside. Mr. Sharpe, who has recently worked up this group for his catalogue, has examined this specimen, and refers it to P. rujigaster. 10. MANUCODIA GOULDI, Gray. P. Z. S. 1878, Manucodia gouldi, Gld. Handb. B. Austr. i. p. 236 ; Sharpe, Cat. of B. P ' iii. p. 181. Manucodia Iceraudreni, Gld. B. Austr. Suppl. pi. 9. Two female specimens. " No. 152. Bower-bird. Eyes orange, bill and legs black. Stomach contained small seeds. For curious loop in the wind-pipe see body. This bird was shot on the island opposite Somerset by Moseley." The colours of the soft parts and contents of the stomach are the same in the other specimen. The two skins sent agree well with Mr. Sharpe's description and differ from Mr. Elliot's figure of the New-Guinea species (M. Jceraudreni) in the points noticed in Gray's original description of the species. The curious conformation of the trachea in M. Tceraudreni has long been known, having been originally described by Lesson. Beccari (Ibis, 1876, p. 252) says he thinks it probable that this is a peculiarity of the male sex. Mr. Murray's notes, however, would seem to contradict this vie\v, unless, as is hardly likely, two species otherwise so nearly allied should differ in this peculiarity. * Count Salvador! suggests that this really = P. megarhynchus of New Guinea. 38 ON BIRDS COLLECTED BY THE 'CHALLENGER/ 11. PTILOEHIS ALBERTI, Elliot. Ptiloris alberti, Ell. P. Z. S. 1871, p. 583 ; id. Mon. Farad, pi. xxiv. ; Sharpe, Cat. of B. iii. p. 156. Ptiloris magnificus, Gould, B. Austr. Suppl. pi. 51. Craspedophora magnified, id. Handb. i. p. 595. Six skins of this Rifle-bird from Cape York, four of which are adult males in full plumage, and one a female. Another, marked $ , is either a young male corning into fall plumage, or more probably an aged female assuming male plumage, as some of the feathers on the top of the head show traces of a metallic blue colour, as also does the shoulder, whilst more or fewer of the primaries on each side have acquired the black colour characteristic of the male. Of Nos. 164 and 165 it is noted in Mr. Murray's journal : " Males : eyes, bill, and legs black, soles of the feet yellow. The stomach contained a red fruit, with a large stone about the size of a pea. Some parts of insects in 165." Of No. 184, " Female : eyes black ; stomach contained ants and grubs." Besides the specimens sent, several others seem to have been procured. An interesting account of the shooting of these specimens will be found in Lord George Camp- bell's 'Log Letters from the ' Challenger,' ' p. 185. 12. MIMETA VIEIDIS (Lath.). Mimeta viridis, Gld. Handb. B. Austr. i. p. 462. Oriolus viridis, G-ld. B. Austr. iv. pi. 13; Sharpe, Cat. of B. iii. p. 212. A young male from Cape York, agreeing generally with Mr. Sharpe's description (I. c.) of the young bird. " No. 169, rf . Thrush." 13. MIMETA FLAVOCI^CTUS (King). Mimeta flavocincta, Gld. Handb. B. Austr. i. p. 466. P. Z. S. 1878, Oriolus flavocinctus, Gld. B. of Austr. iv. pi. 14. p. 124. Oriolus flavicinctus, Sharpe, Cat. of B. iii. p. 206. An adult male from Cape York, agreeing with skins in Mr. Godman's collection, collected by Cockerell. " No. 189, <$ . Shrike. Eyes red, bill red, feet bluish." 14. SPHECOTHEEES FLAYIYEJSTTBIS, Gld. Sphecotheres flaviventris, Gld. B. of Austr. Suppl. pi. 37 ; id. Handb. B. Austr. i. p. 468 ; Sharpe, Cat. of B. iii. p. 225. Three specimens from Cape York, of which two are males in full plumage, agreeing with specimens in Mr. Godman's collection. " Nos. 172, 173. Eyes black, feet flesh, bill black." The third specimen is marked female (" No. 174, $> . Eyes black, bill horn, legs brownish. Stomach contained insects and seeds "), but, as there are traces of yellow on the breast, is more probably a young male assuming the adult ON BIRDS COLLECTED BY THE 'CHALLENGER.' 39 plumage. The orbits in this bird (in the skin) are dark brownish black, not yellow, though this is probably accidental. In other points, too, particularly in the plainly striated under tail-coverts, and the colouring of the upper surface of the head, the specimen hardly agrees with Mr. Gould's description of the female S.flaviventris. Mr. G ould, to whom I have shown it, would give no decided opinion on it ; but Mr. Sharpe has examined it, and pronounces it to be of this species. 15. PTILOTIS CHKY.SOTIS, Lewin. Ptilotis chrysotis, Gld. B. of Austr. iv. pi. 32. Ptilotis lewinii, Swains. ; Gld. Handb. B. Austr. i. p. 503. Three skins of this common Australian species, which is not recorded from N. Australia in the 4 Handbook ' two from Cape York, one from Cape-York Island. "No. ]50, $. Eyes brown, legs slate, bill black. Stomach contained insects." "No. 157, Dut quite impermeable. The aorta p. 430. gives off an innominate artery, which is only an inch long and then divides into right brachial and right and left carotids. The left brachial is given off immediately after the innominate. This agrees with the descriptions of E. indicus as given by Hunter, Owen, Vulpian and Phili- peaux, Watson, and Miall and Greenwood. On the other hand, Cuvier and Mayer found three trunks, namely two brachials and a common carotid. I found no " arteria thyroidea inferior simplex " coming off from the point of division of the two carotids, such as is figured by Mayer (I. c. pi. 11. fig. 3) and Watson (Journ. Anat. & Phys. vi. pi. vi. fig. 1). The weight of the heart and great vessels, cut short and cleaned of blood, was 7 lb. There was no os cordis ; and the same was the case in Perrault's specimen ; nor is any such bone recorded in E. indicus by recent anatomists. Respiratory System. The lungs were very simple in form, each lung being undivided and bluntly triangular in general outline, the Jett being shorter and broader. In the undistended state they measured as follows : Eight lung 23 inches long by 12 broad, left 21 inches by 14. I found no accessory lobe on the right side, such as has been observed by some anatomists in E. indicus. There is no extra bronchus. The trachea is short, measuring about a foot in length, and not quite two inches in external diameter. It is composed of 28 rings, which are nearly complete, leaving hardly any space behind between their ends. * Perrault gives 3 feet by 7 inches. ON THE ANATOMY OF THE AFEICAN ELEPHANT. 57 They vary considerably in size in different parts of their circumference. The first three rings, as in E. indicus, are truncated obliquely behind, the space so formed being covered in by the body of the cricoid cartilage. The larynx (fig. 6) is of considerable size. The epiglottis, when covered by its soft parts, is short, thick, and rounded. The thyroid consists of two rhoinboidal wings, 4 inches long, and 3| deep, which are united in front superiorly for about one inch, the deep and narrow notch left between the remaining part of the wings being filled up by connec- tive tissue. The superior cornua are short and scarcely project. The posterior are about one inch long, and are directed downwards and for- wards in close proximity with the body of the thyroid cartilage, to which they are attached by connective tissue. The postero-inferior angle of the thyroid cartilage also develops an articular facet ; and this is enclosed with that of the posterior cornu, in the common capsule of the crico- thyroid articulation. The cricoid (see fig. 6) is of the usual type. Its anterior part is 1 inch deep, the posterior (somewhat pentagonal) part 2 inches. The processes for articulation with the thyroid stand out in a step -like way, and are more or less clearly divided into two facets, corresponding to the double articulating surfaces of the thyroid. The arytaenoids (see fig. 6) are vertically elongated. Each measures P. Z. S. 1879, p. 431. Larynx of African Elephant (about half nat. size) viewed somewhat obliquely from behind. The thyroid cartilage has been removed, a, points to the double facet of the crico-thyroid ariculation. about 2f inches long by 1| broad. They have a conspicuous, vertically p - z - s - 1879, directed, raised spine-like process, and a large notch behind the supero- posterior angle. The cartilage of each side articulates with its fellow both above and below this notch. The processus vocalis is short and 58 ON THE ANATOMY OF THE AFRICAN ELEPHANT. P. Z. S. 1879, blunt. The true vocal cords are well-marked and thick elastic folds, 2| p. 431. inches long. The false vocal cords hardly exist. Between the two is a slight laryngeal pouch, which extends backwards a little way, as in the Indian Elephant (Miall and Greenwood, 1. c. p. 76). The muscles of the larynx closely agree with those described by the last-named anatomists. The superior fibres of the crico-arytcenoideus posticus run transversely across in the interval left above by the more inferior, diverging fibres of that muscle. Urino-genital System. The kidneys lie in the usual position. Their shape is an irregular oval. The following details refer to the single kidney (right) which I preserved for further examination. The length is 10 inches, the breadth about 6. The hilus is not marginal, but lies about 1 inch from the side ; its length is 4| inches. The weight of the kidney is 3 Ib. The kidney is indistinctly divided into eight lobes, which are of varying size and shape ; one lobe is scarcely visible on the hilar surface. These lobes are essentially distinct, each consisting of a cortical and medullary part, not, however, very clearly marked off from each other. The Malpighian corpuscles are clearly visible. Perrault's figure of the kidney (I. c. pi. 20) is too elongated and shows no lobes. The number of lobes in the kidney of E. indicus has been variously stated at from two to eight or nine. The suprarenal bodies resemble those of the Indian species. The ureters open into the bladder by semilunar slits about 2 inches from its orifice. The neck of the bladder is short and thick. The female organs are formed on precisely the same type as those of the Indian species*, consisting of a long urino-genital passage (" the common vagina, which is common to the urine and penis " of Hunter), P. Z. S. 1879 a secondary vagina (" the proper, or rather uncommon, vagina, which p. 432. the penis cannot enter "), a corpus uteri, with two horns, and Fallopian tubes and ovaries. The ovaries lie in pouches of peritoneum, attached by peritoneal folds to the kidneys : the one I examined resembled in form those figured by Mayer in the Indian species. It was a little over an inch long, and generally smooth, with only a few small lobular pro- cesses and erupted Grraafian follicles near the line of attachment to the peritoneal pouch. The latter is continuous with the opening of the Fallopian tube, and is of considerable size : its walls are thickened by muscular fibres, prolonged into it apparently from the Fallopian tubes. The tubes are of small calibre, of the size of a crow-quill, about 3 or 4 inches long, and, after a tortuous course, open into the cornua uteri at the side of that tube, as well shown in Mayer's figure (1. c. pi. vi. fig. 2). The two cornua are about \ inch across at their commencement, and * Cf. Hunter, ' Observations,' &c. ii. p. 175 ; Mayer, /. c. p. 37, t, vi. ; Owen, Anat. Vert. iii. p. 692; Miall and Greenwood, I. c. p. 62, pi. iv. ON THE ANATOMY OF THE AFRICAN ELEPHANT. Fig. 7. 59 Corn P. Z. S. 1879, p. 433. . Uterus and vagina of African Elephant (about half natural size), viewed from behind. b The vagina (To^.) and urino- genital canal (u.g.} have been laid open from behind. Carn.ut. Cornua uteri cut short above. Ut. True uterus, formed by the coalescence of the two cornua, but not marked off externally from the conjoined cornua by any constriction, o.u. Above this is the valve-like structure corre- sponding to the Os uteri. Ur. Prominence on which the urethra opens ; above it are seen the Malpighian canals ; below the letters is the papilla-like free point (vide fig. 8). Ves. Bladder. b. Section of the conjoined uterine cornua, half the natural size, to show the distinctness of the two tubes internally at this point. 60 ON THE ANATOMY OF THE AFRICAN ELEPHANT. have very thick muscular and elastic walls. For the last 4| inches of the course of the cornua they are united together (as seen in fig. 7, p. 59) into a single tube, which is about 1 inch across at the point of junction. This tube is externally single ; but nevertheless, on cutting it across, the two comparatively small cavities of the cornua are seen lying beside one another, but separated by a considerable septum. Without any difference in the external calibre of the tube, the two cornua open together into a common cavity 2| inches long, w r hich is the true " corpus uteri." At their opening each cornu admits a large knitting-needle. There is no valve of any kind at the opening. Both cornua and corpus are lined by smooth, longitudinally plaited, mucous membrane. A similar arrangement to that here described would seem to be indicated by Perrault's description : " Ces comes, au lieu de s'ecarter et de se separer comme elles font ordi- nairement, etoient jointes 1'une centre 1'autre, montant jusqu'au hauteur d'un pied, et n'etant separes que par une cloison mitoyenne ; ensuite elles se separent en deux branches." In his example (nearly or quite adult) each horn measured 2 feet 8 inches, and was 1| inch across at the commencement. The female genital organs he pictures on pi. 21 : this shows the conjoined cornua, which are separate till near their end, as seen in section. The next part of the genital organs is the dilated, sac-like, " secondary " or " uncommon," vagina. This is about 5| inches long, and is lined by smooth mucous membrane, with slightly raised longitudinal folds, run- ning from the opening into it of the corpus uteri. This opening is small, only admitting the tip of the little finger, and is provided behind with an irregularly bilobed thick valve of mucous membrane. This constriction and valve undoubtedly represent the " os uteri." Perrault describes this " secondary vagina " as the " corps ovale ;" in his specimen it measured 18 inches by 6 inches, and was smooth and polished within. It is well shown in his figure (1. c. pi. 21) ; but the " valvule frangee aux embou- chures des cornes de la matrice " is not quite like the valve in my specimen. In the text he says, "Deux trous au dedans... etoient entoures par un appendice de la membrane interne... en maniere de la f range ou de pavilion." It would appear, then, that in his animal there was no " corpus T> v a IQ^O uteri," such as that which exists in mine, but that the two cornua r. L. o. lo/y, p. 434. opened separately into the "corps ovale " ( = secondary vagina). Mayer apparently (I. c. pi. 6, p. 38) found a similar arrangement in E. indicus. Hunter, Owen, and Miall and Greenwood all indicate an arrangement like that which obtained in mine*. * In a specimen (2776 A) in the College of Surgeons of the uterus &c. of E. indic-us, the " corpus uteri " is very much more capacious than in my (young) specimen, is about 7 inches long, and is only separated off from the " secondary vagina " by a pro- minent zonary fold of mucous membrane. The calibres of these two chambers are about the same. ON THE ANATOMY OP THE AFRICAN ELEPHANT. 61 Fig. 8. Opening of urethra (U) into the urino-genital canal, about natural size (somewhat diagrammatic). The walls of the urino-genital canal are cut close round the urethral eminence. M.C. Malpighian canals ; below (anterior to) the letters is seen the constriction separating the vagina from the urino-genital canal ; on the top of the urethral eminence is seen the small free point ; below it is the cul-de-sac of the urino-genital canal. N.B. In the natural position the lower parts of the figure are anterior, the upper parts posterior. The secondary vagina, which lies behind the neck of the bladder, is separated by a constriction, leaving only a very small opening, from the urino-genital chamber, which is marked off by the livid blue colour of its mucous membrane from the parts already described. On each side of this median constriction lies a small obliquely-placed slit, about | inch long, and admitting a probe for about the same distance into the small sacs (canals of Malpighi), of which they are the openings. Exactly the same arrangement occurs in the Indian Elephant. There is no trace of p. z. S. 1879. any hymen-like organ dividing this median constriction into two, such P- 435 - as noticed by Miall and Greenwood (I. c. pi. iv. fig. 3, h). This point about corresponds with the entrance of the genital organs into the pelvis. Perrault describes and figures (pi. 22) in his example two " valves sigmoides,'*' which guarded the " orifice interne de la matrice," and also a " rebord qui s'avan9oit au-devant du col de la matrice de la longueur d'environ deux pouces." "What the two sigmoid valves are I do not see, as in his figure he indicates the two Malpighian canals as well. The " rebord " probably corresponds to the tumid rounded eminence (fig. 8, S'iprd) about one inch long, terminating above and behind in a little point, 62 ON THE GENUS LATH AMUS. on which the urethra opens by a somewhat narrow aperture, just below and in front of the opening into the secondary vagina*. In front of this eminence the urine-genital canal, as the remaining part of these organs may be called, is produced into a small cul-de-sac. The total length of this canal is about 20 inchest ; the clitoris, which resembles the same organ in E. indicus, and which has similar relations to the urino- genital canal, is about 15 inches from the attachment of its crura to the pelvis to its extremity. The glans clitoridis is about two inches long, rounded anteriorly, flattened and grooved posteriorly, where it is in contact with the urino-genital canal. There is a well-marked preputial-like reversion of the integuments round the glans, as in E. indic.us. The brain was removed with but little injury ; but its description must be deferred till some future occasion. As will be seen from the foregoing account, but little difference, on the whole, exists in the visceral anatomy of the only two remaiDing species of Proboscideans. What differences there are chiefly relate to the stomach, liver, and female organs ; but, till more specimens of E. africanus have been dissected, it is impossible to say- how many of the points above noticed are due to individual peculiarities or those of age and the like. There appears, therefore, little ground, from an anatomical point of view, to separate Loxodon as a genus from EuelepJias. P.z.S. 1879, 15. ON THE SYSTEMATIC POSITION OP THE GENUS P- 166. LATHAMUS OP LESSONJ. (Plate I.) IN their paper on Australian birds in the Linnean Society's Trans- actions for 1828 (vol. xv. p. 74), Messrs. Vigors and Horsfield established a genus Nanodes, of which the Psittacus discolor of Shaw was made the type, and full generic characters were given. Besides Nanodes discolor, three other species (those now generally known as Melopsittacus undulatus, EupTiema pulcJiella. and Platycercus venustus) were included in the genus, which was considered by its authors to be allied to Pezoporus * This raised part, on which is the opening of the urethra, is probably identical with the " Klappe" figured by Mayer (1. c. pi. vi. fig. 1) as existing between the two orifices of the bladder and vagina. t In Perrault's adult example the length was 3 feet 6 inches. \ Proc. Zool. Soc. 1879, pp. 166-174, PI. XVI. Eead Feb. 18, 1879. White's Voyage, pi. 263 (1790). For the synonymy of the species, see Finsch, Papag. ii. p. 863. ON THE GENUS LATHAMUS. 63 and Platycercus, and as connecting these Australian forms with the South- American Psittacarce ( = Conurus auct.). Nanodes having been already used by Schonherr for a genus of Rhynchophorous Coleoptera*, Lessont substituted for this name that of Lathamus, including under that head four other species (one a Euphema, one a Cyanorhamphus, and two Tri- choglossi, as now understood), remarking that Swainson " a parfaitement etabli ses caracteres " in his ' Zoological Illustrations ' J, where, however, E. pulchella is considered the type of the genus . As will be seen from the species associated with it, all these authors were evidently puzzled P. Z. S. 1879, by the characters of this peculiar little Parrakeet; and the same seems P* ' to have been the case with all subsequent naturalists who have treated of it. The majority, however, seem to have considered that it had Tricho- glossine affinities. Thus Bonaparte || included Lathamus as " dernier des Trichoglossiens;" and Grould, likewise acknowledging the validity of the genus, places it amongst the Trichoglossidae. He says : " Having had ample oppor- tunities of observing the bird in a state of nature, I concur in the pro- priety of separating it into a distinct genus ; in its whole economy it is most closely allied to the Trichoglossi, and in no degree related to the EuphemcB " (Handb. B. Austr. ii. p. 89). Dr. Finsch, in his great work on Parrots^, after a careful examination of its peculiarities, came to the conclusion that these were not sufficient to justify its separation as a distinct genus, and included it as a Trichoylossus. More lately, the same position (*. e. that of a member of the family Trichoglossidse) has been assigned to it by Gray**, Sclatertt, Wallace^, and others. On the other hand, Sundevall in his ' Tentamen ' placed it in his fourth family " Platycercini" remarking, " Haec species, plerumque cum sp. Tricho- glossinis (Ps. concinno &c.) consociata, vera tamen est species Platycercina, maxilla inferiori tumida, &c., Euphemce maxime affinis." In his paper on the anatomy of the Parrots, Prof. Garrod|| || shows that Lathamus differs from Lorius and its allies in having a superficial left carotid, a feature common to it and Platycercus, Psephotus, &c., from which, however, * Schonh. Curcul. Disp. Metb. p. 322 (1826). t Traite d'Orn. p. 205 (1831). J 2nd series, vol. i. part 5, no. 21 (1829). Swainson, however, in his ' Classification of Birds ' (vol. ii. p. 304, 1837), makes Lathamus a member of bis " subfamily Platycercinae," in wbicb he also includes Coracopsis, Pezoporus, Platycercus, and Calopsitta, with the remark that it is a " sub- typical" form. || Comptes Eend. xliv. p. 536 (1857). f Pap. ii. p. 863 (1868). ** Trichoglossus, c. Nanodes, gen. no. 2047, Hand-1. B. ii. p. 156 (1870). tt List Vert. 6th ed. p. 269 (1877). J$ Geogr. Distrib. Animals, ii. p. 327. Methodi Naturalis Avium disponendarum Tentamen, p. 71 (1872). HI! P. Z. S. 1874, p. 586. 64 01V THE GENUS LATHAMUS. it differs in the possession of a furcula *. He further says : " It may at first seem very heretical to remove Laihamus from the Loriinee, the brush-tongue being considered characteristic of that subfamily. To the unbiased student, however, the brush-tongue is a character not more im- portant than several of those that have been above considered The character of the papillae is somewhat different in Lathamus from what it is in Lorius, they being blunter and shorter in the former genus than in the latter." Having undertaken at Prof. Grarrod's suggestion an investigation of the pterylosis of the Parrots, the results of which I hope to communicate to this Society at no distant date, Lathamus was one oc the first forms I examined ; and I at once saw that its pterylosis confirmed the relationship of this form to the Platycercinae already insisted on by Sundevall and Garrod. From this I was led to an examination of some other parts of P. Z. S. 1879, its structure ; and I propose to lay the results of my inquiries before the p. 168. Society to-night, in order to establish the view that Lathamus must be removed from the brush-tongued Trichoglossinse, with which it has been so generally associated, and must be considered a (no doubt aberrant) member of the Platycercine group. The pterylosis of this form having first struck my attention, I will describe this in the first instance, the more so as, as far as I know, no description of this part of the structure of the bird in question has yet been published. I may perhaps anticipate part of my paper on the pterylosis of the Psittaci in general, and point out briefly the general characters of the distribution of the feathering in these birds, so as to enable the reader without any further trouble to appreciate the points of distinction in this respect between Lathamus and the other species with which I have compared it. As will be evident from the figures (Plate I. figs. 1-6), the tracts of contour-feathers in a Parrot may be arranged as follows : On the upper surface of the body, continuous in front with the feathering of the top and sides of the head, is a long narrow tract, the " superior tract," which divides behind in the interscapular region in a fork-like manner, forming the " scapular fork." Behind this, occupying the hinder part of the back and pelvis, is another, more or less Y-shaped tract, with the " handle " (which is usually short) of the fork placed close to the posterior extremity of the trunk, whilst the more lengthy " arms " of the Y are more anterior and run in, in front, between the corresponding ones of the " scapular fork/' usually becoming very feebly feathered in so doing. This tract may be called the " dorso-lumbar " fork. Scattered more irregularly and * M. Blanchard, indeed, says (Compt. Rend. 1857, xliv. p. 521) that Lathamus has no furcula ; but this bone is present, though small and weak, in the specimens I have seen : cf. also Owen, Cat. Ost. Ser. E. C. S. i. p. 279 (1853). ON THE GENUS LATHAMUS. 65 diffusely over the sides of the pelvis, and external to the last-named tract, is the " lumbar feathering," which passes posteriorly on each side into the narrower but more distinct " femoral tracts." These are con- tinued onto the legs as far as the tarsi as the " crural tracts," clothing the legs in a trouser-like way. On the inferior surface, on each side, is a continuous tract, running from the upper part of the neck (where it may or may not unite with its fellow of the opposite side), over the breast and abdomen, to the anus. This " inferior tract," besides one or two small branches running towards the humerus and patagium (the first and second " humeral tracts "), gives off, at about the commencement of the sternum, a more or less separate and well-marked external branch, the " outer pectoral " tract, which runs down more or less parallel to the main part of the inferior tract for a little way, but ceases before the thighs. Amongst the various species of Psittaci I have examined, well marked differences in some of these tracts occur, more particularly in the ar- rangement of the " dorso-lumbar fork," and the greater or lesser develop- ment of a distinct " outer pectoral " branch to the inferior tract. In Lathamus discolor (PI. I. figs. 1, 2) the inferior tract of each side starts from about the angle of the jaw, and does not unite with its fellow. On the sternum it is about eight or nine feathers broad at the widest part, the feathering being rather strong and not close. As in P. Z. S. 1879, most Parrots, there are two humeral tracts. The space on the carina P* " sterui between the inferior tracts of the two sides is not wide. There is a well-marked outer pectoral tract, about 1 inch long, distinguished by its rather stronger and closer feathering. It is quite separate from the main part of the inferior tract, the space between the two tracts being about as broad as the latter tract itself. The outer pectoral has the appearance of being somewhat dilated at its free end, owing to the presence of a few irregularly placed and small feathers lying to the outside of its termination. The main part of the inferior tract is rather narrow, with its rows of four and five feathers each separated by rather consider- able spaces. The scapular fork is rather long, the tracts being narrow and moderately strongly feathered. The dorso-lumbar fork is elongated ; each arm is of nearly the same length and breadth throughout, beginning a little outside the scapular fork, with the part inside the arms of the latter represented only (as usual in the Psittaci) by one or two rows of small feathers, placed singly or in pairs. Each arm is composed of about fourteen rows of feathers (counting to the junction with its fellow), the rows being four feathers wide, rather close together, and of about the same width as the space between the tracts. There is some tendency in some of the anterior rows towards a dilatation of the tract, one or two of the rows being five . 66 ON THE GENUS LATHAMUS. feathers wide. In the more anterior parts of each arm the most internal feather of each row is often placed in front of and at an angle with the other feathers composing it, and so comes to stand between two rows of three feathers each ; so that at first each tract looks as if made up of rows of three (or four) feathers alternating with single feathers. This tendency to a 3.1.3 arrangement, however, disappears in the more pos- terior parts of the tracts, the four feathers of each row there standing in a direct line with one another. The two arms unite to form the " handle " at about three quarters their entire length ; after the junction the tract narrows rather rapidly towards the tail. The dorso-lumbar fork is throughout quite distinct from the lumbar feathering, which is very weak and diffuse. In all the truly Platycercine* forms that I have examined namely Platycercus eocimius and pennantii, Psephotus hcematogaster (four specimens) and P. hcematonotus, Pyrrhulopsis splendens and P. personata, Cyano- rhamphus auriceps and 0. novce-zealandice the disposition of the outer pectoral tract and dorso-lumbar fork resembles essentially that of Lathamus. In all the outer pectoral is a distinct, more closely feathered, and rather narrowish tract, clearly separated throughout from the main part. In Cyanorhamphus this tract is distinctly hook-like, dilated at the end. In all the same lengtht, and uniformity in strength and width, of P. Z. S. 1879, the arms of the dorso-lumbar tract is observable, the inclosed space being p. 170. O f about the same width as either of the tracts inclosing it, no tendency to a dilatation of the arms at their junction (though there is some in front) being present, and the rows of feathers in front having a more or less clear 3.1.3 arrangement. The lumbar feathering is always very weak ; so that the boundaries of the dorso-lumbar fork are very clearly defined. Lathamus, however, differs from the above-mentioned forms a little by its longer and not so widely divaricated scapular fork, and by the greater breadth of its inferior tract on the sternum, thereby causing a correspond- ing diminution in the breadth of the carinal space. The general agree- ment, however, of the pterylosis in the two types will, I think, at once be evident from the figure of Lathamus (PI. I. figs. 1, 2), and that of Platycercus pennantii (PL I. figs. 3, 4), which I have represented next to it for the sake of comparison. If now we turn to the TrichoglossinseJ (see PI. I. figs. 5, 6), in which so many naturalists have included Lathamus, we shall find im- portant and well-marked differences in the two tracts mentioned above, * I. e. excluding Aprosmictus, Potyteles, Euphema, Pezoporus, &c. t In Pe. pennanti and in the two species of Pyrrhulopsis I counted fourteen, in C. auriceps thirteen, in Ps. hcematonotus thirteen, and in Ps. hcematogaster eleven rows of feathers in the arms of this tract to their junction. | Of these I have examined the pterylosis in Eos rvhra, Trichoglossi ornatus, hesma- todes, swainsoni, concinnus (two specimens), and pusillus, and Coriphilus fringilloceus. ON THE GENUS LATHAMUS. 67 though the general character of the pterylosis remains the same in all*. The outer pectoral tract is never so narrow and distinct here as it is in Lathamv.s and its allies ; it is usually almost triangular in shape, and so tolerably broad, shorter, and not so divergent, the interspace between it and the main tract being much narrower, and frequently with a few scattered feathers in it uniting the two tracts together. The inferior tract on the breast is always much broader, and the carinal space narrower. Still better-marked characters between the two groups are to be seen in the disposition of the dorso-lumbar fork. This in all the TrichoglossinaB is extremely weak in front, the tracts not getting at all strongly feathered till some way (in T. cincinnus j inch) from the ends of the scapular fork. Each arm is much shorter (in all the forms I count about eight rows of feathers to the junction), wider and more diffusely feathered than in the Platycercinae, and becomes dilated and more strongly feathered towards its junction with its fellow, which takes place further from the tail than in the other group. The united tract is strongly feathered and rather broad at first, but narrows rapidly again towards the tail. Figs. 5 and 6, PL I. represent the pterylosis of Trichoglossus concinnus (a bird a little larger than the " Swift Parrakeet "), and show the dif- ferences between the two groups, which, if somewhat slight, are neverthe- less easily appreciable after a little study, and are as well marked as any others I have as yet observed in the pterylosis of this order. Several points in the external characters of Lathamus show that it has P. Z. S. 1879, in fact no particular relationship to the TrichoglossinaB. The shape of P- ^ 2 - the upper mandible, with a small but distinct tooth, is obviously (see fig. 1, p. 68) much nearer to that of Psepnotus (fig. 3) than it is to that of a Lory (fig. 5). The same story is told still more plainly by its maxilla, which has none of the laterally compressed, elongate, and pointed form characteristic of the Lories, and which induced Sundevall to divide all Parrots into two groups " Psittaci proprii " and " Psittaci orthognathi," the latter including only the Lories and Nestor, and cha- racterized by having the "maxilla inferior recta, angusta, altitudine longior." In Lathamus the maxilla is short and deep, with a broad and rounded anterior margin. These differences will be seen by a glance at figures 5 and 1, representing the heads of a Trichoglossus (concinnus) and of Lathamus. In all the TrichoglossinaB I have examined, the cere is rather narrow from before backwards, the anterior margin only sinuate, and the nostrils elongated and ovate, with their long axis directed forwards and inwards, and so somewhat transversely to the direction of the beak (fig. 5, p. 68). This is very evident in the living birds, and is also to be made out in * I have as yet been unable to confirm Nitzsch's observation (Pterylogr., Eng. edit, p. 100) that in Lorius garrulus and L. domicella the inferior tracts are continuous over the lower surface of the neck. F2 68 ON THE GENUS LATHAMUS. P. Z. S. 1879, p. 171. Fig. 1. Head of Lathamus discolor. Fig. 2. Foot of ditto. Fig. 3. Head of Psephotus hcematogaster. Fig. 4. Foot of ditto. Fig. 5. Head of Trichoglossus concinnus. Fig. 6. Foot of ditto. P. Z. S. 1879, skins. p. 172. In Lathamus, however, and the Platycercinae generally, the cere g jnuc larger, with the anterior border on each side mearly semicircular ; and the nostrils are oval and directed upwards, more nearly parallel with the culmen (see figs. 1 and 3). In the small size of the nude orbital ring Lathamus agrees with the PlatycercinsB rather than with the Lories, in which it is of fair size and rather conspicuous in the living birds. In the shape of the wings, no doubt, Lathamus is somewhat aberrant, and nearer the Lories than the Platycerci. This is, however, so obviously an adaptive modification, due to the swift flight and arboreal habits of ON THE GENUS LATHAMUS. 69 both these birds as compared with the more ground-loving mode of life of the Platycerci, that no stress can be laid on it as a taxonomic character. The rounded end of the wing-feathers, however, of Lathamus still point to its Platycercine affinities. Its feet, too, though not typically Platy- cercine, differ from those of the TrichoglossinsD (cf. figs. 2 and 6, p. 68) by their more elongated and slender tarsi and toes, with the latter not so much flattened and fitted for grasping branches, &c., as are those of the Lories, and with the claws not so strong and longer, particularly that on the third digit. In both these points more resemblance to the Platycerci is shown (cf. fig. 4, p. 68, foot of Psephotus Ticematog aster), though the different modes of life * have here again induced a certain amount of change from the form observed in the truly terrestrial Platycerci. A thorough study of the osteology of the Parrots has yet to be made ; and till that is done it is perhaps somewhat premature to generalize. P. Z. S. 1879, Nevertheless, having examined somewhat carefully a considerable number P* ^' of the skeletons of the two groups with which Lathamus has been generally associated, I have, I believe, been able to detect certain dif- ferences which will help us in referring the bird at present under dis- cussion to its proper place. First, as regards the skull. This, in all the Trichoglossinse, is remark- able for its somewhat depressed form and the lateral compression and elongation of the upper and lower jaws, the mandible when deprived of its horny sheath showing even more clearly the peculiar shape of the* lower jaw in these birds, first pointed out by Sundevall and already alluded to above (PI. I. fig. 7). In the Platycerci the skull is less depressed above and much shorter in proportion, and the mandible is nbt pointed, but has its symphysial portion wide, deep from above down- wards, and somewhat truncated. The same is the case in Lathamus (PI. I. fig. 8). In the Lories the lengthening of the beak has led to a similar elonga- tion in the anterior limb of the palatine bones, so that this part is as long as, or longer than, the posterior one; and the latter is considerably shorter than the pterygoids. In the Platycerci the anterior part of the palatines is not so elongated ; but, on the contrary, the posterior limb is somewhat lengthened, and, in fact, nearly as long as the pterygoids, Here, again, Lathamus agrees more with the Platycerci. In the Lories (PL I. fig. 9, Eos rubra) the anteorbital processes are much larger and better-developed than in the Platycerci, where the * Mr. Gould says (' Handb. B. Austr.' ii. p. 89) : " In its actions and manners it is closely allied to the Trichoglossi, but diners from them in some few particulars, which are more perceptible in captivity than in a state of nature. It has neither the musky smell nor the jumping motions of the Trichoglossi. I have never observed it alight ou the ground, or elsewhere than among the branches." 70 ON THE GENUS LATHAMUS. hinder margin of these parts, as seen from above, is not very far from being on a level with the cranio-rostral suture, and so causes the orbits to take up a larger part of the surface of the skull (in a view from above) than in the other group. The same is the case in Lathamus * (PL I. fig. 10). The retention of the furcula is no doubt associated with the rapidity of flight of this bird, whilst in the more slowly moving Platycerd it has disappeared almost entirely. As we already know from M. Blanchard's researches (Ann. Sci. Nat. Zool. xi. pp. 84-85, 1859), but little assistance as regards classification can be gained in this group from a study of the sternum. The pelvis, however, has been of more use to me. In the Lories this is always elongated and narrow in proportion, the preacetabular part being particularly elongated, and the iliac fossae on each side for the attachment of the gluteal muscles being deeper and P. Z. S. 1879, more extensive. In the Platycerd and Lathamus the pelvis is wider, the p. 174. preacetabular part much shorter, and the iliac fossae shallower and smaller. These differences will be visible from the figures which I exhibit (PI. I. figs. 11, 12), in which are shown respectively the pelvis of Lathamus and of Lorius tricolor. As regards internal anatomy, little can be said of any important characters, except the difference in the disposition of the carotid arteries in the two groups, first pointed out by Prof. Garrod, and already men- tioned above. The nature of the tongue in Lathamus requires reexami- nation, as also does the coloration of the eyes, this presenting very marked characteristics in all those Trichoglossinse I have been able to examine alive (of the genera Lorius, Eos, Chalcopsitta, and Trichoglossus), and being quite unlike that prevalent in the Platycerd and most other Parrots. In coloration Lathamus is no doubt aberrant, but is no more clearly related, as far as I can see, to one group rather than the other. The external rectrices being blue is perhaps a hint of its Platycercine relations. To conclude, the more important characters of Lathamus, i. e. pterylosis and superficial left carotid, beaks, nostrils, cere, feet, skull, and pelvis, all point to a near relationship to Psephotus, Platycercus, and allied genera. * Bonaparte (Compt. Rend. xliv. p. 536, 1857), following Owen (Cat. Osteol. Series E. C. S. 1853, p. 279, no. 1451), says that in Lathamus the orbit is completed below by the junction of the lacrymal with the " mastoid." This is certainly not the case in a skull lent to me by Professor Garrod, and, if true, would be an anomaly for any member of either of the above-mentioned groups. In the specimen referred to in the Museum of the College of Surgeons (no. 1451) it appeared to me on examination that there was in reality no bony union between the two bones, which were connected simply by ligament. PI. I P.Z-.S.1879.P1.XVL fig 7 FUjf.f '''/ 10. Fuj.11. If f STRUCTURE OT 1 .LA.THA.. Hanharc imp. ON THE GENUS MYZOMELA. 71 The abnormal tufted tongue, the retention of the furcula, and the sharp pointed wings may be regarded as adaptations to its tree- and flower- loving modes of life, and not as due to any consanguinity with the TrichoglossinaB. Lathamus may be a more or less modified remnant of a group that branched off from the common stock with the progenitors of the more typical Platycerci, and of which all the others have become extinct (perhaps due to the competition with the more specialized Trichoglos- sinae) ; or it may be a member of the Platycercine group that has become specialized to modes of life like those of the true Lories and Lorikeets, and so has come to resemble them in some few superficial particulars. EXPLANATION OF PLATE I. Figs. 1, 2. Back and side views of Lathamus discolor, showing pterylosis. 3, 4. The same of Platycercus pennantii. 5, 6. The same of Trichoglossus concinnus. 7. Mandible, deprived of horny sheath, ofEosrubra. 8. The same of Lathamus discolor. 9. Skull, seen from above, of Eos rubra. 10. The same of Lathamus discolor. 11. Pelvis of Lorius tricolor. 12. The same of Lathamus discolor. 16. A SYNOPSIS OF THE MELIPHAGINE GENUS P. z. s. 1879, MYZOMELA, WITH DESCRIPTIONS OF TWO NEW p>266 ' SPECIES.* (Plates II. & III.) THE genus Myzomela t was instituted by Messrs. Vigors and Horsfield in their paper on Australian birds in the Linnean Society's ' Transactions ' for 1826 (vol. xv. p. 316, note), Myzomela sanguinolenta, Lath, (for M. cardinalis, Grm. apud Vig. & Horsf. I. c., is clearly not that species, but the smaller Australian one), being the type. Lesson (Traite d'Orn. p. 298) in 1831 established a " sous-genre " PhylidonyriS) in which were included Certhia sanguinolenta, Ginnyris rubrater, and Cinrtyris eques; but this name must be, as he himself observes, regarded as merely a synonym of Myzomela. Beichenbach in 1851 (Handb. d. spec. Orn. p. 283) made Cinnyris eques the type of a new genus Cosmeteira, which he included amongst the Nectariniidae, its dull colours, with no metallic gloss, being apparently the chief reason for the separation. This species, however, in tongue, bill, feet, and, in fact, in all points is a true Myzomela, though it has * Proc. Zool. Soc. 1879, pp. 256-279, Pis. XXIV. & XXV. Bead Mar. 4, 1879. t pi>, I suck in, /ieAi, honey; hence Myzomela. 72 ON THE GENUS MYZOMELA. been included amongst the Nectariniidse till within the last few years by most writers. On similar grounds of divergent coloration, Bonaparte separated M. pectoralis * under the name Cissomela (C. R. xxxviii. p. 264, 1854) ; but as no generic characters whatever are given, this name falls to the ground, even if any structural differences in the bird exist, which as yet I have been unable to discover. Myzomela is characterized by its Meliphagine tongue, rather short, narrow, and slender curved bill, which is depressed and broadened at the base, rounded and compressed anteriorly, and there finely serrulated on its cutting-margins. The nostrils are linear and curved, extending for almost one third of the length of the bill, and covered in by a con- spicuous opercular membrane. The wings are moderately long, the P. Z. S. 1879, " first " t primary short, the 3rd to 5th longest and subequal, the 6th p. 257. longer than the 7th, which about equals the 2nd. The tarsi are about as long as the bill, rather slender, and covered with 6-7 scales in front, the lower ones being the smallest and transverse. The 2nd and 4th toes are yery slender, about equal in length, and shorter than the 3rd. The hallux is unusually stout for the size of the bird. The tail has 12 feathers, is short, and nearly square. Most of the species have more or less red in their plumage ; but this colour is altogether absent in some, and becomes only a slight tint, con- fined to the margins of the feathers, particularly of the head, wings, and tail, in others. As yet our knowledge of the phases and changes of plumage is by no means perfect. In one group (e. g. in M. sanguinolenta and its allies, including M. nigrita) the females seem to retain throughout life the brown plumage of immaturity ; whilst in others (e. g. M. nigri- ventrw, obscura, &c.) the adults of each sex are similar. In most cases the first plumage seems to be nearly uniform brown, lighter beneath, with the wing-coverts lighter at the edges, and the quills margined externally with olive-yellow. Throughout the group there is seen a great * Although Bonaparte expressly states " Myzomela nigra, Gould, est pour moi le type du nouveau genre Cissomela," yet it is evident from his description, " Subtus cum uropygio alba, torque pectorali nigro," that M. pectoralis was intended ! t I hare here adopted the system of notation for the remiges generally in use amongst ornithologists. But would it not be better, as is usually done in other cases of serially-repeated homologous organs, to begin counting from the proximal rather than from the distal end of the series ? At present, if a bird, for instance a Passerine, be said to have a " long first primary," two things may be meant : either that the bird has only nine primaries, the true " first " (or tenth) being absent, and the (morpho- logically) "second" (or ninth) being of the ordinary length (as, e.g., a Finch, or Drepanis) ; or that there are ten primaries, with the "first" (tenth) fully developed, as is the case in the " Formicarioid " Passeres of Wallace. This ambiguity would be avoided by counting the feathers from the end nearest the humerus ; for any Passerine with a long " tenth " primary could then only be a " Formicarioid." ON THE GENUS MYZOMELA. 73 tendency to retain these markings on the wings, as likewise a white margin on the inner web of the primaries. The eggs seem to be generally whitish or buff, spotted with darker, red or yellow. According to Gilbert (Gould, Handb. B. A. i. p. 558) 31. niyra, like many other species of Meliphagidae, lays only two eggs. The nests are small and cup-shaped, rather flirnsily constructed of grass- stems, hair, spiders' webs, &c., and often placed in the fork of a tree or bush. In their habits the Myzomelce seem to resemble the other smaller Honeysuckers, frequenting flowering shrubs and trees, not apparently so much for the sake of the nectar of the flowers, as for the insects attracted thereby. But one or two species of this genus, which is perhaps most nearly allied to Acanihorliynchus, but distinguishable by its longer beak and different coloration, were known to the older authors. Bonaparte, in his ' Conspectus ' (p. 394, 1850), enumerates 9, one of which, however (Certhia sanguined, Gmel.), is a Drepanis, whilst M. eques is omitted. Gray (Hand-1. B. i. p. 153, 1869) gives 17, though here again M. eques is omitted, being included as " Cosmeteira eques " amongst the Necta- riniidae (no. 1337). In the present paper 26 species, including two new ones, are recognized as distinct, besides one other which remains doubtful. Of these 26 species, 24 are known to me autoptically. Of the two which I have not seen, one (M. lafargii) is unique in the Paris Museum, the other (M. rubro-tincta) has lately been described from specimens at Ley den by Count Salvadori. The collection in the British Museum, that made by the * Challenger/ and the specimens in the collections of Mr. Sclater and Messrs. Salvin and Godman have formed the basis of my present paper. In addition to these I have to thank Canon Tristram, F.R.S., Dr. A. B. Mejer, and Count Salvadori for the very liberal way in which they have lent me valuable series of specimens. To the two latter, in particular, I am p. z. S. 1879, indebted for sending over to me the types of the species described by P- 258 ' them from New Guinea and its islands, and several others which I should not otherwise have been able to examine, and for their kind permission to figure any of them. Count Salvadori, too, has sent me some very valuable notes as to the range &c. of the Papuan species ; whilst to M. Oustalet I am much obliged for information on the type specimen of M. lafargii and on some other points. The following table will assist in the determination of the 26 valid species. It, however, only holds good for adult birds, and in many cases only for the males, our present imperfect knowledge of many of the species making a table that would have included all stages alike an impossibility. 74 ON THE GENUS MYZOMELA. A. Oorpore rubro ornato, aut unicolori. a. Corpore subtus plus minusve olivaceo-griseo aut albicante. a. Fronte coccinea. b. Torque pectoral! nullo. c. Capite et dorso concoloribus. * . J 1. sanquinolenta. I Ahsfusco-mgris; abdomine flavido-gnseo \ 2 caledonica. [ Alls olivaceo-f uscis ; abdomine griseo-flavido 3. chloroptera. c 7 . Capite rubro; dorso fusco 4. adolphince. b'. Torque pectorali fusco. J" Abdomine albicante 5. boieei. I Abdomine fusco-griseo 6. erythrocephala. a'. Fronte nigra. d. Capite supra macula rubra ornato. e. Gula rubra. (" Gutture summo nigricante 7. vulnerata. I Gutture croceo-flavo 8. jugularis. e'. Gula nigra 9. lafargii. d\ Capite supra omnino nigricante 10. sclateri. (3. Corpore subtus dorso concolori. a. Corpore nigro. f Subalaribus albis 11. nigrita. I Subalaribus nigris 12. pammel&na. a'. Corpore griseo-brunneo. b. Stria gulari coccinea 13. eques. b'. Stria gulari nulla. rCapite solum rubro tincto 14. obscura. I Alia et cauda rubro tinctis 15. simplex. f 16. rubrotincta. [Corpore, ahs et cauda rubro tmctis { ^ rubro . brunne ^ a". Corpore rubro 18. cruentata. y. Corpore subtus nigro et rubro vario. a. Gula coccinea. b. Capite toto coccineo. c. Abdomine rubro ; crisso nigro 19. rubratra. c'. Abdomine et crisso nigris. f 20. niqriventris. d. Pectore coccineo 1 ., ,. 7 . I 21. cardinalis. d'. Pectore nigro 22. lifuensis. b'. Capite supra nigro 23. chermesina. a'. Capite toto nigro 24. rosenbergi. B. Corpore nigro alboque vario. f Gula uropygioque nigris 25. nigra. iGula uropygioque albis 26. pectoralis. P. Z. S. 1879, 1- MYZOMELA SANGUINOLENTA.. p. 259. ? Scarlet Creeper, Lath. Gen. Syn. i. pt. 2, p. 740 (1782). ? Certhia rubra, Ghnel. S. N. i. p. 479 (1788). Sanguineous Creeper, Lath Gen. Syn. Suppl. ii. p. 167, t. 130 (1801). Certhia sanguinolenta, Lath. Ind. Orn. Suppl. p. xxxvii (1801). Cochineal Creeper, Lath. Gen. Syn. Suppl. ii. p. 167 (1801). Certhia dibapha, Lath. Ind. Orn. Suppl. p. xxxvii (1801). ON THE GENUS MYZOMELA. 75 Red-rumped Creeper, Lath. Gen. Syn. Suppl. ii. p. 169 (1801). Certhia eryihropygia, Lath. Ind. Orn. Suppl. p. xxxviii (1801). Certhia australasice, Leach, Zool. Misc. i. p. 30, t. 11 (1814). Myzomela cardinalis, V. & H. (nee Gm.) Linn. Trans, xv. p. 316 (1826). Myzomela sanguinolenta, Gld. B. A. iv. pi. 63 ; id. Handb. B. A. i. p. 555. cJ ad. capite, dor so cum uropygio, pectore et lateribus abdominis coccineis ; macula anteoculari, alis caudaque nigris ; alarum lectricibus conspicue albido, remigibus olivaceo-griseo limbatis ; abdomine sordide flavido ; subcaudalibus griseo alboque variis ; rostro nigro, pedibus corneis. Long. al. 2'4, caud. 1*6, rostr. 0*45, tars. 0'5 (poll. Angl.). $ sordide griseo-brunnea, subtus dilutior ; dorso et uropygio rufescenti tinctis ; alis caudaque fuscis, remigibus olivaceo, tectricibus alarum pallide brunneo marginatis. Hob. in Australia. The phases of plumage in this species, the type of the genus (for M. cardinalis, apud Vig. & Horsf . I. s. c., is this bird), seem to have caused some confusion amongst the older authors. It seems to me that in all probability Latham's " Scarlet Creeper/' on which Gmelin founded Certhia rubra in his edition of the ' Systema Naturae,' really applies to this species, the description " lower part of belly and vent white," together with the size (** of a Wren ") and the locality (" from some part of the South Seas ") quite coinciding with this bird, and not at all with M. cardinalis, of which, in his Ind. Oru. (i. p. 290, 1790), Latham treated it as being the female. Besides this, Latham bestowed at least three other Latin names (each with its equivalent vernacular) on this little bird. Myzomela sanguinolenta is perhaps most nearly allied to M. chloro- ptera, which differs, however, as below pointed out. Only the males possess the beautiful red plumage ; and in these, if not quite adult, the variegation of each breast-feather, which is grey at the base, then paler, and red only at the tip, produces the somewhat mottled appearance of the red underparts. According to Mr. Gould, the irides are " dark brown." Myzomela sanguinolenta is the commonest species of Myzomela in Australia, and is familiarly known to the colonists as the "Little Soldier." Mr. Bamsay, in his list of Australian Birds (Proc. Linn. Soc. N. S. ~W. ii. 1877), records it from Rockingham Bay, Port Denison, the "Wide-Bay District, the Richmond- and Clarence-Biver Districts, New P. Z. S. 1879, S. Wales, the interior, Victoria, and S. Australia ; so that it ranges over p * the greater part of Eastern Australia. Mr. Ramsay has given us a good account of the habits and nesting of this species near Sydney, where it is a summer visitor, arriving in October and November, in * The Ibis' for 1865 (p. 304). 76 ON THE GENUS MTZOMELA. 2. MTZOMELA CALEDOFICA, n. sp. Myzomela sanguinolenta (ex Nova Caledonia) auct. c? jprcecedenti simillima, sed tectricibus alarum marginibus albidis carens. Hob. in Nova Caledonia. Mus. H. B. Tristram. The Myzomela from New Caledonia, although no doubt very closely allied to the preceding Australian species, is, I think, fairly entitled to rank as a distinct species ; and I have therefore separated it under the above name. My attention was first directed to this form by a speci- men kindly lent me by Canon Tristram, and shot by Mr. Layard near Noumea. This bird, a fully-plum aged male, differs from a considerable number of Australian specimens with which I have compared it, in the almost entire absence of the conspicuous greyish- white margins to the feathers of the wing-coverts, so that they are nearly entirely black, with only a trace of olive-colour at the margins. Besides this, the red colour of the body is hardly so bright, and extends a little further down on the abdomen, and the margins to the quills are more of an olive-yellow. The size is about the same (wing 2-25), Australian specimens varying a little in this respect. Canon Tristram writes me that he has six specimens of the New- Caledonian bird, and that the differences which I pointed out to him are constant in the series. Mr. Layard gives the following notes as to the soft parts on the label of his specimen : " Beak black, legs brown-black, iris brown." Mr. Layard also met with a Myzomela, which he referred to M. san- guinolenta (Ibis, 1878, p. 280), in the New Hebrides, on the islands of Vate, Api, and Mallikollo, and remarks that a specimen procured is identical with the New-Caledonian bird ; so that it seems probable that M. caledonica may extend its range as far as these islands ; but specimens to show this are as yet wanting. 3. MTZOMELA CHLOEOPTEEA. (Plate II. fig. 1.) Myzomela chloroptera, Wald. Ann. N. H. 4th ser. ix. p. 399 (1872) ; Salvad. Ann. Mus. Civ. Gen. vii. p. 662 (1875). cJ ad. capite, dorso uropygioque, cum pectore, coccineis ; corpore subtus griseo-flavido ; alis caudaque fuscis, remigibus et tectricibus alarum olivaceo limbatis, subalaribus et margine interna remigum albis, alee flexura flavo-albida ; macula anteoculari nigra ; rostro nigricante pedibus obscure cornels. Long. al. 2'2, caud. 1'5, rostr. *55, tars. '50 (poll. Angl.~). Hab. in insula Celebes. This Myzomela, the westernmost of the whole genus, was described by the late Lord Tweeddale from imperfect specimens collected by Dr. ^Z. S. 1879, Meyer at Menado, where it has also been obtained by Bruijn's col- p. 261. lectors ; and from one of these specimens, kindly lent me by Count PHI. P.Z.S.1879P1.XXIV. J.Smi tilth. l.MYZOMELA CHLOROPTERA. 2.. RUBROBRUNNEA 3. ADOPHINzE Ha,nhant imp. ON THE GENUS MYZOMELA. 77 Salvador!, the figure is taken. As yet, I believe, it has only occurred near Menado; and the young and female remain unknown, or at least undescribed. Myzomela chloroptera resembles the Australian M. sanguinolenta, but is a smaller bird, and also differs in the smaller extent of the red on the chest, and in that colour being more intense, the abdomen yellower, and the wings and tail not so black. The black anteocular spot is less conspicuous. In his original description Lord Tweeddale remarks that this bird nearly resembles plate 54 of the ' Oiseaux Dores,' vol. ii., representing " L'Heorotaire ecarlate " from the " South Seas," taken from a drawing of a bird in the Leverian Museum. The figure certainly corresponds very fairly with this species, but, from the locality given, is probably intended for the Australian one (M. sanguinolenta). 4. MYZOMELA ADOLPHXN^. (Plate II. fig. 3.) Myzomela adolphince, Salvad. Ann. Mus. Civ. Gen. vii. p. 946 (1875). cJ pallio, dorso superiore, alis caudaque olivaceo-fuscis, tectricibus alarum, remigibus et rectricibus externe subtiliter olivaceo limbatis ; capite uropygioque coccineis ; macula anteoculari nigra ; corpore subtus flavido-albido, pectore grisescenti lavato ; subalaribus et remigum mar- gine interna albis ; rostro nigricante, pedibus corneis. Long. tot. circa 3*5, al. 2'2, caud. 1'5, rostr. *45, tars. '5 (poll. Angl.). 5 minor ; femince Myzomelse boiaei similis. Hab. in montibus Arfak. This is one of the numerous discoveries of Beccari and Bruijn in the Arfak Mountains, and only a few specimens have as yet been obtained. Count Salvadori writes (1. s. c.) : " This species resembles M. erythro- cepliala of Grould, but differs from it in its much smaller dimensions, by the very slight olive tint of the back, and by the lower parts being not grey-brown, but whitish, very slightly tinged with yellowish on the breast and abdomen." The female resembles that of the Banda species (M. boicei), but differs as pointed out under that species (vide infra). The figure (PI. II. fig. 3) represents an adult male, one of the types of this species, most obligingly lent me by Count Salvadori. 5. MYZOMEIA BCXLEI. Myzomela boiei, Sal. Mull. Verh., Land- en Volkenk. p. 172 (1839-44) ; id. Verb., Zool. Aves, p. 66, t. 10. fig. 1, 2. Myzomela jugularis, Peale, U.S. Expl. Exped. p. 151, t. 41. f. 2 (1848) ; Cassin, U.S. Expl. Exped. p. 176, t. 12. f . 2 (jr.) (1858) ; H. & F. Orn. Centr.-Pol. p. 54, t. 7. figs. 1, 2 (ad. et jr.). Myzomela solitaria, Hombr. & Jacq. Voy. Pole Sud, Zool. iii. p. 99, Atlas, t. 22. f. 6 (1853). Ad. fusco-nigricans, subtus Jlavescenti-albida, mento, gula, maculaque magna occipitali cum uropygio coccineis ; gutture croceo-Jlavo ; remigi- bus, primis duobus exceptis, et tectricibus alarum majoribus flavido marginatis ; rectricibus, duabus mediis exceptis, tectricibusque alee minoribus nonnullis ad apicem albis ; rostro nigro ; pedibus cornels. Long. al. 2'45, caud. 1-6, rostr. *6, tarsi '55 (poll. Angl.). Jr. macula occipitali nulla, gutture sordide flavo, et uropygio brunneo- olivaceo distinguenda. Hab. in insulis Vitiensibus. This Myzomela hardly admits of being mistaken for any other species. It is perhaps most nearly related to M. lafargii of the Solomon Islands, but is at once distinguishable from that species by the red throat and orange-yellow chest, besides other differences. The red of the throat is separated from the yellow of the chest by a distinct though narrow black line. The red on the back appears last, that on the chin first. In not fully plumaged birds the rump and lower back are olivaceous. The sexes when adult are nearly alike, the female being only distinguishable by the colours being less bright. Very often, too, though not always, the red occipital spot is absent in the female. Mr. Murray records the iris as " black," Mr. Layard as " brown," the legs being " verditer " and " dark livid " in the living bird, with the soles of the feet yellow. This bird is entirely confined to the Pijis, where, according to Mr. Layard's list (Ibis, 1876, p. 391), it is found in all the larger islands of that group * ; and in addition to the islands enumerated by him, speci- * See also P. Z. S. 1875, p. 431, for an interesting account of its habits. ON THE GENUS MYZOMELA. 81 mens from Matuku are in the British Museum (Rayner). Its occur- rence in the Samoan group has not yet been confirmed (cf. Whitmee, Ibis, 1875, p. 447). Hombron and Jacqumot indicated their " Myzomele solitaire" as being from the "lies Salomon " with some doubt; and, relying on them, Mr. Sclater included " M. solitaria " in his list of Solomon-Island Birds (P. Z. S. 1809, p. 124), where, however, only M. lafargii, so far as is yet known, occurs. 9. MYZOMELA LAFAKGII. Myzomela lafargii, Hombr. & Jacq. Yoy. Pole Sud, Zool. iii. p. 98, t. 22. f. 5 (1853). Oorjpore supra cum capite, gutture et pectore superiore nigris ; occipite coccineo ; abdomine flavido-olivaceo ; alis caudaque nigris, remigibus P- z - S- 1879, olivaceo-limbatis, subalaribus albis; rostro nigro, pedibus plumbeis. Sab. in insulis Salomonis. This species was obtained by the French Expedition to the South Pole ; and the type specimen in the Paris Museum remains, I believe, unique in Europe. M. Oustalet, to whom I wrote for information about it, kindly replies to me, on comparing it with the figure in the Atlas to the ' Yoyage : ' " Je trouve dans celle-ci quelques inexactitudes. Les proportions de 1'oiseau ont ete un peu exagerees : le noir de la gorge a ete trop etendu et trop marque. L'oiseau type est plus petit, et il a le haut de la gorge seulement noir, le bas, vers la poitrine, etant un peu mele de jaune verdatre." M. lafargii is somewhat allied to M. jugularis of the Fijis, but differs from the latter in having the red confined to the top of the head, and in the throat and chest being black. 10. MYZOMELA SCLATERI, sp. n. (Plate III. fig. 2.) cJ corpore supra, alis caudaque jusco-nigricantibus, capite saturatiore, plumis dorsi inferioris apice flavidis ; remigibus, alarum tectricibus et rectricibus externe olivaceo-flavo limbatis ; gula splendide coccinea ; corpore subtusgriseo-flavido, gutture sordidiore ; subalaribus et margine interna remigum albis ; rostro nigro, pedibus obscuris. Long. tot. circa 4'5, al. 3*65, caud. 1*7, rostr. *6, tars. *55 (poll. Angl.). Hab. in Nova Britannia. A few weeks ago Mr. Sclater, after whom I propose to name this new species, lent me for examination a single specimen of it, marked " male/' which he had recently received in a letter together with two Pachycephalce, from the Rev. G. Brown, C.M.Z.S., of the Wesleyan Mission at present established on the Duke-of-York Islands. The exact locality given on the label is " Palakiiru Island, New-Britain coast." I have not been able to find Palakiiru Island on any map ; but it is probably only an islet lying close to the shores of the larger island. At first I had some doubts as to this individual being adult ; but now G 82 ON THE GENUS MYZOMELA. from the absence of red feathers on any other part, and from the singu- larly bright and shining colour of those on the throat, I have little doubt that it has very nearly or quite attained its full plumage. Myzomela sclateri hardly admits of being compared with any other species of the group, the entirely dark upperside and the red being confined to the throat, rendering it quite unlike any species yet known to us. P.Z.S. 1879, p. 266. 1858, p. 173; Salvador!, Sitzungsber. Wien. 11. MYZOMELA NIGEITA. Myzomela nigrita, G. E. Gray, P. Z. S. RZ.S. 1878, p. 97. Myzomela erythrocephala, Meyer (nee Gould), Akad. Ixx. p. 204 (1874). Myzomela meyeri, Salvadori, Ann. Mus. Civ. Gen. vii. p. 947 (1875). cJ nitenti-niger, subalaribus et remigum margine interna albis ; rostro nigro, pedibus cornels. $ griseo-brunnea, subtus dilutior ; f rente gulaque rubro lavatis ; remi- gibus eocterne olivaceis. Hob. in Nova Guinea occidentali et insulis vicinis. This Myzomela, conspicuous for the almost entirely black plumage of the adult male, was first described by the late Mr. Gray from specimens collected in the Aru Islands by Wallace, where it was obtained again during the recent voyage of the ' Challenger/ It also occurs on the mainland of the north-western peninsula of New Guinea, at Dorey (Wallace) and Eubi (Meyer"), and in the islands of Jobi and Miosnom (Meyer and Beccari), the birds from the mainland and these islands being considerably bigger than those from Aru. This is particularly the case with those from Jobi and Miosnom, so that Count Salvador! is inclined to separate them as a new species. But, as the following table will show, considerable differences in the measurements of this species occur in various localities ; so that at present I consider it better to retain all forms under one name. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Wing. Aru? 2-2 "Wokan 2-4 Aru 2-35 Eubi 2-5 Dorey 2-4 Miosnom 2*7 Jobi 2-5 jr. Miosnom 2*6 jr. Eubi 2-3 cj jr. Eubi 2-3 $ . Eubi 2-1 $. Aru? 2-1 Beak (from fore- Tail, head). Tarsi. 1-65 -55 -5 1-65 55 5 1-7 5 2-0 65 5 1-8 65 2-0 2-0 1-7 65 67 67 5 ^ "M.pluto," 5 >> Salvador!, 53 J inlitt. 1-8 6 53 1-8 58 5 1-65 55 45 1-4 55 45 ON THE GENUS MYZOMELA. 83 The male of this species resembles that of Myzomela pammelcena from the Admiralty Islands, but differs as below specified. The female retains more of the normal colouring of the group, and approaches those of M. boicei and M. adolphince. The young birds resemble the female, the red on the head in the young males being obtained before any indi- cation of the black plumage. Dr. Meyer obtained only females and young of this bird, and referred these with considerable hesitation to M. erythrocephala of Gould, a very different species. Count Salvadori saw that this was a mistake, and proposed the name meyeri for the specimens collected by Dr. Meyer. But on subsequently examining the birds at Dresden, he found that in reality they were the young and females of the present species, the female having been only briefly indicated in Gray's original description. Mr. Murray notes of a male from "Wokan, Aru Islands, that the eyes p. z. S. 1879, are " hazel," the " bill and feet black." P- 267. 12. MYZOMELA Myzomela pammelcena, Sclat. P. Z. S. 1877, p. 553. cJ ad. nigerrimus, remigum marginibus internis cineraceo-albidis, rostra pedibusque nigris. Long. tot. circa 5, al. 2*7, caud. 2, rostr. *65, tars. 65 (poll. Angl.). *Fun. prcecedenti similis, sed omnino sordidior, abdomine et subcaudalibus rufo-tinctis, et subalaribus albis distincta. Hob. in insulis Admiralitatis. Two specimens, an adult male and a young bird, of this Myzomela were obtained during the stay of the ' Challenger ' at Nares Harbour, Admiralty Islands. It is closely allied to Myzomela nigrita of the Aru Islands and JNew Guinea; but the adult male of the new species differs from the more western one by its black under wing-coverts (although these are white in the young bird), dirty white margins to the remiges, and longer and stouter feet and tarsi. In size it exceeds any specimens I have seen of M. nigrita from the Aru Islands, but is equalled in length of wing and tail by the larger birds from the islands and shores of Geel- vink Bay. Mr. Murray marks the irides of the adult bird as " hazel-brown." 13. MYZOMELA EQUES. Cinnyris eques, Less. Yoy. Coq. p. 679, t. 31. fig. 1 (1826). Nectarinia eques, Miill. & Schleg. Verhand. p. 62 (1839-1844). Cosmeteira eques, Meyer, Sitzungsber. Wien. Akad. Ixx. pp. 215-217 (1874). Cosmeteira minima, Wald. Ibis, 1870, p. 50 ( ? ). Omnino cinerascenti-brunnea, subtus dilutior ; stria gulari nitide cocdnea ; G2 84 ON THE GENUS MYZOMELA. rostro pedibusque nigro-corneis. Long. al. 3, caud. 2'5, tarsi '6 (poll. Angl.). ( <5 ex Nova Guinea.) Hob. in Nova Guinea et insulis vicinis. Although generally placed amongst the Nectariniidae, this species in structure and coloration is a true Myzomela, allied to the Australian M. obscura, from which it is at once distinguished by its bright red gular streak. The sexes are similar ; but the females are considerably smaller than the males ; and on one of these from Mysol the late Lord Tweed- dale founded his species C. minima. Dr. Meyer describes (I. s. c.) the young as having the forehead and top of the head tinged with reddish an interesting fact, as showing in the young bird a style of coloration not retained in the adult, but occur- ring in other members of the genus, and therefore probably a more primi- tive character. This species is widely distributed over New Guinea, occurring at Dorey (Wallace and Meyer), Eubi, Passim (Meyer), Sorong (Mus. Lugd., fide Salvadori), and Wa Samson (Beccari) ; and D'Albertis found it on the Fly River. It also occurs in Waigiou (Lesson, Wallace, and Bern- P Z S 1879 ste ^ n } an d ^7 S 1 (Wallace and Hoedt). Count Salvadori has lent me p. 268. specimens from Salwatti, and says that in the Leyden Museum there is one said to be from Ceram (Moens), but that this locality, as well as Gilolo (Forsteri), are in all probability errors. 14. MYZOMELA OBSCTJEA. Myzomela obscura, Gould, P. Z. S. 1842, p. 136 ; id. B. A. iv. pi. 67 ; id. Handb. i. p. 559. Ptilotis fumata, " Miill. Mus. Lugd., ex Nova Guinea," Bp. Consp. i. p. 392 (1853). Omnino griseo-brunnea, subtus pallidior, capite vinaceo tincto ; remigi- bus externe subtilissime griseo limbatis ; alis caudaque subtus griseis, remigum margine interna albida ; rostro pedibusque nigro-corneis. Long. al. 2'7, caud. 2'2, rostr. -6, tars. '6 (poll. Angl.). Hob. in Australia septentrionali et Nova Guinea. This plainly-coloured Honey-eater was first described by Mr. Gould from specimens obtained at Port Essington by Gilbert. It seems to have rather a wide range over the northern parts of Australia, occurring at Port Darwin (Masters), Cape York (' Challenger '), and in the north of Queensland "as far south as the Mary river" (Ramsay). D'Albertis found it at Naiabui and on the Fly Biver ; and there are specimens from the river Utanata in the Leyden Museum the originals of Bonaparte's " Ptilotis fumata" (cf. Salvadori, Ann. Mus. Civ. Gen. xii. p. 334, 1878). The sexes are similar. I have not seen young birds. The iris has been variously recorded as " red " (Gould), " brown " (Murray), and "black" (VAlbertis). ON THE GENUS MYZOMELA. 85 15. MYZOMELA SIMPLEX. Myzomela simplex, G. E. Gray, P. Z. S. 1860, p. 349. Sordide griseo-brunnea, subtus dilutior ; remigibus et rectricibus rubido limbatis ; margine internet remigum albida ; rostro pedibusque corneis, his pallidioribus. Long. tot. 5*2, al. 2'3, caud. 2, rostr. '5, tarsi '6 (poll. Angl.). Hob. in Halmahera et insulis adjacentibus. This plainly-coloured Myzomela was first discovered by "Wallace in the island of Batchian, and it also occurs in most of the other islands of the Halmahera group of the Moluccas, but is replaced on Obi by the nearly allied Myzomela rubrotincta. Count Salvadori informs me that he has seen " many specimens in the Leyden Museum from Gilolo (Bernstein), Tidore (Bernstein, Von Rosenberg), and Dammar (Bernstein). A specimen from Ternate (Bruijn) is in Turati's collection. A single specimen from Morty in the Museum of Leyden is much darker than the others." This species is allied to M. rubrobrunnea and M. rubrotincta, but differs from them in the less extent of the red colour, which is confined to the margins of the quills and tail-feathers. The sexes are probably similar in colour ; I have not seen the young bird. 16. MYZOMELA BFBBOTINCTA. P.Z S 1879 Myzomela rubrotincta, Salvad. Ann. Mus. Civ. Genov. xii. p. 344 (1878). p * 269t " Brunnea, dorso, alis et cauda pulcherrime rubro tinctis ; pectore, ab- domine et subcaudalibus obsoletius rubro tinctis. Long. tot. '120 m., dice -067, caud. -048, rostri -020, tars. -020." .Hob. " in ins. Obi (Bernstein)" Salvad. I. c. This species has recently been described by Count Salvadori from five specimens two males and three females the two sexes are similar in the Leyden Museum. He says it " resembles M. simplex of Gray from Halmahera, in which only the remiges and rectrices (and not all the parts between the head and neck) are margined with red, and in which the red colour is very indistinct." 17. MYZOMELA BUBBOBBTJNNEA. (Plate II. fig. 2.) Myzomela rubrobrunnea, Meyer, Sitzungsber. Ak. in Wien, Ixx. p. 203 (1874). c? grisescenti-brunneus, subtus dilutior, capite saturatiore, plumis plus minusve vinaceo limbatis ; dorso inferiore et uropygio, cum marginibus externis remigum et rectricum vinaceo-rubris ; alis cauddque subtus griseis ; margine interna remigum albida ; rostro pedibusque nigro- corneis. Long, tota circa 4, alee 2'4, caud. 1'8, rostr. '6, tars. '55 (pott. Angl.). $ mari similis, sed color ibus minus intensis et paullo minor. Bab. in insula Mysore. f 86 ON THE GENUS MYZOMELA. Dr. Meyer first discovered this beautiful species of Myzomela, during his travels in and about New Guinea in 1873. He obtained only two specimens, both males, at Kordo, the chief settlement in the island of Mysore in Geelvink Bay. Beccari obtained others in the same island, to which it is apparently confined ; and from one of his specimens, a fine male, kindly lent me by Count Salvador!, the figure is taken. This species resembles M. simplex and M. rubrotincta of the Moluccas, but differs from both in the red margins to the feathers being continued over a larger part of the bird. 18. MYZOMELA. CRTJENTATA. Myzomela cruentata, Meyer, Sitzungsber. Ak. Wien, Ixx. i. p. 202 (1874) ; Gould, B. New G-uin. pi. pt. v. Myzomela coccinea, Eamsay, Proc. L. S. N. S. W. ii. p. 106 (1877) ? (Ex insulis Ducis Eboraci.) Myzomela erythrina, Eamsay, Proc. L. S. N. S. W. ii. p. 107 (1877) ? (Ex Nova Hibernia.) c? corpore omnino chermesino, uropygio splendidiore, plumis ad basin nigris ; alis rubricantibus, plumis externe rubris ; remigibus fuscis, primis duobus exceptis, rubro limbatis ; rectricibus rubido-griseis, ex- P. Z. S. 1879, terne rubro marginatis ; alis caudaque subtus griseis ; rostro pedibusque P- 27 - nigris. Long. tot. circa 4, al. 2'2, caud. 1-5, rostr. -55, tars. *5 (poll. Angl.). Sab. in montibus Arfak NOVSB Guinea. This very beautiful Myzomela, at once distinguished from all others of this group yet described by its uniformly red colour, was first obtained by Dr. Meyer, in the Arfak Mountains, in 1873. Only one specimen, an adult male, was procured; and this and another specimen, likewise a male and nearly or quite adult, procured by Bruijn's collectors in the same locality, and now in the Genoa Museum, are, I believe, the only examples yet brought to Europe of this splendid little bird. A short time ago Mr. E. P. Eamsay, of the Sydney Museum, described two new species of Myzomela, both remarkable for their nearly uniform red coloration. One is indicated as a female and from the Duke-of-York Islands (M. coccinea) ; the other, a young male (M. erythrina), is from. New Ireland. Of it Mr. Eamsay says : "This species is smaller than the preceding, and the bill is comparatively stronger and stouter ; other- wise I should be inclined to consider it the young of the former." From his description it is evidently a young bird ; and after having carefully compared both it and that of the other species with Dr. Meyer's and Count Salvadori's specimens, I have come to the conclusion that both M. coccinea and erythrina are probably referable to M. cruentata. If this is so, it would seem, provided Eainsay's specimens are correctly sexed, that the adults of this species are nearly or quite similar in coloration. The ON THE GENUS MYZOMELA. 87 species probably has a wide range through New Guinea eastward of the Arfak Mountains. 19. MYZOMELA ETJBEATEA. Cinnyris rubrater, Less. Yoy. Coquille, Zool. p. 678 (1826) ; id. Man. ii. p. 55 (1828) ; Kittlitz, Kupf. Yog. t. 8. fig. 1 (1832). Myzomela rubratra, Bp. C. E. xxxviii, p. 263 (1854) ; Hartl. P. Z. S. 1868, p. 5; Hartl. & Einsch, P.Z.8. 1872, p. 94; Finscb, Journ. God. Mus. xii. p. 26 (1876). Myzomela major, Bp. C. E. xxxviii. p. 263 (1854). (Ins. Carol.) Myzomela sanguinolenta, pt., Gray (nee Lath.), Gen. B. i. p. 118 ; Bp. Consp. i. p. 394 (1850). Ad. coccinea, alis, cauda, crisso et subcaudalibus nigricanlibus ; alis caudaque subtus griseis, remigum margine internet albida ; rostra ni- gricante, pedibus cornels. Long. al. 2*95, caud. 2*3, rostr. "65, tars. 75 (poll. Angl.}. Jr. olivaceo-brunnea, remigibus externe olivaceis ; subalaribus obscuris. Hob. in insulis Pelewensibus, Marianis, et Carolinis. This species belongs to the group of M. cardinalis, nigriventris, and chermesina, but is at once distinguished from all of these by the greater extent of the red colour in the adult, only the vent and under tail-coverts being black. Myzomela major was founded by Bonaparte on specimens of this bird from the Caroline Islands, and characterized as " Similis M. rubratrse, sed p. z. S. 1879, major et percoccinea." But any such difference in size is not constant, and P- 27*' Dr. Hartlaub says (I. c.) that Pelew birds are as large as Caroline ones. The young bird is nearly uniformly dark olive-brown, and gradually attains its full plumage by the gradual appearance of the red on various parts of its body. M. rubratra is remarkable for its wide range over the archipelagos of the North-eastern Pacific. Lesson found it on the island of Ualan in the east of the Caroline group (his assertion that it was also found in the Philippines by M. Dussumier being of course erroneous), as did Kittlitz, who gives an interesting account of the habits of this species as observed by him on this island and the Marianne Island of Guam (Denkwiird. ein. Eeise, i. pp. 364 and 381, 1858). Kubary found it on Ponape in the east, and on Yap and the Mackenzie Islands in the west, of the Carolines ; so that it is probably found all over that archipelago. Specimens from these islands are in the Godeffroy Museum ; likewise examples from the Pelews (or Palaos). Gray, in his Catalogue of Pacific birds, gives " Island of Yanicoro " with a query ; but in all probability this is a mistake, for as yet no Myzomela has been found there. 20. MYZOMELA JQGEIVENTEIS. Myzomela nigriventris, Peale, U.S. Expl. Exped. p. 150, pi. 41. f. 2 88 ON THE GENUS MYZOMELA. (1848) ; Cassin, U.S. Expl. Exped. p. 175, pi. 12. f. i. (1858) ; H. & E. Orn. Centralpolyn. p. 56, t. 7. f. 3 and 4 Cad. and jr.). Myzomela rubratra, Hartl. (nee Lesson), Wiegm. Arch. 1852, p. 1.30 (ex Samoa). Myzomela cardinalis, Hartl. (nee. Gmel.), "Wiegm. Arch. 1852, p. 109. " Myzomela arnouxi, Verr.," Bonaparte, C. B. xxxviii. p. 263 (1854). Ad. capite, dorso uropygioque cum pectore fulgido-coccineis, plumis ad basin nigris ; corpore subtus, macula anteoculari, alis caudaque nigris ; remigibus interne albidis ; rostro pedibusque nigris. Long. al. 2*75, caud. 1'8, rostr. '65, tars. *7 (poll. Angl.). Jr. olivaceo-fusca, subtus dilutior et flavido lavata ; uropygio rubro tincto ; remigibus olivaceo-limbatis ; subalaribus et margine interna remigum albis. Hab. in insulis Samoensibus. This species is very closely allied to M. cardinalis, which it replaces in the Samoa group. The differences between the two I have pointed out under the last-named species. Erom M. rubratra, with which it was at first confounded, both these species differ in the black flanks and belly, these in M. rubratra being red, only the vent and under tail-coverts being black, whilst the red on the chest in all three of these species easily separates them from M. lifuensis. M nigriventris is confined to the Samoan Islands, its reported occur- rence in the Eijis being erroneous (cf. Layard, Ibis, 1876, p. 391) and P. Z. S. 1879, founded on a mistake of Dr. Graffe. It is apparently rather a common p ' " bird in the Samoan group, occurring both on Savaii and Upolu. 21. MYZOMELA CABDINALIS. Cardinal Creeper, Lath. Gen. Syn. i. pt, 2, p. 733, pi. 33. f. 2 (1782). Certhia cardinalis, Gm. S. N. i. p. 472 (1788) ; Lath. Ind. Orn. i. p. 290 (1790). Cardinal Honey-eater, Lath. Nat. Hist. ir. p. 199, pi. 71. f. 2 (1822). Myzomela cardinalis, Gray, B. Trop. Isl. p. 10 (1859) ; Tristram, Ibis, 1876, p. 261. Myzomela melanogastra, Bp. C. E. xxxviii. p. 263 (1854). Ad. capite, dorso uropygioque cum pectore superiore coccineis, plumis ad basin nigris; macula anteoculari, alis caudaque nigris, Ms nitore nonnullo metallico ; corpore subtus f uliginoso-nigro ; remigum margine interna albida ; rostro pedibusque nigris. Long. al. 2' 9, caud. 2'], rostr. *7, tars. '75 (poll. Angl.). Jr. Myz. nigriventri similis, sed supra magis brunnea, et subtus dilutior ; dorso uropygioque castaneo-brunneis, nee rubris. Hab. in JSTovis Hebridibus. This Honey-eater, one of the few of this genus known to the older ON THE GENUS MYZOMELA. 89 authors, is very nearly allied to M. nigriventris of the Samoan group, which it replaces in the New Hebrides. The adult bird (I agree with Messrs. Hartlaub and Einsch in con- sidering that in this section of the group the sexes are nearly similar) is distinguished from M. nigriventris by the scarlet of the upper parts and chest being duller, and extending not quite so far down on the chest. The black of the lower parts is less intense, being tinged with brownish ; the white margin to the remiges internally is more distinct ; and the bill is stouter. It is also a slightly larger bird. The young bird is paler and browner above (not so much dark brown as greyish brown), and paler and yellower below ; the rump and back are washed with chestnut-brown. Judging from the series of specimens I have seen, the red colour in this species seems to appear first on the head, and not on the back as in M. nigriventris. The remiges, as usual in the young of this genus, are externally lined with olive-yellow. From M. lifuensis this species may be distinguished by* its larger size and by the red extending on to the breast. Latham's description and figure clearly apply to this bird, not to M. lifuensis. The irides are marked " black " or " dark brown." Latham describes this bird from the island of Tanna, where, he says, it is called " Kuyameta " and is common, sucking the juices of flowers ; and I have seen specimens collected on that island by Mr. Layard. There are specimens in the British Museum from Erromango and Anei- teum (Owning); and Canon Tristram has received it from the latter island, as well as from Tanna and Aniwa. It thus seems to be confined rather to the southern portion of the New-Hebridean archipelago, being replaced in the north by M. caledonica ? and M. chermesina. 22. MYZOMELA LIFFENSIS. p 2 s. 1879 Myzomela lifuensis, E. L. and L. C. Layard, Ibis, 1878, p. 258. p * m <5 capite, dorso uropygioque coccineis ; alis, cauda et corpore subtus toto cum macula anteoculari fuliginoso-nigris ; alis caudaque nitore non- nullo metallico ; remigum margine interna albida ; rostra nigro,pedibus nigro-corneis. Long, iota circa 4'2, al. 2-5, caud. 1-75, rostr. -55, tars. '68 (poll. Angl.). Hob. in Lifu, ex insulis " Loyalty " dictis. Canon Tristram having kindly submitted to me two skins (now in his collection, both marked " males " and adult) collected by the Messrs. Layard, who first indicated this species, I can give a more complete account of it, and say that it is certainly a very good species. It is nearly allied to M. nigriventris and M. cardinalis of the Samoas and New Hebrides respectively, more particularly to the last, but is at once dis- tinguished from both by the red below not extending beyond the head, the breast being sooty-black like all the rest of the lower parts. It is 90 ON THE GENUS MYZOMELA. also a considerably smaller bird ; the bill is shorter and more slender ; the tarsi are not so stout, and the claws smaller. From Myzomela erythro- cephala it is easily distinguishable by the uniform black of the lower parts. Mr. Layard notes the " beak black, legs very dark brown, iris dark brown," and food " insects." Both specimens were obtained at Hepenehe, the chief town in the island of Lifu, the largest of the Loyalty Islands. Whether M. erythrocephala of Marie's list (Ibis, 1877, p. 362) is this bird, remains uncertain ; as yet, M. caledonica is the only Myzomela certainly known to be found on New Caledonia itself. 23. MYZOMELA CHERMESINA. (Plate III. fig. 1.) Myzomela chermesina, Gray & Mitch. G. B. i. pi. 38 (1840) (fig. mala) ; Gray, Cat. B. Trop. Isl. p. 11 (1859) ; Forbes, P. Z. S. 1878, p. 352. c? ad. fusco-nigricans, alis caudaque nitore nonnullo metallico ; mento, gula, pectore lateribusque abdominis, cum dorso uropygioque nitide coccineis, plumis ad \asin nigris ; subalaribus nigris, remigum pogonio interno griseo ; rostro nigro, pedibus brunneo-corneis. Long, iota circa 4|, oil. 3, caud. 2, rostri ||, tarsi (poll. Angl.). Hob. in insulis Pacificis E-otumah et Mallikollo. This species was first figured by Messrs. Gray and Mitchell in their * Genera of Birds ; ' but no description was given, the species being only mentioned in the list of the species of Myzomela ; nor was any habitat indicated. Bonaparte, and Gray later on, in his ' Hand-list ' (vol. i. no. 1989), gave "New Guinea?" as the locality, without any apparent reason for so doing. The bird was never recognized agciin till last year, when Mr. Sclater received two specimens, an adult male and a nearly adult female*, from the Eev. G. Brown, C.M.Z.S., of the Wesleyan P. Z. S. 1879, Mission, together with some other birds, from the small island of Eotu- p. 274. mah, north of the Fijis. Fortunately Gray's type is still in existence in the gallery of the British Museum ; and on comparing the birds from Eotumah with it, it was at once evident that they were of the same species, though Gray's figure represents a bird with a uniformly scarlet underside. About the same time Mr. Sharpe got a specimen (from which the figure is taken) of the same bird, apparently identical in every respect, from the island of Mallikollo (in my paper, 1. c., by a mistake I wrote Erromango) in the New Hebrides, where it was obtained by Mr. Wykeham Perry, H.M.S. ' Pearl.' The species thus has a wide range, though I believe the above-mentioned four specimens (which are all nearly or quite adult) are as yet the only ones of this bird ever brought to Europe. The female is similar to the male in colour, but a little duller ( present bird (cf. Ibis, 1. c. p. 286). As regards the systematic position of Mascarinus duboisi, the available material is so scanty that we shall probably never (for the bird is certainly extinct) be able to arrive at any satisfactory conclusion about it. In the form of the beak, the feathered nostrils and lores, the narrow orbital ring, and the structure of its feet, it more resembles the genera Tanygnathus and Palceornis than any of the African genera of Parrots now existing (Psittacus, Coracopsis, PceocepTialus, arid Ayapornis) ; and the forms of the wings and tail point to a similar conclusion. In its general coloration it is decidedly aberrant ; but the fact of its beak being red is also a con- firmation of its Palaeornithine affinities, Prof. Garrod having shown (P. Z. S. 1874, p. 598) that none* but species with normal carotids (a group including Palceornis, Tanygnathus, &c., but not Coracopsis, Psittacus, and Pceocephalus) have their beaks so coloured. We already know that in both Mauritius and Rodriguez a very different genus t of Parrots existed in each island, along with a species of Palceornis, and therefore there is no primd facie reason against a similar state of things having also been the case in Bourbon. On the other hand there is no evidence that Coracopsis ever occurred in a state of nature on any of these three islands. To briefly recapitulate, then, I submit : (1) That the "Perroquet mascarin" of Brisson belongs to a genus, Mascarinus, distinct from Coracopsis. (2) That, failing any older name that can with propriety be applied to it, it may be termed Mascarinus duboisi. (3) That, so far as can be judged from the material that exists, Mas- carinus is allied rather to such Palaeornithine genera as Palceornis and Tanygnathus than to Psittacus, Coracopsis, or allied forms. Cambridge, May 8, 1879. nudis rubris," about which Hasselquist, in his very lengthy description, says nothing so that they are probably only a paraphrase of Brisson's phrase " oculorum ambitu nvdo, coccineo" suggest that his diagnosis was compounded by grafting on part of Brisson's diagnosis an abstract of Hasselquist's description. * Pionus corallinus is the only exception to the above rule that I have yet met with. t Lophopsittacus and Necropsittacus. H2 100 ON BUTTERFLIES OBSERVED IN EntM.M.xv. 18. NOTES ON BUTTERFLIES OBSERVED IN THE P . 275 (1879). VALAIS OF SWITZERLAND IN 1878.* DURING the early part of last summer, in company with my friend, Mr. P. L. Sclater, F.R.S., and party, I spent about three weeks in the Valais of Switzerland, and, during that time, we devoted a considerable part of our energies to butterfly-catching. As the two valleys where we spent the greater part of our time are rather out of the beaten track of tourists, and have possibly not been visited by English entomologists before, I think a few notes on our captures may be worth inserting in this Maga- zine. The valleys visited, the Vals d'Herens and d'Anniviers, are two of the lateral valleys which run from the main chain of the Pennine Alps into the Rhone Valley, debouching into it between the better known valleys of Chamounix and Visp. In the Val d'Herens, we made Evoleua (about 4,500 ft. above the sea) our head quarters, staying there a week, and making excursions thence higher up the valley, including a two days' stay at Arolla, at the head of the valley of the same name, a most lovely spot (about 6,500 ft.), close to the Glacier d' Arolla. and overshadowed by many magnificent snowy Ent. M. M xv peaks, such as Mont Collon, the Dent d'Herens, and others. Unf ortu- p. 276 (1879). nately, the weather at this period was not propitious, but one beautiful day in the Val d Arolla showed what might have been done with better weather. From the Val d'Herens we went over the Col du Torrent (about 8,000 ft.) into the Val d'Anniviers, where we spent a week at Vissoye (about 4,500), and from there, after an ascent of the Bella Tola (10,000 ft.), Sclater descending the other side to the Turtman Thai, returned to Sierre, whence I returned home, via Brieg and Paris, whilst Sclater con- tinued over the Furka and thence home by Lucerne, obtaining thus a few additional species. The time (from about June 25th to July 15th) was probably, on the whole, as good as any we could have chosen, though certainly too early for the highest ranging species. The weather was only moderately good, fine and wet days being in about equal proportion. Nevertheless, we managed to see or capture in that time 83 species of butterflies, and brought back about 600 specimens of these, as well as about 150 Hetero- cera, more than double that number of Coleoptera and Hemiptera. In the present notes, however, I only notice the Rhopalocera, not having as yet determined fully the other groups. I may perhaps mention, to show the abundance of butterflies in the Alps under favourable circumstances (i. e., on a fine day), that one day (July 6th), at Vissoye, we captured (or saw) no less than 45 distinct species, and on two or three other days, * Ent. Month. Mag. xv. pp. 275-278 (1879). THE VALAIS OF SWITZERLAND. 101 both there and in the Val d'Arolla, the number observed exceeded 40. In the following list I have only enumerated the more interesting species, and have followed Dr. Staudinger's catalogue (1871) throughout. Papilio machaon only a few specimens, occurring singly in various localities, but never very high up. Parnassius apollo common about Vissoye, also a few at Evolena and in the Arolla Valley ; this species flies slowly and steadily, but, if alarmed, goes off at a great pace, and is not then easily caught : P. delius this species occurred, flying with the last, in the Val d'Arolla, and was rather abundant on the slopes of the Col du Torrent, above the Val d'Herens ; both it and P. apollo vary much in the size, number, and intensity of the red ocelli and spots ; this species is not found so low as P. apollo, not, according to our experience, much below 6,000 ft. : P. mnemosyne this species occurred with the last in the Val d'Arolla, and on the Col du Torrent ; the curious ovisac, with which, in this genus, the females are provided, is, in P. mnemosyne, much larger than in the other two, and of a white colour; all the three species frequent the wetter slopes on the sides of the valleys : the occurrence of three species of this genus together in exactly the same locality, as was the case at one point in the Val d'Arolla, is certainly very remarkable, considering the resemblance in habits of the different forms. Aporia cratceyi one of the most abundant butterflies in Switzerland, and extend- ing some height up. Pieris napi, ab. $ bryonia* a few in the Val d'Arolla, and on the Col du Torrent and Furka, with males of the ordinary form ; specimens vary considerably in darkness : P. callidice not common in the Arolla Valley, especially at Arolla itself, and a few on the Col du Torrent, Purka, and Bella Tola at high elevations (8,000 ft. or so) ; the flight of this species resembles that of a Colias, more than that of our ordinary English whites. Euchloe belia, var. simplonia this species occurred with the last in the Val d'Arolla, and was rather numerous in the meadows outside the inn there. LeucopJiasia sinapis verv common at Evolena and elsewhere. Colias phicomone a few of this mountain species in the valley at Evolena, but commoner on the slopes around and higher up the valley towards Arolla ; also on the MM Furka and Bella Tola. Tliecla rubi two or three specimens at various p. 277 (1879). places ; on the Bella Tola at about 6,500 ft. Polyommatus virgaurece males very common in the valley at Vissoye, also in the Turtman Thai and near Zinal, but I only saw a single female ; one male from Vissoye is remarkable for being of a yellow colour above: P. hippothoe, var. eurybia this alpine form of liippothoe, from which it differs in the duller colours of the male, and in the female being almost unicolorous brown above, and greyer beneath, occurred sparingly at Evolena, more commonly at Vissoye, where it occurred with P. virgaurece, and also on the St. Gothard route, between Andermatt and Amsteg : P. dorilis, var. two specimens, one at Evolena, the other on the Bella Tola ; 102 ON BUTTERFLIES OBSERVED IN of the alpine form of dorilis, which differs from the type in being unicolo- rous above, and with no redness beneath. Lyccena argyrotoxus (cegon) this blue swarmed in the lower part of the Val d'Arolla, rising in crowds from the muddy water in the roads, &c. P. argus I secured two specimens from amongst the crowds of the last, and doubtless passed over many more ; will somebody explain why, in two such closely allied species, one (cegon) should possess, the other (P. argus) want, the " horn- stachel " on the fore tibiae : P. eumedon rather common at Evolena, and one on the Furka. L. escheri two specimens, one in the upper part of the Val d'Anniviers, coming down from the Col du Torrent, and another at Evolena : L. corydon Evolena, Zinal, &c., and one, remarkable for the brightness of its blue, which approaches that of bellargus (adonis), at about 6,500 ft. on the Bella Tola : L. Jiylas several at Vissoye, and also occurred at Evolena and in the Val d'Arolla, but only one female : L. damon only one, a male, above Evolena : L. minima (alsus) very common ; I was surprised to see our little " Bedford blue," so local an insect in England, in the Alps at elevations of 6-7,000 ft., as at Arolla and on the Col du Torrent ; also on the Eurka : L. semiargus (acis) common at Evolena, Vissoye, Arolla, &c. : L. alcon rather common at Evolena, and one at Vissoye : L. arion occurred at every place we visited. Limenitis populi a single specimen of this fine species in the river-valley at Vissoye sitting on the poplars, and safely secured by Sclater. Melitcea cynthia Sclater secured a single male of this species on the Eurka ; we were probably too early for it elsewhere : M. aurinia, var. merope a single specimen at Arolla, one in the Turtman Thai, and a fine series from the Eurka : M. ri-nxia Vissoye, Evolena, &c., common : M. plioebe Evolena, Val d'Arolla, and common about Vissoye and at Sierre. M. didyma common at Vissoye and Sierre : M. dictynna fairlv common at Evolena, Vissoye, Arolla, &c. : M. athalia common, Evolena, Vissoye, Arolla, &c. : (M. parthenie one at Lausanne). Argynnis pales one at Arolla, and tolerably numerous at high elevations (7-8,000 ft.) on the Col du Torrent and Bella Tola ; also occurred or the Eurka, and between Andermatt and Amsteg ; we did not see the var. arsilache : A. dia one at Sierre : A. amathusia fairly common at Vissoye and Evolena, in damp places along the valleys : A. ino with the last: A. laiJionia one near Vissoye: A. niobe, var. eris one at Vissoye, others on the St. Gothard route ; we did not see paphia or adippe : A. aglaia rather common. Erebia epiphron (the form helamus ?) sparingly in the higher parts of the Vals d'Herens and d'Anniviers ; also on the Bella Tola, Eurka, and the St. Gothard route : E. ceto very common in the valley at Evolena, and a few elsewhere : E. stygne this species also abounded at Evolena, but occurred higher than the last, frequenting, not the meadows near the river, but the rocky slopes at the side of the valley; also sparingly at Arolla, Zinal, in the Val d'Anniviers, and on the St. Gothard ; THE VALAIS OF SWITZERLAND. 103 nearly every specimen is a male : E. gladalis, var. alecto a single speci- En ^- M - M. xv. men of an Erebia which I saw at the end of the Arolla Glacier (about 7,000 ft.), and, after a hard run over the stones of the terminal moraine, secured, turns out to belong to this species : E. lappona this species had the highest range of any, according to our experience ; it occurred tolerably common about the higher slopes of the Col du Torrent, and also on the top of the pass (about 8,000 ft.) ; in descending from the Bella Tola I caught one specimen on the snow at about 9,000 ft., and saw others at nearly the same elevation ; it was common on the Eurka : E. tyndarus this species we only met with on the Col du Torrent, at from about 6,500 ft. upwards ; it was not common : E. ligea ? I believe that several Erebias, from Evolena and Vissoye, are referable to the true liyea: E. euryale being common at Vissoye and Evolena, in the valleys, and a few in Arolla Valley and the Eurka. Chionobas aello I caught two specimens in the Val d' Arolla, at about 6,000 ft., and afterwards another in the Val d'Auniviers at the foot of the Col du Torrent, at about the same elevation ; Sclater found it commonly on the Furka, and secured both sexes. Satyrus hermione this fine species was common about Vissoye, particularly frequenting the slopes of some dry rocky ground in the valley, exposed to the sun ; it has a sailing sibylla-]ikQ flight ; also in the Turtman Thai : S. actcea, var. cordula this species was also common at Vissoye, occurring in the same localities as the last, but nearly all caught were males. Pararge mcera this is one of the commonest butterflies in the Swiss valleys, at moderate heights; it loves to rest on the surfaces of rock overhanging the roads, and on walls, &c., exposed to the sun, starting out moderately like a grayling, the grey under-surface matching well with its chosen haunts : P. hiera we were apparently too late for this species, only securing a few worn specimens in the Val d'Herens, between Evolena and Arolla. Ccenonympha satyrion this pretty little " heath " was common at Arolla on damp ground, but also occurred more sparingly around ArolJa, in the Turtman Thai, on the Col du Torrent, and on the St. Grothard route ; the males were much commoner than the more brightly coloured females. Syricihus earthami not common ; a few in the Val d'Anniviers, at Arolla, Vissoye, and Zinal : S. alveus ? Vissoye and on the Bella Tola, but I am not sure that the specimens are rightly determined, owing to the difficulty of this group : S. cacalice sparingly at Vissoye and Evolena, but tolerably common at higher elevations on the Col du Torrent, the Bella Tola, and Eurka : S. sao one at Vissoye. Nisoniades tages abundant, flying over wet ground, at Arolla. W. A. EOEBES, St. John's College, Cambridge : IQth February, 1879. 104 GLACIAL PERIOD AND GEOGRAPHICAL DISTRIBUTION. Nature/ 19. THE GLACIAL PERIOD AND GEOGRAPHICAL Feb. 20, 1879. DISTRIBUTION.* PEOF. ASA GRAY, in his very interesting lecture on the distribution of the forest trees of the northern temperate region ('Nature,' vol. xix. p. 327), after pointing out the remarkable differences that exist between the forests of the eastern and western sides both of North America and the Old World, suggests that the great poverty of the European as com- pared with the Japan-Manchurian region in this respect was caused by the Mediterranean cutting off the retreat of the flora which then occu- pied Europe, as it retired, at the approach of the glacial epoch, before the ice from the north. This explanation derives considerable support from some other facts in geographical distribution. The most character- istic Alpine and Arctic butterflies of the Palsearctic region belong to the three genera, Parnassius, Chionobas, and Erebia. Of Parnassius, Dr. Staudinger in his latest catalogue (1871) enumerates fourteen Palsearctic species, of which three occur in North and Central Europe, ranging as far south as the Balkans, but always in or near high lands, about a dozen occur in temperate Asia, ranging as far east as the Amur, and probably as many in North America, where they also are truly Alpine butterflies. Of Chionobas one species (C. aello, confined to the Alps) occurs in Central Europe, whilst six or seven others range from Lapland over Russia and Siberia, Mongolia, &c., to the Amur, and there are numerous species in Arctic and Alpine North America. Of Erebia there are forty- five Palsearctic species enumerated by Staudinger, and of these no less than twenty-five occur in the central Alpine chains of Europe. The genus likewise ranges all over temperate Asia, going as far south as the Himalayas and Moupin, and in North America is repre- sented by a dozen or more species. Now, though an Erebia (E. tyndarus, var.) occurs as far south in Europe as the Sierra Nevada, not a single species of any of these three genera occurs in North Africa, although the Atlas Mountains would seem eminently well suited for such Alpine insects. In this case, then, it seems clear that the same cause the barrier of the Mediterranean which in the case of the miocene flora of Europe prevented any further retreat south, has operated to prevent any similar southerly spread amongst the victorious invaders from the north which pressed on the retiring host. With regard to the general similarity in facies and richness between the East American and East Asiatic tree-flora, certain facts pointing in the same direction will at once occur to the zoologist. Thus the Meno- * Nature, Feb. 20, 1879. ON THE UAKARI MONKEYS. 105 pomas of the Ohio and Alleghany have their only near relations in the gigantic Sieboldias o north-east Asia, one species of these occurring in Japan, the other being one of Pere David's discoveries in Moupin. Similarly with the genus Polyodon amongst ganoids. Only two species of this genus are at present known, P. folium, inhabiting the Mississippi, P. (jladius, the Tang-tse-kiang. The recent discovery of at least two species of Scaphirhynclius in Turkestan makes it probable that ere long species of that Americo-Asian genus will be found in the Chinese rivers as well. The parallelism in the case of the salamanders is particularly interesting, when one remembers the celebrated Andrias scheuchzeri of the (Eningen beds, and it tends to favour the view that at that time practical identity in the forms of animals and plants reigned throughout the northern temperate zone. W. A. FORBES, Cambridge, 14tfi February, 1879. P.S. The reported discovery ( 4 Nature,' vol. xix. p. 351) of a true alligator in the Yang-tse-kiang, will, if confirmed, add a still more remark- able case to those mentioned above. 20. ON THE EXTERNAL CHARACTERS AND ANATOMY P.z.S.1880, OF THE RED UAKARI MONKEY (BRACHYURUS p ' 627 ' RUBICUNDUS) -, WITH REMARKS ON THE OTHER SPECIES OF THAT GENUS.* (Plates IV.- VI.) ON May 24th, 1879, the Society purchased a female specimen of a red- haired short-tailed American Monkey, which on its arrival was somewhat doubtfully entered as an example of Bracliyurus rubicundus of Isidore Geoffrey f. This animal lived in fair health till April 22nd last, when it died without any premonitory symptoms. On dissection, both lungs and liver, so frequently the seat of disease in Monkeys kept in captivity, were found to be perfectly healthy ; indeed, with the exception of a little inflam- mation of the stomach and small intestines, and a slight intussusception of the transverse colon, no morbid appearances whatever were found. The death of this animal has enabled me to give that further notice of it promised on its arrival (vide Mr. Sclater's monthly report, infra at.), as well as to give some notes on its anatomy. For though, as might have been expected, Bracliyurus differs in no essential respect from its allies, the great rarity of Monkeys of this genus in captivity makes it advisable to record any facts concerning the anatomy of its soft parts. In * Proe. Zool. Soc. 1880, pp. 627-6-47, Pis. LXI.-LXIIL Bead Nov. 30, 1880. t P. Z S. 1871). p. 551. 106 ON THE UAKAEI MONKEYS. particular, the brain of this genus of Monkeys being hitherto unknown, the description of it will fill up one of the few gaps till now left in our knowledge of this organ amongst the Primates. Our specimen of Brachyurus was a female, not yet adult, though perhaps nearly full-grown. All the teeth are in place, but the canines, both above and below, have not yet finished cutting, and the epiphyses of the bones are still unanchylosed. As regards the name of our animal, I may at once state that, Mr. Blanford having been kind enough to take the skin to Paris for com- parison with the types of Brachyurus rubicundus in the gallery of the Jardin des Plantes, no doubt remains that it really belongs to that species. The specimen from which Isidore Geoff roy's figure * was taken is still . extant in Paris ; and the apparent shortness of its tail, reproduced in the figure, is due in all probability to the " make " of the skin, the skin of P. Z. S. 1880, the tail having apparently shrunk much after the removal of the bones p. 028. inside. Other specimens in the Paris Museum, Mr. Blanford informs me, have tails of about the same length as ours, while they closely correspond in other respects, the amount of grey on the crown of the head varying in different specimens. The accompanying plate (Plate IV.), taken from a sketch made by Mr. Wolf shortly after the animal's arrival, will give a more correct impression of this Monkey than the figures hitherto published of it. The weight of our specimen, considerably emaciated, was 2 Ib. 11 oz. The following measurements were taken on the body before being skinned or otherwise interfered with : Total length (measured in a straight line from super- i nc h eg ciliary ridges, over head and body, to tip of tail) . . 21 Length of tail, including hairs 6'5 Fleshy part of tail 5-65 Length of head, from occipital prominence to glabellum 2-65 From occiput to anterior margin of upper lip, in a straight line 3'65 Breadth of face (just above the eyes, from outer margin of orbits) 2-0 Breadth of nasal septum O75 Breadth of mouth v . . 1*2 Length of arm 5'5 Length of forearm 5-0 Length (extreme) of manus : 3*5 Length of thigh 6-5 Length of leg 5-75 Length (extreme) of pes (plantar surface) 5-25 * Arch. d. Mus. v. pi. 30. ON THE UAKARI MONKEYS. 107 The face, chin, and sides of the head, as far as the ears, are nearly naked, with only a scanty covering of hairs. The nose and the interval between the eyes (measuring ^ inch) are very- nearly bare, there being here only a very few most minute hairs, only seen in a side light. Along the superciliary ridges are a few long, forwardly directed, black hairs, white at the base, and slightly curved forwards. There are a few similar hairs also on each side of the face between the nostrils and the angle of the mouth, as well as on the chin. The muzzle, which is somewhat truncated, and the chin are sparsely covered with short bristly white hairs, directed more or less downwards aud forwards. The naked skin of the sides of the head was in life flesh-colour, the naked ears being slightly redder. The face and muzzle were, as will be seen from Mr. Wolf's sketch (Plate IV.), during life bright vermilion- red, so red, indeed, as to give the animal the appearance of being painted ; but the amount of this bright red varied much from time to time, depen- ding, apparently, both on the animal's health and on its emotions. Mr. Bartlett tells me he has seen the animal flush up, as it were, in a P. Z.S. 1880, moment a brilliant red all over the naked parts of its face, and as soon P- 629 - become pale again when the disturbing cause had subsided. After death the brilliant red colour was confined to the region of the nasal openings and the interval between them and the upper lip *. The eyelashes are represented only by very fine silky minute hairs. The irides were light hazel-brown, the sclerotic white. These points may be well seen in the accompanying figure of the head (Plate V.), of the natural size, drawn soon after death by Mr. Smitf. The ears are of a somewhat squared shape much more so thaii in a specimen of Pitkecia satanas I was able soon afterwards to examine in the flesh with the angles rounded off. There is no lobule ; and both tragus and antitragus are little developed. The helix has a small recurved flap above, lying over the top of the helical fossa. They are quite naked. At a point about corresponding with the top of the occiput there is a parting of the hairs of the head, these radiating forwards, outwards, and backwards from this point the long red hair which covers the sides of the head, passes over and behind the ears, passing outwards and then forwards, whilst the hairs of the back of the head and neck pass backwards. The short silky grey and \vhite hairs covering the top of the head pass directly forwards. These are very fine, and closely appressed to the scalp ; white for the greater part of their length, they become black at the tips ; towards the sides and front of the scalp they become tinged with reddish, so * Of. Bates, ' Naturalist on the Amazons,' ii. p. 310. t In this figure the extent of the red colour of the face has been restored, partly from memory and notes taken from the living animal, and partly from the sketch m ide bv Mr. Wolf. 108 ON THE UAKARI MONKEYS. P. Z. S. 1880, p. 630. p. 629. Hand of Red Uakari. Palmar aspect ; natural size. P. Z.j3.^1880, gradually passing into the red of the sides of the head. This reddish tinge is produced by the presence on these hairs of a rufous zone between the white of their bases and the black of the tips ; a few, however, are black throughout. These short hairs almost entirely disappear a little behind the superciliary hairs. The hairy covering of the scalp ceases along a line between the top of the ears and the top of the orbits. Below this limit the sides of the head are only very sparsely covered with rather long, fine, forwardly-directed hairs, which are mostly rufous, paler at the base, and black-tipped ; here, as elsewhere, however, some are quite black. The skin round the angles of the mouth is, for a small area, almost completely naked. The posterior border of the lower jaw, on the contrary, as well as the sides of the throat, are covered by long ON THE UAKARI MONKEYS. Fig. 2. 109 P.Z.S. 1880, p. 631. Foot of Eed Uakari. Plantar aspect ; natural size. rich chestnut largely black-tipped hairs, which are directed forwards ; these run as far as the symphysis, and form a sort of whiskers. The hairs of the back of the bead, nape, and neck are paler in colour than those on the rest of the body, being pale fulvous, many having, however, black tips, whilst a few are entirely of that colour. The rest of the body is covered with very long, fine, backwardly-directed hairs of a bright rich chestnut colour, as usual more or less black-tipped, with a sprinkling of quite black ones. P. Z. S. 1880, p. 629. 110 ON THE UAKARI MONKEYS. P.Z.S. 1880, In general colour and texture the coat of the Uakari greatly resembles p ' ' that of the Orang, as was noticed by many who saw the animal alive. These red hairs are continued on to the limbs and tail, the hair being particularly long on the arms, about the shoulders (forming here a sort of cape over the back and shoulders), and along the posterior border of the thigh and leg there being a wide patagial-like expansion of skin behind the knee, between the thigh and lower part of the limb. Some P. Z. S. 1880, of the hairs on the back measure over 4 inches ; and those on the limbs p. 632. are rom 3_gi i nc h es j n length*. The tail is pretty uniformly covered with moderately long hairs, and is in no degree, in the fresh state, bushy. It is not flattened or bare beneath. On the forearms the hairs converge on the posterior margin to the elbow, being directed backwards and more or less upwards as the elbow is approached. On the anterior margin, however, all the hairs are directed downwards and forwards, so that a parting of the hairs runs down here from the elbow to the wrist. On the posterior limbs this is not observable. The lower surface of the body is much less hairy than above ; and there is a well-defined median parting running along it from the thorax to the abdomen. The greatest length of the hand (fig. 1, p. 108) is 3-5 inches, the breadth across the knuckles about 1*35. The fourth digit is the longest (2-15), being 0-15 inch longer than the third ; the second and fifth are about equal (1*75). The thumb is, as in other Cebidse, directed in the same plane as the remaining digits ; it measures 1/15 inch in length, reaching slightly beyond the first phalanx of the second digit. The interdigital membrane is slight. The nails are compressed, and rather elongated, particularly on the fifth digit ; that of the pollex is shorter, and more compressed and "nail "-like. The palmar surface is nude; the dorsal sparingly hairy, the hairs extending on to the fingers. The greatest length of the nude plantar surface is 5*25 inches ; its breadth, at the base of the hallux, is 1*5 inch. The toes have about the same relations as the fingers, except that the fifth is notably longer than the second. The fourth is about 2-25 inches long. The hallux measures 1*25 inch, and has a broad oval nail, slightly compressed ; the nails of the other digits are much compressed, slightly curved, and rather claw- like. The inferior aspects of the hand and feet, of the natural size, are represented in the accompanying figures (fig. 1, p. 108, and fig. 2, p. 109)t. * One of the characters of Dr. Gray's genus " Ouakaria" is "Fur short, silky" ! (Oat. Monk. p. 61). f In fig. 2 the second digit of the foot is seen to be twice bent abruptly on itself. This position was constant after death, and if altered it was always returned to. I am not, however, prepared to say that it is natural, as I never observed it in the living animal. The other digits showed no trace of it. ON THE UAKARI MONKEYS. Ill The axillary folds are well marked ; and the axillae are quite nude, as is a space continuous with them on the inner aspect of the arm for nearly one third its length. The teats, two in number, are situated about 1| inch from the middle line, just on a line between the ends of the axillary folds *, about 1-25 inch from the apex of the axilla. The umbilicus is represented by a slight, scarcely perceptible slit-like mark 4| inches from the pubic symphysis. The anus is a transverse slit; the vulva, which is provided with a short clitoris, is vertical. The P. Z.S. 1880, perinaeum measures about -35 of an inch. The ischiatic prominences, p> perinaeum, and root of the tail are covered by greyish skin forming a rhomboidal space, about 1*5 inch broad and high ; under the tail there is a slight hollow, with a raised fold of skin at each side. The skin of the cheeks is thin and smooth throughout inside. The hard palate has about ten slightly curved (lunate) ridges on each side, best marked anteriorly, and not meeting mesially. The first two lie between the canines, the last on the level of the last molar. The more posterior ones are faint and irregular, and straighter ; the two most anterior the strongest and most curved. In front of the most anterior are two small slits, one on each side of the centre, directed antero- posteriorly, and lying in a line with the inner margin of the median incisor. The tongue is elongate and parallel-sided, being bluntly squared off at the tip. In front of the palato-glossal folds, which are well developed, it is covered, above and on the sides, with filiform papillae ; below it is smooth. The fungiform papillae are numerous, and distributed over the sides and tip of the tongue in front of the circumvallate papillae ; of these there are four, arranged in the usual reversed ^-shape, the extra one lying on the right side. In Pithecia satanas I found three only. There is a " Mayer's organ " of about 15 slits, in the usual position in front of the palato-glossal folds. There is a frenum linguae, and a smooth, fleshy, well-developed sublingua, bifid apically, with the duct of the submaxillary glands opening on the two papillae behind this. The uvula is blunt and feebly developed. All the salivary glands are well developed. The parotid is large, measuring 2 inches across at its greatest development ; it occupies part of the "anterior triangle," sending a lobe up and behind the auditory meatus ; it then runs forward over the masseter muscle, the superior border coinciding with the zygoma, as far as its anterior border, where on one side there is a small down- wardly directed lobule developed. Below it extends far into the fossa behind the jaw, and is in contact beneath with the submaxillary. The duct opens opposite the last premolar. * In a female of Mycetes seniculus, examined some years ago by Prof. Garrod and myself, the mammae were found to be situated in the axillae. 112 ON THE UAKARI MONKEYS. The submaxillary glands are also large ; in contact with the parotids above, they nearly meet each other below. A few small accessory lobules appear superficially towards the anterior part. The glands are covered to a large extent by the jaw, running up on the deep side of the ascending ramus of the mandible, and covering there the digastric muscle ; at the angle of the jaw they appear superficially. The sub- lingual glands, well developed, extend back in the floor of the mouth for 1 inch behind the subllngua. On opening the abdomen, the great length and narrowness of the abdominal cavity are striking. The caecum occupies super ficuilly nearly all the posterior part of the abdominal cavity, filling up thus nearly one third of the whole. Behind, it rests on the bladder, covering the rectum ; its apex, directed downwards, lies in the right iliac region. The descending colon is quite superficial and lengthy, as is the ascen- ding, which is also superficial, except in the middle ; the transverse, P.Z. S. 1880, on the contrary, is very short*. (It was partly intussuscepted.) The p. 634. stomach was visible in the left hypochondriac region, the liver appearing all across the abdomen. The great omentum did not cover any of the viscera as now exposed ; it was attached only to the upper part of the ascending colon, for about 2 inches. It contained no fat, the animal being, it is to be remembered, considerably emaciated. The stomach is of the usual Simian form, with a globular cardiac cul-de-sac, and fairly distinct tubular pyloric part ; it measured 3 inches in length by 1 J deep. The pyloric constriction is distinct ; and towards that part the walls become thicker. Internally there is a distinct thick ridge on the lesser curvature, f inch to the right of the oesophagus, dividing off the pyloric part, which is quite smooth internally, whilst the mucous membrane of the cardiac part has a few irregular, slightly developed rugae. The length of the intestines is as follows : inches. Small intestine 103-5 Large 19-0 Caecum (distended) 6-0 I append (p. 113) a few measurements of the alimentary canal of other species of Cebine Monkeys for comparison. P. Z. S. 1880, As far as can be judged from the few examples given in this table, p. 635. Brachyurus rubicundus apparently has a greater absolute, and even greater relative, length of intestines and caecum than any other New- world Monkey, including even the considerably larger Lagoihrix. This * In Pithecia satanas the transverse colon hardly exists, the descending colon being bent sharply back upon the ascending. The caecum lay altogether to the right of the descending colon and rectum ; the latter, therefore, was not hidden by it. ON THE UAKARI MONKEYS. 113 Name. Sex. Length of Authority. pij 1 M ^1 s~ 02 i! j) Caecum. Aieles geoffroyi (nearly adult) in. 15 13 is 13 11 12-5 16 15-5 in, 95 99 925 58 56-5 102 37 29-5 48 42 50 52-5 68 108 103-5 in. 17 15 13 11 8-5 19 12 12 12 11-5 22 12-75 9 125 19 in. 3-5 4 2-75 2 1-75 4 2 1-5 2 2-5 4-5 2 2-75 2-75 6 Prof. Flower *. A. H. G-arrod (MS.). W. A. F. Prof. Flower *. W. A. F. A. H. Garrod (MS.). W'.'A. F. A. H. Garrod (MS.). W. A. F. 3 rof. Flower*. W. A. F. W. A. F. W. A. F. W. A. F. (young) . Cebus capucinus d 1 S c? 3 2 d 1 2 Lagothrix humboldti Nyctipithecus vociferans. . rufipes Callithrix amicta Pithecia monachus (nearly adult) (young) (not adult) . . . Brachyurus rubicundus (not adult) P. Z. S. 1880, p. 634. would seem to indicate that in its native forests Brachyurus is more of a p. z. g. 1880 vegetarian than its allies. There are no valvulae conniventes. Peyer's patches are almost con- fined to the last yard of the ileum, there being 14 large and well- defined ones in that space, the largest | inch long. There is a large one close to the ileo-caecal aperture. Higher up only a few can be found ; and they disappear in the jejunum. The caecum is in no degree sac- culated, neither is the colon. The former, which is of considerably larger calibre, is cylindrical, blunt, and curved on itself when distended into more than a circle, with a well-developed median peritoneal fraenum. The general form and proportions of the liver may be seen in figs. 3 & 4 (pp. 114, 115), representing respectively the diaphragmatic and visceral (superior and inferior) aspects of that organ. Regarding its form in the Cebidse we are told by Prof. Flower, in his lectures on the digestive system of the Mammalia (Med. Times and Gaz., May 4, 1872, p. 509), that " all the members of this group in which he has been able to describe the liver, agree in the depth to which the lateral fissures cut up the organ into its four principal lobes, and also in the great develop- ment of the caudate lobe, which is the principal character by which they can at once be distinguished from the Old-world families." In both these respects, as will be seen by an inspection of the figures, the liver of Brachyurus is perfectly Cebine. The two lateral fissures extend very nearly, at least on the superior surface, back to the posterior (attached) margin of the liver. The four principal lobes are very distinct ; the left p. 635. * Taken from his lectures, Med. Times &c.. May 4, 1872. 114 OJS T THE UAKARI MONKEYS. P.Z.S. 1880, p. 636. central is markedly smaller, as is often the case in the Cebidae, than the three others, u hich are here all very nearly equal. The left lateral lobe is decidedly thin along its outer margin ; the other lobes are thick, and of simple form ; the right lateral is much longer anterio-posteriorly than trans- versely. The caudate is large and square ; on its visceral surface it is marked by a conspicuous diagonal ridge running from the entrance of the vena cava towards its postero- external angle. It develops two well-marked though small fissures, but is otherwise simple. The right lateral lobe appears internally to it, when viewed from below. The Spigelian is an elongated, somewhat clavate thickening, which is not free, but is most de- veloped on the left side. The umbilical fissure is well-marked, extending for about one third of the total median depth of the liver. There is no trace Fig. 3. V.C.&YH V.C. Liver of Bed Uakari, from above. About natural size. P. Z. S. 1880, of a cystic notch ; and the gall-bladder, which is large, does not reach by p. 635. some little distance the anterior (free) margin of the liver. It lies very superficially, and, as in most of the CebidaB (though not in Cebus itself or in Ateles), lies very close to, and almost in, the umbilical fissure. An accessory lobule, developed at the internal angle of the left central ON THE UAKATU MONKEYS, 115 lobe, helps in large part, on this side, to form a shallow cystic fossa. A second similar, but smaller, partly free lobule is also developed by the P. Z. S. 1880, left lateral lobe at its antero-internal angle, The vena cava inferior is almost entirely bridged over by hepatic tissue between the Spigelian and caudate lobes. The development or otherwise of this bridge, however, is an unimportant feature in the liver of the Primates, and is largely an individual structure. On the whole, this liver of Brachyurus rubicundus more resembles that of Gallithrios amicta in its structure and proportions than that of any other species I am acquainted with, though I have not seen that of Pithecia monachus *, which, judging from Prof. Flower's description t, must much resemble that of Brachyurus. In two specimens I have by P.Z.S. 1880, p. 637. The same, from below. E.G. Right central lobe. L.O. Left central lobe. E.L. Right lateral lobe, L.L. Left lateral lobe. C. Caudate lobe. Sp. Spigelian lobe. u.f. Umbilical fissure. r.l.f, U.f. Right and left lateral fissures. G.B. Gall-bladder. V.P. Vena porta. V.H. Hepatic vein. V.C. Vena cava inferior, s.l. Suspensory ligament, l.r. Ligamentum rotundum. * P.S. Jan. 27, 1881. The recent death of a specimen of this ppeoies allows me to confirm the resemblance pf its liver to that of Brachyurus. t P. Z. S. 1862, p. 332, and Med. Times, &o. s. c. p. 508. 12 116 ON THE UAKARI MONKEYS. me of the liver of P. satanas, the caudate lobe is triangular and forked instead of square, and the left lateral lobe is proportionally smaller P. Z. S. 1880, than in Brachyurus ; this develops moreover a very marked, subtri- p. bd7. angular, inwardly directed process at its antero-internal angle, which is not seen in the other genus. The pancreas is fairly compact, and is, at its greatest extent, 3 inches long ; its duct opens, with that of the bile-duct, \ inch from the pylorus. About J an inch from the common aperture of these in the duodenum is a smaller opening, apparently that of a secondary pan- creatic duct. The spleen is elongated and flattened, and attached to the left of the greater curvature of the stomach ; it is of a red colour, and 2 J inches long. The great vessels are given off as in Man and the other Primates. The external and internal iliacs are given off from a common trunk ; the caudal artery is small. The lungs have two lobes on the left and three on the right side *, 638 besides an azygos lobe, which lies behind the heart in a transverse direction, and is triangular in shape. The left lower lobe is the largest, the right lower the next. The cricoid cartilage is deeply notched above. The vocal cords are well developed, *35 of an inch long ; the ventricles have the shape of watch-pockets, extending downwards and inwards. There are no laryn- geal pouches. The epiglottis is notched above. There are about 27 tracheal rings. The kidneys are smooth and rather elongated in shape; the supra- renal bodies are well developed and oval. The ureters open '75 inch from the neck of the bladder. The ovaries measure *45 inch, the Fallopian tubes 1 inch. The uterus is single and smooth within ; it is '75 inch long. The vagina is double that length ; and the clitoris is small and flattened. As regards the osteology, already briefly described by Gervais (" Ee- marques oste'ologiques sur les genres Brachyure et Callitriche ") in Castelnau's 'Expedition dans 1'Amerique duSud' [Paris, 1855, Ana- tomic, pp. 93-99], there is not much of interest to add. The vertebral formula of my specimen of Brachyurus is C. 7, D. 13, L. 6, S. 4, C. 15 or 16t. Gervais gives C. 7, D. 14, L. 6, S. 4, C. 17 for B. rubicundus -, the * In a 5 Pithecia satanas the right lobe was only two-lobed, the extra superior lobe being only indicated and not cut off. t In my specimen the few minute terminal caudal vertebrae have unfortunately been laid aside, and cannot now be found. There were not, however, more than three or four of them, which, added to tbe twelve that remain, give the above numbers. ON THE UAKARI MONKEYS. 117 British Museum Catalogue 7, 13, 6, 3, 18, for both B. calvus and B. melanocephalus. In the latter skeleton (" OuaTcaria spixii" 806 6) I counted, however, nineteen or twenty caudal vertebrae, the first five of which bear transverse processes, whilst the terminal ones are very minute and styliform. The carpus has an os centrale, and the humerus a well-marked supra- condylar foramen. The clavicles are well-developed, and strongly curved sigmoidally. The manubrium sterni is broad. I may, however, take this opportunity of pointing out a useful means of discriminating, in most cases, between the skulls of the Platyrrhine and Catarrhine Monkeys, in addition to the well-known differences in their dentition and in the form of their external auditory meatus. In nearly every skull of a New-world Monkey, it will be found that the parietal and malar bones are in contact with each other, for a more or less considerable extent, on the side walls of the skull (vide fig. 5). In the Old-world Monkeys, on the contrary, this contact never (with the exception named below) takes place, the frontal and alisphenoid bones articulating with each other, and so cutting off the connexion between the parietals and malars (vide fig. 6). In the skulls of the genus Hylobates that I have examined this isthmus is very narrow, so that the parietals and malars approach much nearer each other than is usually the case in the Catarrhini ; indeed, in one specimen in the College of Surgeons Museum (5027s) the malar and parietal of one side only touch each other for a very short distance, the frontal and ali- p z g sphenoid not meeting. On the other hand, in all the New-world P- 639 Monkeys' skulls that I have examined, the arrangement above described obtains, except in some skulls of the genera Mycetes and Ateles. Thus Fig. 5. Part of side walls of skull of a New-world Monkey (Brachyurus rubicundus). The parietal (Pa) and malar (Ma) articulate, as in other members of this group. 118 ON THE UAKAKI MONKEYS. Fig. a The same parts in an Old-world Monkey (Cercopithecus pyrrhonotus), showing the parietal and malar separated from each other by the intervention of the frontal (Fr), alisphenoid (Al\ and squamosal (Sq), which are shaded obliquely. of five skulls of the first genus in the Royal College of Surgeons Mu- seum, in one the sutures are invisible on account of age, whilst in the P. Z. S. 1880, remaining four the union takes place in one only on both sides, and not p. 540. a j. a jj ' n fa Q remaui i n g three. In the same collection, a single skull of an Aides (471 7 a) also shows no trace of this union. In all the remaining genera, so far as I have yet seen, the rule holds good. I was first struck with the arrangement here described when ex- amining the collection of Monkeys' skulls in the Cambridge Museum ; and finding that there was no exception whatever, either there or in the skulls belonging to the Prosector's department, I examined the entire collection of unmounted skulls in the College of Surgeons Museum (in- cluding nearly every known genus of Monkey), with the results already mentioned. The character is at all events worth knowing for practical purposes, even if of no greater scientific value. This, of course, must be left open for more extensive examination *. The brain of Brachyurus rubicundus is represented in the accompanying figures (figs. 7-10, pp. 120, 121), which give views of its superior, in- * P.8. Jan. 27, 1881. My views have been both confirmed and anticipated by Dr. Gustav Joseph, in a paper in the first volume of the ' Morphologisches Jahrbuch ' (i. pp. 453-65, Taf. xv.). Whilst my paper was going through the press, Prof. Flower was kind enough to call my attention to this paper, as well as to another by the same author in the ' Bericht der Schlesischen Gesellschaft ' which, as yet, I have not been able to see both being referred to in a recently published ethnological paper (in Eussian) by Demetrius Arnoutchine, which also, apparently, contains some more information on the same subject. ON THE UAKARI MONKEYS. 119 ferior, external, and internal aspects, of the natural size, drawn after the organ had been hardened in spirit for a short time. The total length of the hemispheres is 2-3 inches, their greatest breadth 1*8 inch, whilst the vertical depth is about 1'25 inch. Viewed from above, the hemispheres have a fairly rounded contour, and the cerebellum, does not project beyond their posterior margin, though it appears above in the middle line between the somewhat cut-away inner margins of the occipital lobe. From the side, the hemispheres are seen to be but slightly arched. The occipital lobe is well developed, and the orbital surfaces but little excavated. The temporal lobes are also well developed. The hemispheres possess the most important sulci characterizing the Simian brain well developed ; as regards their complexity, they stand between Ateles, Cebus, and Lagothrix, on the one side, and Callithrix, Mycetes, Piihecia, &c., on the other. The Sylvian fissue * (s.) is well developed, running upwards and back- wards to end, *3 inch from the middle line of the hemispheres, a little in advance of the spot where the temp oro- occipital sulcus (scissure occipitale externe) first appears externally. It is at first slightly concave forwards, p.z.S. 1880, then convex. At about two thirds of its course it is joined by the well- P- 641 - marked antero-temporal sulcus (ci.t.) (scissure parallele, Gratiolet; premier sillon temporal, Broca), which commences near the lower margin of the temporal lobe. This is a condition not found in Ateles, LagotJirix, Mycetes, or Pithecia monachus ; it is represented by Gratiolet as existing in Cebus capucinus (Mem. Plis. Oer. Atlas, pi. x. figs. 7 & 8), and likewise occurs in Cynocephalus, Macacus, and other Old-world genera. Embracing the upper extremity of the Sylvian fissure is a somewhat Y-shaped sulcus, the " stem " of the T being short and springing from the median line, whilst the two arms are much longer and run down- wards and outwards, and in the case of the anterior one forwards as well, on to the external surface of the hemispheres, appearing there one on each side of the upper part of the Sylvian fissure (fig. 9). The posterior of these is, no doubt, the aforesaid temporo-occipital sulcus (t.o.) (scissure perpendiculaire externe of Gratiolet, scissure occipitale externe of Broca). The anterior limb no doubt corresponds to the anterior part of * In the following description of the sulci, &c., I have in the main followed the nomenclature proposed by Prof. Huxley in his valuable paper on the brain of Ateles paniscus (P. Z. S. 1861, pp. 247-260, pi. xxix.), and adopted by Prof. Flower in his de- scriptions of the brains of Mycetes seniculus (P. Z. S. 1864, pp. 335-338, pi. xxix.) and Pithecia monachus (P. Z. S. 1862, pp. 328-331). The late Dr. Paul Broca has more recently written an elaborate article on the subject of cerebral nomenclature (" Nomen- clature Cerebrale, denomination des divisions et subdivisions des hemispheres et dee anfractuosites de leur surface," Kevue d'Anthropologie, (2) i. 1878, pp. 193-236). In this he endeavours to limit more strictly than has hitherto been done the terms used by various writers on the structure of the brain, and to introduce a uniform nomen- clature. I have, where necessary, added his names in brackets after those here used. 120 P.Z.S. 1880, p. 642. ON THE UAKAEI MONKEYS. Fi. 7, Brain of Bed TJakari, seen from above. Fig. 8, Bight half of the same, seen from below. P.Z.S. 1880, the " sulcus (4) bounding the upper border of the angular gyrus, having p. 641. the form of a broad pointed arch," described by Prof. Flower in Pithecia monachus (I. c. p. 330). A comparatively slight modification of the con- dition of these parts in that species as represented in his fig. 1 would bring about that \vhich obtains in Bracliyurus, which also is represented in Gratiolet's figures of Cebus capucinus and C. apella (1. c. pi. x. figs. 7 All). This " supraangular " (a .a.) sulcus defines anteriorly the well- ON THE UAKARI MONKEYS. Fig. 9. 121 P. Z. 8. 1880, p. 643. The same, left half, seen from the outside. Fig. 10. The same, from the inside ; the cerebellum and medulla have been removed. All the figures of the natural size. s. Sylyian fissure ; p.p. Postero-parietal ; s.f. Supero-frontal ; s.a. Supra-angular ; a.t. Antero-temporal ; t.o. Temporo-occipital ; c.m. Calloso-marginal ; c. Calca- rine ; cl. Collateral ; o.p. Occipito-parietal. developed angular gyrus. The union of the temporo-occipital and ' g^ occipito-parietal (the two parts of the scissure occipitale of Broca) sulci divides off perfectly the occipital and parietal lobes ; so that there is here no such superficial " pli de passage " between these two lobes as exists in Cebus capucinus and Pithecia satanas. In P. monachus there is, in addition, a second, more superior passage-fold between these two lobes (Flower, I. c.)*. In Cebus apella (according to Gratiolet's figure, I. c. pi. 10. figs. 12, 12 bis\ as in Brachyurus, the "pli de passage " is concealed, so that superficially the two lobes appear perfectly distinct. Anterior to the supraangular sulcus, the fissure of Rolando (p.p.) (postero-parietal, Huxley) is seen as a well-developed sulcus forming a * As regards this and sundry other differences between Pithecia monachus and P. satanas, it must be remembered that the two are not perhaps strictly congeneric, P. satanas having been separated, with P. ckiropotes, as a genus Chiropotes. 122 ONTHEUAKARI MONKEYS. sigmoid curve extending very nearly to the middle line. Anterior to this is the superofrontal sulcus (s.f.), also curved, though roughly parallel with the external border of the hemispheres. The orbital surface of the hemispheres is also marked by a somewhat H-shaped complex sulcus (indsure en-H.). The occipital lobe is almost entirely smooth ; below a sulcus is observable, curving upwards, and ending in a bifurcate manner in front of the lower termination of the occipito-temporal sulcus. Pos- P.Z. S. 3880 teri 01 to tn i s * 8 a ver J mu ch smaller, short sulcus, easily seen internally, p. 642. but only just appearing on the inferior margin of the lobe externally. The internal surface of the hemispheres has the ordinary sulci well displayed. The calloso-marginal sulcus (c.m) (scissure sous-frontale), somewhat broken up anteriorly, inclines posteriorly towards the margin P.Z. S. 1880, of the hemispheres, but does not reach it by about *1 inch. The oceipito- p. 643. parietal fissure (o.p) (scissure perpendiculcdre interne) is distinct, inclined first backwards, and then abruptly bent forwards for a small distance. Between its termination and the posterior end of the corpus callosum is a faintly impressed triradiate mark. The calcarine sulcus (c) is well de- veloped, and terminates posteriorly by its characteristic fork ; the point of division is *15 inch from the margin of the brain ; and the upper arm is slightly longer than the lower. The sulcus is continued forwards with an upward concavity to the end of the corpus callosum, where it passes into the dentate sulcus *. At a point *3 inch in front of its bifurcation P. Z. S. 1 880, it is joined by the collateral sulcus t (cZ), which passes forwards and p. 644. slightly outwards, so appearing on the external face of the temporal lobe (fig. 9). The temporal lobe has, in addition, a slight impression anterior and internal to this, but is otherwise quite smooth below. The corpus callosum is well developed ; it is nearly 1 inch in total length ; the prsecallosal part of the cerebrum is 0'45 inch, the postcallosal 0*9 inch in length. The cerebellum is well developed; the superior vermis is 0-85 inch in * According to Broca (Revue d'Anthrop. viii.p. 470, 1878) the bridging convolution between the callosal and uncinate convolutions, which here interrupts the apparent continuation of the calcarine into the dentate sulcus, is nearly always present in Mon- keys, although it may be deep and in some cases extremely delicate also. He considers, however, that it is wanting in the Hapalidse. It is certainly present in Sracht/urus, though small and deep. Prof. Flower says of Mycetes (I. c. p. 337), " the dentate sulcus is continuous with the calcarine." On the other hand, in Ateles (cf. Huxley, I. c. p. 255, fig. 2) it is superficial and obvious. Broca also figures its existence in Lago- thrix (I. c. p. 471, fig. 31). t This junction, at least superficially, of the calcarine and collateral sulci I also find exists in Cebus capucinus (cf. also Gratiolet's figure, Atlas, pi. x. fig. 9) and Pithecia satanas in Ateles and Nyctipithecus, &c. , however, it does not exist, as far as I have seen, nor do I find any allusion to such a condition existing at all in any of the memoirs already quoted on the brain of Primates. ON THE UAKAEI MONKEYS. 123 length ; with the lateral lobes it is altogether 1*25 inch transversely. The flocculi are large, and the pons Varolii well developed. The medulla has distinct olivary bodies. In the general characters of its brain, as will be thus seen, Brachyurus approaches most nearly the genera Cebus and Piihecia (including Chiro- potes), and especially the type found in Cebus apella. With Ateles and Lagothricc it has no close relationship ; and the same may be said as regards Mycetes ; it also departs widely in the greater complication and development of the brain from Callithrix, Chrysoihrix, and the smaller Cebidse generally. Eeviewing the facts already stated as to the structure of Brachyurus rubicundus, it is evident that Brachyurus is a perfectly good genus, more or less intimately related to a number of the larger Cebine forms, but nevertheless characterized by a combination of characters peculiarly its own. A relationship to Mycetes, suggested by its external appearance and the form of the lower jaw, is not borne out by its visceral anatomy, the brain and liver both pronouncing decisively against the idea, besides other peculiarities. With Ateles and Lagothrioc it has no particular features in common, but it undoubtedly approaches Cebus in the structure of its brain ; and it is with this genus and Piihecia (including Chiropotes) that it has probably the nearest affinities. The institution of a genus Ouakaria for the reception of these short- tailed Monkeys by Dr. Gray (P. Z. S. 1849, p. 9) seems to me unneces- sary. As he there remarks, the genus Brachyurus, as originally proposed by Spix (Sim. et Vespert. Bras. p. 11), contained two species, B. israelita (I. c. pi. vii.) and B. oualcary (pi. viii.). The former of these is now generally referred to the genus Piihecia, standing as Piihecia chiropotes (cf. Sclater, P. Z. S. 1871, p. 228). Having examined skulls of all the known species o Brachyurus, as well as of Pithecia satanas, which is p merely a representative form of P. chiropotes, I am unable to agree with 'p. 645. Dr. Gray (I. c.) as to Spix " having evidently described the teeth &c. of his first species in his generic character," for I find the characters there given apply equally well to both the forms under consideration. Indeed, as the " character essentialis " of the genus Brachyurus is " Cauda non volubili, abbreviate^" B. israelita, in which that organ is of the normal length, can in no way be considered the type of the genus. With certain other points, too, of the " descriptio " there given, B. oualcary corresponds better than the first species. I therefore agree with Isidore Geoffrey (Exped. Am. Sud, Mammif . p. 18) in retaining the generic name Brachy- urus, of which Oualcaria thus becomes a synonym *. * The name Brachyurus has also been used, but erroneously (cf. Sclater, 'Ibis,' 1877, p. 260), for the Ant-Thrushes (Pitta). Mr. Alston ha?, since this paper was read,' 124 ON THE UAKARI MONKEYS. As regards the species of this genus, there are three well-marked ones, each inhabiting a distinct geographical area ; of each of these I have seen skins and skulls. They may be arranged as follows : a. Facie nigra. 1. BEACHYUEUS MELANOCEPHALTJS. (Plate VI.) Simia melanocepliala (Cacajao), Humboldt, Rec. pi. xxix. p. 317 (1811). Pithecia melanocephala, Geoffr. Ann. Mus. xix. p. 117 (1812). Brachyurus ouakary, Spix, Sim. et Vesp. Bras. p. 12, pi. viii. (1823). OuaTcaria spixii, Gray, P. Z. S. 1849, p. 10, fig. Ouakaria melanocephala, Gray, Cat. Monkeys &c. p. 62 (1870). Pithecia melanocephala, Schlegel, Mus. P.-B. vii. p. 227 (1876). Ater, dorso lateribusque cum brachiis posticis plus minusve castaneo- brunneis. Hob. Forests traversed by the Casiquiare and Rio Negro (Humboldt) ; forests between the Solimoes and lea (Spix} ; Marabitanas, Rio Negro, and Moura, Rio Branco (Natter er, fide Pelzeln apud Schlegel, I. c.). This species is at once distinguishable by its black face, as well as by the black hands, feet, &c. It is the most northern form of the three, and apparently the most wide-spreading also. No doubt it is the " black- faced, grey-haired " species heard of, but not obtained, by Mr. Bates as being found " 180 miles from the mouth of the Japura " (Nat. Amaz. ii. p. 313). Of this species we have, at different times, had two specimens living P.Z.S. 1880, in the Gardens (vide Sclater, P. Z. S. 1870, p. 1). From the first of p. 646. these the water-colour drawing by Richter in the Society's possession, which is here reproduced on a diminished scale (Plate VI.), was taken. In all probability it is the skin and skeleton of this individual which are now preserved in the British Museum. [The stuffed skin is marked " Zool. Soc.'s collection," the skeleton 806 &.] b. Facie rubra. 2. BRACHYTJEUS CALYUS. Brachyurus calvus, Isid. Geoffr. C. R. xxiv. p. 576 (1847), et Arch. Mus. v. p. 560 ; Exped. Arner. Sud, Mammif. p. 17, pi. 4. fig. 1 (1855). Oualcaria calvus (sic), Gray, P. Z. S. 1849, p. 10. pointed out to me that the same term also had been proposed some years previously to Spix by Fischer (' Zoognosia.' i. p. 24) for a genus of Eodentia (1813). As, however, this name has never, I believe, been adopted for use in that group, it seems to me quite unnecessary, on that ground, to reject the name for the TJakaris. Purists will have, I suppose, to adopt Lesson's barbarous term, proposed as a subgenus, Cacajao (Species des Mammiferes, p. 181, 1840). PI IV. P. Z.S.1880P1.LXI. lith. ,. imp B.RACHYURUS EUBICUNDUS D.PLLXtl. J.Smib.liLh. HEAD OF BRACHYURUS x; CG CC ON THE UAKARI MONKEYS. 125 OuaJcaria calva, Gray, Cat. Monk. p. 62 (1870). Pithecia calva, Schl. Mus. P.-B. vii. p. 228 (1876). Pithecia alba, Schl. Mus. P.-B. vii. p. 229 (1876). Corpore fulvido-albicante, subtus saturatiore. Hob. Opposite Fonteboa (Castelnau $ Deville) ; banks of the Japura delta, west of its mouth {Bates). According to Castelnau (I. c. p. 567), B. calvus is confined to the forests lying on the north bank of the Amazons, between the rivers Putumayo (or lea) and Japura. The locality " Para," given to the species by its discoverer M. Lisboa, and also marked on the mounted specimen in the British Museum, is of course a mistake, as already pointed out by Schlegel (1. c. p. 226). Mr. Bates's notes on this species and the next, and their distribution, are well known to naturalists. On his short description and the figure in the second edition of the * Naturalist,' Prof. Schlegel has attempted to found a fourth species, " Pithecia alba." But, in the first place, the short de- scription given, as well as the locality, suit B. calvus quite well ; and, secondly, there is a specimen in the British Museum, purchased of Stevens, which in all probability was one collected by Mr. Bates himself, and is quite the same as three other specimens of that species. 3. BRACHYUBUS RUBICUNDUS. (Plates IV., V.) Brachyurus rubicundus, Isid. Geoffr. & Dev. C. B. xxvii.p. 498(1848); Geoffr. Arch. Mus. v. p. 564, pi. 30 ; Exped. Am. Sud, Mamm. p. ] 9, pi. 4. fig. 2 (head). OuaJcaria rubicunda, Gray, Cat. Monk. p. 62 (1870). Pithecia rubicunda, Schlegel, Mus. P.-B. vii. p. 228 (1876). Corpore castaneo-rufo, collo pallidiore. Hab. Forests on the north bank of the Amazons opposite Olivenca, not passing eastwards of the lea (Castelnau). The exact westward extension of this species still remains unknown. The young specimen seen at Fonteboa by Bates (I. c. p. 313) and by him referred to this species, was more probably B. calvus, as we know, from Geoffroy and Castelnau's account, that the young of B. rubicundus resembles in coloration the adult, and is not paler. Opposite Fonteboa, p z S 1880 moreover, is exactly the locality where the French expedition obtained P- 647. B. calvus, and is well within the limits assigned to that species by Castelnau. B. rubicundus is the western representative of B. calvus, which it very closely resembles, but can at once be distinguished by its very different coloration, being nearly all over of a rich deep chestnut, only paler on the neck, instead of the pale sandy- white, slightly rufous below and on the inside of the limbs, of the last species. The fact of the series of individuals of each of these species obtained by Castelnau and Deville differing to no 126 ON THE UAKAEI MONKEYS. Fig. 11. Map of part of the basin of the Amazons, to show the distribution of the Uakari Monkeys. Supposed area of B. melanocephalus. B. calvus, B. rubicundm. important extent amongst themselves, as well as their different ranges, clearly show that, so far as our knowledge yet goes, there is no reason whatever for considering B. calvus an albino form of B. rubricundus moreover in B. calvus a considerable number of hairs on the back &c. are black throughout, just as in B. rubicundus ; this would hardly be the case in an albino. Gray's arrangement of these species in the 4 Catalogue of Monkeys ' has already been dealt with by Schlegel : " II semble, d'ailleurs, que ce savant se soit forme une idee a lui propre de ces animaux, puisque, apres avoir mis les OuaJcaria rubicunda et calva dans une catee;orie-a part, laquelle porte en tete : ' pelage blanchatre ou rougeatre,' il ajoute : ' albinos de Y OuaJcaria melanocephala.' On avouera que, ceci est trop fort pour le directeur d'un des plus grands Stablissements de Zoologie" (Mus. P.-B. p. 229.) ON ANTILOCAPRA AMERICANA. 127 21. ON THE CAUSE OF DEATH OF A LEOPARD.* P. z.^1880, MR. W. A. FORBES exhibited a small fragment of bone which had caused the death of a Leopard (Felis pardus) in the Society's Menagerie on April 20, under the following circumstances : For about a week previous to its death the animal, a fine adult male, had refused food, and, having been separated from its companions, was noticed by the keeper to be apparently suffering from some intestinal obstruction. The animal was in good condition and very fat. On opening the abdominal cavity after death, about a gallon of an opaque, dirty-red-coloured, chyly-looking fluid was found in it. There was a large clot of indurated faeces in the large intestine. In addition, near the commencement of the jejunum, was found a small bolus of straw that had been swallowed, as is often done by these animals in the absence of grass. In this a triangular splinter of bone, about 1J inch long by 1 inch high, with a very sharp edge, had become impacted firmly, so much so as to perforate the walls of the intestine, and to project outside into the abdominal cavity for about J of an inch. The movements of the animal, or the peristaltic action of the intestines, had caused this sharply projecting angle of the bone to cut through the intestinal walls for the distance of some 2 inches. Through this wound the juices of the stomach and intestinal canal, together with the fluid swallowed by the animal, had apparently leaked, and had given rise to the accumulation of fluid in the abdominal cavity which had caused death. 22. ON ANTILOCAPRA AMERICANA, f RZ.S.1880, p. 540. MR. "W. A. FORBES exhibited some drawings of the horns of the Prong- buck (Antilocapra americana), and made the following remarks : " Many of those here present to-night will doubtless remember the surprise created amongst naturalists by Mr. Bartlett's announcement, in 1865, of the shedding of the horns of the Prongbuck. The first surprise that this statement created having passed away, the deciduous nature of the horns of Antilocapra americana seemed in a fair way of being accepted as one of the commonplaces of zoology. About two years ago, however, the celebrated American zoologist Prof. E. D. Cope appended the following editorial note to a short account of this animal * Proc. Zool. Soc. 1880, p. 358. Read May 4, 1880. t Ibid. 1880, pp. 540-543. Read Nov. 16, 1880. 128 ON ANTILOCAPRA AMERICANA. published in the * American Naturalist ' (xii. 1878, p. 557) by a Mr. E. W. Endlich : 'After several years' familiarity with the Prong-horned Antelope in a wild state, I may say I have never met with an undoubted case of shedding of the horn-sheath. Shed horn-sheaths are not common Fig. 1. P. Z. S. 1880, p. 541. Head of Prongbuck, showing the new pair of horns the clay after the shedding of the old ones : reduced. where these animals abound, as they would be were the phenomenon usual. Their appearance on the animal at times indicate that they may be shed ; and I suppose the evidence is sufficient that the shedding occurs. But it is not periodical, or even frequent. 1 " Fortunately, we have not had long to wait before being able to again test the accuracy of Mr. Bartlett's original position. " On December 4, the Society purchased a pair of Pronghorns, the male of which was nearly or quite adult, and had ' apparently lately shed his horns, as the pair which he bears were quite soft when he arrived ' (c/. P. Z. S. 1880, p. 23). In confirmation of this, Mr. Bartlett tells me that his horns also had then no trace of the ' prong,' which subsequently ON ANTILOCAPRA AMERICANA. 129 grew in the ordinary position. Our male animal lived in good health and condition through the summer, and grew a good (though by no means large) pair of horns. "During the night of October 18-1 9th last, these horns were shed ; but no trace of them could be found, after the most careful search ; so that in all probability they were either devoured by the animal itself, or carried away by some predatory rat, or visitor ! Fig. 2. Horn of Prongbuck, drawn the day after the shedding of the old horns : nat. size. " Mr. Smit's drawings that I now exhibit were made on the spot, under my supervision, the day after the shedding (October 20). " Fig. 1 shows the general form and size of the new horns. From it it is obvious that any person who was not acquainted with the mature horns of the animal would not for a moment suspect that any shedding had taken place. Fig. 2 shows one of the new horns, drawn of half the P. Z. S. 1880, natural size, so far as the movements of the animal allow this to be done. P Ma> The base of the newly uncovered horn is thicker and larger than the top part, and is of a different texture, being greyer and pretty thickly covered with long, whitish, closely appressed hairs. The integument is rather soft and decidedly warm to the touch ; and growth is evidently going on here at a rapid pace. The top part, about one inch long, is smoother and blacker, though nearly white at the tip. It is nearly glabrous, with only a very few small hairs, and has the appearance and touch of ordinary horn. It is separated from the basal ' pedicel ' part by 130 ON ANTILOCAPRA AMERICANA. a slight constriction, and is movable on this part in a slight degree from side to side. Fig. 3. Horn of Prongbuck, one month after the shedding of the old horns : ^ nat. size. ** Fig. 3 shows the condition of the horns to-day (November 16), exactly four weeks after the shedding took place. As will be seen, the horns have 'grown rapidly, and have already acquired a characteristic inward curve. The hair-covered ' pedicel ' and the black apical part still retain their original character unaltered ; and all the increase of length in the horn is due, as far as I can make out, to the lengthening-out of the ' node ' (to use a botanical term), which is marked off as a slight con- striction on the fresh horn (vide fig. 2). The horn above the ' pedicel ' is still slightly movable on this latter part, which is still markedly warm to the touch, particularly in its upper half, just below the annular con- P. Z. S. 1880, striction which separates the two parts of the horn. It is here, I am p. 543. inclined to believe, that the new formation of horn is going on, the apical, harder part being pushed, by the growth of new matter, further and further away from the pedicel. I may add that the ' snag' or * prong ' is not yet visible, but may be felt as a slight eminence at the base of the * pedicel,' close to the skull, on the anterior margin of the horn. " On making a longitudinal vertical section of a horn of AntUocapra, I find that there is a more or less open canal in the substance of the ON THE STRUCTURE OF NASITERNA. 131 horn, continued for nearly the whole extent of the main part of the horn, but considerably nearer its posterior than its anterior margin. In its upper part are still observable minute blood-vessels, which have become naturally injected. The horn has obviously been formed around this dermal papilla, which, whilst the surrounding parts have become hard and corneous, has remained soft and vascular for a while ; but eventually its tissues have shrivelled up. It will be readily understood how in this way the papilla has gradually been converted into a hollow channel by the growth and elongation of the horn formed originally round it. " In conclusion, I can only express my surprise that Prof. Cope has apparently overlooked Judge Caton's excellent chapter on the Prong- buck in his ' Antelope and Deer of America ' (1877). Judge Caton himself has apparently witnessed the phenomenon several times ; and his account of the growth of the horns (pp. 25-35) agrees very well with my own observations, excepting as regards the point of origin of the * prong/ If the shedding of the horns is * not periodical or even frequent,' it is certainly strange that both his and the Society's speci- mens should all have exhibited it. I may remark that the period of the year assigned for shedding the horns by Mr. Caton also quite corre- sponds with the dates of shedding here (November 7 and October 19)." [P.S. To-day (Dec. 2) the "prong" is still concealed beneath the hairy covering of the pedicel, but is now very easily perceptible to the touch. The hairy covering of the " pedicel " is thicker now than six weeks ago, apparently owing to the growth of new hairs round it. The horn above is still movable on the " pedicel." W. A. F.] 23. ON SOME POINTS IN THE STRUCTURE OF NASITERNA BEARING ON ITS AFFINITIES.* P.Z.S 1880, p. 76. FOR many years the true position in the series of Parrots of this very singular little form, of which about seven species are now known, has been a moot point amongst ornithologists, most authors placing it amongst the Cacatuinse. Although two accounts have been published of some points in the anatomy of Nasiterna pusio first by Mr. Sclater when describing that species t, and secondly by Signor Camerano, in a paper read before the * Proc. Zool. Soe. 1880, pp. 76, 77. Read Feb. 17, 1880. t P. Z. S. 1865, p. 620. K2 132 ON THE STRUCTURE OP NASITERNA, Turin Academy of Sciences *, nothing very definite has resulted from them tending to elucidate this doubtful point. Mr. Sclater was inclined to regard it (1. c. p. 622) as " an aberrant form of the Psittacinae .... unless it can be allowed to stand as the type of a distinct subfamily, which would probably be more correct." At my request, some fifteen months ago, M. Alphonse Milne-Edwards was kind enough to forward to the late Prof. Garrod a specimen (in spirit) of a Nasiterna, probably N. pygmcea, for dissection ; and I now place before the Society a few statements on its structure as recorded in his MS. notes. As in all other Parrots, except in certain species of Oacatua and in Licmetis tenuirostris, there are two carotid arteries in Nasiterna (a fact previously recorded by Camerano), both of which run in the normal manner in the bypapophysial canal. As in all Parrots with the carotids so disposed (except some individuals of Stringops), the ambiens muscle is absent. The furcula is represented only by a rudiment at the upper end ; and the orbital ring is incomplete. As the oil-gland is present, the formula for Nasiterna, adopting the system used by Prof. Garrod in his paper on the anatomy of the Parrots f, will be 2, ,,+, as in Agapornis, String ops, Geopsittacus, and their allies. Pterylographically, I have been able to ascertain that Nasiterna pygmcea agrees generally in the form and disposition of the tracts with such genera as Cyclopsitta, Psittinus, &c., and differs from the Cacatuinaa in the absence of the crest and naked head-space (cf. Sclater, 1. c. p. 622) P. Z. S. 1880, universally present, as far as I have yet seen, in that group, as also in P- 77< the absence of powder-down feathers, very frequently, though not in- variably, present in those birds, though absent in the other Psittaci with "normal" carotids. In the Cockatoos, too, the orbit is completely encircled by bone J, and, as a rule, doubly so (vide P. Z. S. 1874, pi Ixxi.). In Nasiterna, as already stated, it is not so. Of the other " Palseorni- thidae," as defined by Prof. Garrod, the TrichoglossinaB form a well- marked group, characterized by numerous features to which there is no approach in Nasiterna. Its nearest allies must therefore be in the remaining forms of that family, which I propose to call Eclectinse, including all those not either Cacatuine or Trichoglossine, with the exception, perhaps, of the ground- frequenting forms, Stringops, Pezoporus, &c. The spiny tail-feathers of Nasiterna. are, no doubt, very peculiar, and with its curious beak and * Atti Reale Accad. Torin. xiii. 1878, p. 301. t P. Z. S. 1874, p. 595. J This was not t.he case, however, in a specimen of Calyptorhynchus funereus, lately examined by me. ON THE STOMACH IN TANAGERS. 133 diminutive size must always make this a very well-marked genus. But I fail to see in its spiny tail sufficient importance to elevate Nasiterna into a special subfamily, as suggested by Mr. Sclater. Chcetura is not separated on similar grounds from the other Chaeturinsa ; nor has the spatulate tail of Prioniturus been advanced as entitling that genus to form a special subfamily. The anatomy of the small short-tailed genera Cydopsitta, Psittacella, &c. is as yet unknown ; but I believe that it is amongst these forms related, as far as can be judged from external appearance, through this last to Pezoporus, Geopsittacus, &c. that Nasiterna has its nearest allies. Ayapornis and Psittinus are also not very distantly related, though I believe that the loss of its furcula by Ayapornis, in which it resembles Nasiterna, is probably due to independent causes *. That the loss of the furcula is not exclusively correlated with terrestrial habits is shown by its absence in three such essentially arboreal genera as Ayapornis, Nasiterna, and the Neotropical Psittacula. 24. CONTRIBUTIONS TO THE ANATOMY OF PASSE- BINE BIRDS. PART I. ON THE STRUCTURE OF THE STOMACH IN CERTAIN GENERA OF TANA- GERS.f UNDER this heading I propose to continue from time to time, as material may occur, the "Notes on the Anatomy of Passerine Birds," of which the late Prof. Garrod published four parts in the Society's ' Proceed- ings ' +. In the vast majority of Passerine birds the structure of the anterior part of the alimentary canal conforms to the type present in the Fowl that is to say, to an oesophagus, which may or may not be dilated into a crop, succeeds a stomach consisting of two parts : an anterior glandular part, the proventriculus ; and a posterior part, separated off from both proventriculus and duodenum by more or less distinct constrictions the gizzard or ventriculus, of which the muscular walls are always more or * Judging by its behaviour in captivity, Agapornis is extremely sluggish in its movements. t Proc. Zool. Soc. 1880, pp. 143-147. Read Mar. 2, 1880. | Part I. P.Z. S. 1876, p. 506; Part II. 1877, p. 447; Part III. 1877, p. 523; Part IV. 1878, p. 143. 134 ON THE STOMACH IN TANAGEKS. less thickened, and provided with a central tendon on each side (vide fig. i). P. Z. S. 1880, As was first pointed out by Lund, half a century ago, a singular exception to this rule obtains in the Tanagers of the genus Euplwnia *. From his description (quoted below) and figures, it is quite evident that Lund considered that there was, in these birds, an intermediate zone devoid of glands or muscles, between the proventriculus and the P. Z. S. 1880, p. 144. Fig. 1. Stomach of Tachyphonus melaleucus, natural size, undisturbed, and viewed from behind. The liver, cesophagus, and small intestine are also partially represented. P Z S 1880 commencement of the small intestine, and that a small lateral diverticulum p. 144. springing from this zone was also present, representing the true, though rudimentary, gizzard. Lund found, as he believed, this state of things in three species of Euphonia, whilst the normal type of stomach existed * In a pamphlet entitled "De genere Euphones, prsesertim de singular! canalis intestinalis structura in hocce avium genere, autore Dr. Peter Wilhelm Lund," pub- lished at Copenhagen in 1829 (31 pages and 1 plate). This pamphlet being rather scarce, I here give Lund's own words : " 13 (D. 12). Ubi incipit intestinum tenue, ibi conspicitur in externa superflcie angustse illius zonse, quas locum ventriculi occupare videtur, minima qusedam pro- tuberantia, cui intus respondet levis qusedam impressio. Opaca est parvula haic protuberantia ; quare in parietibus fibras musculosas contineat non dubito ; hoc vero, propter minimam ejus molem, decidere vix valebis Ad earn sententiam maxime inclmare animam, ut statuam verum esse hoc ventriculi analogon, haud diffiteor. " 14. Hsec si vera judicetur cogitandi ratio, referas licet structuram hujus avis ad formam avium generalem ; ita enim habebimus, uti solet, musculosum nostrum organon e late re intestini exortum, interque proventriculum et intestinum ten ue positum ; neque a forma generali avium aberrat haec structura, nisi in eo, quod ad minimum quoddam rudimentum reductus est ventriculus : contra, si zonam illam, de qua mentionem fecimus, angustam pellucidamque, quas inter proventriculum et intes- tinum tenue invenitur, analogon esse ventriculi statuamus, constantes duo maximeque essentiales ventriculi avium characteres subriientur, in inusculosa structura ortuquo laterali positi." ON THE STOMACH IN TANAQERS. 135 in sixteen other species of Tanagers which he examined. Lund's de- scription has frequently been copied since in various text-books, and his figures at least three times reproduced *. Mr. Sclater having called my attention to this subject, I have been able, thanks to the resources of the Prosector's department and to the material afforded by Mr. Salvin, to re-examine this question. I have been able repeatedly to dissect specimens of various species of Euphonia both preserved in spirit and quite fresh. I can fully confirm Lund's description in all points, except as regards the presence of a small lateral diverticulum from the alimentary canal, of which I have never been able to find the slightest trace, though I have always carefully looked for it. Fig. 2 will show the structure of this part of the alimentary canal, with the parts as little disturbed as possible, but with the stomach &c. cut open from behind, in a perfectly fresh specimen of Euphonia violacea. As will be seen, between the glandular proventriculus and the villi-covered duodenum a narrow zone is interposed, with its walls in no degree thickened, but thin and membranous, and of rather greater calibre than the adjacent parts, there being no pyloric constriction. Moreover there is none of that approximation of the cardiac and pyloric ends of P. Z.S.I 880, the stomach that obtains in most other birds. There is no trace of p ' Fig. 2. Stomach of Euphonia violacea. A portion of the alimentary canal of Euphonia violacea, twice the natural size, "cut open and seen from behind, to show the proventriculus (p), the narrow zone representing the gizzard (z), and the commencement of the small intestine (sm.i). The liver and spleen are also seen, as is the end of the (Esophagus, which is opened up. * Carus and Otto, Tab. Aiiat. Corap. Illustr. 1835, pt. iv. tab. vi. figs. 4, 5; Wagner, Icones Zool. iii. t, xi. figs. 3, 4, Bruhl, Zoot. nller Thierkl. Lief. iii. pi. ix. fig. 13/t. 136 ON THE STOMACH IN TANAGERS. any external diverticulum to be seen ; and I therefore can only conclude that Lund must have been misled, he, owing to the bad condition of his specimens (a very probable contingency when dissections are made in tropical climates), having mistaken a bit of fat or connective tissue for a diverticulum of the ventriculus, which last there can be no doubt that this non-glandular zone really represents, the muscular walls and hard epithelium of the true Passerine gizzard being almost entirely undeveloped *. 1 have also been able to ascertain that the nearly allied genus Chloro- phonia (at least in C. viridis) is characterized by the same non-develop- ment of a gizzard. On the other hand, all Tanagers yet examined belonging to other than these two genera have stomachs constructed on the normal type. Thus in a specimen of Tachyphonus melaleucus (see fig. 1, p. 134) the characteristic gizzard with the two central tendons is present and well developed, the muscular walls being nearly | inch thick, and the epithelium lining it hard and horny. As might have been expected, considerable variations in the comparative development of these parts occur in different genera. Thus in the thick-billed Pitylus the P Z S 1880 wn l e organ is much more strongly developed than in the more slender- p. 146. billed genera Tanagra, Calliste, &c. Why the genera Euphonia and Chlorophonia alone, as far as it is yet known, of birds should present this structure is an as yet unsolved problem ; I believe they differ in no appreciable degree from other Tanagers in food f or habits. I may also remark that in such genera as Coereba and dEthopyga, feeding chiefly on minute insects and juices of flowers, there is a well-marked gizzard, with muscular walls and hardened epithelium. Subjoined is a list of all those species of Tanagers, 27 in number belonging to 11 genera, in which the condition of the stomach is as yet known. This includes the species mentioned by Lund (L.), as well as those examined by the late Prof. Grarrod (A. H. Gr.) and myself, and the nomenclature is that of the * Nomenclator/ Mr. Sclater having kindly reduced Lund's names to the terms of that list for me. * In confirmation of the above-mentioned view being correct, I may notice that neither Owen (Anat. Vert. ii. p. 106) nor Gadow (Jen. Zeitschr. B. xiii. p. 168, 1879), when mentioning the stomach of Euphonia, describe any lateral diverticulum. Prof. Garrod, in his MS., notes of Euphonia violacea, with characteristic terseness, " No stomach specialized, the intestines apparently continuing from the oesophagus." t Several of the wild specimens of Euphonia that I have dissected have had in their intestines a large number of small round reddish seeds, which are probably, Mr. Salvin tells me, those of a climbing species of Ficus common in the Central-American forests. ON THE STOMACH IN TANAGEES. 137 Tanagers with a normal Tanagers with the stomach stomach. abnormal. Calliste tricolor (L. & W. A. F.). Chlorophonia viridis (W. A. F.). _ f estiva (L. & A. H. G.). Euphonia chlorotica (L.). __ cyaneiventris (L.). - trinitatis (W. A. F.). _ tJioracica (L.). - violacea (L., A. H. G., & -- melanonota (L.). W. ** * ) _ nigriviridis (W. A. F.). - rufiventris (L.). Tanagra episcopus (L.). - pectoralis (W. A. F.). - ornata (L.). (W. A. F.). - saijaca (W. A. F. & A. H. G.). - palmarum (L. & A. H. G.). Rliamphoccelus brasilius (L.). - jacapa (W. A. F.). Pyranga erythromelcena (W. A. F.). Trichothraupis quadricolor (L.). Tachyphonus melaleucus (W. A. F.). - cristatus (L.). - coronatus (L.). Saltator maynus (L.). Cissopis leveriana (W. A. F.). Pltylus fuliginosus (W. A. F.). Two other species not named by Lund. Pipridea melanonota is mentioned by Lund (under the name Tanagra vittata) as one of the species with a normal stomach. On the other hand, according to M. Taczanowski (P. Z. S. 1879, p. 226), Stolzmann found in this bird " la poche stomacale rudimentaire," and consequently p. z. S. 1880, considers that it is nearly allied to the Euphonies. It is to be hoped ? 147 - that we shall know ere long which of these statements is correct. It would be also highly desirable to ascertain the structure of the stomach in the other genera placed near to Euphonia, particularly that of the genus Procnias. I propose on some future occasion to publish a supplementary list describing the condition of the stomach in any other forms that I may have an opportunity of examining. 138 ON THE ANATOMY OF THE EUEYL^EMID^E. P. z. s. 1880, 25. CONTRIBUTIONS TO THE ANATOMY OF PASSERINE p. 380. BIRDS. PART II. ON THE SYRINX AND OTHER POINTS IN THE ANATOMY OP THE EURYLJE- M1DJB* THE true position of the Broadbills or Eurylaemid* in the series of birds, and particularly the question as to their passerine or non-passerine P. Z. S. 1880 affinities, has long been in question amongst systematic ornithologistsf. p. 381. That more intimate knowledge of their structure from which alone any true answer to this question could be given, has been likewise gradually accumulating for many years. Nitzsch, in his great work on Pterylography, published posthumously in 1840, showed that the species examined by him possessed a character- istically Passerine pterylosisj. Johannes Miiller, in 1846, in his classical memoir on the vocal organs of Passeres, remarked that in Corydon sumatranus, the only species of this group examined by him, there were "no muscular fibres on the larynx." Blanchard, in 1859||, showed that Eurylcemus javanicus agreed in its sternal characters with other Passeres, and particularly compared it with the Swallows in this respect. Mr. Sclatei% in 1872, figured the sternum of Cymbirliynclms macro- rhynchus (under the name of Eurylcemus javanicus ; cf. Lord Walden, I. c. p. 370), and stated that in his opinion these birds were truly Passerine. Prof. Grarrod**, in 1877, was enabled, by an examination of dry skins * Prnc. Zool. Soc. 1880, pp. 380-386. Eead May 4, 1880. t For a succinct resum6 of the opinions of ornithologists on this point, see Mr. Sclater's paper in the ' Ibis,' quoted below. J Kay Soc. ed., pp. 76, 77. These were Corydon sumatranus, Calyptomena viridis, Eury- l(gmus javanicus and E. ochromelas, and Cymbirhynchus macrorhynchus. In the three last named Nitzsch describes nine of the remiges as situated " on the hand ;" in all the specimens of this group I have examined, I find there are ten primaries (cf. also Wallace, Ibis, 1874, p. 406, and Sundevall, Tentatnen, p. 61). An examination of the pterylosis in my spirit-specimens has also convinced me of the partial inaccuracy of Nitzseh's figure of that of Cymbirhynchus (pi. iii. fig. 15). The lumbar saddle is here represented as too angular, and the inclosed space, as well as the antero-lateral tracts bounding it, too broad. The postero-lateral tracts also are represented as consisting of but a single row of feathers. In reality, in this species there is a large ephippial space, of an elon- gated oval shape, the whole shape of the saddle being more like that represented by Nitzsch in Cephalopterus (I. c. fig. 10). The tracts behind are two feathers broad. In Calpytomena, judging from skins, there is an acutely -angled rhombic saddle, whilst in Euryl&mus the condition is intermediate. I may add that in E. ochromelas and Cymbirhynchus the neck-feathering of the lower surface is uninterrupted till behind the middle, and that the throat is entirely feather-clad, with no naked symphysial space. Garrod's edition, p. 27. || Ann. Sci. Nat. (4) Zool. vol. xi. p. JL*. t Ibis, 1872, p. 177, &c. ** P. Z. S. 1877, p. 447. ON THE ANATOMY OF THE EURYL^MID^E. 139 of Cymbirliynclius, Calyptomena, and Eurylcemus ochromelas, to show that these species differed singularly from all other Passeres yet examined in that in them the tendon of the flexor longus hallucis sends a strong vinculum to the tendon of thejf&eor digitorum profundus, as in nearly all other non-passerine birds in which a hallux is developed. He also showed at the same time that in these species the palate was truly Passerine, and proposed to divide the order Passeres " into two sections to start with, those with the hallux not free (the Eurykemidae), and those with the hallux independently movable." The following year he was able to add to this account some facts in the anatomy of two other species, Psarisomus (lalliousice and Serilophus rubropygius. These facts included the typical Passerine arrangement of the tendon of the tensor patagii brevis (P. Z. S. 1876, p. 508), the presence of the left carotid only, the normal disposition 'p/382. of the vessels of the thigh, the presence of the femoro-caudal, semiten- dinosus and accessory semitendinosus, and the absence of the ambiens and accessory femoro-caudal muscles. He also called attention to the un- forked condition of the sternum in Psarisomus, this feature resembling that figured previously by Mr. Sclater in Oymbirhynchus. Having lately, through the kindness of Mr. Edward Gerrard, jun., become the possessor of a specimen each of Cymbirhynckus macrorhynchus and Eurylcemus ochromelas from Sarawak, excellently preserved in spirits, I am in the position to supplement the above-mentioned facts in our knowledge of the anatomy of the Eurylcemidce by describing the syrinx and alimentary canal, previously hardly known at all in this group, in these two species, as well as of confirming or modifying previously published statements. As regards osteology, the only point I wish to record is the unforked condition of the manubrium sterni in both species. In this respect they resemble the condition present in Psarisomus, as already noticed by Prof. Garrod. As this feature appears equally in Mr. Sclater's figure of Cymbirhynchus above alluded to, as well as in a specimen of the sternum of that bird in the College of Surgeons, and in Mr. Eyton's figure of Cory don sumatranus (Osteol. Av. pi. 8. fig. 5), it seems probable that it is a regular character of this family of Passeres, though in other families of that group it seems to be an individual or specific characteristic. ~P *7 Q 1 ^ftO Both the species under discussion agree entirely with Psarisomus and pggS Serilophus in the points already noted by Prof. Garrod. I may add that the pectoralis primus is large, as is also the pectoralis secundus, this muscle extending to the end of the sternum, or thereabouts. As in other Passeres, the biceps-slip to the patagium and the evpansor secundariorum are absent. The semimembranosus is slender, but muscular. The gluteus primus is large, covering the biceps ; and the obturator internus is elongatedly oval. As will be seen in fig. 1 (p. 140), in CymbirJiynclius the vinculum in the deep plantar tendons is strong, and has the character of a firm round tendon, 140 ON THE ANATOMY OF THE EURYL^EMID^E. instead of being composed of more or less transversely-directed fibres running between the two tendons, as in many birds where this structure obtains. In Euryloemus ochromelas it is apparently double, there being a second additional slip given off lower down from the hallux-tendon, which joins the tendon of the digital flexor at the point where the latter, splitting up into three, receives the main vinculum. P. Z. S. 1880, Fig. 1. p. 382. Left foot of Cymbirhynchus macrorkynckus, viewed from behind, to show the deep plantar tendons, and the vinculum (y.), which the flexor longus hallucis (/. I. k.) sends to the tendon of the flexor profundus digitorum. The skin has been turned aside, and the superficial flexors removed ; the flexor longiis hallucis has been cut short above and displaced. P. Z. S. 1880, As regards the alimentary canal of these birds, there is nothing p. 683. unusual in its conformation. The tongue is elongatedly cordate, and slightly bifid at the tip. Both it and the palate generally are smooth ; along its posterior sides it is provided, as is frequently the case, with about eighteen small, backwardly directed, spiny processes, that at the angle being much larger than the others. There is no crop developed ; and the proventriculus is zonary : in Cymbirhynclius it is |, in E. ochromelas | inch in vertical depth. The stomach has the character of a not very muscular gizzard, and is lined with hardened brown epithelium ; the left lobe of the liver is the smallest (considerably). The caeca are present, as might have been predicted from the nude oil-gland*, and are truly Passerine in nature, being mere nipples J, or, in the smaller species, y 1 ^ inch long. The following are the intestinal measurements : CymbirTiynchus. Small intestine 7| in., large intestine 1 j, total 9 in. E. ochromelas. 5| f, 6.J in. * Cf. Garrod, P. Z. S. 1874, p. 110. ON THE ANATOMY OF THE EURYL^EMID^E. 141 The nature of the syrinx was the most interesting question to be examined in these specimens, Miiller's short allusion to that of Corydon, quoted above, being all that was known as regards its structure. In Eurylcemus ochromelas the syrinx is less specialized, as regards its cartilaginous constituents, than in Cymbirhynchus, and will therefore here be described first. The tracheal rings have their usual complete form, being notched before and behind to varying extents, and separated only by narrow intervals. The strong sterno-tracheales, the only extrinsic syringeal muscles, are inserted on the last ring but five. Only the last two tracheal rings are modified. The penultimate ring is narrowed and slightly pro- duced downwards in front ; the last is also narrow, and closely apposed to the penultimate, the membranous interval between the two being very much reduced, except in the middle line in front, where it is well ** z< s - developed, the last ring being here notched above. Behind, the ultimate and penultimate rings are united by a vertically disposed median bar. The last ring forms a three-way piece, there being a forwardly-directed narrow pessulm developed from its hinder margin below. The pessulus is apparently membranous, in this specimen at least, except at its base. The first bronchial semirings are still narrower than the last tracheal one and strongly arched, being concave downwards. They are separated by a very slight interval indeed from the last tracheal ring (three-way piece), and are nearly in contact with each other in front in the middle line ; behind they are inturned and somewhat thickened. To the middle of each ring, or a little posteriorly to this point, is attached the lateral tracheal muscle, which is extremely slender and hardly visible when dry. The Fig. 2. Syrinx of Eurylamus ochromelas, x 5, viewed from in front. p, the " pseudo-ring." second bronchial semirings are shorter than either the first or third ; they are slightly deeper than the first pair, but are narrowed behind. Being only slightly concave downwards, a considerable membrane-covered fenestra is left between them and the first pair. What at first looks like a bronchial semiring is interposed between what are here described as the first and second of that category. This pseudo-ring is most 142 ON THE ANATOMY OF THE EURYL^EMID^E. P. Z. S. 1880, p. 385. evident in front (vide fig. 2), but is also visible when dry behind ; it is apparently due to the accumulation of tissue inside, forming the outer boundary of each glottis. The third and fourth rings are slightly concave upwards : they are deeper, especially behind, thinning away towards the front, than those that preceded them. The fifth and other succeeding rings are typical, unmodified bronchial rings, which more and more encroach upon the membrane*, tympaniformis,'a,nd. eventually almost overlap behind. The second and succeeding semirings are more or less incompletely ossified at their ends. In Oymbirhynchus macrorhynchus (figs. 3-5), the syrinx is constructed Fig. 3. Syrinx of CymbirJiynchus, viewed from in front, X 5. Fig. 4. Fig. 5. Fig. 4. The same, X 5, from the side, to show the insertion of the intrinsic syringeal muscle into the middle of the first bronchial semiring. The stcrnotrachcalis is cut short. Fig. 5. The same, X 5, from behind, to show the pessulus. ON THE ANATOMY OF THE EURYLJEMIDJE. 143 on essentially the same plan, with some modifications. The trachea, ** z - s - 1880, below the insertion of the extrinsic muscles (on the last ring but six), appears to be somewhat laterally compressed and diminished in size. In the specimen before me the last few tracheal rings are somewhat irregular in their disposition, which may or may not be due to individual p. Z. S. 1880, variation. The antepenultimate ring is apparently incomplete on the P* ^* right side, or is at least exceedingly reduced (vide fig. 5). The penul- timate ring is narrowed, and is closely apposed to the terminal ring throughout except in front, where there is a well-marked subtriangular fenestra. The last trachea! ring is produced downwards in front, but is apparently incomplete on the left side in front ; so that there the penul- timate ring is next to the first semiring, in consequence of this disposition. The pessulus is a well-marked, forwardly-directed linear process, p. z. S. 1880, formed only by the right half of the last tracheal ring, which is thus, like P- 386. the antepenultimate, incomplete for a part of its course. The first bronchial semirings are much thickened and strongly arched ; they are nearly in contact in front in the middle line, and are separated by but a narrow space from the last tracheal ring ; behind they become incurved and thickened. The second bronchial semiring is the slenderest of all ; it is shorter than the first or third, and slightly concave downwards. There is the same pseudo-ring formed, apparently by accumulation of tissue inside, between it and the first semiring as already described in Eurylcemus, but it is not so apparent here. The third semiring is much longer, being the longest of all the semirings, and in consequence pro- jecting at both extremities. It is nearly straight, narrow, but not so narrow as the second ring, dilated behind and tapering slightly towards the front. The first, second, and third semirings of each side are nearly in contact with each other before and behind. The fourth ring has much the same shape as the third, but is slightly shorter, and more con- cave upwards than that one. The intervals between the second and third and third and fourth rings are somewhat deeper than are the rings themselves. The fifth and succeeding rings take on the usual form of bronchial semirings, gradually becoming more complete. As in Eury- Icemus, the single intrinsic tracheal muscle, which is very slender, is inserted slightly behind the middle of the first bronchial semiring. The Eurylsemidse are therefore, as is evident from this description, Mesomyodian*, in that respect agreeing with most of the other " Formi- carioid " Passeres of Wallace (Ibis, 1874, p. 406). It is probable that the existence of an intrinsic muscle in the syrinx of Corydon sumatranus escaped the notice of Johannes Miiller always supposing that in that species the same essential form of syrinx occurs as in those above described owing to its slenderness. They are not Tracheophone ; and * Garrod, P. Z. S. 1876, p. 517. 144 ON THE STRUCTURE OF PHILEPITTA. in that they possess the sciatic, instead of the femoral artery*, they differ from the Pipridaa and Cotingidae, with which they have so often been associated. From these, too, they differ, as they do from the Tyrannidaa, Pittidse, and Rupicola, in the details of their syrinx as well as in the simple manubriurn sterni and other points. As has already been stated, they differ from all the other Passeres in the retention of a vinculum in the deep flexor tendons of the foot. To the general bearing of these facts on the classification of the Passeres, I hope to return on some future occasion. P. z. s. 1880, 26. CONTRIBUTIONS TO THE ANATOMY OF PASSERINE P . 387. BIRDS. PART III. ON SOME POINTS IN THE STRUCTURE OF PHILEPITTA, AND ITS POSITION AMONGST THE PASSERES. f THE doubt which has hitherto prevailed amongst ornithologists as to the true affinities of the very singular Malagash bird for which Geoff roy founded his genus Philepitta J, makes a knowledge of its anatomy, and particularly of its osteology and syrinx, a desideratum. Its original describer considered this genus most nearly related to Philedon. Bona- parte, in his ' Conspectus ' , referred Philepitta with some doubt to the Starlings (Sturnidae), placing it near Diloplius. The late Mr. Gray, in his Hand-list |1, made it a genus of Pittidse, Pitta being the only other genus of that family recognized by him. Mr. Sharpe in 1870 ^[ suggested that it ought to be regarded as an aberrant genus of the Paradiseidse, forming a subfamily which he pro- posed to call Philepittinae. That neither this position nor those assigned to it by Geoffroy or by Bonaparte can be accepted is evident from the fact that, as shown by Sundevall **, Philepitta possesses a long 10th (" first " tt) primary, at the same time that the tarsus is not bilaminate. The Swedish naturalist last mentioned made his subfamily Paictinae (he having rechristened Philepitta Paictes) the first in the fifth cohort, " Taxaspidece," of his " Oscines Scu- telliplantares" the others being the Thamnophilinae, Myrmornithinae, Hypsibaemoninae, and Scytalopodinae, in which last Menura was also * Cf. Qarrod, P. Z. S. 1876, p. 516. t Proc. Zool. Soc. 1880, pp. 387-391. Read May 4, 1880. \ Mag. Zool., Ois. pi. 3 (1839). Op. cit. p. 422. jj Op. cit. i. p. 297, gen. no. 1094. [ P. Z. S. 1870, p. 397. ** Tentamen, p. 63. tt See P. Z. 8. 1879, p. 256, note. ON THE STRUCTURE OF PHILEPITTA. 145 included a striking illustration of the unsatisfactory results that a classification founded on external characters only always leads to. More recently, M, Alphonse Milne-Edwards has figured the two known species of Philepitta, as well as the tongue and osteology of P. castanea, in Grandidier's magnificent work on Madagascar *. In this work (the plates only of the part in question having been issued) he places it next to the Nectariniidae, apparently on account of the eye- wattle of the male and the bifid tongue approximating it to such a form of that group as Neodrepanis. Having written to M. Milne-Edwards to ask if he had examined the syrinx or other soft parts of the bird under discussion, he was kind enough to reply by sending me the viscera, including the trachea &c., of a specimen (in all probability P. castanea}, and by generously granting me permission to make any use of them I liked. He also informs me that in the text to the plates he has fully described the osteology. As regards this part of the structure of Philepitia, I only wish to p. remark on and give a figure of the palate, extracted from a skin of P- 388- P. castanea by Prof. Grarrod, M. Milne-Edwards's figure of this (pi. 112. fig. 2 a) being rather indistinct in some important points. As will at once be seen, the vomer is truly Passerine, being split behind and trun- Fig. 1. Palate of Philepitta castanea (nat. size). cated in front ; to its outer and anterior angles are articulated two small nodules of bone, probably corresponding to the " septo-maxillaries " of Prof. Parker. The maxillo-palatines are slender, long, recurved apically, and pointed backwards ; the transpalatines are distinct and slightly curved inwardly, and the palatines tend to diverge behind. In Pitta (cyanura) the vomer is proportionally broader, the maxillo-palatines are much shorter and broader and more transversely directed, and the pala- * Hist. Phys. Nat. ot Polit. de Madagascar, tome iii. Oiseaux, Atlas ii. l rc partie, pis. 109-112. 146 ON THE STKUCTURE OF PHILEPITTA. tines are nearly parallel to each other throughout*. In the Eury- laemidaD f the maxillo-palatines, though slender, are nearly transverse to the axis of the skull, and the " transpalatines " tend to become obsolete. Judging from M. Milne-Ed war ds's figure (I. c. pi. 112. fig. 3), the manubrium sterni is but slightly bifid, therein approaching that of the Eurylaemidse. As regards other points, in its pterylosis Philepitta, which was one of the few important forms unexamined by Nitzsch, is perfectly Passerine. There is a longish oval ephippial saddle, with a large space, much as in some of the Eurylsemidae (vide supra, p. 138) ; in Pitta, according to Nitzsch, the saddle is undivided. But Philepitta differs from the Eury- IsemidsB, and agrees with all other Passeres, in the absence of any vinculum in the deep plantar tendons, as was ascertained by Prof. Garrod from the examination of a skin, and recorded by him in MS. P. Z. S. 1880, As regards the alimentary canal, there is nothing peculiar. The p. 389. tongue, in the specimen forwarded from Paris, was removed ; but, as we know from M. Milne-Edwards's figure, it is triangular and bifid at the end. There is no crop developed ; and the zonary proventriculus is half an inch deep. The stomach is a strong gizzard, rather elongated in shape, with thick and considerably plicated epithelium. The liver is unequilobed, the left lobe being half the size of the right ; it has a gall- bladder. The total length of the intestines is seven inches, of which the last half -inch is large intestine. The caeca are truly passerine, being mere nipples, and rather widely separated. The syrinx of Philepitta being hitherto entirely unknown, I herewith give a description and figures of it. Fig. 2. Fig. 3 Fig. 4. Fig. 2. Syrinx of Philepitta, from before. Fig. 3. The same, from behind. Fig. 4. The same, from the right side. (These figures are magnified about 4 times.) * . Cf. Prof. Parker's fig. of Pitta melanocephala, Trans. Zool. Soc. ix. pi. Ivi. figs. 6, 7. In this species the " transpalatine " processes are far less developed than in P. cyanura. f Figures of the palates of Euryl&mus ochromelas and Catyptomena viridis are given in Prof. Garrod's paper, P. Z. S. 1877, p. 449. ON THE STRUCTURE OF PHILEPITTA. 147 The trachea is slightly laterally compressed below ; the rings, which are complete, are somewhat irregular in shape, owing to the greater or less development of the notchings on their borders. In front, of the last few rings preceding the terminal one, two or more are joined together by vertically directed bars, which makes it difficult to count their number with exactitude. Behind, however, they are all free. The terminal tracheal ring is narrow laterally, and closely opposed to the first bronchial semirings ; in front and behind it is produced downwards triangularly, and behind bears a well-developed forwardly directed narrow pessulus. As seen from behind, therefore, the terminal tracheal ring has somewhat the shape of an arrow-head. The narrow sterno- tracheales are inserted on about the last ring but six. The first bronchial semirings are thickened, and very much arched, being concave downwards. As seen from the side (fig. 4) they are more strongly convex anteriorly than posteriorly. The second and third semirings are very slender indeed, closely approximate, much shorter and much less concave downward, so that a large membranous fenestra is left between them and the first semirings. The fourth and fifth semirings are also slender, but less so than the last two : they are slightly concave upwards, so that a second, though shallower, fenestra is formed between them and the P- z < S. 1880, second and third pairs. The fifth semirings are slightly dilated at their p ' extremities, where they are in close proximity, before and behind, with the first four pairs. The sixth and succeeding bronchial rings take on the ordinary character, being deeper than those that precede them, and gradually becoming more complete, till the fifteenth pair are nearly perfect. On one side, the left, in this specimen, the eighth and ninth semirings are partially fused externally. The lateral muscles of the trachea, after the insertion of the sterno- tracheales, become excessively thin, so that it is difficult to make out accurately their exact extent. They apparently fan out, so as to be nearly in contact with each other before and behind, and are inserted on to the first bronchial semirings for the greater part (as far as I can make out) of their lateral surface, the tips, however, being quite free from muscular fibres. Philepitta is therefore perfectly Mesomyodian, as Prof. Garrod pre- dicted would probably be the case *. There are thus three families of Mesomyodian Passeres in the Old World the Pittidse, the Philepittidae, and the Eurylaemidse. All agree in the possession of a broncho-tracheal syrinx, in that respect agreeing with the Cotingidae (including Rupicola), Pipridse, and Tyrannidas of the New "World, and differing from the Tracheophone families, which are all, as is well known, American. Philepitta differs in the details of its * Voice-organs of Passeres, p. 68. L2 148 ON THE STRUCTURE OF PHILEPITTA. P. Z. S. 1880, p. 391. syrinx from all the other " Haploophone " Pas seres. In Pitta (cf. P. Z. S. 1876, pi. 53. figs. 1-6) the bronchial semirings are much less modified, being nearly entirely simple semirings ; the lateral muscle, too, is slender and not spread out as in Philepitta. This fact, taken with others, as the scutellation of the tarsi, osteology, &c., justifies, in my mind, the establishment, as has been done by Sundevall under the name Paictidae, of a separate family for the bird under consideration *. The Eurylae- midse differ in their retention of a plantar vinculum (cf. Grarrod, P. Z. S. 1877, and supra, p. 139), as well as in the structure of their feet and other points. In the form of their syrinx, however, they approach Philepitta perhaps more nearly than any form yet described, though in them too the lateral muscle remains slender and unexpanded. The peculiarities of the EuryleBuridsa, and especially their oft-spoken-of retention of the plantar vinculum, are sufficient, I think, to justify their forming a main division of Pas seres by themselves, as suggested by Prof. Garrod f, which may be termed DESMODACTYLI in distinction from the other or ELETJTHEEODACTYLI. Prof. Garrod's arrangement of Passeres J may therefore be modified as follows : a i. DESMODAOTYLI. (The plantar vinculum retained ; manubriura sterni not forked.) ii. ELEUTHEKODACTYLI. (The plantar vinculum lost; manubrium generally strongly forked.) A. Mesomyodi. HETEROMEEI. HOMOSOMERI . Haploophonae. Tracheophonse. B. Acromyodi. New World. Pipridcs. Cotingida. Rupicola. Dendrocolaptidce . Pteroptochidee. Old World. Euryl&midce. Philepittida. Pittidce. ABNORMALES. Atrichiida. Menuridce. NORMALES. * The general myology and vascular system of Philepitta are still, it must be remembered, unknown, but are in all probability perfectly Passerine. t Voice-organs of Passeres, p. 73, and P. Z. S. 1877, p. 449. | P. Z. S. 1876, p. 518. I place Philepitta only provisionally amongst the Homceomeri, presuming that, as in all Passeres but the Pipridae and Cotingidse (minus Rupicola), the artery of the leg is sciatic. ON LEPTOSOMA DISCOLOR. 149 Till more material has been examined, it is impossible to say whether or not some of the points in the above classification fairly express the affinities of the various groups treated of. This appears to me parti- cularly the case as regards the primary division of the Mesornyodi into Hetero- and Hoinoeomeri, depending as it does on the presence of the femoral or the sciatic artery respectively. The pseudo-schizorhinal character of the skull also in some of the Tracheophonse * may necessitate an ultimate arrangement of that group different from that here adopted (taken from Messrs. Sclater and Salvin's ' Nomenclator '). As regards the Passeres whose anatomy still remains unknown, the forms that most require examination are Phyiotoma t and OocyrJiamphus of the New, and Orihonyx and Melampitta of the Old World. The last may be, as suggested by Mr. Gould J, a link between Pitta and PUlepitta ; Count Salvadori, on the other hand, is inclined to regard it as a Timeliine and therefore a normal (Oscinine) Acroinyodian form. It is also highly desirable to obtain some knowledge of the soft parts of some of the larger forms usually placed amongst the Cotingidae, especially PtilocJiloris and Phcenicocercus (placed by Sundevall with Rupicola), as well as of Gymnoderus, Querula, Ceplialopterus, &c. 27. ON THE ANATOMY OF LEPTOSOMA DISCOLOR.\\ p.z.s.1880 p. 465. IT is to the liberality of my friend Prof. A. Newton that I am indebted for the opportunity of dissecting a female example of this bird, the most peculiar, perhaps, with the exception of Mesites, of all the anomalous forms that Madagascar produces. Till the past year or two our know- ledge of the structure of Leptosoma was almost confined to its skin and certain parts of its skeleton. Mr. Sclater, in this Society's ' Proceedings ' (1865, pp. 682-689 ; also in Nitzsch's ' Pterylography,' Eay Soc. ed. App. ii. p. 158), has already given us an account of the different views that have at various times been held by ornithologists as to the position of this peculiar form ; and he was also the first to point out the existence in it of powder-down P. s. Z. 1880, patches, as well as other of its peculiarities. Since then I am unaware P- 466< * Cf. Garrod, P. Z. S. 1877, p. 452, &o. t Very imperfectly described by Eyton and Eydoux and Souleyet, of. Job. Miiller, Stimmorgane, &c., p. 8. | B. New Guinea, pt. ii. (1876). Ann. Mus. Oiv. Gen. x. p. 147. j| Proc. Zool. Soc. 1880, pp. 464-475. Bead June 15, 1880. 150 ON LEPTOSOMA DISCOLOR. of any thing more having been done to elucidate its structure till 1878. In M. Grandidier's magnificent work on Madagascar*, in the plates of the Atlas devoted to the birds, M. A. Milne-Edwards has figured the entire skeleton, together with separate views of the bones, as well as the tongue and alimentary canal, and has likewise given pictures of the bird when plucked, showing the external nares, the position and shape of the powder-down patches, and its naked oil-gland. In reply to my inquiries on the subject, M. Milne-Edwards kindly replied that he intended to describe in full the osteology of Leptosoma, together with that of AtdorniS) Brachypteracias, &c., of which figures are given also in the Fig. 1. Eight foot of Leptosoma (nat. size), seen from before, to show the disposition of the toes. (The fourth toe is slightly removed outwards, to better show its position.) above-named work, in the text, but that, as regards other points, only an explanation of the plates was to be given. I have therefore thought it would be of interest to bring before the Society some additional notes on its pterylosis and soft parts, derived from my examination of Prof. Newton's specimen. Before proceeding further, I should like to call attention to the struc- ture of the feet in Leptosoma, which has already been accurately described by Mr. Sclater (I. c. p. 688). They are in no way " zygodactyle," in the sense in which that term is applied to the feet of such birds as the * Hist. Phys. Nat. et Pol. Madag., Zool., Ois. pis. 85-88. ON LEPTOSOMA DISCOLOR. 151 Cuckoos, Parrots, or Toucans. In this spirit-preserved specimen it is P. Z. S. 188'0, easily demonstrable that the fourth digit cannot naturally be placed in a P' ' * really reversed position, like that of the above-named birds. While the second and third toes look directly backwards, the hallux looks inwards and forwards, and the fourth toe inwards and slightly backwards at its apex, there being, as it were, a slight twist in its axis *. However much the fourth toe is bent backwards (and this is only done by the exercise of some little force), its plantar surface always looks more or less inwards. The presently-to-be-described arrangement of the deep plantar tendons also confirms the view here taken as to Leptosoma not being a true zygo- dactyle bird. Pterylosis. As regards Leptosoma, Nitzsch only noted the presence of an aftershaft and 12 rectrices, he only having been able to examine a stuffed specimen. Mr. Sclater, in his above-mentioned paper, besides describing the two characteristic lumbar powder-down patches of this bird, briefly alludes to the pterylosis, which " appears nearly similar to that assigned by Nitzsch to Coracias and Eurystomus" These features are diagrammatically represented in a woodcut (fig. 5, I. c.). The following is a more detailed description : The inferior tract divides about 1 inch behind the junction of the rami of mandible the (badly) so-called " chin-angle" from which it starts as a narrow, single tract f. Between this tract and the manclibular rami, extending as far as the angle of the jaw, a narrow naked space is left ; at this point the inferior tract becomes continuous with the feathering of the head above, so that here the neck, except for the narrow median ventral apterium, is continuously feathered. This continuous feathering extends downwards to about f inch above the shoulder, when, the inferior and dorsal tracts diverging, the lateral neck-space is formed. The inferior tracts diverge gradually as they approach the breast, and then run parallel to each other over the pectoral muscles and abdomen to the sides of the vent, leaving a rather wide bare carinal space, with a few scattered down-feathers. As the inferior tract emerges on the breast, it gives off a branch to the anterior margin of the patagium ; and this at first is dilated somewhat, so that the space between it and the main tract is feathered. The broad humeral tract is also connected with the inferior tract where the latter gives off this patagial branch. In the lower part of the neck the inferior tract is about 8 feathers broad, on the breast * This disposition of the fourth toe makes Leptosoma, at first sight, look as if it had three toes anteriorly directed, and no doubt accounts for Mr. Sharpe entirely omitting any notice of its peculiar feet in his paper on the Coraciidae (of. Ibis, 1871, pp. 187, 285). t In Coracias garrula the naked median space left between the halves of the inferior tract extends quite up to the symphysis, so that the inferior tract is double from the commencement. 152 ON LEPTOSOMA DISCOLOK, about 6, and on the abdomen only 2. About the middle of the sternum the outer pectoral tract, which is about 4 feathers wide and slightly P. Z. S. 1880, stronger than the main tract, is given off ; it is not very divergent, but is dilated terminally, and develops a recurrent hook, which, however, is not very distinct. There is a circlet of feathers round the vent, and a short tract of feathers behind it, on each side of the fleshy part of the tail, continuing the direction of, though quite separate from, the main inferior tract of its side. The feathering of the head is continuous, and from it the anterior moiety of the dorsal tract runs, being anteriorly continuous at the sides, as already noted, with the inferior tracts, along the dorsal median line of the neck, as a rather broad, thickly feathered band, which forms a strong interscapular fork, just as in Coracias and the Parrots, the ends of the fork lying about j inch anterior to the posterior extremities of the two scapulae. The posterior moiety has also a forked form, the two arms enclosing a fairly broad naked median space, and only uniting about 1 inch in front of the oil-gland, the united tract ao formed ceasing altogether about J inch in front of that organ. This posterior fork is very narrow anteriorly, not more than two feathers wide ; indeed, for the first two or three rows each arm consists of only one feather in each row, and the two arms run in between the forks of the anterior moiety, just as in the Parrots, Coracias, and some other birds. Posteriorly the fork widens, and becomes connected closely with the scattered contour- feathers which are found outside it, over the space between the dorsal tract proper and the lumbar powder-down patches, so that on the rump the dorsal tract appears to consist of five or six rows of feathers on each side of the median line. There is a very strongly feathered and broad band of feathers over the knee, being the anterior end of the lumbar tract of its side; this tract is quite distinct from all others but the crural, which are much weaker and clothe the leg as far as the " ankle." The powder-down patches, one on each side, lie between the posterior portion of the dorsal tract and the lumbar tracts. They form elongated patches, extending forwards over the femur as far as the sartorius muscle, and backwards to within inch of the vent ; their dorsal border is parallel to the dorsal tract, the ventral to the lumbar ones. On the inside of the skin they are conspicuous as dark grey patches, formed by the closely aggregated insertion of the feathers of which they are com- posed, these lying at a less angle with the skin than the contour-feathers. Nitzsch* has described the pterylosis in Coracias garrula and C. indica, with figures of that of the former, and in Eurystomus gularis. I have examined the first-named species in the flesh, and also a skin of Atelornis crossleyi. In all essential respects, as will be seen by a comparison of * Pterylogr. (Kay Soc. ed. p. 89). ON LEPTOSOMA DISCOLOR. 153 the above description with Nitzsch's figures of Coracias garrula, Leptosoma is essentially Coraciine, though it differs from all others of that group in its possession of powder-down patches*. In the CuculidsB the dorsal tract, though it divides between the shoul- P.Z. S. 1880, ders, is perfectly continuous throughout, enclosing an elongated oval ? " space (vide Nitzsch's figures of Cuculus canorus and Centropus rufipennis, 1. c. pi. iv. figs. 12 & 14). In the Cuculidae too, as is well known, the after shaft is absent and there are but 10 rectrices. I may remark that in the possession of an inte/*scapular dorsal fork the Coraciidae and Lepto- soma form an exception to Prof. Garrod's generalization f that when " the dorsal tract develops a fork between the shoulder-blades a bird is homalogODatous." Visceral Anatomy. The mucous membrane of the palate and mouth is smooth throughout, except along the margins of the nasal aperture, where it develops three or four small blunt retro verted tubercle-like papillae on each side, and also external to this on each side along a line parallel to the axis of the palatine bones, where there is a similar short row of small papillae. The tongue is tapering and elongated in shape ; its length is 1| inch. The basal part, which alone is fleshy, and supported by the hyoid bones, is of a triangularly sagittate shape, about | inch long, and provided at its postero-exterual agles with a few minute, blunt, retroverted papillae ; it is prolonged forwards into a horny lamina, which is strongly concave above and forms the greater part of the tongue ; at its apex the part, which is of a slightly tapering shape, is apparently entire . This tongue closely resembles that of Coracias, and differs from that of such of the Cuculidae as I have examined in wanting the w r ell-developed retroverted spines that are always present on the posterior part of the lateral margins in those birds . The oesophagus is capacious at first, but rapidly narrows ; it develops no crop. The proventriculus is zonary, being j inch deep. The stomach is globose and not strongly muscular : there is a distinct pyloric bulb in- * I may here mention that Atelornis crossleyi differs as regards its pterylosis but slightly from the Coraciine type. It has the same interrupted dorsal tract, each half having a furcate form ; but here the interscapular fork is very short and narrow, and does not enclose the anterior part of the posterior fork, which has a long stem or " handle." There are the same strong lumbar tracts. Below, the outer pectoral tract, given off on the middle of the breast, is only indicated by an enlargement of the main tract, and is not at all free. There are no traces of powder-down patches. t P.Z. S. 1878, p. 931. | The tongue of Leptosoma has been figured by Mr. Sclater (/. c. p. 688), and also by M. A. Milne-Edwards (/. s. c. pi. 88. fig. 1). Vide also the figures of the tongues of Coua gigas (pi. 63. fig. 1) and C. olivacciceps (pi. 64. figs. 1, 2) in Grandidier's work. 154 ON LEPTOSOMA. DISCOLOR. P.Z.S. 1880, p. 470. dicated externally at the commencement of the duodenum. Internally it is lined with rather soft epithelium, which is concentrically striated. In the present example the stomach contained hairs, apparently of lepi- dopterous larvae, and the horny jaws and other hard parts of insects : many of the smaller hairs had become impacted in the soft lining of the stomach, so that this at first sight appeared to be villous. The same appearance has often been described in our common Cuckoo *. The intestines in all measure 12 inches, of which 2| are " large ; " they are not markedly capacious. The caeca f are long and cylindrical in shape, largest apically, and slightly tapering towards their bases ; they measure respectively 2| and 2| inches. The liver has the left lobe much the smallest ; there is a distinct gall-bladder. There is thus nothing striking or characteristic about the alimentary canal. In the possession of large cylindrical caeca, Leptosoma agrees with both Coraciidae (including Bracliypteracias and Geobiastes} and Cueulidse, as also in most of the other points noted. In the Cuckoos, however, the gall-bladder is said to be absent as a rule J. Fig. 2. Wing-muscles of Leptoswnd. Termination of the tensor patagii brevis (t.p.br.} in Leptosoma. e.m.r., the fleshy belly of the superficial layer of the extensor metacarpi radialis longior muscle ; t, the tubercle on the humerus, whence it arises ; above it the humerus. P, the pata- gium, its dorsal layer having been removed to show the muscles, &c. * Cf. Hunter's Essays and Observations, ii. p. 285 &c. t Figured, with other parts of the intestinal canal, by M. Milne-Edwards, I. c. pi. 88. \ Owen, Anat. Vert. ii. p. 177. Gadow also states its absence in Cuculus. Hunter, on the other hand, found it, though " very small," in C. canorus (1. s. c. p. 285). Ac- cording to the plates in Grandidier's work, Coua gigas has a gall-bladder (pi. 63) ; so has Geobiastes sqiiamigera (pi. 99. fig. 2). ON LEPTOSOMA DISCOLOR. 155 Myology, fyc. The first pectoral is big ; the second extends at least halfway down the sternum ; the third is not represented. There is no biceps-slip to the patagium,^as is the case in all " Anomalogonatse " except the Caprimulgidae. In none of these points does Leptosoma differ from the Coraciidae or Cuculidoe. The expansor secundariorum is present and well developed ; its proximal end is T-shaped (" ciconiiform" Garrod), the sternal part of the tendon being attached to that bone at the junction of the costal process with the body near the coracoid groove ; it therefore resembles the same muscle in the Coraciidae. In no other birds amongst P. Z. S. 1880, the Anomalogonatae is this muscle present. In the Cuculidse this muscle p ' ' is present, but its terminal tendon is not T-shaped, the sternal moiety being undeveloped. The arrangement of the termination of the tensor patagii brevis is re- presented in the accompanying figure (fig. 2, p. 154). The main tendon (t.p.br.) runs on to the ulnar side of the arm, and there becomes fused with the fascia covering the muscles. Before doing so, however, it crosses the superficial tendon of origin of the extensor metacarpi radialis longior (e.w.r.), which springs from the humeral tubercle, and becomes firmly blended with it. It likewise sends off, distally, a special slip of tendon which joins the same tendon of that muscle more externally (wristward). This is much the same arrangement as in the Coraciidse, as described and figured by Prof. Garrod (P. Z. S. 1876, p. 511, pi. 49. fig. 1), except that in those birds the tendon of the tensor patagii brevis is split into two quite separate halves ; if these were united together, an arrangement would be arrived at practically identical with that of Leptosoma. In the Cuculidae the condition of things is quite different, as in them the " undivided tendon runs on to the ulnar superficial fascia without auy complication " (I. c. p. 512). Of the leg-muscles, the glutens primus is present, though small, only slightly overlapping the biceps, and with its fleshy part not reaching the innominate, to which it is attached only by fascia. The ambiens is ab- sent ; the femoro-caudal is very large, but lacks the accessory head, as in all Anomalogonatae. Both the semitendinosus and its accessory are well developed, as is the semimembranosus. The biceps mem, as usual, passes through a tendinous loop. The obturator externus is well developed, and the obturator internus is of a very elongated oval shape. The formula of Leptosoma is therefore . A . X . T, exactly the same as that of the Coraciidae and the greater number of Anomalogonatous birds. In the Cuculidse the ambiens is always present and well developed, and the accessory femoro-caudal usually so *, giving a formula of + . A . (5\ X . T . * It is absent only in Cuculus, Chrysococcyx, and Cacomantis. [ Garrod 's MSS.] 156 OJS T LEPTOSOMA DISCOLOR. p. 472. Leptosoma is therefore clearly not Cuculine. In the Cuctilidae, too, the obturator internus is triangular in shape, as in the Gallinae and their allies ; in Leptosoma, as already stated, as in Coradas, it is oval. The anomalous arrangement of the toes in Leptosoma made me very anxious to observe the disposition of its deep plantar tendons, these, in all " zygodactyle " Anomalogonatous birds, being arranged in a manner quite unique amongst birds and entirely different from that which obtains in the even-toed Homalogonatous birds (i. e. the Psittaci, Cuculidae, and Musophagidaa) *. But in Leptosoma neither of these conditions occurs ; on the contrary, the disposition of its plantar tendons is exactly that found in many birds P. Z. S. 1880, with feet of the ordinary structure. This condition is diagrammatically represented in fig. 3 ; as will there be seen, the tendon of ihejle&or longus liallucis (f.l.hS) joins the tendon of i\Q fleocor profundus digitorum(f.p.d.) on the outer side, some little way above the phalanges, and completely blends with it. From the single compound tendon so formed the small slip to the hallux is given off, on the inner side, just before the common tendon splits up for distribution to the three other digits. This is exactly the same condition as that found by Prof. Garrod in Coracias garrula, and by myself in Atelornis crossleyi (in a skin). It differs completely from that found in the Psittacidae, Cuculidae, and Musophagida} on the one hand, and that of the Galbulidae, Bueconida), Fig. ?. f.p.d. Diagram of the arrangement of the deep plantar tendons in Leptosoma. f. 1. h., the flexor longus hallucis ; f.p. d., the flexor profundus digitorum. and Picidaa and their allies on the other. Therefore this fact, when taken in conjunction with the statements already made as to the natural position * Vide Garrod, P. Z. S. 1875, p. 345. ON LEPTOSOMA DISCOLOR. 157 of the fourth digit in Leptosoma, shows that there are no real grounds for calling Leptosoma a " zygodactyle " bird *. As regards other points, it may be mentioned that the vessels and nerves of the thigh are normal; that is to say, the sciatic nerve and artery and. the femoral vein are all present in their normal position. There are two carotid arteries present, both of them being unusually small, the left particularly so. They run up in the usual converging way, springing from the vertebral arteries into the hypapophysial canal of the neck, and there become so closely applied to each other that it is impossible to dissect them away as can usually be done in birds. As far P. Z. S. 1880, as I can make out, they do not, however, fuse, but are continued up to p * the head and there diverge. In OpistJiocomus t Prof. Garrod found a somewhat similar condition, though he says nothing about the vessels being minute. In Leptosoma they have the appearance of white fibrous cords, and they may possibly be, like the carotids of Bucorvus J, no longer functional as blood-channels. But satisfactorily to decide this, as well as the ultimate termination of these carotids, fresh or injected specimens will be necessary. In both the CuculidaB and CoraciidsD there are two equisized carotids, which are as free as usual. As regards the vocal organs, there are present but one pair of extrinsic muscles, which diverge to be attached to the " costal processes " of the sternum. The syrinx possesses a single pair of intrinsic muscles, as Fig. 4. Fig. 5. Fig. 4. The syrinx of Leptosoma seen from in front, the muscles of the left side having been removed. Fig. 5. The same, from behind. (Both are twice the natural size.) usual. This organ having been previously unknown in Leptosoma, I here take the opportunity of describing and figuring it. The tracheal rings, which, as usual, interlock with each other for the * In Podargus cuvieri, where the outer toe is reversed in perching, and in Coitus, where the toes are directed at various times in very different ways, the "same blended" distribution of the deep plantar tendons obtains. t P.Z.8. 1879, p.112. J Vide Mr. Ottley's paper on this bird, P. Z. S. 1879, pp. 461-467. 158 OX LEPTOSOMA DISCOLOR. greater length of the trachea, are well ossified, and only separated by narrow intervals. They gradually narrow as they approach the thorax, the last two being the narrowest of all. The penultimate tracheal ring is produced downwards in a triangular way behind, as is the terminal one in front ; behind, this last ring bears the anteriorly directed narrow pessulus, which intervenes behind between the inturned ends of the first pair of bronchial semirings, but in front does not appear, stopping short before it reaches the anterior surface of the bifurcating trachea. Like the tracheal rings, the first three bronchial semirings are well ossified, and separated from each other only by very narrow interannular intervals. They are nearly straight, with only a very slight concavity upwards, and increase in depth as they descend. In front the semirings P. Z. S. 1880, of opposite sides are separated from each other by a small notch ; behind P* ' 4 they are less and less complete as they go downwards. The first semi- rings are posteriorly closely applied to, though separate from, the pessulus, and are apparently continued on, as cartilaginous rings, posteriorly, so as to form complete or nearly rings. The posterior ends of the second and third semirings where they appear behind are widely separated from their fellows of the other side. The fourth and succeeding bronchial rings are all cartilaginous. Of these the fourth is the largest, being nearly straight, and slightly more prominent than the others. To its middle, rather towards its posterior margin, is attached the (single) intrinsic syringeal muscle. The rings succeeding the fourth ring rapidly become more and more complete, at the same time that the bronchus becomes less capacious, the whole tube tapering away from this ring as it approaches the lung. The fifth and sixth semirings are more slender than those that succeed them, and are slightly concave upwards. The remaining ones are straighter and deeper. Except between the fourth and fifth, and fifth and sixth semirings, the interannular intervals are exceedingly narrow. This syrinx does not show much similarity of form to that of Coracias garrula, the only one of the family of Coraciidae that I have been able to examine as regards this point. At the same time it does not much resemble that of any Cuckoo I am acquainted with. Reviewing the facts already stated, it is clear that the affinities of Leptosoma to the Cuculidae are very remote, whilst, on the contrary, its relations to the Coraciidae are quite the reverse. The subjoined tabular statement (p. 159) of the principal points in the structure of the three just named groups will perhaps render this additionally clear. In common with both Cuculidae and Coraciidae, Leptosoma possesses a nude oil-gland and long caeca, two carotids, and the f em oro- caudal, semi- tendinosus, and accessory semitendinosus muscles. Wherever there is any difference, Leptosoma resembles the Coraciidae ; and the same story is told by the pterylosis and tensor-patagii arrangement. ON TWO PLOCEINE BIRDS. 159 , 2 M Aftersh. T3 Id ?* Ambieiis. 'i * 1 ?; f4 4 'a A .j 'a i d 1 *1 H | J _rt jS o Carotids. Q Cuculidae ... 10 - + + + + + peculiar triangular 2 + Leptosoma... 12 + - - + - + + ciconiine oval 2 + Coraciidae ... 12 + - + - + + ciconiine oval 2 + Nevertheless, both in the syrinx and in these last two points, as well as in some others, especially osteological ones *, Leptosoma is not quite typically Coraciine, and it may therefore be retained, as was proposed by p.z. S. 1883, Mr. Sclater, as the type of a peculiar family, Leptosomidae. This should P- *^' be placed in the series of Passeriform Anomalogonatous birds as denned by Prof. Grarrodt, next to the Coraciidse, with which its relations are most intimate. Indeed it is possible that, when the anatomy of the allied genera, Brachyptercicias, Geobiastes, and Atelornis^ becomes fully known, the truth of Mr. Sharpe's proposition , that Leptosoma should be relegated to the position merely of a subfamily of the Coraciidae, may be established. 28. ON TWO RARE PLOCEINE BIRDS NOW OR LATELY P.z.s.1880, LIVING IN THE SOCIETY'S MENAGERIE. || p ' 475 ' (Plate VII.) 1. VlDUA SPLEKDENS. (Plate VII. fig. 1.) Vidua splendens, R/eichen. Orn. Centralbl. 1879, p. 114. On the 17th of July, 1878, Mr. Archibald Brown presented to the Society, with some other birds, a specimen of a small Weaver-bird, which being then " out of colour," was entered on the list of additions as Vidua principalis, the common and well-known "Pin-tailed Whydah bird." Last summer this bird had assumed an entirely blue-black plumage, like that of Hypochera nitens, also a common cage-bird. But I was struck by the appearance of the beak and feet, these being of a bright coral-red colour, whereas in H. nitens they are only pale flesh-coloured. The Vide Grandidier's work, Atlas, Ois. pis. 85, 86. t P.Z.S. 1874, p. 119, and 1. c. 1878, p. 99. \ The osteology of these genera, with some other points, is figured in Grandidier*g work on pla. 97-99, 101, 102, 103a. Ibis, 1871, p. 187. || Proc. Zool. Soc. 1880, pp. 475-477 PL XL VII. Read June 15, 1880. 160 ON TWO PLOCEINE BIRDS, tail-feathers, too, were slightly tipped with white, and the two central ones became gradually slightly more lengthened than the others, and so projected beyond them. The accompanying figure (Plate VII. fig. 1) shows the appearance of this bird at that time, as sketched from life by Mr. Smit. Unfortunately it died on the 29th of March in the present year, being then in very poor plumage, as it was moulting ; 011 dissection it proved to be a male. Thinking I had here a new species of Hypocliera to deal with, I took the skin with me, on a late visit to Berlin, to show to Drs. Hartlaub, Cabanis, and E/eichenow. The latter gentleman speedily recognized this bird as the young male of a species he had lately described from E. Africa as Vidua splendens (Orn. Centralbl. 1879, p. 114). Of this only a single specimen was collected at Kibaradja, E. Africa, by Dr. Fischer, and is now in the Berlin Museum. A sketch from this bird is reproduced in the distant figure of the accompanying Plate ; as will be seen from it, the male bird, when fully adult, possesses enormously elon- P Z S 1880 g a ^ e ^ rectrices, the two of each side fitting into each other, so that at p. 476. first sight there only appear to be two on each side. The excess of these four central tail-feathers over the other rectrices is, in this specimen, nearly 6 inches : in the young male that lived in our Gardens, there are only two lengthened feathers, which project only to the extent of i inch. If our bird had lived, the other two tail-feathers would, no doubt, have been duly developed, the birds in other respects being nearly similar. In our specimen all the rectrices, including the two central ones, are slightly tipped with white ; the eyes were very dark red-brown. Vidua splendens is perhaps most like Vidua hypocherina of Verreaux*; from that species it can be at once distinguished by the absence of the white, elongated rump-feathers. Vidua (Hypochera) nitens is also entirely blue-black, but has no lengthened rectrices, and, moreover, has the feet and beak flesh- colour ; in V. principalis the beak is also bright red, but the feet are fleshy, besides many other differences. The discovery of this bird renders, in my opinion, the retention of the genus Hypochera, founded by Bona- parte t in 1850, for Fringilla nitens unnecessary the males of V. splen- dens and V. principalis, as well as probably V. hypocherina $ when in winter plumage, as well as the females and young males at all times, being indistinguishable by any characters, structural or otherwise, from that section of the group including V. nitens and V. nif/errima (Hypocliera nigerrima, Sharpe, P. Z. S. 1871, p. 133), in which the male has, in nuptial plumage, no elongated rectrices. * Vidua hypocherina, J. and E. Verreaux, Eev. et Mag. Zool. 1856, p. 260, t. 16. t 0. R xxxi. p. 434. J Vidua superciliosa (Vieill. Gal. Ois. pi. 61) I only know by the plate and de- scriptions ; it is said to have only two elongated rectrices ; as there are said to be ten of the ordinary length, this statement is therefore probably correct. PI. VII P.Z.S.1830.FLXLVII. J .SimlliLH. Hannar'L irno l.VIDUA SPLENDENS. 2.PYTELIA WIENERI ON TWO PLOCEINE BIRDS. 161 Our specimen was said to be from the " east coast of Africa," a fact rendered probable by the arrival along with, it of specimens of Euplectes nigriventris *, a truly eastern species. 2. P r TELIA WIENEEI. (Plate VII. fig. 2.) Pytelia wieneri, Finsch, Gef. Welt, Aug. 9, 1877. Pytelia cinereigula, Cab. Orn. Centralb. Dec. 1, 1877, et Journ. fiir Orn. 1878, p. 101. In the ' Gefiederte Welt ' (6th Jahrg. no. 32, p. 317) for Aug. 9th, 1877, Dr. Finsch described as new, under the above title, a species of Pytelia, of which Mr. A. F. Wiener, F.Z.S., had purchased four living specimens in London, supposed to be from " Australia." On June 18th, 1879, Mr. Wiener presented one of these specimens to the Society, which is still (May 31) living in the Parrot-house in good health. From it the accompanying drawing has been taken (Plate VII. fig. 2). In the ' Ornithologisches Centralblatt ' for Dec. 1, 1877 (p. 182), Dr. Cabanis described a Pytelia f cinereigula, of which there had been two p. z. S. 1880, specimens lately received at the Berlin Museum from East Africa. One P- 477. of these had been collected at Zanzibar by Dr. Fischer, the second at Mombassa by Drs. Hildebrandt and von Kalkreuth. During my late visit to Berlin I at once recognized in this species Finsch's Pytelia wieneri ; and by the kindness of Drs. Cabanis and Reichenow I was allowed to bring back with me to London a third skin of the same bird, still more lately received, and collected in Angola, at Malange. A comparison of this with our living bird has quite confirmed the opinion I had already arrived at, so that Cabanis's name must yield to Finsch's . The Aus- tralian habitat is, of course, a mistake, Pytelia being an entirely African form. Of the red-beaked section of Pytelia, to which it belongs, P. wie- neri can only be confused with P. melba and its ally (or geographical form) P. citerior. The differences between these and the bird under con- sideration have already been pointed out by Drs. Finsch and Cabanis in their descriptions ; suffice it to say that P. wieneri is at once, inter alia, distinguished from these by its very different markings below, and also by the red of the chin and throat being separated from the greenish-yellow of the lower parts by the interposition of a grey band. In our living bird the beak is bright red and the feet pink ; the irides are dark red. * Cassin, J. Ac. Phil. 1849, p. 242, pi. xxxi. fig. 1. Erroneously entered in the register (P. Z. S. 1878, p. 1008) as E. oryx. Cf. List Vert. 1st supplem. 1879, p. 65. t This description is reproduced in the Journ. fiir Orn. 1878, p. 101. I may here remark that, in my opinion, Pytelia, though perhaps a " nonsense name," is sufficiently " like Latin " to be retained, and not replaced by " Zonogastris," or altered into " Pytilia" as proposed by Dr. Cabanis (L c. p. 100). \ I also found a single specimen of this bird, with no precise locality, in the Museum at Hamburg. M 162 ON A SPECIMEN OF DENHAM's BUSTARD. p.z.s.1880, 29. NOTE ON A SPECIMEN OF DENHAJVTS BUSTARD P- 477 - (EUPODOTIS DENHAMI)* THE interest attached to the existence, or otherwise, of special mecha- nisms connected with the habit of " showing off " in the males of the Otididse, together with the fact of the subject of the present note being of a species rarely seen in captivity, so that some time may elapse before a further opportunity of examination offers itself, must be my excuse for this short and imperfect notice. On March 20, 1872, two specimens of Eupodotis denliami, from W. Africa, I believe, the first and only ones of this species possessed by the Society, were presented by Governor Ussher and C. D. O'Connor, Esq. Of these one lived in good health in the Gardens for many years, dying on May 12 last, after having been attacked by a companion hen of Otis tarda that was in the same enclosure with it. Having never observed any signs of " showing off " in this bird, I had always considered it to be a female. This surmise, however, proved incorrect, for on dissection it turned out to be a male. P. Z. S. 1880, On examining the mouth there was no trace of any sublingual or gular p. 478. pouch ; on the contrary, the frcenum linguae was well developed in its ordinary position. But the oesophagus, for the greater part of its course in the neck, though stopping short considerably of its entrance into the thoracic cavity, was much dilated, so that by blowing it up with a blow- pipe a large distention of the neck took place, confined, however, to the upper two thirds, or thereabouts, of the neck the oesophagus, which, as usual in the Otididse, develops no crop, being in the rest of its course, till it entered the stomach, of very much smaller calibre. On inquiring of Mr. Bartlett and the keeper, J. Church, whether they had ever witnessed any display on the part of this bird, they both told me that during the last two summers (1878, 1879) it bad " shown off." But this display did not take place in the same way as in E. australis, as depicted and described by Dr. Murie (P. Z. S. 1868, pp. 474, 475, pi. xxxvi.) ; for there was none of that downward distention of the oesophagus, and consequent trailing of it on the ground, that is so marked a feature in the showing-off of that species ; on the contrary, the distention of the O3sophagus during display in E. denhami is lateral, the neck being im- mensely puffed out on both sides in a globular way, and so resembling when seen from in front, to use Mr. Bartlett's words, " a lady's muff." In E. australis, it must be remembered, there is a similar dilatation of the oesophagus, extending, however, in that species, over a larger extent of its course, so that " before dissection, by filling its cavity with air, the * Proc. Zool. Soe. 1880, pp. 477, 478. Bead June 15, 1880. ON THE DIGESTIVE SYSTEM OF BIRDS. 163 lower portion of the dilated O3sophagus protruded downwards considerably in front of the symphysis furculce, and formed the depending portion of the sac which was so conspicuous in the living animal " (Garrod, P. Z. S. 1874, p. 473). 30. REMARKS ON DR. GADOW'S PAPERS ON THE ibis, 1880, p. 234. DIGESTIVE SYSTEM OP BIRDS.* DR. HANS GADOW, who is already known as a worker at the anatomy of birds from his papers in the ' Journal fur Ornithologie/ has lately published in the ' Jenaische Zeitschrift ' an elaborate article on the anatomy of the digestive organs of birds t. This paper, in two parts, extends over more than 140 pages, and is illustrated with nine plates. In the first part Dr. Gadow gives a descriptive account (without histological details) of the alimentary canal and its appendages tongue, liver, pancreas, cseca, &c., as well as of the kidneys, which he strangely includes with these organs in the various orders of birds, in large part based upon his own investigations. I cannot always agree with Dr. Gadow in his grouping of the various forms together, as, e. g., including such different forms as Auks, Penguins, and Grebes in the same order, " Pygopodes." For this part of his work Dr. Gadow seems to have examined a large number of birds ; but it is to be regretted that he has apparently had no opportunity of investigating some of the most interesting forms, such as the Tinamidae and Turnicidse. Had Dr. Gadow been acquainted with the structure of the former group, he would not, I think, have insisted ^ a > 1880, so strongly on the absolute isolation of the " Ratitaa " from all other living birds +. Dr. Gadow justly regrets the small amount of attention that has been paid of late years to the anatomy of birds, and particularly, as he says, to the digestive system. But he seems to be unacquainted with the work done lately in this country by Prof. Garrod, as in the list of papers quoted by him but two of that anatomist's are mentioned. Hence no account is given of some of the most peculiar variations that are * Ibis, 1880, pp. 234-237. t " Versuch einer vergleichenden .Anatomie des Verdauungssystemes der Vogel," I. TheH, Jen. Zeitschr. f. Wissenschaft., Band xiii. Heft 1, pp. 92-171 (1879) ; II. Theil, torn. cit. Heft 3, p. 339 &c. (1879). \ "Dass aber .... Uebergange der 4 noch lebenden Ratitenfamilien unter einander und auch gu den Carinaten gdnzlich feklen" torn. cit. p. 107. M 2 164 ON THE DIGESTIVE SYSTEM OF BIRDS. known to occur in the alimentary canal of birds, of, for instance, the peculiar proventriculus and caeca of Chauna (though Dr. Crisp's paper on this bird is quoted), and of the extraordinary stomachs of the species of: Plotus. Xo allusion is made to the tongue of Nestor ; and the old statements as to the universal absence of a gall-bladder, or, at all events, its only exceptional presence as an individual variety, in the Parrots and Pigeons, are repeated. In the second part Dr. Gadow commences with the different forms of the same organs throughout the series of birds. In a tabular statement of the correlation of the nature of the caeca to the nature of the food, Phcenicopterus is included as one of the " Eleisch u. Fische " eaters. But, according to Mr. Salvin and other authorities, the major part of the food of these birds consists of the vegetable matter that grows at the bottom of the lagoons which they frequent (vide Dresser, B. of Eur. pts. 75, 76). The length and width of the alimentary canal, the relative size of its various parts, the influence of the food on the canal as a whole, and the variations in its length in birds of the same species, both young and old, are then discussed. The concluding part of the paper is devoted to a description of the disposition of the convolutions of the intestines ; and this is decidedly the most novel part of Dr. Gadovv's work, previous accounts of this subject being very meagre. Ibis, 1880, Excluding the Ratita?, Dr. Gadow distinguishes three chief types of p. 236. intestinal arrangement. These he calls Orthoccela, Plagioccela (or Plagio- brochi\ and Cycloccela. In the first group (Orthoccelci), in which, he includes the Pygopodes, Steganopodes, Anseres, Tubinares, Erodii, Alectorides, and Rallida?, the chief folds, which are from five to eight in number, are straight and parallel to one another and to the long axis of the body. In the Plagiocoela, which includes only the Easores, the two middle of the four chief folds form more or less horseshoe-shaped loops at their extremities, and the general direction of the intestines always forms, more or less, an angle with the long axis of the body. In the Cycloccela one or more of the chief folds are spirally coiled round their ends. This division includes some of the Grallse, the Pelargi, Laridae, Psittaci, Raptores, Columbae, the Coccygomorphae and Pici in part, the Cypselomorphae, and the Passeres. It is further subdivided into the Telogyri, in which only the terminal part of the fold is coiled, and the Hologyri, in which the whole fold is so disposed, these latter, again, being further divided into Progyri, Mesogyri, Amphigyri, and Polygyri. But, judging from the forms associated together under some of these heads, the groups so named are eminently artificial. Thus, the Eaptores are divided amongst the first three, and under the Mesogyri are included forms as various as Astur, Melierax, Halcyon, and Phcenicopterus. ON BUTTERFLY-COLLECTING IN THE ALPS. 165 The paper concludes with the inevitable phylogenetic table, showing Dr. Gadow's ideas of the lines of descent amongst birds. He holds that all the highest forms of each subdivision belong to the Hologyri or Mesogyri, the more primitive ones being Ortlioccela or Playioccela. The figures in the plates are chiefly devoted to showing, in a more or less diagrammatic way, the various types of intestinal convolution described in the text, and will be found very useful in elucidating Dr. Gadow's views. In conclusion, it seems to me that, as it is a well-known fact that individuals of the same species vary, sometimes very greatly, in the Ibis, 1880, length of their intestines, the stowing away of a greater or less amount P- 237 ' of gut in a given space, the abdominal cavity, becomes simply a mechanical problem, and therefore that there is less help in forming a sound view of the mutual affinities of birds to be derived from the facts in this direction described by Dr. Gadow than from many other points, more complicated, and therefore less easily altered, in the structure of birds. 31. THREE WEEKS' BUTTERFLY-COLLECTING IN EntM.M.xvi. THE ALPS.* P. 256 (1880). THE following is an account of a short trip in the Alps of Dauphine and Piedmont made last summer by myself, in company with Messrs. Salvin and Godman, and Capt. Elwes. Our object was quite as much to enjoy a change and breathe fresh air, as to catch butterflies, though we devoted most of our time to the latter pursuit. We left London on June 22nd, and reached it again on the llth of July, so that we were only about three weeks, and as we got over a good deal of ground in that time, rarely staying more than one night in a place, a large part of our trip was spent in travelling. Our route was as follows : from Chambery we drove, by St. Laurent du Pont, a village close to the famous monastery of La Grand Chartreuse, to Voiron, and thence by rail to Grenoble. From there we proceeded to Bourg d'Oisans, and next day over the Col du Lautaret a driving pass about 6800 ft. high to Briancon. Mr. McLachlan t had made known to us before starting his * Ent. Month. Mag. xvi. pp. 256-259 (1880). t "I visited this part of the Alps of Dauphin^ as far as the Col du Lautaret, in tne beginning of July, 1876, in company with M. Constant (then of Autun, now of Cannes), who joined me at Grenoble, and a botanist from the neighbourhood of Bordeaux. This 166 ON BUTTERFLY-COLLECTING IN THE ALPS. experiences some years ago of this part of the Dauphine Alps ; but unfortunately we were too early for Lepidoptera, the snow lying still thickly about the top of the pass above La Grave, where, indeed, we narrowly escaped being carried away by a small avalanche a catastrophe that happened to a small cart that had preceded us by about an hour. At Briancon, although over 4000 ft. above the sea, we got for a while into a more southern fauna, as evidenced by the occurrence of such forms as Melitcea dejone, and the beautiful yellow " orange-tip " Anthocharis Ent.M. M.xvi. euphenoides. The Mediterranean fauna would, therefore, seem to extend up the valley of the Durance quite into the Alpine district. From Briancon we drove by Mount Grenevre, a pass of about 6000 ft., over the frontier to Oulx, a small village (at an elevation of 3500 ft.) on the Mount Cenis district had long been known to members of the Alpine Club, and possessed the peculiar attraction of a mountain (La Meije, over 13,000 ft.) that had, up to that time, baffled all attempts to scale it (it has since been several times successfully ascended), in addition to many other inducements for mountaineers of the more amateur class ; it was also well known to botanists as a paradise for rare alpine plants, and it supplies (through its adventurous and migratory inhabitants) many of the horticultural establishments of Europe (and even of America) with them, either in the form of seeds or roots. French entomologists had also visited it ; but it had rarely seen an English net ; yet there are probably few districts in Europe so favourable for a Lepidopterist ; it is not favourable for a Neuropterist, owing to most of the streams having their source in glaciers. It has the advantage of a magnificent military road, a wonderful piece of civil-engineering. British tourist-entomologists should decidedly make its intimate acquaintance. It is easy of access. From Grenoble to the summit of the Col du Lautaret is about 50 English miles by diligence and mail. Grenoble can be reached from London in about 27 hours (on my return I left that city at 3.15 P.M., and was at home in my study before 7 P.M. next day). The end of June is too early, even in an ordinary season, and in such a season as 1879 was a month too early. I would recommend entomological tourists (not pressed for time, nor wanting to go over too much ground) to stay first at Bourg d'Oisans, where there is a comfortable inn, kept by an obliging old Frenchman, M. Martin ("Hotel de Milan"). Afterwards they should push on to the Col du Lautaret, where there appears to be good accommodation at the Hospice on the summit (subsidized by government, as a refuge for wayfarers in the long winter months). My head-quarters were at Bourg d'Oisans and La Grave, the latter at the foot of La Meije. But I think (for an entomologist) the Hospice is preferable to La Grave. This latter is a miserable village with a poor inn, offering no special inducements, excepting to Alpine climbers : the sleeping-quarters were over the stable (which is, perhaps, cleared out once a year), the food was indifferent, the charges not moderate ; and, moreover, newly-arrived strangers are liable to an indisposition (already alluded to several times in the records of - mountaineering), that may place them (as it did me and one of my companions) hors de combat, and take several days to shake off. (The water, and the sudden change of temperature from the excessively hot experiences of Bourg d'Oisans were both blamed for this; but there has been no report from an official sanitary inspector !). Any British entomologist who is not specially connected with water insects should visit this district ; and even the exception I have made would, perhaps, not hold good in the autumn months, when the glaciers have discharged their annual surplus." E. MC-LACHLAN. ON BUTTEJftFLY-COLLECTING IN THE ALPS. 167 railway, between Bardoneche and Susa, and after a day there, proceeded to Turin. Spending the Sunday there, we, after a good baking, were glad to get away early next morning, and travel by rail to Arona, and then up the lake by steamer to Baveno. Baveno being hot and crowded, we left next day, and drove up the Yal Anzasca to Ponte Grande, a charming village about 2500 ft. above the sea, with a lovely view of Monte Rosa. Finding good quarters here, we stayed several days (from July 1-5). The Val Anzasca is a good example of an Italian alpine valley, and proved like- wise very productive in insects, though the weather was not as fine as it might have been. We only had one really fine day, July 3rd, and on that Mr. Salvin and I working down the valley towards Vogogna, saw or caught fifty-two species of butterflies, not a bad day's work for one morning between 8 A.M and 2 P.M. In this valley below Ponte Grande alpine and southern species were curiously interblended, as evidenced in such forms as Neptis and Libyfhea occurring with Parnassius and other mountain insects. From Ponte Grande we went further up the valley to Macugnaga, and after spending a day there, over the Monte Moro pass (about 9000 ft.), and down the Saas Thai to Saas, and eventually Visp. After this, except for an hour or two near Bienne, on our way home, we had no occasion to use our nets. The total number of species of Rhopalocera seen or caught by us during the trip was 103, and, had the weather been finer, this number would, doubtless, have been increased. We altogether missed numbers of common Alpine species, as owing to the unusual amount of snow that had fallen during the winter, the season was extremely backward, so that had we started a fortnight later, our " bag " would, no doubt, have been correspondingly increased. A list of some of the more uncommon species we obtained is appended. Papilio podalirius : Chambery, Col du Lautaret, Briancon, &c. Parnassius apollo : Val Anzasca, Col du Lautaret, &c. P. mnemosyne \ Val Anzasca, above Ponte, Macugnaga. Pirn's napi, var. bryonice : several near Macugnaga. Anthocharis belia, var. simplonia : rather common towards, and on, the top of the Col du Lautaret. A. euphenoides : this truly Mediterranean species occurred, but not commonly, at Briancon ; one specimen was seen at an elevation of about 5000 ft., on the road towards Mont Genevre. All seen were males. Leucophasia duponcheli : Oulx (?) and Briancon. At the time we did EntM.M.xvi. not distinguish this from the common species, so only got two or three P- 258 ( 188 )- specimens. According to M. Bellier de la Chavignerie (Ann. Soc. Ent. France, 1869, p. 514), this species is, in France, almost confined to the lower parts of the Basses Alpes, and the neighbourhood of Digne, and Aix in Provence. Colias edusa, var. helice : Oulx, Chambery, &c. 168 ON BUTTERFLY-COLLECTING IN THE ALPS. Theda ilicis: round bushes in the Val Anzasca. This and T. rubi were the only " hairstreaks " met with. Polyommatus virgaurece : Val Anzasca, not numerous. P. Tiippotlwe, var. eurybia : upper parts of Yal Anzasca, and near Macugnaga. P. alci- phron, var. gordius: this beautiful "copper" was abundant in the Yal Anzasca, flying about, and settling on, the flowers by the sides of the road. The males were by far the most numerous; a few were also caught at Baveno and near Briancon. P. dorilis: near Chambery, Briancon, Val Anzasca, &c. ; the males commoner. The alpine form (subalpina, Speyer) occurred at Macugnaga. Lyccena argyrotoxus (=cegori) : St. Laurent du Pont, &c., common. L. argus : very common in the Val Anzasca ; also on the Col du Lautaret, at Oulx, and Baveno ; most of our specimens are referable to the form cegidion (Meissner). L. orion: not uncommon, flying over the road, particularly where muddy, in the lower parts of the Val Anzasca, but local, and generally worn. L. baton : one at Bourg d'Oisans, and a few at Briancon and Macugnaga. L. eros : Oulx, and more commonly in the Saas Thai, above Stalden. L. icarus, ab. icarhms : Oulx. L. eumedon : Oulx, and Saas Thai, above Stalden ; nowhere common. L. escheri : Chambery, near Bourg d'Oisans, Oulx, and Stalden, singly. L. Tiylas ; common at Oulx, Saas Thai. L. sebrus : Col du Lautaret, Oulx, males only ; we probably passed this species over as the next in many cases. L. semiar.gus : common at Briancon, Oulx, &c. L. cyllarus : rather common on the Col du Lautaret, and about Briancon, Baveno, and Val Anzasca. L. alcon : Oulx, a few. L. arion : Col du Lautaret, Oulx, Saas Thai, &c. Nemeobius luclna : Chambery, Val Anzasca, &c. Libythea celtis: I saw, and caught, a single specimen of this S. European species in the Val Anzasca, below Poute Grande. This was the only one seen. Apatura ilia, var. clytie : a single specimen of this species was seen, but not secured, on the roadside near Baveno. Limenitis populi : a fine female near Ponte Grande ; we saw another higher up the valley, but failed to catch it. L. Camilla : I caught a single specimen at Oulx, the only one we saw. Ent. M.M. xvi. Neptis lueilla : we got two or three specimens, only in the Val Anzasca, p. 259 (1880). O f this species. This must be nearly its most western habitat. Melitcea pTioebe : Briancon, &c., very common in the Val Auzasca. M. didyma : Chambery and Val Anzasca. M. dictynna : Briancon, Macugnaga. M. dejone : a few specimens at Briancon ; this species is, according to Dr. Staudinger, confined to Spain and the South of France. M. athalia : abundant nearly everywhere in suitable localities ; in swarms in the Val Auzasca, with M. phoebe and others. M. parthenie : St. Laurent du Pont. ON BUTTERFLY-COLLECTING IN THE ALPS. 169 Aryynnis amathusia : common at Oulx, Dear Macugnaga. A. thore : Godman caught a single specimen of this rather scarce species in the Val Anzasca, above Ponte Grande. A. lathonia : common in the Val Anzasca; this species seems fond of settling on the dusty roads, and has a peculiar jerking flight, unlike the other species of Aryynnis. A niobe : a single specimen of the typical silvery- spotted form at Briancon; curiously enough, we saw nothing of eris, which is usually the commoner of the two. Erebia melampus: Val Auzasca and Saas Thai. E. epipJiron, var. cassiope: Saas. E. ceto: Col du Lautaret, Val Anzasca, and more commonly near Macugnaga. E. medusa, Val Anzasca, Macugnaga, &c. E. stygne : Col du Lautaret, Briancon. E. evicts : on the Col du Lautaret near La Grave, but mostly worn ; also near Macugnaga. E. euryale : Val Anzasca. (Eneis aello : not very uncommon near Macugnaga, and also caught in the Saas Thai, between Stalden and Saas. Satyrus dlcyone : Val Anzasca and Saas Thai, common near Stalden. S. semele, near Stalden. S. actcea, var. cordula : a few near Stalden. Pararge mcera : common everywhere in the alpine valleys. P. hiera : Oulx, Val Anzasca, nowhere abundant. P. achine (=dejanira) : two specimens in a wood near Bienne. C&nonympha arcania, var. darwiniana : Chambery, Baveno, and jommon in the Val Anzasca. The alpine form satyrion occurred sparingly in the Saas Thai. Spilothyrus althcece : we got two specimens of this rather scarce species in the Val Anzasca. S. lavaterce : this skipper was not uncommon one one hot day flying over the road in the Val Anzasca, but it was very lively and difficult to catch ; we subsequently saw it again in the Saas Thai, above Stalden. Syricthus cartliami: Briancon, Oulx, and Saas Thai. S. sao : near Chambery, Briancon, Oulx. S. alveus : Saas Thai and Oulx. Hesperia tJiaumas : Ponte Grande, Saas Thai. H. lineola : Saas Thai and Val Anzasca. Carterocephalus pdlcemon (=paniscus) : a single specimen caught by Salvin near Chambery. February 23rd, 1880. 170 OX THE MALE GENERATIVE ORGANS R fP- B - As * oc - 32. ON A LITTLE-KNOWN CRANIAL DIFFERENCE 1881, p. 718. BETWEEN THE CATARRHINE AND PLATYR- RHINE MONKEYS.* BESIDES the well-known difference in the dentition, and in the form of the external auditory meatus, in the monkeys of the old and new worlds, there is a difference in the formation of the bony walls of the temporal fossa which in nearly every case suffices to distinguish at once the skull of a member of one of these groups from that of one of the other. As independently discovered by the author (P. Z. S. 1880, p. 639) and Dr. Gustav Joseph (' Morphologisches Jahrbuch,' i. pp. 453-465), in the Platyrrhine monkeys the parietal bone is prolonged forwards to meet the malar, there being a well-marked suture usually between the two, the frontal being in consequence altogether excluded, superficially at least, from articulating with the squamosal and alisphenoid. In the Catarrhine monkeys, on the other hand, as also in man, the parietal does not reach the malar, there being an isthmus between the two bones formed by the articulation of the frontal with the alisphenoid. T.z.s.1881, 33. ON THE MALE GENERATIVE ORGANS OF THE xi. p. 107. SUMATRAN RHINOCEROS (CERATORHINUS SU- MATRENSIS.^ ON two occasions the late Prof. Garrod had opportunities of dissecting the Sumatran two-horned Rhinoceros ; and his notes on their anatomy will be found duly recorded in the Society's publications J. Both his specimens were females. On March 20, 1879, the Society received on approval a fully adult male of this animal, being, I believe, the first individual of that sex * Rep. Brit. Assoc. 1881, p. 718. t Trans. Zool. Soc. vol. xi. pt. iv. pp. 107-109, PI. XX. (1881). Eead March 16th, 1880. Prof. Garrod had the drawings which accompany this paper made by Mr. Smit from the animal whilst still fresh, with the object of laying some notes on the subject before the Society. Unfortunately I have been unable to find any such amongst his numerous MS. papers. He also requested me to make notes and measurements of the male organs for him with the like object; and from these sources I have drawn up the present paper. The glans penis is now preserved in the College of Surgeons. t Proc. Zool. Soc. 1873, p. 92, and Trans. Zool. Soc. x. p. 411 (1878). OF THE SUMATRAN RHINOCEROS. 171 brought alive to Europe. Unfortunately it died on the 5th of April following, the post-mortem examination showing evidence of dropsy, as well as tubercle in the lungs and spleen. The skin and skeleton of this specimen are now in the British Museum. Prof. Owen, in his account of the anatomy of Rhinoceros indicus (Trans. Zool. Soc. iv. pp. 31-58), has described and figured the male organs of that species ; and the present account will fill up the corre- sponding blank that has as yet existed as regards these parts in Cerato- rhimis sumatranus. As was to be expected, the two genera closely conform with each other in all main points, with some considerable differences in matters of detail. As in R. indicus, there was no scrotum ; each testis measured 4| inches long by 2 broad at the widest part. The epididymis was of the same length as the testis. The vasa deferentia were 29 1 inches long by J inch broad ; unlike these ducts in the Indian species, they were not dilated terminally. The vesiculsB seminales resembled in shape those described by Owen : they were 7J inches long, and 1 inch across at the broadest part. The right vesicula had two, the left four, narrow ducts, l|-2 inches long, which joined the vasa deferentia just before these entered the urethra. The verumontauum is short and rounded, | inch long and 1 inch broad. The openings of the ejaculatory ducts were very minute ; a larger pore, which was the only representative of a vesicula prostatica, lay close above. The prostate was of a roughly triangular shape, 2 inches long by 5 T. Z. S. 1881, inches across, and had the same structure as in R. indicus, the glands opening by numerous pores on each side of the verumontanum in a well- marked sinus prostaticus. Cowper's glands were large (3| inches by 2) and oval; their ducts opened by pores 1| inch in front of those of the ejaculatory ducts. The urethra measured in all, in the unerected state, about 23| inches, of which \ inch was " prostatic," 3 inches " membranous," and the rest " spongy." The glans penis is a long and tapering cylinder, provided at the end with a second, somewhat mushroom- or trumpet-shaped expansion, nearly in the centre of which is the opening of the urethra. It thus conforms closely with the same organ in R. indicus. But, as will be seen from the drawings, it is provided, in addition, with two large oblong-oval lobes, of the same colour and substance as the rest of the glans, which are free for the greater part of their length, and only attached to the rest of the glans at their bases. These lobes lie on the sides of the dorsum of the penis, and are closely approximated at their bases, as represented in fig. 2. In fig. 1 they are spread out 172 GENERATIVE ORGANS OF THE SUMATRAN RHINOCEROS. artificially, so as to show better their extent and attached bases. The total length of the glans, to the reflection of the prepuce, was 7 inches, the trumpet-like terminal part being 1 inch long, and 1 inch trans- versely. The lobes of the glans measured 2| inches long by 1J inch across. In R. indiwu, according to Prof. Owen (I. c. p. 51), " on each side of the base of the glans, and. rather towards its under part, there is a longitudinal thick oblong ridge or lobe, 3^ inches in length, and 8 lines in basal thickness ; the thick rounded free border of each lobe inclines downwards." Prof. Owen's figure is reproduced in outline, of the original size, in fig. 3, to show the differences thus indicated. By the kindness of Prof. Mower I have been enabled to examine the penis of an Indian Rhinoceros preserved in the stores of the College of Surgeons, and which is probably the same specimen as that dissected and described by Prof. Owen, with whose description and figures it closely corresponds. The lobes, however, seem to me to be (as also indicated in his figures) rather on the upper than on the under part of the penis, as they lie, in fact, on each side of the dorsum a little removed from the middle line, as also is the case in Ceratorhinus. They are about 1| inch in height at the centre, diminishing towards each end till they become undistinguishable from the rest of the glans. Ceratorhinus therefore differs from restricted Rhinoceros in the greater size and development of the lobes, which have now ceased to be mere elevations or ridges attached throughout their length to the body of the glans, but have become freely projecting lobes attached only by their bases *. In R. indicus, too, the terminal part of - S ' 1 QQ 81 ' the glans is more slender, being longer in proportion to its depth, and its apical expansion narrower across in proportion to its height (J inch to 1-J), with its margins, moreover, somewhat crinkled. It is, in conclusion interesting to observe that the distinctness of the two genera Rhinoceros and Ceratorhinus, as shown by other characters external, cranial, and visceral is confirmed by these differences in the sexual organs. * I may mention that Prof. Flower also found for me in the stores of the College of Surgeons a detached glans penis of a Rhinoceros exactly like that now described. Its history is somewhat uncertain ; but it was probably sent over, along with other viscera of animals, by Sir Stamford Raffles when Governor of Java. There can be no doubt that it belongs to a species of Ceratorhinus. ON THE ANATOMY OF THE KOALA. 173 34. ON SOME POINTS IN THE ANATOMY OF THE p - z - KOALA (PHASCOLARCTOS CINEREUS}* ON April 28th of last year (1880), as already recorded in the Society's Proceedings t, the Society purchased for its collection the first living Koala (Phascolarctos cinereus) ever brought to Europe. The animal, a young female, continued to do well and thrive after its arrival at the Gardens, and on a diet of Eucalyptus-leaves, which were substituted after a while for the dried ones on which it had been kept alive during the voyage and the first part of its stay in this country, became daily in better condition and more active. Being a pet animal, accustomed to being caressed, it was thought better not to put it in a cage ; so a room for its use was fitted up in the Superintendent's office. Here, under the charge of a special attendant, it slept, perched upon the branches of a P- Z. S. 1881, tree erected for its use, by day, whilst at night it wandered about the room. Very unfortunately, on the night of the 14th of June it was accidentally killed, whilst thus roaming about at night, by getting caught between the top and bottom of a fixed washing-stand, which had been allowed to remain in the room. It had apparently climbed up this and brought down on its neck the heavy lid. Nobody being near, and in spite of evidently determined struggles on its own part, it failed to relieve itself, and so was found dead in the morning from asphyxia. The death of this animal, so unfortunate for visitors to the Society's Gardens, has given me the opportunity of putting on record some additional facts concerning the anatomy of the soft parts of this species. Mr. W. Martin, in this Society's 'Proceedings' for 1836 1, has described already some of the most striking features of the animal's organization ; and in Prof. Owen's ' Anatomy of Vertebrates ' (vol. iii.) a few additional facts concerning it are also recorded. More recently Mr. A. H. Young has described and figured the male reproductive organs (Journ. Anat. Phys. xiii. pp. 305-317, pi. xviii.). All these anatomists, however, had only spirit-preserved specimens to work on ; a few additional observations from the fresh specimen may therefore be worth putting on record, and the liver, brain, and female reproductive organs described in particular, these important parts of the system having been only imperfectly, or not at all, described by my predecessors in this field. The following dimensions were taken on the body of the animal : * Proc. Zool. Soc. 1881, pp. 180-195. Read Jan. 18, 1881. t P. Z. S. 1880, p. 355; where a woodcut of it, from the pencil of Mr. Wood, is given. t P.Z.S. 1836, pp. 109-113. 174 ON THE ANATOMY OP THE KOALA. inches, mi Him. Total length, from tip of nose to end of body .... 17-0 432 Length of eye .............................. 1-0 25 ear (greatest) ...................... 2-15 55 head ....................... ..... 4-0 100 nude muzzle ...................... 1-4 35 chin ............................ 2-65 67 Breadth across muzzle ............ .......... 1-0 25 of mouth .......................... 1-3 33 Distance between cloaca and mammae .......... 1-25 32 The tail is a mere stump above the cloaca, which latter is well defined by a well-marked circular marginal fold of the integuments. The hallux has no trace of a nail. The skin is generally flesh-coloured ; but the soles of the manus and pes, together with the naked " muffle," are black. The skin of the large and hairy ears is flesh-coloured. There is a narrow naked ring round the eyes ; and the irides are brown. The pupil is a vertical oval. The nostrils are transversely oblique, the nasal septum measuring 3| millims. The upper lip is split ; but the split does not quite, when the surrounding parts are expanded, reach the nostrils. The skin is sparsely covered with hairs between the rami of the mandible ; for nearly two inches behind it the skin is absolutely naked ; and on the 182 1 81 ' S ^ es ( runnul S U P t war( ls the angle of the mouth) it is nearly so, a patch of black hairs being developed just behind the mouth on the lower and outer surface of this bare space. The marsupial pouch in this young specimen is very imperfectly developed. It appears as a small, oval, nearly naked space, measuring about 0'8 inch both across and antero-posteriorly, with a well-marked bounding-fold of integument on each side; inside which is a smaller, secondary one. The hairs of the surrounding parts more or less radiate from this nude space, which lies between the epipubes (or so-called " marsupial bones "). The skin covering it is pinkish. The teats are two* in number, 15 millims. apart, and are situated at the posterior and inner angles of the bounding-folds ; they are covered by fur. The lips of the pouch, it may be noted, look as much downwards as forwards. In an adult $ Koala, 20J inches long, preserved in spirit, the pouch is much better developed its antero-posterior extent being about 1*85 inch, whilst the breadth of the aperture is 1*4 inch. It admits (my) three median fingers, and extends widely outwards into the groins, as far as the skin-fold between the knee and trunk. The teats, two in number, are situated behind, on a level with the posterior margin of the pouch's * Prof. Owen (Anat. Vert. iii. p. 769) describes Pkascolarctos as having four ["two on each side"] mammary glands. ON THE ANATOMY OP THE KOALA. 175 mouth. The skin lining the pouch, except just around the ventral opening of the pouch, is smooth throughout. The mucous membrane of the cheeks is smooth throughout ; the skin is attached to the gum opposite the first palatal ridge, and again opposite the posterior border of the first premolar. Between these two attach- ments there is formed a sort of cheek-pouch, defined by a distinct sphincter, and capable of receiving the end of the little finger. This pouch extends upwards on the side of the skull, occupying the somewhat oval space that exists, in the macerated skull, in front of the zygoma ; it is lined by smooth, white, mucous membrane*. The palate presents 9 irregular raised ridges, best marked anteriorly. There is no uvula, and the narrow fauces are smooth. The tongue quite fills up the space between the gums. It is parallel-sided and elongated, but rounded off and thinner in front. It has a single, small, circumvallate papilla behind ; the fungiform papillae are distributed chiefly along the sides of the upper surface. The salivary glands are well developed. The sublingual (which is not mentioned by Martin in his description) is a long, narrow, and thin gland, somewhat foliaceous at the extremity, and about 2-7 inches in extent, lying deeply along the inner margin of the lower jaw. The long duct of the submaxillary gland pierces it. I could find no subzygomatic gland, as described by him (1. c. p. 112). On opening the abdominal cavity the stomach is visible in the epigastric p.z. s. 1881, and left hypochondriac regions, the pylorus being directed towards the P- 183. right side ; and it is there in contact with the gall-bladder. The liver does not appear. The commencement of the transverse colon is visible, running downwards towards the left, below, but parallel with, the greater curvature of the stomach. The great omentum is atttached to the transverse colon in the right hypochondrium, and does not cover the mass of the viscera. The greater part of the rest of the abdominal cavity is occupied by the great, longitudinally plicated, folds of the caecum and caecum-like ascending colon, a few folds of the small intestine appearing between the transverse colon and a great fold, apparently the caecum, which runs transversely across the middle of the abdominal cavity. On turning back these great superficial folds the end of the caecum is seen passing downwards to the left of the rectum, behind the uteri and bladder, to terminate, deep in the pelvic cavity, close to the cloaca! The descending colon, which is narrow and of the ordinary appearance, is very long, and is arranged on a broad mesocolon to the right of the vertebral column, forming here a series of loose loops, which, however, are not closely coiled together on each other as in Ruminants. The * These pouches are also, I find, described by Owen (Anat. Vert. iii. p. 385). They also exist, though less well defined by a sphincter, in the Wombat ; but I cannot find them in the other Phalangers I have examined. 176 ON THE ANATOMY OF THE KOALA. right kidney lies superficially to the liver. The duodenal loop passes downwards and to the right, and overlies the right kidney, but passes under the ascending colon just here. The stomach is cylindrical and sac-like. Its length, moderately distended, is about 3$ inches; its greatest depth, opposite the pyloric constriction, 1| inch. There is a well-marked cardiac fundus to the left of the oesophagus, and the pyloric part is slightly bent on the cardiac part ; this latter is marked off internally by a distinct fold of the mucous membrane, which is smooth and pale, with some slight traces of rugae in the cardiac fundus. The most marked peculiarity of the Koala's stomach is its possession, as is well known, of a special gland-patch, similar to that found in the Beaver* and "Wombat f. This gland-patch forms a slight elevation externally on the lesser curvature of the stomach, just on the pyloric side of the entrance of the oesophagus. It is somewhat saddle-shaped, with a transverse extent of 1*4 inch. Internally it forms an eminence about the size of a florin, which includes the entrance of the oesophagus. The mucous membrane on the gland-patch, around the entrance of the oesophagus, is red and vascular ; elsewhere in the stomach, as already stated, it is quite pale. The openings of the gland-patch are about 30 in number, of varying sizes, and irregularly arranged over the eminence. Some of the openings of the gland are complicated, several smaller openings debouching into a larger one ; and the area occupied by the openings is not symmetrical. The general appearance of this patch is well represented by Sir Everard Home's figure (I. c.) of that of the Wombat. In this latter animal the general structure and form of the P Z S 1881 stomach are also very like that here described ; but it is more globular, p. 184. and therefore less cylindrical in shape, the cardiac and pyloric openings being more approximated. The small intestine is villous, but otherwise smooth. It is not sacculated, and when spread out, after being cut, is O5 inch across. At its commencement it is dilated for about two inches ; there are no Peyer's patches ; its length is 115 inches. The large intestine is very peculiar : for the first 28J inches or so of its length, which forms the ascending colon, it is very capacious, and internally longitudinally corru- gated, like the caecum, which externally it much resembles, the rugse of the interior appearing through the walls of the intestine, and giving it a longitudinally striated appearance. These folds of the mucous membrane, which might be called longitudinal valvulce conniventes, where best developed are about -2 inch in depth ; they are arranged longitudinally and are roughly parallel, though somewhat irregular in extent ; they are * Cf. Owen, Anat. Vert. iii. p. 422. t Home, Phil. Trans. 1808, p. 307, pi. ix. ON THE ANATOMY OF THE KOALA. 177 separated from each other by intervals of about the same extent (O2 inch). At the commencement of the colon, which here, when cut open and spread out, is 3*75 inches broad, and of the caecum, there are about a dozen of these folds very well marked. These continue throughout the ascending, caecum-like, colon ; but where it narrows to form the transverse and descending parts they converge, and become more or less blended with each other, forming linear elevations. They are continued down- wards as far as the rectum, but are reduced by that time to five*. In the caecum, which is also very capacious, the same arrangement of folds obtains till within 18 inches of its apex, when they gradually disappear, the rest of the organ being thence onwards quite smooth internally. The caecum, the curious position of the caput of which has already been described, measures 46*75 inches in length (nearly three times the length of the animal's body!); the large intestine 93-25 inches. In an adult female (20| inches long), preserved in spirit, the following were the intestinal measurements : inches. Small intestine 111-15 Large 160-8 Crccum 66-0 Owen (Anat. Vert. iii. p. 420) gives 92, 125, and 77 inches respectively. On each side at the junction of the ileum and colon is a small patch of three glands. The liver of the Koala is of very remarkable form. It is represented, drawn to scale of |- the natural size, in the accompanying figures (1 and 2, P. Z. S. 1881, pp. 178, 179). All four principal lobes are well developed ; but those p< 185< on the right are far larger than those on the left, the left central being considerably the smallest of these. The umbilical fissure is distinct, extending about halfway across the liver. The right central lobe, which is broad transversely, and forms the largest lobe, is divided very deeply by the large cystic fissure, which extends on the thoracic surface nearly as far back as the umbilical one, and allows the very large and elongated gall-bladder to appear above. Both right and left lateral fissures are also * Mr. Martin describes (1. c. p. Ill) both colon and caecum as sacculated "by a slight longitudinal (mesenteric) band of muscular fibres," with indications of a similar opposite band. I could find no traces of any such sacculation in the fresh Koala examined by me ; nor are they mentioned by Prof. Owen (Anat. Vert. iii. p. 418) It is also to be noticed that Martin does not in any way allude to the existence of the very remarkable folds of the interior of the caecum and colon. In Phalangista and Phascolomys an examination of fresh specimens has completely failed to exhibit any traces in either caecum or colon of the longitudinal folds here described. In the latter genus the colon is capacious at its commencement, and Bacculated transversely, in a way that does not obtain in either Phalangista or Phascolarctos. 178 ON THE ANATOMY OF THE KOALA. P.Z.S. 1881, p. 186. (in this specimen) well developed. The right lateral lobe is large and somewhat oval in shape, but pointed below. The caudate is not present as a free structure, but it is represented by a somewhat squared, diagonally ridged elevation, lying to the right of the inferior cava, and Fig. 1. Liver of Koala, from above ; three fifths the natural size. 1881, broadly attached to the substance of the right lateral lobe. The lower border of this elevation is slightly excavated to receive the corresponding kidney. The Spigelian is represented by a smaller thickening, ending in a pointed and free apex, and lying to the left of the vena cava ; it is united over this by hepatic tissue to the caudate. This liver is further remarkable for the great tendency it has to subdivision, numerous fissures, of varying sizes and depths, being developed along the margin of the chief lobes. Their position and relative size will be better understood ON THE ANATOMY OF THE KOALA. 179 from the figures than from any verbal description. They are more conspicuous on the visceral than the thoracic surface. The right half of the right central lobe has one such notch on its right external border ; the other half 3, on the right internal border; the left central has 4, the Fig. 2. P. Z.S. 1881, p. 187. G.U The same, from below. R. C. Eight central lobe; L.C. left central lobe; R.L. right lateral lobe; L.L. left lateral lobe ; C. caudate lobe ; SP. Spigelian lobe ; G.B. gall-bladder ; G.D. bile-duct; r.l.f. right lateral fissure; l.Lf. left lateral fissure; u.f. umbilical fissure; c.f. cystic fissure; V.C. vena cava inferior-, V.P. vena portce; V.H. hepatic vein. left lateral 3, whilst the right lateral is still more cut up by about 10. Finally, the caudate has 3 of these supplementary fissures. The gall-bladder is remarkably long, projecting far beyond the anterior margin of the liver, and, as already described, appearing superficially. N2 P. Z. S. 1881 p. 185. 180 ON THE ANATOMY OF THE KOALA. It is 2| inches long from its apex to the commencement of its duct opposite the anterior margin of the left central lobe. The free part is connected by a peritoneal investment to the sides of the cystic fissure. In a second liver of Phascolarctos examined (which, however, having been extracted from a spirit-preserved animal, an adult $ , is not so well preserved as might be wished) the same general features obtain. The left lateral fissure, however, is less distinct, as is the caudate ; and the left central lobe is smaller proportionally to the left lateral. The Spigelian wants the 'pointed apex ; and the development of secondary fissures seems to attain an even greater extent*. P. Z. S. 1881, The bile- and pancreatic ducts open into the duodenum 2| inches from the pylorus. The pancreatic duct is dilated terminally into a vesicle, which does not receive the bile-duct, the latter opening alongside the former into the intestine. P. Z. S. 1881, The spleen, as usual in Marsupials, is somewhat triradiate in shape, being broader and forked at one end, tapering and more pointed at the other. Its greatest length is 2 5 inches. There is a lymphatic gland, the size of a pea, outside each marsupial bone, and a pair of similar ones, superficial, on the neck. The axillary glands are large. The heart is of the usual Marsupial type. The right auriculo-ventri- cular valve in membranous, and nearly complete all round the aperture, P. Z. S. 1881, being largest on the right side. It is attached to two, or, in one of my specimens, three columnce carnece, which also decrease in size from right to left. On the side corresponding with the septum the valve is attached, * In Phalangista vulpina the right central lobe is also, as in the Koala, the biggest, and very deeply divided by a cystic fissure. The gall-bladder, however, does not nearly reach the margin of the liver ; the left lateral lobe is much bigger than the right lateral, which is as large as the left central ; the caudate is quite free and narrow. All the lobes are remarkably distinct ; and their margins are quite simple, with no trace of any such fissures as obtain in the Koala. The liver of Cuscus maculatus is formed on a similar principle, though the right central lobe is not bigger than the left lateral, and the gall-bladder reaches to the liver- margin. All the lobes are simple. The caudate and Spigelian, though small, are quite distinct. The livers of Belidens sciureus and B. breviceps, Acrobata pygmcea, and Dromicia (nana ?), though differing among themselves considerably in the relative degree of development of their constituent lobes, all agree in having a distinct and free caudate lobe, as well as a Spigelian, and in no system of secondary sulci attaining any degree of development. In Phascolomys wombat the left lateral lobe is the largest ; the right central is also large ; but the left central is very small, as is the right lateral fissure. The umbilical fissure is distinct, as is the cystic fissure, which allows the gall-bladder, which reaches to the anterior margin of the liver, to appear superficially. There are no distinct caudate or Spigelian lobes, though the former is indicated. There is a tendency particularly on the left lateral and right central lobes, to develop accessory sulci. As in the Koala, too, the small right lateral lobe is pointed below. ON THE ANATOMY OF THE KOALA. 181 not to a columna carnea, but by cJiordce tendinece inserted on the septa 1 wall. There is apparently only a single opening for the coronary veins, just at the entrance of the inferior cava into the auricle. The aorta gives off, in the specimen which died in the Society's Gardens, three vessels from a common trunk, and then the left subclavian, as in Phalanyista and most other Marsupials *. In another specimen, however, the arrangement is as in Man and as in Phascolomys, the left carotid arising independently from the aortic arch. Of the two vence azygos, each opening into the superior cava of its side, the left is much the larger, the right being formed mainly by vessels derived from only the first few intercostal spaces, whilst below these the veins of the right side pass over, behind the aorta, into the leftazygos. This is an arrange- ment I have found in several Marsupials examined, including Phascolomys, Belideus, Cuscm, and Phalangista, though notiu Petrogale orHypsiprymnus. In Phascolomys there exists a commissural branch between the first intercostal vein on the right side going to the left, and the last going to the right, vena azygos. In the Hedgehog, and some other animals according to Prof. Owen (Anat. Vert. iii. p. 553), the right is also smaller than the left azygos, though usually the reverse condition holds ; and in the highest forms, where there is only one vena azygos, it is the right that persists. The external and internal iliac arteries come off separately from the aorta, there being 110 common iliac arteries. This disposition is, I believe, nearly universal t in the Marsupials, but is by no means confined to them, as I have found it in Tamandua, Tapirus, and Hyomoschus, and Prof. "Watson records it in Hyaena crocuta (P. Z. S. 1879, p. 89). The lungs are simple in form. The right side has three, the left two lobes ; the lower lobes of each side being about equal in size, and much larger than the others half as big again as the upper, or two upper, lobes. There is no azygos lobe at all. The female generative organs of Phascolarctos have not been, so far as I have been able to ascertain, hitherto described, though Mr. A. H. Young has lately given us an excellent account, with figures, of the corresponding * P.S. Feb. 11, 1881. In a fresh specimen of Belideus breviceps, which I have just dissected, I find only one trunk arising from the aortic arch ; this splits up iuto 3 branches a left innominate, dividing into the subclavian and carotid branches for that side, a right carotid, and a right subclavian. Moreover, as in no other Marsupial known to me, there is only one anterior cava, the right and left innominate veins joining to form a large trunk, some % inch long, which opens into the auricle. t In a Cuscus maculatus that I dissected I found the abdominal aorta splitting up into four trunks, the right and left external, and the right internal iliacs, whilst from the remaining or median (caudal) one, the left internal iliac was given off some way below the level of the other. 182 ON THE ANATOMY OF THE KOALA. system in the male. In their essential points they differ in no important respect from those of the Wombat *. P. Z.S. 1881, The ovaries are rounded ovals in shape, considerably depressed, and p. 189. measuring about -45 inch along their greatest extent. They are cut up by three or four sinuous fissures ; each of these lobes is further subdivided into ovisacs, which are of large size for a Mammal, though nothing like so big as the large ones figured by Prof. Owen in Phascolomys. The ovaries are enveloped to some extent by the fimbriated ends of the Fallo- pian tubes, and are enclosed, in common with these, in pouches of delicate peritoneum. The fimbriated ends of the oviducts are attached narrowly to the posterior part of the ovary ; they extend hence for about O5 inch to the ostium abdominal*. The Fallopian tubes are a little bent, and are of small calibre, passing gradually into the larger, somewhat fusiform uteri, which, as usual in the Marsupials, are quite separate from each other; muscular, thick- walled, and nearly straight, these open on a prominent, somewhat com- pressed nipple-like eminence, forming the os tincce, by a small pore. The total length of the Fallopian tubes and uteri is about 1*3 inch from the ostium at the commencement of the former. The vagina3 are also two in number, each being bent outwards in a simple curve, and not com- municating with its fellow at any point. The lower part of each vagina is thick-walled, with but a small central cavity which opens into the urino-genital sinus by a small pore, O2 inch above the opening of the vesical urethra. Above they are thin-walled; and from the internal side is developed a blind cul-de-sac, also thin-walled, communicating only with the vagina of its own side and the corresponding uterus, there being a median septum between the two culs-de-sac. No opening from the latter into the urino-genital sinus exists in either specimen I have examined. From the os tincce there is prolonged downwards on each side a slightly elevated fold of the mucous membrane, which separates off the vagina proper from the more medianly placed cul-de-sac. Both vaginae and culs-de-sac are lined by smooth mucous membrane, with slight longitudinal rugae. The two uteri, as well as the vaginas and their appendices, are united together by peritoneum. The two ureters penetrate this to open into the neck of the bladder, beyond the termina- tion of the vaginal culs-de-sac. The length of the vaginae is about 0*65 inch, measured in a straight line ; that of the culs-de-sac about 0*45 inch. The urino-genital sinus is a tube, with moderately thick walls and longitudinally plicated mucous membrane, of 1-3 inch in length. It communicates below by a considerable aperture with the rectum, and the * For description of these see Owen, P. Z. S. 1836, p. 52, and Anat. Vert. iii. p. 680 et seq. OiV THE ANATOMY OF THE KOALA. 183 cloaca so formed is surrounded by a common fold of muscles and integu- ment. A small, flattened, linguiform clitoris, not free at its apex, with two grooves above and about 0'2 inch long, is developed on the anterior Avail of the cloaca, beginning at the level where the rectum and urino- genital canal meet. A second specimen examined an adult female that has been preserved in spirit, and which, judging from the condition of its mammae, has been a mother shows exactly the same relations of these parts as that here described, the only differences being in the sizes of some of the parts, P. Z. S. J881, due, no doubt, to age. The clitoris, however, is free at the apex and p< ' slightly bilobed *. The brain of the Koala is represented of the natural size in the accompanying figures (figs. 3-6, p. 185), of which that representing its superior aspect was taken from the brain before being removed from the cranial cavity, and therefore unaltered by displacement or hardening in spirit. The other three figures are drawn from the brain after hardening in alcohol for some months. The cerebral hemispheres are remarkable for their simple surface, which is broken up by no convolutions. Broadest behind, they taper forwardly, and so are somewhat pyriform in outline when viewed from above. They leave the corpora quadriyemina largely exposed behind ; and in consequence the cerebellum is left entirely uncovered : indeed, w^hen the parts are undisturbed (fig. 3) it is not even in contact with the cerebral hemispheres. The greatest length of the cerebral hemispheres is about 1'2 inch ; their greatest depth about 0*7 inch. Viewed from the side, their superior contour is seen to be but little arched behind, whilst anteriorly it slopes downwards away rather suddenly towards the olfactory * In Phascolomys wombat, in a two-thirds grown female, I can detect no differences of importance whatever from the type here described. The Fallopian tubes are apparently longer, and their fimbriated extremities better developed. The form of the ovaries, and the disposition of the uteri, vaginae, and urino-genital sinus seem to be nearly precisely similar in the two genera. In Phalangista vulpina, on the other hand, considerable differences occur. The Fallopian tubes are shorter in proportion to the uteri, and are more convoluted. The uteri are more distinct from the Fallopian tubes, are more capacious, and strongly curved outwards. Each os tincce projects as a prominent and quite free papilla into a common vaginal chamber, formed by the coalescence and fusion of the two diverticula present in Phascolomys and Phascolarctos. This chamber is capacious, and has only a very slight indication of a median septum left. In Belideus sciureus the Fallopian tubes and uteri resemble those of Phalavgista. The vaginae, however, are much longer and curved on themselves, much as in the Kangaroos. There are apparently two small culs-de-sac ; but the specimen examined does not allow me to say whether or no they unite. In Petaurus ( = Belideus ?), accord- ing to Owen (Anat. Vert. iii. p. 682), where the vagina 1 are also long and curved, the culs-de-sac remain separate. 184 ON THE ANATOMY OF THE KOALA. lobes. These last are not large, and but little exposed ; in fact, in the undisturbed state, they are covered, when viewed from above, by the hemispheres. The temporal lobe is small. Superiorly the hemispheres, save for a few slight vascular impressions, are altogether smooth ; laterally, a well-defined sulcus, running from the temporal lobe forwards, and curved, first upwards and then downwards, is visible. Anteriorly, this separates off the olfactory tract from the side walls of the hemi- spheres. A slight indentation, about halfway along its course, at the top of its upward convexity, may represent a rudimentary Sylvian fissure. Just behind this is a second similar, though smaller, impression. The olfactory ganglion is large, as is the tract. Internally, the characteristic features of the Marsupial brain * are distinct, the corpus callosum being small and indistinct, and the anterior commissure very large. The P. Z. S. 1381, hippocampal sulcus is distinct and deep, strongly curved, and continued p. 192. forwards over the corpus callosum onto the interna] face of the hemi- spheres to a point about 0*15 inch in front of the anterior commissure. Behind is another rather deep, /-shaped sulcus, which appears at both ends on the prominent rounded margin of hemispheres. The corpus Jlmbriatum and fascia dentata are both distinct. The middle (grey) commissure is very large. Of the corpora quadriyemina, the nates are longer (from before backwards) than the testes. The posterior limb of the crucial impression is not as distinct as the fore one. In the cerebellum the vermis is well-developed, as are the lateral lobes and theflocculi, which have the form of projecting, rounded lobes. The pons Varolii is narrow, the anterior pyramids well defined, and the corpora trapezoidea distinct. As compared with Pliascolomys, the principal points of difference in the brain are the more richly convoluted hemispheres a distinct calloso- marginal sulcus being present, as well as others on the external surface and the non-projecting flocculi, of the latter. PJialangista has nearly as simple a brain as the Koala ; but iheflocculi project more. A consideration of some of the facts on the visceral anatomy of the Koala here stated appears to me to throw considerable light on the classification of the Marsupials. Naturalists generally have placed the Koala in, or close to, the Phalangistidse ; whilst the Wombats have been retained as a separate family or section, of equal value with the former group, the Kangaroos being often, indeed, interposed between the twof. "Writing as long ago as 1846, Mr. Gr. E. Waterhouse, in his 'Natural History of the Mammalia' (vol. i.), though in that work keeping the Phascolomyidse separate from the Phalangistidae, evidently * Vide Flower, Phil. Trans. 1865, p. 647. t Cf. Owen, "Classification of the Marsupialia," P. Z. S. 1839, p. 19; Sclater, Key. List of Vertebrate, 7th edition, 1879. ON THE ANATOMY OF THE KOALA. Fig. 5. P. Z. S. 1881, p. 191. Fig. 4. Fig. 3. Fig. 6. Fig. 3. Eight half of Koala's brain, from above, of the natural size ; drawn before removal from the skull. Fig. 4. The same, from below. Fig. 5. The same, from the side. Fig. 6. Left cerebral hemisphere, from the inside, the optic thalamus being cut short. a.c, anterior commissure ; h. hippocampal eulcus. 186 ON THE ANATOMY OF THE KOALA. did so with some hesitation. He says (7. c. p. 16) : " Upon a careful examination of the Wombat, I find so many points in common with the Phdlangista group, that it is so intimately connected with the Koala (which is more clearly an aberrant Phalanger), as indicated by the struc- ture of the stomach and the deficiency in the number of the false molars, and the total absence of tail, that I am inclined to regard the genus Plias- colomys as presenting an aberrant form only of the Phalangistidse. That the thumb should be reduced to a small size in this animal, which differs from others of its (supposed) family in living upon the ground, I am prepared for, since in the Dasyuridse the same thing takes place under similar circumstances. I am also prepared to find in an herbivorous group like the Phalangistidae a difference in the structure of the molar teeth, in haying them rooted in one case and rootless in another, for such happens in other herbivorous groups of the Mammalia." Again, in a note on p. 257 : " With regard to the position of the Wombat and the Koala (Phascolarctos) in a natural position, I may observe, in the first place, the Wombat (cceteris paribus) shows some affinity to the Phalangistidse in the possession of a thumb, which, though short, is very broad and sufficiently distinct. Then, beyond this, we have to add that P. Z. S. 1881, the limbs are equal, the tibia and fibula are widely separated, excepting, p. 193. o f course, at the extremities ; and the stomach is simple *, as in the Phalanger group. On the other hand, we perceive in the Koala an animal possessing all the essential characters of Phalangista, but in which the stomach is provided with a peculiar glandular apparatus, and the tail is wanting, as in the Wombat. The two animals agree, moreover, very closely in the structure of the humerus ; they agree in the non- possession of a patella, in the absence of a liyamentum teres f, and in the outermost of the articular surfaces of the upper extremity of the tibia being continuous with the articular surface of the fibula. The skull of the Koala, as compared with that of a typical Phalangista, differs in having the posterior palatine openings confined to the palatine bone, which is also the case in the Wombat; the lower jaw differs in the greater extent of the symphysis menti ; and, lastly, an approximation to that Rodent-like type of dentition which is exhibited by the Wombat is perceptible in the Koala, in the smaller development of the posterior incisors and canines of the upper jaw, and the total absence of any of those premolars which, in the typical Phalangers, intervene between the * I suppose by this is meant as opposed to the sacculated stomach of the Kangaroos. t As regards these last two characters, it must be observed that the first is a character practically common to all Marsupials, excepting the Peramelidse (cf. Flower, 4 Osteology of Mammalia,' 2nd ed. p. 306). As regards the alleged absence of a liga- mentum teres, I find it perfectly well developed in fresh specimens of both Koala and Wombat; on the femur the depression for it, though not distinct, is traceable. ON THE ANATOMY OF THE KOALA. 187 canine and the five molars of the upper jaw, and the incisor and the corresponding teeth in the lower jaw." Dr. Murie, from his examination of the osteology of the Wombats (P. Z. S. 1867, p. 815), appears also to incline to Mr. Waterhouse's view. In the course of this paper I have already noted several other points of resemblance between the Koala and Wombat, in the presence in both of more or less distinct cheek-pouches, in the absence of a distinct caudate lobe to the liver and the tendency of its lobes to develop addi- tional superficial sulci, and, finally, in the structure of the female repro- ductive organs. In the Wombat, too, the first traces of the syndactyle condition of the pes appears, both externally and also in the structure of the bones. But, to my mind, the most convincing token of their affinity is their possession of the peculiar gastric gland * already referred to and described. In no other Marsupial is there any trace of such a structure visible, whilst in the two forms under consideration its identity is almost precise. That such a unique structure should have been independently developed in two forms unrelated to each other appears to me to be in the highest degree improbable. The main points of divergence from the Phalangers presented by the Wombat are the peculiarities of its dentition, and its extraordinary p.z. S. 1881, caecum (see the description and figure by Prof. Flower, Med. Times and P- 194 - Gazette, Dec. 14, 1872, p. 642). In its teeth being all rootless, as well as in the equality in the number of its incisors, Pliascolomys differs from all other Marsupials. But it is highly probable that this peculiar Bodent- like dentition has been brought about in accordance with its mode of life, and that therefore these features, being adaptive, have in reality less importance in classification than has been assigned to them. Moreover, in a very young Wombat's skull preserved in the Hunterian Museum (1795 D),in which the first three molar teeth only in each jaw have cut the gum and are quite unworn, each lobe of the teeth has two quite distinct, though small cusps ; hence the second and third teeth on each side have four distinct cusps, and the anterior two, as in the Phalangers generally. The caecum is no doubt peculiar, and quite unique amongst Mammalia, any resemblance to the " appendix vermiformis " of the highest Primates being fanciful. If in these points sufficient reason is considered present for elevating the Wombats to the position of a primary group of the Marsupials whether such group be called a tribe or a family is no matter it should not be forgotten that in some features Phascolarctos, too, is nearly as peculiar as Phascolomys itself. These are mainly : the peculiar alisphenoidal bullaof the skull ; the extraordinary complicated liver, with the elongated gall-bladder ; the immensely deve- * It would be interesting to investigate the histological structure of this gland, with the object of determining whether or not the resemblance is more than external. 188 ON THE ANATOMY OF THE KOALA. loped caecum and caecum-like ascending colon, with their longitudinal folds of mucous membrane ; and the absence of an azygos lobe to the lungs, the Wombats agreeing with the Phalangers in possessing one. Hence it appears to me to be a more natural course to keep these three groups together as subdivisions of a larger one, though whether that one be called a family, or made into a larger section, will depend on the value attached to those ideas by different naturalists. Adopting the former as most convenient, they might be defined briefly as follows : PHALANGISTID M . Diprotodont Marsupialia, with clavicles, and not more than six incisors above. The hallux present ; the 2nd and 3rd digits of the pes smaller than the others, and more or less united together by integument. Stomach not sacculated. Caecum present. Grlans penis more or less bilobed ; vaginas provided with median culs-de-sac which may unite. 1. Teeth rooted ; superior incisors 3.3; at least one small additional preinolar on each side above. Tail well developed. No cheek-pouches. Stomach and ascending colon simple. Caecum long, simple. Liver not complicated by secondary sulci, and with distinct caudate and Spigelian lobes. Lungs with an azygos lobe. Vaginal culs-de-sac coalesced (at least in Phalangista). Phalangista, Cuscus, Belideus *, Acrobata *, Dromicia *. P. Z. S. 1881 , 2. PHASCOLAECTIN^E. Teeth rooted ; superior incisors 3.3; additional premolars absent. Tail rudimentary. Distinct cheek-pouches. Stomach with a cardiac gland. Caacum very long; commencing colon caecum-like, both being dilated and provided with numerous longitudinal folds of mucous mem- brane. Liver very much complicated by secondary eulci ; caudate lobe not free; gall-bladder immensely elongated. Lungs with no azygos lobe. Vaginal culs-de-sac free. Phascolarctos. 3. PHASCOLOMYIM. All teeth rootless ; superior incisors 1.1; no additional premolars. Tail and cheek-pouches rudimentary. Stomach as in Phascolarctince. Caecum short, peculiar. Commencing colon transversely sacculated. Liver somewhat complicated by secondary sulci; no distinct caudate lobe. Lungs with an azygos lobe. Vaginal culs-de-sac free. Phascolomys. * For an opportunity of dissecting examples of these genera I am indebted to the liberality of our President. CONTRIBUTIONS TO BIKD-ANATOMY, ETC. 189 35. ON THE CONTRIBUTIONS TO THE ANATOMY AND ibiB.1881, CLASSIFICATION OF BIRDS MADE BY THE LATE PROF. GARROD, F.R.S.* IT having been suggested to me by one of the Editors of this Journal that a concise resume of the ornithological papers of my late friend and predecessor, Prof. A. H. Garrod, F.R.S., would not only form an appro- priate memoir of him, but would also be useful to those ornithologists who are interested in the anatomy of birds and the questions of classi- fication that depend on it, I have endeavoured in the present paper to give a short sketch of the contributions Prof. Garrod made to our know- ledge of, and of his views on these points. In the seven years (1872-1879) during which Prof. Grarrod held the post of Prosector to the Zoological Society, no less than thirty-eight papers from his pen (all, with one exception"!", published in the Zoological Society's 'Proceedings') appeared, dealing with various points in the anatomy or physiology of birds. Of these a complete list will be found Ibis, 1881, in the January number of this Journal for last year:}:. All of these, p ' ' except two, are morphological in nature ; but many of the characters of birds from the physiological side were fully expounded in his series of Fullerian Lectures at the Royal Institution and elsewhere. At the time of his death, Prof. Garrod was also engaged on an article on the mecha- nism of flight ; for his wonderful mechanical skill enabled him to explain and demonstrate this and other physiological problems in a method but rarely to be met with amongst biologists generally. But this, unfortu- nately, he left in an unfinished condition. * Ibis, 1881, pp. 1-32. f " Note on some of the Cranial Peculiarities of the Woodpeckers," Ibis, 1872, p. 357. J In addition to his published papers on birds, Prof. Garrod was engaged, as probably many of the readers of ' The Ibis ' are aware, on a general account of the Anatomy of Birds, to be published in three fasciculi. As originally planned, the first fasciculus of this work was to contain a complete account of the anatomy (not in- cluding the histology) of the common Fowl, as a type of all birds ; the second was to be occupied with a comparative account of the " soft parts " in the different groups ; whilst the third was, I believe, to have been devoted to osteology and a consideration of the results arrived at as regards classification. Of these three fasciculi, the first was nearly completed at the time of his death, and the second left about half done, nearly all the groups of the " Homalogonatous " birds being treated of in it, together with some of the remaining ones. The MS. of both of these portions has been, fortunately for our science, preserved ; and it is my hope some day to complete the work for publication in a form worthy of its original author. " On the Mechanism of the Gizzard in Birds," P. Z. S. 1872, pp. 52o-529 ; " On a Point in the Mechanism of the Bird's Wing," P. Z. S. 1875, pp. 82-84. 190 ON THE LATE PROF. GARROD S CONTRIBUTIONS Ibis, 1881, p. 3. In the present article I propose first to consider those points in the anatomy of birds first brought into notice, or worked out in large groups, by Garrod, and secondly to consider the light thrown by these facts on the correct collocation of various genera, or larger groups, as well as on the arrangement of these latter into groups of a still higher power. But I shall avoid, as far as possible, any comparisons with previously pro- posed classifications, as it is not my wish to enter, in this place, into discussions of that kind. Under each of these headings I shall endeavour, as far as is consistent with clearness and conciseness, to preserve a chronological order. I. On the Conformation of the Nasal Bones*. " In most birds the anterior margin of the nasal bone is concave, with the two cornua directed forwards," these processes being " continuous behind with the body of the bone and with one another, there being no interruption of any kind between them. Such a condition is found in Otis and the Gallinse proper ; and birds possessing the bone so constructed may be termed holorhinal : in them a transverse straight line, drawn on the skull from the most backward point of the external nasal aperture of one side to that of the other, always passes in front of the Fig. 1. Fig. 2. Skull of Gallus domesticus (from P. Z. S. 1873, p. 35, fig. 9). Skull of Larus argentatus (from P. Z. S. 1873, p. 34, fig. 2). * " On the Value in Classification of a Peculiarity in the Anterior Margin of the Nasal Bones in certain Birds," P. Z. S. 1873, pp. 33-38. TO BIRD-ANATOMY AND CLASSIFICATION. 191 posterior terminations of the nasal processes of the praemaxillae." This simply concave nature of the posterior margin of the osseous external nares, as well as the relations of the extremities of the nasal bones to those of the nasal processes of the praemaxillae, is shown in the subjoined figure of the Fowl's skull (see fig. 1). In a large number of birds, however, the condition of things is Ibis, 1881, different, as will be evident from an inspection of a similar view of a P e 4 * Gull's skull (Larus argentatus). Here (see fig. 2) the posterior margin of the osseous nares has a distinctly slit-like or triangular form, instead of being simply concave ; hence the birds presenting this peculiarity, which varies to some extent in the degree of its development in different forms, may be called " schizorhinal" In most of these schizorhinal forms the line joining the posterior extremities of the nostrils passes behind, instead of in front of, the ends of the nasal processes of the praemaxillae. When the beak becomes shortened and broad at the base, however, as, e. the main muscle, to be inserted with it into the femur, but nearer the head of that bone. As already mentioned, some birds possess all these six muscles, but no bird is known which does not possess at least one. It is a convenient plan to designate the four last-named muscles by the letters X, T, A, B. Thus a Fowl possessing all four would have a formula A.B.X.Y, ihe> femoro- caudal, accessory femoro-caudal, semitendinosus, and accessory semitendinosus being all present. On the other hand, a Hawk or Owl, possessing only one, the femoro-caudal, would have as formula A. Of these four muscles there are sixteen possible combinations ; but of these only eight are actually found in birds t. Furthermore, if we denote the presence or absence of the ambiens muscle by the signs + or , we obviously get sixteen combinations again, though of these only fifteen have, as yet, been observed. For the sake of brevity it will be useful to remember the above four letters, to save the repetition of the full names. To the bearing of these muscles in the classification of birds, I intend to revert at a subsequent part of * " On the Anatomy of Chauna derbiana," &c., P. Z. S. 1876, pp 189-200. t P.Z.S. 1874, p. 111. 198 ON THE LATE PROF. GARROD'S CONTRIBUTIONS this paper, only stating here that species of the same genus and nearly allied genera have, nearly in every case, the same myological formula, and that \-D. families peculiarities involving more than one change are rare, farther differences indicating a more remote relationship. V. The Distribution of the Deep Plantar Tendons*. In all birds, whatever number of toes they have, there are two deep flexor muscles of the toes, the fleshy bellies of which are situated between the knee and the " ankle," whilst their tendons run along the posterior aspect of the tarso-metatarse. One, arising from behind the external condyle of the femur, is the flexor longus hallucis ; the other, arising from the posterior aspect of the tibia and from the fibula, is the flexor perforans digitorum, so called because its terminal tendons per- Ibis, 1881, f orate those of the more superficial flexor perforatus (the muscle, it will p. 13. j^ remembered, joined by the tendon of the ambiens). Below the " ankle " the tendons of these muscles run along the tarso-metatarse ; whatever their ultimate distribution, they may be easily identified in this region, the^m^r longus hallucis being always external to, or super- ficial of , the flexor perforans (or both). In all the Passeres, as already noticed by Sundevall (except in the Eurylaemidse, vide infra), as well as in Upupa epops and perhaps one or two Ardeine birds, these two tendons are quite independent of each other, so that if the flexor hallucis be artificially pulled no flexion (closing) of the other digits takes place. This arrangement is represented in fig. 6. In all other birds, however, the two tendons, during some part of their course in the tarso-metatarse, are more or less intimately connected together by a fibrous band or vin- Ibis, 1881, culum, or may even completely blend. It is to a consideration of the p. 14. varying arrangements produced that this paper of Prof. G-arrod's is in the main dedicated ; and some of the most important modifications may be noticed here. In a large number of birds the type presented by the common Fowl obtains. Here the flexor perforans supplies digits II-IY, and t~he flexor hallucis only digit I (the hallux) ; this, as it crosses the tendon of the flexor perforans, sends down a strong fibrous vinculum (vide fig. 7, V). The proportions of the vinculum to the main tendon (that distributed to the hallux) vary greatly, as it may or may not be greater than the hallucial portion. In some Birds of Prey a vinculum may be combined with a special slip of tendon to digit n, or it may be nearly entirely distributed to that digit. In the Cathartidae a quite different arrangement, next to be described, obtains. Here, and in many other birds, particularly amongst the Anomalogonatce, the two tendons blend completely, and the * " On the Disposition of the Deep Plantar Tendons in different Birds," P. Z. S. 1875, pp. 339-348. TO BIRD-ANATOMY AND CLASSIFICATION. 199 Fig. 6. Fig. 7. Ibis, 1881, p. 13. A typical Passerine foot Foot of Gallus banfciva (from P. Z. S. 1875, p. 347, fig. 9). (from P. Z. S. 1875, p. 341, fig. 1). V, Vinculum. tendinous slip to the hallux comes off from the blended tendon, IMs,'1881, apparently springing from the inner side. A slight modification of this produces an arrangement by which the hallucial slip seems to come off from the inner side of the flexor perforans tendon in its upper part, before it has been joined by the flexor hallucis. This last-named con- dition obtains in such birds as Momotus, Merops, and Dacelo. "When the hallux is absent, as well as in Struthio (where only two digits are present), the two tendons fuse completely in the leg, and the compound tendon is distributed in the usual way to the three (or two) digits. In many birds with a hallux, when there is no long flexor to that digit, the slip to it is extremely small ; and in some cases it is alto- gether absent. In the Trogonidse, as might have been expected from the well-known peculiarity of their feet, an equally peculiar arrangement of the plantar tendons obtains (L c. p. 345, fig. 6). By far the most interesting feature, however, brought out by Prof. Garrod's investigations into this subject, is the discovery of the existence 200 ON THE LATE PROF. GARROD*S CONTRIBUTIONS of two entirely different types of plantar arrangement in the so-called Ibis, 1881, " zygodactyle " birds, as well as the fact that this diversity of type P- !5. exactly coincides with the two groups of birds so marked out being respectively " Homalo-" and " Anomalogonatous." Thus, in the Parrots, Cuckoos, and Musophagid, which are all Homalogonatous, possessing (at least typically) the ambiens muscle, the plantar tendons are distributed in exactly the same way as in the common Fowl, the flexor perforans supplying digits II-IY, and the flexor hallucis digit I alone (neglecting the vinculum). In all the Anomalogonatous zygodactyle birds (which all lack the ambiens and accessory femoro-caudal muscles), namely the Picidae, Capitonidse, and their allies, BucconidsB and Galbulidse, an entirely unique arrangement is found ; for in those birds the flexor longus hallucis splits up into three parts, supplying digits II and iv as well as the Tiallux, whilst the flexor perforans digitorum is distributed to the third digit alone. These differences in the two types will be clearly seen by com- paring fig. 8 (Crotophaga sulcirostris) and fig. 9 (Megalcema asiatica}. Fig. 8. Fig. 9. Ibis, 1881, p. 16. Foot of Crotophaga sulcirostris. Foot of Megalcema asiatica. (From P. Z. S. 1875, p. 346, figs. 7 & 8.) VI. The Method of Insertion of the Tensor patagii brevis Muscle. In the patagial membrane of the wing in most birds there are two muscles present, the fleshy bellies of which arise chiefly from the scapular TO BIRD-ANATOMY AND CLASSIFICATION. 201 extremity of the furcula, whilst their tendons run between the two layers of membrane of which the patagium is composed. These muscles are the tensor patagii longus and the tensor patagii brevis. Arising in common, or in close proximity to each other, the tendon of the tensor patagii longus forms the more or less stiffened, though flexible, anterior border of the patagium ; the tendon of the shorter of the two muscles runs more or less parallel with the humerus, ending near the elbow-joint. In different birds its insertion takes place in different ways, the tendon in some cases simply running straight on to the ulnar side, and there becoming blended with the general fascia of the part, whilst in others it becomes united, more or less intimately, with the tendinous origin, springing from a tubercle on the humerus, of a muscle lying on the radial side, the extensor metacarpi radialis longior. In some cases additional slips from the main tendon are given off, the arrangement sometimes thus getting very complicated (e. g., in the Trogonidae). As a rule, every natural family of birds has a characteristic arrangement of these tendons ; so that their disposition often affords great help in classi- fication. This is the case, for example, in the Passeres ; throughout the whole of that immense group one arrangement, only slightly masked in one or two aberrant forms, which can be easily recognized and is most characteristic, obtains. The same is the case with the very closely allied groups of Picidae, Eamphastidae, Capitonidae, and Indicatoridae. These points were first worked out by Prof. Garrod in the first of his papers on Passerine birds * ; and the arrangements in nearly all the families of his " Anomalogonatous birds " are there described and in several cases figured. In the Homalogonatse, too, they offer nearly as well-marked peculiarities, though, unfortunately, his purpose of describing and figuring Ibis, 1881, their arrangement in these birds was never carried out. In his MSS. and p ' 17t drawings, however, he has recorded their condition in very many groups. VII. Certain other Muscles. In his paper on Chauna (supra, p. 197), Prof. Garrod, for the first time, called attention to the value in classification of certain other myological facts. These are : (1) The presence or absence of the '* cxpansor secundariorum " muscle. This is a slender muscle which arises from the last few (generally two or three) secondary quills and has a peculiarly long and slender tendon, which, running superficially posterior to the humerus, together with the axillary vessels and nerves, is inserted into the thorax in different ways in different birds. One common arrangement is that found well deve- loped in the Storks, and hence called " ciconiiform." Here the proximal part of the tendon is T-shaped, it splitting into two parts, one being * P. Z. S. 1876, pp. 506-512, pis. xlviii.-li. 202 ON THE LATE PROF. GARROD's CONTRIBUTIONS inserted into the coracoid near the junction of this bone with the sternum, the other into the scapula close to the coraco-scapular articulation. In other cases the tendon may end by joining the coraco-brachialis brevis muscle, so as to appear to be part of that muscle, as in the Fowl and most of the Gallinaceous birds ; or the tendons of the two sides may join in the middle line, as in most of the Anseres ; or other and more complicated arrangements, duly described in the paper above referred to, may obtain. The only Anomalogonatous birds in which this muscle is present are the Coraciidae. A table (1. c. p. 199) records the nature of this muscle in the families of Homalogonatous birds. In many it is quite absent. (2) The presence or absence of a biceps slip to the patagium of the wing. In many birds there is given off from the anterior margin of the biceps muscle of the arm a distinct and most peculiar muscular slip, which joins the patagial membrane of the wing. Its presence or absence is a very constant character amongst closely allied birds. (3) The area of origin of the obturator internus muscle. Ibis, 1881, This muscle, arising in the interior of the pelvis, is inserted proximad P' 18 - of the obturator eocternus on the head of the femur. In most birds its shape, as seen in the pelvis, is more or less oval ; but in some, as in the Gallinse and Eails, it is distinctly triangular. Both its character and the presence or absence of a bicipital slip are recorded in the majority of the Avian families, in the table already mentioned in the account of the expansor secundariorum. Besides these, a few other myological peculiarities insisted on by Prof. Garrod in various papers may here be named, such as the presence of an additional secondary femoro-caudal muscle in Apteryx, and the presence in it, as well as in the other " Struthious " birds and the Crypturi, of a muscular slip to the accessory femoro-caudal above the exit of the sciatic nerve and artery * ; the occasional complete absence of the semi- membranosus in some of the Grebes t ; and the double condition of the great pectoral muscle in the Storks, Steganopodes, Petrels, and their allies J. VIII. The Conformation of the Trachea and Syrinx. The curious contortions of the trachea, and other peculiarities of its structure, in various birds, such as the Cranes, Spoonbill, and Ducks, have long been known to ornithologists ; and, as far as concerns this portion of the subject, Prof. Garrod's notes on these structures in various * P. Z. S. 1873, pp. 643, 644, fig. 6. t P.Z.S. 1873, pp. 629, 642. | P. Z. S. 1874, p. 120, and 1876, p. 340. TO BIRD-ANATOMY AND CLASSIFICATION. 203 forms * are simply confirmations of, or additions to, our previous know- ledge of the subject. But, so far as I am aware, little or no attention had ever previously been paid to the details of modification in the cartilaginous or ossified tracheal and bronchial structures concerned in Ibis, 1881, the formation of the hard framework of the lower larynx, or " syrinx." Garrod's investigations into this subject therefore mark a new line of departure ; and it is exceedingly to be regretted that only one part of his notes on the subject were completed before his death, these forming his paper " On the Conformation of the Thoracic Extremity of the Trachea in the Class Aves. Part I. The Gallinae"t, the last contribution of his pen to ornithological science. There is every reason to believe that this line of research, when prosecuted further, will lead to most valuable results as a means of separating, on anatomical grounds, allied genera or families of birds*. Of his investigations of the lower larynx of the Passeres I propose to speak later, under that head. In the remaining part of this paper I propose to consider the results arrived at, from the consideration of these and other anatomical features, by Prof. Garrod as to the relationships of various obscure forms of birds, and also to describe certain remarkable peculiarities of others as first discovered by him. In these remarks, as before, I shall, for convenience* sake, follow, as nearly as possible, a chronological order, reserving, however, till the last any general views on the classification of birds as a whole. 1. Strutkio. In this paper, written in conjunction with Mr. Frank Darwin, the principal point of interest is the discovery, or, at all events, first notice, of a peculiar nodule of bone lying on the centre of the pubis, and, in some respects, similar to the " marsupial " bone of the Impla- Ibis, 1881, cental Mammalia and its corresponding fibrous representative in certain p ' Carnivora. * " On the Form of the Lower Larynx in certain Species of Ducks," P. Z. S. 1875, pp. 151-156 (the species described are Sarcidiornis melanonota,Rhodonessa caryophyllacea, and Metopiana peposaca) ; " On the Form of the Trachea in certain Species of Storks and Spoonbills," P. Z. S. 1875, pp. 297-301 ( Tantalus ibis and Platalea ajaja) ; " On the Trachea of Tantalus loculator and of Vanellus cayennensis" P. Z. S. 1878, pp. 625-629. t P. Z. S. 1879, pp. 354-380, figs. 1-35. \ A second part of Garrod's notes on the trachea, describing that of the Cuculidae, I found in a nearly complete state amongst his MSS., as well as a very considerable mass of drawings and notes on this subject in other groups, it having been the special object of his study up to within a very few weeks of his death. Indeed, during all his last illness, when too weak to attend to larger and less convenient objects, he con- tinued to work away with all his old enthusiasm and energy at the windpipes of birds, especially those from the extensive collection of Procellariidse &c. made by H.M.S. ' Challenger.' " Notes on an Ostrich lately living in the Society's Collection," P. Z. S. 1872, pp. 356-363. 204 ON THE LATE PROF. GARROD's CONTRIBUTIONS 2. Heteralocha*. In 1872 the true affinities of the Huia bird were quite uncertain, many authorities placing it amongst the Hoopoes (Upupidae). From ail examination of the specimen that lived in the Zoological Society's Garden's, Prof. Garrod was enabled to show that it was truly Passerine, and not only so, but in many respects so peculiarly Sturnine'f as to entitle it to a place " at the head of the family." 3. jSteatornis%. Besides the myology and visceral anatomy, the pterylosis, skull, and syrinx (which is "bronchial") are particularly described and figured. The result arrived at, as regards the systematic position of Steatornis, is that it must form a family by itself, with strong affinities to the Owls, Caprimulgidse, and Coraciidee and their allies. 4. Columbce. In these papers the number of the rectrices (varying from twelve to twenty in different genera), together with the presence or absence of the oil-gland, of the caeca, and of the ambiens muscle, is recorded in a large number of forms, and, from the various combinations of characters so obtained, an attempt is made to divide up the group of Pigeons, in which are included the Pteroclidae, in a more natural way than has hitherto been done. Attention is also called to the charac- teristic form of the humerus in these birds, as well as to the peculiar form of the gizzard in the genus Ptilopus, there being here four crushing- pads, instead of two as in all other birds, including even Treron. Ibis, 1881 I n CarpopTiaga latrans the stomach is even more peculiar ; for here p. 21. the epithelial lining of the gizzard is developed into about two dozen horny conical processes, like the tubercles of a Cidaris or similar Sea- urchin. A similar condition has been described by MM. Verreaux and Des Mars in G. goliath of New Caledonia || ; but no other species of Garpophaga yet examined shows any trace of such a structure. 5. Psittaci^. In these papers, as in the last, the condition of certain structures is recorded in a large number of forms, and from the * "Notes on the Anatomy of the Huia Bird (Heteralocha gouldi)," P.Z.S. 1872, pp. 643-647. f Mr. Sharpe must, we fear, have overlooked thia paper of Prof. GarrocVs, as he still retains Heteralocha amongst the Corvidae (Cat. Birds, iii. p. 143), quite in opposition to the conclusions above stated. J " On some Points in the Anatomy of Steatornis" P. Z. S. 1873, pp. 526-535. "On some Points in the Anatomy of the Columbce" P. Z. S. 1874, pp. 249-259; " Notes on two Pigeons, lanthcenas leucol&ma and Erythrwnas pulcherrima," P. Z. S. 1875, p. 367; "Note on the Gizzard and other Organs of Carpophaga latrans" P. Z. S. 1878, pp. 102-105. || This has lately been described at greater length and figured by M. Viallanes (Ann. Sc. Nat. ser. 6, vii. art. no. 12). ^f " On Points in the Anatomy of the Parrots which bear on the Classification of the Suborder," P. Z. S. 1874, pp. 586-598, pis. Ixx., Ixxi. ; "Notes on the Anatomy of certain Parrots," P. Z. S. 1876, pp, 691, 692 ; " Note on the Absence or Presence of a Gall-bladder in the Family of Parrots," P. Z. S. 1877, p. 793. TO BIRD-ANATOMY AND CLASSIFICATION. 205 combinations of characters so obtained a scheme of classification is sketched out, each group having assigned to it a formula stating its most essential characters. The presence or absence of the ambiens muscle, of a furcula, and of an oil-gland are the points here laid stress on, together with the condition of the carotid arteries. Of these there may be either two normally situated, or only one, the left (Cacatua and Licmetis tenuirostris), or two, the left of which, instead of running with its fellow in the hypapophysial canal, as already explained, runs superficially up the neck with the left vagus nerve and jugular vein. This last condition, which obtains in no other birds, is considered, as I think rightly, sufficiently important to divide off as a main group of Psittaci all those possessing it a group including all the American Parrots, together with the Platycercidae (including Laihamus), Nestor, Dasyptilus, and the African Parrots, other than Agapornis and Palceornis, of the Old- World forms. The further subdivision of these groups is effected in the way already indicated. In a supplementary note attention is called to the probably invariable presence of a gall-bladder in the Cacatuince, though this organ has not been found in any other Parrots. In a preceding paper* Prof. Garrod has described and figured the tongue Ibis, 1881, of Nestor, which, as he shows, is peculiar, and not like that of the Lories, p * 22> with which it has often been associated. 6. Otididce^. In Eupodotis australis there is not, as had been supposed, and even stated +, by previous observers, a gular pouch, such as has been seen in Otis tarda. On the contrary, the oesophagus is highly distensile, and so produces the singular appearance of the males of this bird when excited during the breeding- season. In a young male specimen of Otis tarda examined, there is also no gular pouch present ; but the frcenum linguce was double ; and it is suggested that the pouch which has been found in the males of that species is due to a rupture and distention of the mucous membrane between this duplicate feenum, owing to the inflation of the air-passages during the period of display. 7. Chauna. The pterylosis, visceral anatomy, myology, and cranial and other characters of the Derbyan Screamer are here fully described. The very peculiar nature of the alimentary canal, in the glandular parts of the proventriculus forming, not a zone, but a patch, as well as in its possession of long sacculated caeca, without any spiral valve, which open into a special division of the intestine situated between the colon and * " Note on the Tongue of the Psittacine genus Nestor," P. Z. S. 1872, pp. 787-789. t "On the 'Showing-off' of the Australian Bustard," P. Z. S. 1874, pp. 471-473; "Further Note on the Mechanism of the 'Show-off' in Bustards," P. Z. S. 1874, pp. 673, 674. J Ibis, 1862, p. 114. " On the Anatomy of Chauna derbiana, and on the Systematic Position of the Screamers (Palamedeidae)," P.Z.S. 1876, pp. 182-200, pis. xii.-xv. 206 ON THE LATE PROF. GARROD'S CONTRIBUTIONS the ileum, is particularly worthy of notice, Chauna* being absolutely unique amongst birds in this combination of characters. In its ptery- losis and syrinx, too, it is very peculiar. As a result of his examination, Prof. Grarrod concludes that the Palamedem cannot be placed amongst the Anseres, as had been done by Professors Parker and Huxley, but Ibis, 1881, must form an independent group of birds, having probably " sprung from P' the primary avian stock as an independent offshoot at much the same time as did most of the other important families." 8. Aramus^. In the schizorhinal character of its skull, as well as in the presence of occipital foramina, Aramus resembles Grus, Ibis, Platalea, and the Limicolae, and not the Ballidae, with which it had generally been associated previously; and this collocation is quite confirmed by its myology and visceral anatomy. 9. Plotus$. In the first paper the anatomy of P. anhinga is described at considerable length particular stress being laid on the curious arrangement of the cervical vertebrae which makes this bird literally unable to carry its neck straight, and the correspondingly modified muscular system. Other points described are the presence of but one carotid artery, and of but one caecum the latter a feature previously unknown in any Steganopodous bird, though constant in the Herons. The most interesting feature, however, of Plotus anhinga is its very extraordinary stomach, previously only partially described by Macgillivray. The proventriculus, instead of forming a zone or patch, is here developed into a special sac-like diverticulum, which projects from the gizzard externally in a way quite unlike that of any other bird. Moreover the pyloric compartment of the stomach, which is present in a less complete form in Pelecanus, the Herodiones, Ealconidae, &c., here develops a covering of hairs, " a peculiarity which, as far as I know, is found only in one other bird, namely Caihartes aura." This very extraordinary stomach is figured on plate xxviii., and is certainly, as far as yet known, unique amongst birds. In Plotus levaillanti the same features in its anatomy generally are present as in P. anhinga. But there are, as usual, two small caeca : and its stomach differs considerably ; for here Ibis, 1881, there is no proventricular gland-pouch, but this organ, as in some other p. 24. birds, assumes the form of two separate patches. The second (pyloric) compartment of the stomach is also present, in a well- developed form, and is also hair-clad. But here another difference presents itself ; for " the hairy epithelium surrounding the pyloric orifice .... is produced * The anatomy of the soft parts of Palamedea is still, I believe, almost unknown. t " On the Anatomy of Aramus scolopaceus" P. Z. S. 1876, pp. 275-277. J "" Notes on the Anatomy of Plotus anhinga" P. Z. S. 1876, pp. 335-345, pis. xxvi.- xxviii. ; " Note on Points in the Anatomy of Levaillant's Darter (Plotus levaillanti}" P. Z. 8. 1878, pp. 679-681. TO BIRD-ANATOMY AND CLASSIFICATION. 207 into a considerable conical hair-covered process, projecting into the second stomach, and evidently acting as a valve to close the pylorus when necessary." These differences in two species so closely allied in all other points show that, though in nine cases out of ten similarity of external characters predicates similarity of internal structure, nevertheless in the tenth the correspondence breaks down, and that, too, without any obvious differences in mode of life, food, &c. The parallel, pointed out by Prof. Garrod, presented by these two species of Plotus with the two living genera of Sirenia (Manatus and Halicore), as regards the modification of their gastric gland- structures, is particularly interesting. 10. Coliidce*. The skull of Colius is desmognathous, and has no vomer, as in Alcedo. The viscera and myology do not bear out the idea of any relationship to the Parrots or Musophagidae ; on the contrary, these birds are truly Anomalogonatous, and are most nearly related perhaps to the Alcedinidse and Bucerotidse. Nevertheless their combina- tion of characters fully substantiates their claim to form a separate family, Coliidae. 11. Thinocoridcet. These birds, in their schizorhinal skull, and in many other features, visceral and myological, resemble most some of the more aberrant forms of Limicoline birds, such as Cursorius and Olareola. Attention is also drawn in this paper to the very extensive variations in the form of the vomer in various Charadriiform birds, it being (so far from always " tapering to a point anteriorly," as it should [?] do in these " schizognathous " birds) in several forms extraordinarily broad or even widely emarginate anteriorly ! 12. Momotidce $. The colic caeca being absent, at the same time that, ibi a , 1881, except in Momotus, the oil-gland is tufted, the Momotidae must be placed P- 25- amongst the Piciform series of Anomalogonatae, close to the Todidae, and not with the Coraciidae amongst the Pas serif ormes. The syrinx and some other points in their anatomy are also described. 13. Megacephalon . A short paper describing the pterylosis (hitherto almost unknown in the Megapodidae), syrinx, and other points in this peculiar form, which is perfectly gallinaceous. 14. Indicator \\. In its pterylosis, visceral anatomy, myology, and osteology, Indicator closely approaches the Picidae, Capitonidae, and their allies, and is in no respect Cuculine. Its vomer is large and strongly bifurcate anteriorly, as in the Capitoninae ; of the latter some are * " Notes on the Anatomy of the Colies," P. Z. S. 1876, pp. 416-420. t " Notes on the Anatomy and Systematic Position of the Genera Thinocorus and Attagis? P. Z. S. 1877, pp. 413-418. \ " On the Systematic Position of the Momotidae," P. Z. S. 1878, pp. 100-102. " On the Anatomy of the Maleo," P. Z. S. 1878, pp. 629-631. II " Notes on the Anatomy of Indicator major," P. Z. S. 1878, pp. 930-935. 208 ON THE LATE PROP. GARROD*S CONTRIBUTIONS desmognathous, others, as in Indicator, not. The truncated vomer of Ramphastos is also figured. In conclusion it is suggested that Indicator should form but a subfamily, to be comprised, together with the Bamphastinae and Capitoninse, in a larger group, the Capitonidae. 15. Opisihocomus* '. Opisihocomus is a true Homalogonatous bird, having both the ambiens and accessory femoro-caudal muscles ; it cannot, therefore, have any thing to do with the Passeres. It is, perhaps, most nearly related to the Gallinae, but, at the same time, can hardly be in- cluded with them ; it is also not far from the CuculidaD and Musophagidse, helping thus to fill up the gap that now exists between these latter families and the Gallinaceous birds. Ibis, 1881, 16. Passeres t. To define by anatomical characters supergeneric groups p * 26 ' in the immense mass of Passerine birds was always a favourite object with Prof. Garrod ; and the four papers quoted above J are the published results of his efforts at a solution of the difficulties that have always attended the classification of this group. It was whilst working at Pas- serine birds that the classificatory value of the mode of termination of the tendon of the tensor patagii brevis muscle, already alluded to (supra, p. 200), first attracted his attention. The presence also in certain Pas- serine birds, the Cotingidae and Pipridse, of a femoral instead of a sciatic artery has also been mentioned. A slight exception, too, to Sundevall's generalization about the independent muscular supply of the Jiallux in Passerine birds (suprd, p. 198) was found by him to exist in the Eurylae- midse (P. Z. S. 1877, p. 447). But the most novel fact pointed out by Prof. Garrod as regards these birds is that they may be divided into two main groups, according as to whether the intrinsic muscles of the syrinx are inserted into the ends or into the middle of the bronchial semirings. The former group, called by him Acromyodi, includes all the ordinary singing- birds with four or five pairs of muscles, the Oscines, together with two aberrant Australian groups, formed by the genera Menura and Atrichia. In these the number of intrinsic muscles is reduced to three and two pairs respectively ; but they are still inserted into the tips of the semi- rings . * " Notes on Points in the Anatomy of the Hoatzin," P. Z. S. 1879, pp. 100-114. t " On some Anatomical Characters which bear upon the Major Divisions of the Passerine Birds :" Part L, P. Z. S. 1876, pp. 506-519, pis. xlviii.-liii. ; Part II., P. Z. S. 1877, pp. 447-452; Part III., P.Z. S. 1877, pp. 523-526, pi. liii.; Part IV., P.Z.S. 1878, p. 143. | Together with his appendix to the English edition of Johannes Miiller's ' Stimm- organe der Passerinen.' I may here remark that I cannot at all agree with Mr. Sclater's view on the position of these two genera, which form his group " Pseudoscines " (Ibis, 1880, p. 345). By placing Atrichia and Menura away from the other Acromyodian Passeres, and inter- polating the Mesomyodian ones, the important fact is ignored that, in their possession TO BIRD-ANATOMY AND CLASSIFICATION. 209 The other group, the Mesomyodi, have the intrinsic muscles (which Ibis, 1881, are usually reduced to a single pair, one on each side) inserted into the middle of the rings. In them, too, the tenth (" first ") primary is always more or less long, and the tarsus, with trifling exceptions, not " bila- minate." They nearly correspond to the " Formicarioid " Passeres of Wallace, except that Wallace included in that group the Acromyodian, though in some respects aberrant, Menura. The Mesomyodi include all the Tracheophone Passeres, together with the Pipridse, CotingidaB, Tyran- nidaD, Pittidae, and a few smaller groups. A further division of these two main groups is given by Prof. G-arrod in the first paper quoted (t. c. p. 518), the smaller divisions being based on one or other of the other characters already noticed. A considerable number of the previously unknown syringes of Passerine birds were described by him for the first time in one or other of the above communications, amongst which those of Pitta and Atrichia are particularly noticeable. The peculiar form of the nasal bones in certain of the Tracheophonae, so that these birds are to this extent " schizorhinal," has already been mentioned when speaking of that character, as has also the paper on Heteralocha. It is much to be regretted that Prof. G-arrod did not live to make public his maturer v r iews on the difficult subject of the general classi- fication of birds. The only published scheme of any such classification is to be found in part ii. of his paper on the Thigh-Muscles (supra, p. 194) ; and it is within my knowledge that he had already seen reason to deviate in some respects from the arrangement there adopted. Nevertheless I Ibis, 1881, think I may say he was satisfied to the last as to the naturalness of the ^' ' two main groups into which he there divided birds, the " Homalogonatao " and the " Anomalogonatae." It is often assumed thab this division rests only upon a single character, namely the presence or absence of the ambiens muscle. As a matter of fact this is not the case ; for the ambiens muscle is absent in many birds that are ranked amongst the Homalogonatae. What Prof. Garrod says is this : " The oft T named of an " Acromyodian " syrinx, these birds depart essentially from the typical avian " Mesomyodian " structure, the one which there cannot be the slightest doubt is the more primitive form. "The much more important osteological characters" in which these two forms are said to diverge from the other Passeres are, as far as I am aware, two only ; and these, moreover, are individual peculiarities of each genus, and by no means common to the two forms in Menura the curved posterior margin of the sternum, in Atrichia the absence of clavicles (Garrod, P. Z. S. 1876, p. 516). As is now well known, Prof. Huxley's original description of the skull in Menura (P. Z. S. 1867, p. 472) was founded on a misconception of its structure, apparently owing to the im- perfection of his specimen (cf. Parker, Trans. Z. S. ix. pp. 307, 308). Moreover the "most anomalous forms of Passerine birds yet known," or at all events the most generalized, are, according to the views of Garrod and myself, certainly the Eurylamidfe, which last therefore, and not the " Pseudoscines," should be placed at the end, in a descending scale, of the Passerine series. 210 ON THE LATE PROF. GARROD*S CONTRIBUTIONS amliens muscle is, in my mind, the key to the whole," and that, not because of its own intrinsic importance, but because its presence is always associated with peculiarities in other parts never found in any Anoma- logonatous bird. That the same combinations of three or four different characters should have arisen independently in different birds is so ex- tremely improbable, that we can hardly ascribe these similarities in com- binations of characters to any other cause than to blood-relationship, the expression of which is now unanimously accepted as the true end of all biological classifications. " The facts disclosed by a study of the myology of birds do not, without extraneous assistance, place the families in their true relationship to one another. Because the same muscles are present in two families of birds, it cannot therefore be said that their kinship is extremely close, or the reverse .... It is therefore necessary to look around to find, if possible, myological characters which have some definite relation to equally well- marked pterylographic, visceral, or osteological peculiarities " (P. Z. S. 1874, p. 114). As already insisted on, it was in this combination of characters that Prof. Garrod trusted to find the true " key " to the question. To return to facts, no bird which is " Anomalogonatous " has ever an accessory femoro-caudal muscle ; that is, the letter B never enters its formula. Again, no bird that is Anomalogonatous has ever a tufted oil- gland and caeca, though this combination is nearly always found in the Homalogonatous birds. So much so is this the case, that there are only Ibis, 1881, nuie groups of Homalogonatous birds that have not this mentioned P- 29. combination *, and in two of these the exception is caused by the entire absence of the oil-gland (in the OtididaB and the Struthiones). Of the remaining seven, in which the caeca and a tufted oil-gland are not correlated, I find, by tabulation, that four always have the ambiens muscle developed, whilst the remaining three have it present in at least some of their members. Of these latter, moreover, two have the accessory femoro-caudal (B) always present ; so that it is only certain Parrots which have a formula like that of any anomalogonatous bird (for it must be remembered that in several Parrots the ambiens muscle is present). This fact will be made clearer by the accompanying table, containing the names of the three main groups of the Anomalogonatae, Ibis, 1881, followed by those of the seven groups above mentioned as the exceptions P- 30. to the combination of caeca and a tufted oil-gland amongst the Homalo- gonatae. Thus, of the whole series of Homalogonatous birds, not one, except certain Parrots (the most specialized, no doubt, of all that series), has * One or two genera in various families, as e. g. Cancroma, have also lost the oil- gland tuft. As, however, all the allied genera retain it, these slight exceptions in no way invalidate Prof. Garrod's argument. TO BIRD-ANATOMY AND CLASSIFICATION. 211 Anibiens. Accessory femoro- caudal. Tuft to oil- gland. Caaca. Anomalogoiiat 639 ' POSITION OF THE J AC ANAS IT having lately been my good fortune to dissect two specimens of Parrajacana, from Pernambuco, and an eviscerated specimen of Meto- pidius africanus, as well as to examine skeletons and skins of some other species of this group, a few notes on their anatomy may be acceptable to the Society, the more so on account of the very considerable difference of opinion that has hitherto existed amongst systematic ornithologists as to the true relationships of this group. Two main views on this subject have been put forward, one placing the Jac,anas near the Eails (Kallidae), the other asserting that they are, essentially, modified Plovers. The former of these views has been maintained by Jerdon :, Sundevall , and Milne-Edwards [|, to mention only some of the most recent ornithological writers of importance, as well as by the illustrious Nitzsch in his classical memoir on the pterylography of birds If. The latter view has been * Garrod, P. Z. S. 1876, p. 517. t Proc. Zool. Soc. 1881, pp. 639-647. Eead May 17, 1881. J Birds of India, iii. p. 707. Tentamen, p. 130 (1872). H Oiseaux Fossiles, ii. p. 110 (1869-71). If Pterylography, Bay Society's edition, p. 126. POSITION OF THE JACANAS. 221 adopted by Messrs. Sclater and Salvin in their Nomenclator,' where the Parridje are interpolated between the (Edicnemidae and the Charadriidse as members of the order " LimicolaB " *. The late Prof. Garrod, in his P. 1 paper on the nasal bones of birds t, says that " Parra should be removed to the Charadriomorphse " from the Eallidse, on account of the schizo- rhinal nature of its skull, as represented in the figure of that of Parra (Hydralector) cristata on p. 34 of his paper. In his subsequent paper on the muscles of the thigh in birds J, Parra (i. e. Metopidius) africana is placed amongst the " Grallae," with the other Charadriine or Scolopacine forms, and not included in the Eallidae. It will be my object in the present paper to still further strengthen this latter view of the affinities of the Parridse . Pterylosis. Nitzsch, in his * Pterylography/ places Rallm, Or ex, Porphyrio, and Parra as members of a group of the Fulicariae, characterized by the narrow form of the tracts, by the presence of a distinct outer branch to the inferior tract, and by the dorsal tract being " neither interrupted nor strikingly weakened" between the shoulder-blades. He says (I. c. p. 126) : " The first three [genera] have twelve tail-feathers, and exactly the same pterylosis as that figured as occurring in Rattus aquaticus. In Parra, of which I have examined all the four principal species (sinensis [i. e. Hydropliasianus chirurgus], cenea, africana, and jassana), I found only ten tail-feathers, and a remarkable narrowing of the bands of the dorsal tract close behind the shoulder-blades ; whilst, on the other hand, * Op. cit. p. 142 (London, 1873). The term " Limicola " was, I believe, originally used by Nitzsch (Pterylogr. p. 194) to include the birds now included in the " families " Charadriidse and Scolopacidae, together with some aberrant forms, such as Dromas, Cursorius, Thinocorus, &c. By Messrs. Sclater and Salvin its use is still further ex- tended to include the (Edicnemidse, Parridae, and Chionididae in addition. Lastly, Prof. Garrod used it (P. Z. S. 1874, p. 122, &c.) as a term for all the non-columbine " Charadriiformes," including in it, besides Nitzsch's groups, the Cranes, Auks, Gulls, and, presumably, the Turnicidse, Rhinochetidae, Plataleidas, and Parridse as well. In fact, Garrod's restricted " family " Charadriidae corresponds pretty nearly to the whole of Nitzsch's " LimicolaB sen Scolopacinse." To obviate further confusion, the term "Limicolae " should be restricted to the group mentioned by Nitzsch ; and I propose to substitute, as a name for the non-columbine Charadriiformes (the " Limicolaa " of Garrod) the word " Pluviales," to correspond with the other division, " Columbse " (including the Colurnbidse and Pteroclidse), of that great group. t "On the Value in Classification of a Peculiarity in the Anterior Margin of the Nasal Bones of certain Birds," P. Z. S. 1873, pp. 33-78. | P. Z. S. 1873, pp. 626-644. Besides Metopidius africanus, Prof. Garrod dissected a specimen of Hydrophasianus chirurgus ; and some MS. notes of his on that species I have incorporated in what follows.- 222 ON THE ANATOMY AND SYSTEMATIC the hindmost, or pelvic portion of it, was dilated. This genus has also weaker lumbar tracts ; and these are united with the uropygial portion of the dorsal tract by sparse contour-feathers." As Mtzsch himself later on says that the pterylosis of the "Lhnicolse" closely approaches that of the Eallinae, and is but little modified from that type, the evidence from pterylosis of the Ealline affinities of the Jaganas is not very strong. In their possession of well-marked firm rectrices, in the weakness of the lumbar tracts, and in the tendency to a division of the dorsal tract into an anterior and a posterior fork, the Parridse differ from the typical Eallidae, and approach the Limicoline type. The same relationship is indicated by the inner, or main, pectoral tract, though very narrow, consisting, at least at its commencement, of two or three rows of feathers in the Parridae, as well as in the Charadriida3 ; whereas in the typical Eallidse, according to Nitzsch, it issues from the branch as only a single row of feathers. Visceral Anatomy, <$fc. The tongue is long and narrow in shape, thin, and of horny con- sistency. Its apex is slightly notched, and its base spinulose ; for the greater part of its length it is strongly concave. The oesophagus develops P.Z. S. 1881, no crop ; and the proventriculus is zonary. There is a muscular gizzard, p. 641. lined by a rather thick and hard epithelium. The contents of the stomach, in one of the specimens examined, consisted of small seeds mixed with vegetable debris and small fragments of stone. The right liver-lobe is elongated, and twice as large as the left ; and there is a well-developed gall-bladder. The ceeca are lateral in position, and closely approximated to the intestine, which makes them difficult to see. They are mere nipples '2 inch long *. In this respect Parra differs from all the Eallidae (except Porzana notata) which have been examined, as in all of these the caeca are long, sometimes very long. Of the Pluvialine birds, only the Plataleidae and Ehinochetidae, with Sterna, Larus, and one or two others, have such short caeca. The intestines measure, in these two specimens of Parra jacana small intestine 12-3 and 13'2 inches, large intestine 1-1 and 1*0 inch respectively. In Parra jacana and in Hydrophasianus africanus, as also in Metopidius africanus, as already recorded by Grarrod t, there are two carotid arteries. This is the number found in all the Eallidae, and in most of the Pluviales, excepting the Turnicidas and Arctica alle, according to Grarrod J, where there is only the left developed. * In Hydrophasianus chirurgus there is a strong gizzard, and the left liver-lobe is smaller than the right ; the caeca measure "15 inch, the whole length of the intestines being 12 inches. Garrod's MSS. t P. Z. S. 1873, p. 469. | L. c. pp. 469, 470. POSITION OP THE JAANAS. 223 Myology. Parra jacana resembles P. (Metopidius) africana, as recorded by Garrod*, in possessing the ambiens, femoro-caudal, accessory femoro-caudal, semi- tendinosus, and accessory semitendinosus muscles, all well developed t. Their formula is therefore AB . XY J. In both these species the gluteus primus is well developed, covering the biceps superficially towards the median line ; the gluteus quintus is also well developed. As in the Eallidae, and the Gruidse and Eurypyga, amongst the Pluviales, the area of origin of the obturator internus is triangular, as it is also in Hydrophasianus ; in the Pluvialine birds generally it is oval. The two deep flexor tendons of the foot are not at all ossified, but completely blend together some way up the leg in Metopidius, in fact, just below the joint. There is no slip at all to the hallux, as was also found to be the case in Parra africana || and Hydrophasianus by Prof. Grarrod. This is the more remarkable on account of the very large size of the hallux in all these birds. A special tendinous slip to that digit is very frequently present in birds which have a very insignificant hallux indeed ; and I know of no other case of a bird with such a large hallux as that of the Parridae lacking the tendon. This fact would seem to indicate that the Parridaa may have been developed from some form with a more normal-sized foot and a small hallux, which had no special long flexor, the great size of their feet p z g having been developed in accordance with their peculiar habits. p. 642. In the anterior extremity the second pectoral arises from nearly the whole length of the sternum ; in all three genera the third pectoral is wanting. The expansor secundariorum is strong and T-shaped, as in all Ralline and many Pluvialine birds. In Parra jacana (as in Hydro- phasianus, according to Grarrod) there is a distinct biceps slip to the patagium, as in all the EallidaB, the Charadriidae, Gruidae, and many other Pluvialine birds. In Metopidius africanus it is apparently absent, the absence being probably correlated with the peculiar expanded form of the radius (to be hereafter described). In the wing- membrane the tensor patagii brevis presents a peculiar arrangement, the tendon being completely divided into two portions an inner, more slender, and an outer, stronger one. The former runs on to the fibrous tissue near the superficial origin of the extensor metacarpi radialis longior, and there stops; the latter continues over this last muscle to the ulnar side of the arm, where it is lost in the fibrous covering-tissue adjacent. Before crossing, however, it sends off a short, special wrist ward slip to the superficial tendon of origin of the * P.Z.S. 1873, p.641. t In Hydrophasianus all these five muscles are also present. I Cf. Garrod, P. Z. S. 1874, p. 123. Garrod, P. Z. S. 1876, p. 199. || p. z. S. 1875, p. 348. 224 ON THE ANATOMY AND SYSTEMATIC metacarpal extensor, as in many other groups of birds. There is also a thin fibrous expansion given off just before this to the teudon of the tensor patagii longus, and the tissue of the patagium generally, as in many Pluvialine birds *. This splitting up of the tensor patagii brevis tendon into two distinct slips, the external one in turn giving off a special wristward slip, occurs in many Pluvialine birds (e. g. in Namenius arquatus, Totanus calidris, Machetes pugnax, Himantopus nigi-icollis, ThinocoruSj Attagis), but never in the Bails, where the tendon is always much more simple, not being divided into two separate parts, or giving off a wristward slip. In fact, in most Eallidae it runs quite simply, as a narrow straight tendon, on to the origin of the extensor metacarpi muscle, and there stops. The trachea is provided with the usual pair of sterno-tracheal muscles ; and the lower larynx, which is of simple structure, has also only a single pair of intrinsic muscles. Osteology. From a consideration of the pterylographic, visceral, and myological features only of the Parridae, perhaps no very definite conclusion as to their affinities could be drawn. But their osteological characters, in this case, leave no doubt as to their real position. All the skulls of ParridsD which I have examined, including those of Parrce jacana a,nd.gymnostoma, Metopidii indicus, africanus and albinucha, and HydropJiasianus chirurgus, like that of Hydralector cristata figured by Grarrodt. are strongly schizorhinal, therein differing completely from that of the Bails, and P. Z. S. 1881 resembling that of the Pigeons, Plovers, and their allies (the " Cha- p. 643. radriiformes " of Garrod ) only amongst Homalogonatous birds. Fig. 1. Skull of Parra jacana, from below; natural size. There are well-developed basipterygoid processes, which are always absent in the Eails, though of very frequent occurrence amongst the " Pluviales," occurring in all the Charadriinso and Scolopacinae I have examined. In Parra jacana and Metopidius albinucJia, the long, narrow, slightly * In Hydrophasianus much the same arrangement of the tensor patagii brevis obtains, to judge from a small drawing in Garrod's MS. t P. Z. S. 1873, p. 34, fig. 5. \ P. Z. S. 1874, p. 117. POSITION OF THE JA^ANAS. 225 decurved vomer is emarginate apically, as in certain Charadriidaj * (see fig. 1). In the Ballidae it is, I believe, always sharp at the point. The maxillopalatine processes are rather slender and directed back- wards; they have the form of concavo-convex lamellae, are not at all swollen, and do not unite by some way in the middle line, the vomer appearing between and (when the skull is viewed from the palatal aspect) below them. There is no ossified internasal septum, nor any ossification of the narial cartilages. The lachrymal is small, ankylosed with the naso- frontal region of the skull above, and with the " pars plana " below. On the posterior aspect of the skull there are no traces of the occipital fontanelles, which are found in so many of the birds related to the Plovers. The supraorbital impressions for the nasal glands, which are so conspicuous in most Plovers, the Gulls, Auks, and many other birds, are absent in the Parridae. The combinations depending on the presence or absence of basi- pterygoid processes, of occipital foramina, and of impressions on the top of the skull for the supraorbital glands, coincide, as may be seen from the following Table (p. 226), pretty accurately, with hardly an exception, with the chief groups of the Pluviales (the web-footed Laridae and Alcidae being omitted as irrelevant to our present purpose) as determined by other characters. In the Table + and - represent respectively the presence or absence of the structure indicated. In the Plataleidae and Gruidae the nasal glands occupy the truncated edge of the cranium above the orbits, and hardly appear on its upper surface : this condition I have indicated by the use of the double sign ( + ). The drawing (fig. 2, p. 227) of the sternum of Metopidius albinucTia will P. Z. S. 1881, show how unlike it is to that of the Kallidse. In the latter group the p ' 645 ' sternum is always peculiar in that the xiphoid processes exceed in length the body of the sternum, which tapers to a point posteriorly, and from which they are separated by very long and well-marked triangular notches. The carina sterni also is less well developed ; and the clavicles are weaker and straighter, being less convex forwards, than in the ParridaB. The sternum and clavicles of Parr a and Metopidius in general form, on the other hand, resemble closely the type found in some of the Pluvialine birds (e. g. Thinocorus, Attagis). The pelvis, again, of the Bails presents certain well-marked pecu- liarities. If that of Rallus aquaticus be taken as a typical form, it will be found that the ilia are long and narrow, and but little expanded in their preacetabular part. The postacetabular portion of the pelvis is but little bent down on the preacetabular part ; and the ischia and pubes * Cf. Garrod, P. Z. S. 1877, p. 417, figf. 2-4. 226 ON THE ANATOMY AND SYSTEMATIC P. Z. S. 1881, p. 644. PLATALEID^E. Ibis rubra 4- leaajaja leucorodia GRUID.E. Grus cinerea 4- americana australasiana 4- Aramus scolopaceus 4- CHARADRIID^E. Numenius arquatus 4- phseopus 4- Recurvirostra avocetta 4- Tringa canutus cinclus JEgialites hiaticula Eudromias morinellus 4- Vanellus (cristatus ?) 4- Machetes pugnax 4-* Hsematopus ostralegus 4- Himantopus nigricollis Limosa melanura 4- PARRID^S. Metopidius albinucha indicus Parra jacana Hydrophasianus chirurgus.. Turnix lepurana Hemipodius varius Dromas ardeolat Chionis alba Attagisgayi Thinocorus rumicivorus dorbignyanus Cursorius gallicus Glareola pratincola RHINOCHETID.B. Rhinochetus j ubatus Eurypyga helias MESITID.B. Mesites unicolor J Occipital foramina. Basiptery- goid pro- lupraorbi- al impres- sions. * Sometimes ossified. t There is a perfect skeleton of this peculiar form in the Cambridge University Museum, which I have examined. J I only know the cranial characters of this bird from the plate illustrating M. Milne-Edwards's memoir (Ann. Sci. Nat. [6] vol. vii. art. no. 6). POSITION OF THE JAANAS. 227 are but little everted. The ischia are united by broad bony plates to p - Z ' 6 ^ 5 1881 ' about the three most posterior " sacral " vertebra ; between these plates and the expanded part of the ilia above are well-developed and deep fossa?, occupied, in the fresh state, by the posterior portion of the kidneys. Viewed from above, the well-marked " postacetabular " ridge, which divides off the dorsal from the lateral aspect of the pelvis, running Fig. 2. Sternum and shoulder-girdle of Metopidius albmucha, viewed laterally ; natural size. from just behind the antitrochanteric eminence to the posterior spine of the ilium, presents, a little behind those two points, a strongly projecting process. The greatest breadth of the postacetabular part of the pelvis is therefore here, and not at the more anteriorly-situated prominence, close to the antitrochanter. Viewed from the side, this ridge forms a sort of overlapping roof to the slightly excavated external pelvic fossa. The genera Ocydromus, Aramides, Fulica, and Porphyrio do not essentially depart from this type. In Parr a and Metopidius* the ilia are wider and more expanded anteriorly. The postacetabular ridge has hardly any median projection ; P. Z. S. 1881, and the pelvis is widest, dorsally, just behind the antitrochanter s. The P* ^* plates of bone between the ischia and sacrum are narrower, and the posterior part of the renal fosssD less well developed, and more open, in consequence. In all these points these forms thus approach the Limi- coline birds. There is one other point of interest in the osteology of the Parridse. This is the extraordinary form assumed by the radius in some of tha genera. In birds, as a rule, the ulna is -a stouter bone than the radius, this last being almost universally a slender cylindrical bone. In Meto- pidius africanus, as already noticed by M. A.Milne-Edwards t, as well as * Milne-Edwards has also described tke difference of the pelvis in the Jasanas as compared with that of the true Ballidse : cf, ' Oiseaux Fossiles,' ii. p. 123. t ' Oiseaux Fossiles,' ii. p. 134. Q2 228 ON THE ANATOMY OF THE JAANAS. in M. albinucna and in M. indicus (as I have been able to ascertain by extracting the wing-bones from a skin), the radius presents the form shown in the drawing (fig. 3), being dilated and flattened into a sub- Fig. 3. Wing-bones of Metopidius albinucka, to show the peculiarly modifie d radius ; natural size. triangular lamellar-like expansion for its distal half. Its superior surface is slightly grooved posteriorly for the tendon of the extensor metacarpi radialis longior muscle. This dilated portion forms the margin of the patagial space for its distal portion. A considerable portion of the marginal tendon of the tensor patagii longus is inserted into the radius at the angle of the bone ; the main tendon, however, continues in a groove on the inferior aspect of that bone, a little behind the border, to its ordinary insertion. About half of the peculiar flattened radius is left bare of muscle above, the extensor metacarpi, as already stated, playing over its lower half. Below, the flattened area is largely covered by the fibres of the pronator radii superficialis, which extend up nearly to the margin of the bone ; below this is the pronator radii profundus, which likewise has an extensive insertion into the lower part of the bone. The margin of the bone, where it is superficial, is slightly roughened ; and no doubt the peculiar form of radius is associated with the quarrelsome habits of these birds, this dilated and somewhat scimitar-shaped bone being probably capable of inflicting a very severe downward blow. In Parrajacana and P. gymnostoma the radius presents the ordinary form ; and the same is the case in Hydrophasianus chirurgus. P. Z. S. 1881, In these two genera, it is to be observed, the metacarpal "spur" is much more developed and sharp than in the species of Metopidius, where it is small and blunt ; so that there is a correlation apparently between a sharp spur and a simple radius, and a blunt spur and flattened radius. In Hydralector gallinacea there is a blunt spur, with, so far as I can make out from a skin, a flattened radius. The " claw " or " spur " of the wing of the Jaganas has, it may be ob- served, no relation whatever to the " claw " or nail of the pollex, which is also present, though small, in all the three genera I have examined. The " spur," in Parrajacana at least, consists of an external, translucent, ON THALASSIDKOMA NEREIS. 229 yellow epidermic layer, which invests a central core of compact fibrous tissue, this in turn being supported by a bony projection developed at the radial side of the first metacarpal. As regards the position of the Parridaa in the group Pluviales, it appears to me that they form a well-marked family, with no very obvious relation- ships to any of the other families of that group, approaching, however, perhaps most nearly to the CharadriidaB, from which they are easily distinguishable by the absence of supraorbital glands and occipital foramina, by their enormously elongated toes, by the number of rectrices, and other points. A brief definition of the Parridae may be given as follows : Charadriiform birds, with ten rectrices, short caeca, and a tufted oil- gland ; with the ambiens, accessory femoro-caudal, and accessory semi- tendinosus muscles developed, and with the obturator internus triangular ; with a two-notched sternum, and with the digits, including the hallux, greatly developed ; with the skull provided with basipterygoid processes, but lacking occipital foramina and supraorbital gland-impressions. 40. ON THE PETREL CALLED THALASSIDROMA NEREIS P. z. s. 1881, BY GOULD, AND ITS AFFINITIES.* p ' 735 ' IN this Society's Proceedings for the year 1840, the late Mr. Gould described a "beautiful fairy-like " new species of Stormy Petrel from Bass's Straits, which he called Thalassidroma nereis (torn. cit. p. 178), under which name it is figured in the last volume of the * Birds of Australia/ Dr. Elliott Coues, in his revision of the family Procellariidaa t, treating of the species under the name Procellaria nereis, says : " I have had the pleasure of examining Mr. Gould's types of this species from Bass's Straits, Australia, now in the collection of the Philadelphia Academy. It is a beautiful little species, quite unlike any other known Stormy Petrel. In form it comes nearer to Procellaria pelagica than to any other species ; and it is probably congeneric with it, though it differs some- what $ in the proportion of the tarsus and toes, and very widely in its pattern of coloration The proportions of the tibia and tarsus differ from those of pelagica in the greater comparative length of the former." * Proc. Zool. Soc. 1881, pp. 735-737. Bead June 21, 1881. t Proceedings of the Academy of Natural Sciences of Philadelphia, 1864, p. 81. The italics are mine. W. A. F. 230 ON THALASSIDROMA NEREIS. Amongst the Petrels mentioned at various times by the late Prof. Garrod as having been examined by him, a species several times occurs which is doubtfully named " Procellaria (or Thalassidroma) fregata ? " * The specimens dissected by him are now before me, and have been identified by Mr. Salvin as being really referable to the Procellaria nereis of Gould, an example of which, from the Falkland Islands, is now in the museum of Messrs. Salvin and Godman. A careful examination of the three spirit-specimens of this bird, as well as of the skin mentioned, has convinced me that this species is not referable to the true genus Procellaria as represented by Procellaria pelagica, and is in fact in no way related to that group of Petrels, but has its nearest allies in the flat- clawed genera Oceanites, Fregetta, and Pelagodroma. In his paper on the muscles of the thigh in Birds t the late Prof. Garrod divided the Nasutae, or Petrels, into two groups, the " Storm- Petrels " and the Eulmaridae, the former group differing from the latter in that they possess the accessory semitendinosus muscle (T), but lack intestinal caeca. In the Fulmaridae, on the other hand, the accessory semitendinosus muscle is absent, but caeca are present. The species of Storm-Petrels on which this generalization was based are called, with doubt J, "Procellaria pelagica and P. fregata" the latter being the P. Z. S. 1881, species now identified by Mr. Salvin as P. nereis. As regards the first- p. 736. named species, there can be little or no doubt that the bird really dissected by Prof. Garrod, and called by him " 'Procellaria pelagica" was Wilson's Petrel (Oceanites oceanicus), as in this bird there are no caeca , at the same time that the accessory semitendinosus muscle is present. The true Procellaria pelagica (of which I have lately dissected two perfectly fresh examples) agrees with the Fulmaridse, as defined by Prof. Garrod, in having cseca ||, but no accessory head to the semitendinosus ; and Cymochorea leucorrlioa agrees in both these points with Procellaria pelagica. The so-called " Procellaria nereis " of Gould is therefore obviously not a true Procellaria at all ; and this view is confirmed by other characters, such as the shape of its nostrils, the elongated tarsi, which are much longer than the mid toe U" and covered anteriorly with transversely * Cf. P. Z. S. 1873, pp. 470 and 641. t P. Z. S. 1874, p. 122. J P. Z. S. 1873, p. 641. Cf. also Macgillivray, in Audubon's ' Ornithological Biography/ v. p. 646. || Cf. Macgillivray, I.e. p. 313 ; also Wagner in Naumann's ' Vogel Deutschlands,' x. p. 556. ^[ In Procellaria pelagica and Cymochorea leucorrlioa the tarso-metatarse is not longer, and may be shorter, than the 3rd toe. As against 21 '5 and 21 '5, and 23 and 26 millitn. in the two first-mentioned genera, in the so-called Procellaria nereis the lengths of the two are respectively 34 and 26 millims. ON THALASSIDKOMA NEKEIS. 231 arranged scutella *, the very minute hallux, and the lamellar, concave form of the claws. It belongs, in fact, to the group of Oceanites, Fregetta, and Pelagodroma, but is not exactly congeneric with any of them. I propose therefore to make it the type of a new genus, to be called Garrodia, in memory of my lamented friend A. H. G-arrod, not only as a token of my personal esteem for, and indebtedness to him, but also as some slight recognition of the thanks ornithologists generally owe him for the additions he made to our knowledge of the anatomy of birds. The genus Garrodia may be shortly defined as follows : GABBODIA. Genus ex ordine Tubinarium OCEANIT^ maxime affine, tarsis pro diyitis longioribus et antice scutellatis, necnon margine sterni pos- teriore integro distinguendum. Type Procellaria nereis, Grould. Garrodia is perhaps most closely allied to Oceanites, as already stated, but differs from that genus in having the tarso-metatarsi covered anteriorly with a series of transverse scutella instead of being " entire," in their slightly greater proportional length as compared with the third toe t, in the even more minute hallux, and in the more flattened and lamellar form of the claws. The sternum too is posteriorly entire, whereas in Oceanites oceanicus it is slightly notched. The coloration of the two genera is also quite different. From Fregetta, Garrodia may be easily distinguished by the very different proportions and forms of the nails and feet in that genus, and from Pelagodroma by its much shorter feet and entire tail. These four genera Oceanites, Garrodia, Pelagodroma, and Fregetta ^' Z ' form a very well-marked family of the Tubinares, which may be called Oceanitidae, as distinguished from the remainder of the group, or Fulmaridae of Prof. Garrod. Anatomically, these four genera agree together, and differ from the Fulmaridse (on nearly all the genera of which, including Diomedea and Puffinuria, I have notes), in the two important characters already mentioned the absence of caeca and the presence of the accessory semitendinosus muscle. Externally they may be at once recognized by their peculiar elongated tarsi, lamellar nails, and by nev naving more than -10 secondaries, Procellaria and Puffinuria having 13, and the remaining Fulmaridae more (in Diomedea, according to Nitzsch, as many as 40). My family Oceanitidae, in fact, corresponds to Bonaparte's section " ** Unguibus depressis " of his Procellaridae J, and toCoues's " second group'* of the similarly-named section in his ' Review ', * In Procellaria pelagica the tarsi are pretty uniformly covered with somewhat irregular hexagonal scutes. t In a specimen of Oceanites oceanicus (in spirit) the middle toe measures 29 millim. in one of Garrodia the length is 26 millim. The length of the inetatarse in both is 34 millim. J Consp. Av. ii. p. 197 (1857). Op. cit. p. 74, where characters for it are given. 232 ON THE TRACHEA IN THE " RAT1TE " BIRDS. with the addition, in each case, of Garrodia, included by both authors in the restricted genus Procellaria. Being now engaged in a report, for the Voyage of H.M.S. ' Challenger,' on the anatomy of the Petrels collected during that expedition, I propose to reserve further details of the differences and characters of these two groups, and of the genera composing them, till that occasion. P. z. s.1881, 41. ON THE CONFORMATION OF THE THORACIC END P- 778. TRACHEA m THE RATITE" BIRDS.* IN the present communication I propose to follow out the line of work developed by the late Prof. Grarrod in his paper on the trachea of the Gallinse t, by describing in detail the structure of the bifurcating trachea in the " Eatite " birds. So far as I am aware, no proper description of this structure in the birds in question has ever been given, though the statement, apparently originally due to Meckel , that in them " there is no lower larynx," has been very generally followed and copied, even in the latest text-books on the subject . Prof. Owen has briefly described the bifurcating trachea in the Ostrich || and Apteryx ^[ ; and his accounts, as far as they go, are accurate enough. More recently E. Alix has very briefly mentioned some peculiarities of this part in the Ehea; and his account will be found quoted below. P. Z. S. 1881, Struthio camelus (figs. 1, 2), on account of its size and simple structure, p. 779. m ay be described here first. The trachea, inferior to the insertion of the sterno-tracheales, slightly narrows, having above the antepenultimate P. Z. S. 1881, ring a diameter of about one inch. The tracheal rings are here, as p. 780. elsewhere, entire simple rings, of an average depth of about *15 inch, and are separated only by very slight interannular intervals. The trachea is slightly compressed and posteriorly carinated for about the last 7 rings. The last ring but four is somewhat produced downwards in the middle line, both anteriorly and posteriorly; it is, in consequence, narrower * Proc. Zool. Soc. 1881, p. 778-788. Read June 21, 1881. t " On the Conformation of the Thoracic Extremity of the Trachea in the Class Aves. Part I. The Galling," P. Z. S. 1879, pp. 354-380. \ ' Trait< general d'Anatomie comparee,' x. p. 571 (1838). Cf. Huxley's 'Anatomy of Vertebrates,' p. 313; Macalister, 'Morphology of Vertebrates,' p. 161. || ' Catalogue of the Physiological Series of the Museum of the Royal College of Surgeons,' ii. p. 1C3, prep. 1159 (1834). ^[ Trans. Zool. Soc ii. p. 279. ON THE TRACHEA IN THE " EATITE " BIRDS. Fig. 1. 233 P.Z.S. 1881, p. 779. Bifurcating trachea of Struthio camelus, from before. b. Section of wall of trachea, from behind, to show the vocal cord formed by the thickening of the mucous membrane of the interior. I. II., first two bronchial semirings. 1, 2, &c., last trachea! rings. Here, and elsewhere in these figures, o indicates the last, oo the penultimate, ooo the antepenultimate tracheal rings. Fig. 2. The same, from behind. About natural size. laterally than elsewhere. The antepenultimate ring presents the same P. Z. S. 1881, features more strongly developed. In two of the four specimens exa- p ' ' mined it sends down a small pessulif orm process of cartilage in the middle line behind, filling the chink left between the posterior extremities of the two next (incomplete) rings. The penultimate ring is narrower and 234 ON THE TKACHEA IN THE " RATITE " BIRDS. more cylindrical than its predecessors ; it is also wider transversely, and incomplete behind in the middle line, its extremities, however, being closely approximated to each other. The last tracheal ring is still wider transversely, and more cylindrical ; and it too is incomplete posteriorly, to a greater extent than its predecessor ; viewed from the side it is convex upwards, as are its few immediate predecessors in a less degree. The interannular intervals between all these rings are, when undisturbed, mere chinks filled up by dense fibrous and elastic tissues. There is no trace of a pessulus, though the last tracheal ring is slightly produced downwards in front. The first bronchial semiring, on each side, is narrow and cylindrical, strongest anteriorly, and somewhat attenuated posteriorly. It is separated only by a narrow interval from the last tracheal ring. The second and third rings are similar, but are more slender and lengthy ; they are convex downwards, but very slightly so ; hence the interannular intervals are small here also. Their anterior ends are very slightly inturned, impinging but to a small extent on the membrana tympaniformis, which completes the bronchial tubes internally, and, in consequence of the absence of any three-way piece, passes con- tinuously from one bronchus to the other, so closing the tracheal tube inferiorly. The fourth, fifth, and succeeding bronchial rings are similar in character ; but their ends, which tend to be dilated posteriorly, are successively more and more incurved to about the tenth. Nowhere are the bronchial rings complete. There is, at most, only a trace of a membrana semilunaris, in the form of a very feeble, scarcely raised, antero-posteriorly directed fold of mucous membrane. Internally, the mucous membrane of the interior is greatly thickened, forming a vocal cord, in the region of the last three tracheal rings and first two bronchial semirings (vide fig. 1, 6). There is no trace of any intrinsic voice-muscle ; and the lateral tra- cheal muscles stop at the point of insertion of the sterno-tracheales. The genus Apteryx, in the simplicity of the structure of its lower larynx, stands on the same level as Struthio. In Apteryx mantelli (figs. 3, 4) about the last dozen tracheal rings are quite simple in form, with narrow interannular intervals, and no anterior and posterior notching. The penultimate ring is produced slightly downwards, in a triangular way, both anteriorly and posteriorly. The last ring is also produced down- RZ. S. 1881, wards anteriorly, but is incomplete in the middle line behind; it is p. 781. slightly wider and stronger than the preceding rings. The first two bronchial semirings on each side closely resemble it in form ; the first semirings of opposite sides are almost in contact at both extremities, the next pair being more widely separated at those joints. There is only a narrow space between the last tracheal and the first bronchial ring. The ON THE TRACHEA IN THE " RATITE " BIRDS. 235 Fig. 3. Bifurcating trachea of Apteryx mantelli, from before. Fig. 4. \ The same, from behind. About twice the natural size. succeeding bronchial semirings are perfectly simple, rather deep and stout pieces of cartilage, separated by narrow interannular spaces, and com- pleted internally by a broad membrana tympaniformis ; nowhere do they become complete circles. As in Struthio, there is no pessulus, and no intrinsic muscle. There is a slight antero-posteriorly directed vertical fold of mucous membrane between the two bronchial apertures internally, and also a feebly developed vocal cord on the external wall of the bronchi, where they diverge from the trachea. The lateral muscles stop some way before the end of the trachea, at the place where the sterno-tracheales are inserted. Apteryx australis, A. haasti, and A. oweni have all been examined by me, and all agree closely in their tracheal structure with A. mantelli. My specimens of A. australis and A. oweni (two) agree together in having the last three tracheal rings incomplete posteriorly ; whilst in A. mantelli and A. Jiaasti, of which I have seen only single specimens, the last ring alone is complete. In the Casuariidse we meet with peculiarities in the structure of the P.Z.S. 1881 p. 782. 236 ON THE TEACHEA IN THE " EATITE " BIRDS. bifurcating trachea not existing in the other " BatitaB." In Casuarius galeatus (figs. 5, 6, p. 237) the trachea is somewhat dilated for the ter- minal inch or so of its extent, transversely and also posteriorly. The last tracheal rings (for a number varying in different specimens, in the specimen in question 12, in another 23) are incomplete in the middle line behind, though the posterior ends are closely approximated together *. These rings are tolerably uniform in breadth posteriorly, tapering only somewhat at their extremities; the interannular intervals are mere chinks. Anteriorly, however, the rings being dilated in the median line and attenuated laterally, the intervals are better developed. The last 5 or 6 rings are more and more curved downwards anteriorly, whilst their posterior moieties are somewhat dila.ted, their ends gradually rece- ding more and more from each other in the middle line. There is no trace of a pessulus. The first bronchial semirings much resemble the last tracheal rings, their anterior extremities being closely approximated together, and their posterior ones dilated, and somewhat pointed, terminally. The second, third, and fourth bronchial semirings are simple cartilaginous hoops, tolerably deep, separated only by narrow intervals, and, as usual, com- pleted internally by a membrana tympaniformis. The fifth, sixth, and seventh are similar but longer semirings, the sixth and seventh being dilated anteriorly. The succeeding rings are similar but quite simple hoops, never forming anywhere complete circles. There is no trace of any intrinsic muscles. The sterno-tracheales are inserted on the trachea at about the twelfth ring from the last. Ante- riorly they expand on the tracheal wall, and are in contact with each other over the middle line, as is also the case in Dromceus, though not in the other three genera. A small part of each muscle runs to be inserted into the posterior wall of the trachea near the margin. The lateral muscle of the trachea passes between these two portions of the sterno- trachealis of its side, but does not pass down further than the com- mencement of the tracheal tympanum. * Between the extremities of each imperfect tracheal ring runs a short band of connective and elastic tissue, with the fibres running transversely. These extend the whole length of the trachea, and when well developed have the appearance of a longi- tudinal band running along the middle line of the tube posteriorly. By the contraction of these fibres, the ends of the tracheal rings, where these are incomplete, or their more slender middle portions where perfect, are drawn together, and pressed into the interior of the tube, so forming what at first sight looks very much like a longitudinal, though incomplete, tracheal septum, such as is found in some Procellariidas and other birds. In consequence of this structure, a transverse section of the tracheal tympanum posteriorly presents two strong convexities separated by a median concavity. ON THE TRACHEA IN THE " EATITE " BIRDS. Fig. 5. 237 P.Z.S. 1881, p. 783. Bifurcating trachea of Casuarius galeatus, from before. Fig. 6. The same, from behind. About natural size. The portion of the stemo-tracheales muscles at their insertion is represented in each figure. As compared with Struthio and Apteryx, all the tracheal and bronchial p. z. S. 1881, rings are much less firm and more cartilaginous in Casuarius, as also P- 784. in Dromceus. The membrana tympaniformis completing the trachea below is a simple membrane, passing continuously from one bronchus to the other, with no intervening pessulus. 238 ON THE TRACHEA IN THE "RATITE" BIRDS. Internally, a very slight thickening of this membrane in an antero- posterior direction, at the bifurcation of the tube, may be seen ; but there is nothing that can be properly called a membrana semilunaris present. The external vocal cord, on the other hand, situated over the first two bronchial semirings, is very well-developed, with a sharply- defined margin. On the internal wall of the bronchus there is, in addition, a second, much slighter and less prominent fold, slightly concave forwards, running somewhat obliquely backwards and downwards, and supported by the anterior ends of the third and sixth bronchial semirings. Besides Casuarius galeatus, I have examined trachea of CC. beccarii, bennetti, uni-dppendiculatus (2 specimens), and westermani ; and in none of these species can I detect any difference of importance from the arrange- ment I have described above. In the last-named species (an adult specimen) there is a considerable amount of ossification in the last tracheal and first bronchial rings. The mucous fold on the internal walls of the bronchi varies much in development in different specimens. In adults there is a great accumulation of the fibrous and elastic tissues of the mucous membrane in the region of the tracheal tympanum . Of Dromceus novce-nollandice I have only, as yet, been able to examine one trachea, and that too from a young specimen. This closely resembles that of Casuarius ; but the number of imperfect tracheal rings seems to be considerably smaller, in the specimen in question only the last three being incomplete behind. The third and fourth bronchial semirings are considerably stronger than the first two and the immediately preceding tracheal rings. There is no pessulus; but the membrana semilunaris, especially posteriorly, seems to be better developed. The insertion of the sterno-tracheales is as in Casuarius. The lateral tracheal muscles extend down to within about 1-5 inch of the end of the trachea. In the genus Rhea (as represented by R. americana and R. macro- rhyncha) a very different condition of things occurs, there being a highly- specialized and peculiar syrinx, provided with a pair of intrinsic muscles *. P. Z. S. 1881, In Rhea americana (figs. 7, 8) the average diameter of the trachea p. 785. * I had observed the peculiar syrinx of Rhea some months before I met with M. Alix's short paper on this bird in the " Bulletin " of the Society Philomatique for 1874 (p. 38), in which he points out, for the first time, the fact that Rhea possesses a true syrinx. His account, which I here reproduce, runs as follows: "II y a chez le Nandou un larynx inferieur. Les cordes vocales sont placees a 1'origine des bronches, dont les premiers anneaux sont incomplets, en sort que la paroi interne qui leur corre- spond a 1'aspect d'une membrane tympaniforme. Le reste des bronches est forme par des anneaux complete." This description is, as will be seen, very incomplete, and, in the last statement, incorrect. ON THE TRACHEA IN THE " RATITE " BIRDS. 239 interiorly is about 1 inch, and it is somewhat compressed from before backwards. The cartilaginous tracheal rings are complete behind, and Fig. 7. Bifurcating trachea of Ehea americana, from before. Fig. 8. The same, from behind. About the natural size. The intrinsic muscle has been removed on the right side. In fig. 8 the fibrous band running down the middle of the trachea posteriorly, as described above in Casuarius, is also represented. closely approximated to each other. The last four tracheal rings are soldered together to form a cartilaginous box, the constituents of which are marked out by the interannular sutures, which are only interrupted in the middle line anteriorly and posteriorly. The exact number of tracheal P. Z. S. 1881, rings which are fused to form this box varies in different specimens from p * 786> 240 ON THE TRACHEA IN THE " EAT1TE " BIRDS. four to six ; in some cases it is, apparently, formed by four rings on one side and five on another. The lowest tracheal ring is strongly concave downwards, but in front in the middle line is transversely truncated. Posteriorly the tympanic box is deeply and widely notched. There is a distinct, narrow, cartilaginous pessulus, which runs from behind forwards, connecting the anterior and posterior walls of this box, and interrupting, in the mid line, the continuity of the membrana tympaniformis, which completes the bronchial walls internally. The first bronchial semiring is nearly straight, and of cylindrical form. It is closely connected at its extremities with the last tracheal ring; but between these points is a pretty wide, lunate, interannular interval. The anterior ends of these first semirings are inturned con- siderably, but do not meet each other, or the pessulus ; posteriorly they do not extend inwards, by some way, as far as the preceding or succeeding rings. The second bronchial semiring is similar in form, but dilated slightly behind ; it is closely approximated to the preceding ring in the greater part of its extent, but is anteriorly strongly curved downwards (in a somewhat sinuous way), so that here a considerable space is left between the two semirings in question. The third and fourth rings are considerably dilated, and produced inwards, posteriorly ; anteriorly, in the particular specimen figured, they are fused into a comparatively narrow ring. The fifth, sixth, and succeeding semirings are quite simple incomplete hoops of cartilage, which become smaller and less complete internally as they approach the lungs. Between the pessulus in the middle line and the tracheal box and first four bronchial semirings, is spread a membrane, completing the tracheal and bronchial walls at the bifurcation of the tube. This membrana tympaniformis is thinnest posteriorly ; but in the middle part of its extent, over a triangular area the base of the triangle resting on the pessulus, whilst the apex is at the posterior end of the fourth bronchial semiring it is thickened and of a fibro-cartilaginous consistency. From the sides of about the last ten tracheal rings a single thin but broad band of muscle arises, the fibres of which, running downwards and forwards, are inserted into the anterior halves of the first five bronchial semirings and the membranes between them, extending, in some speci- mens, almost to the anterior tips of the first two semirings. This muscle is therefore in all respects a true intrinsic syringeal one. The lateral tracheal muscle stops anteriorly to the origin of the intrinsic one. Internally, there is a well-marked, sharp-bordered membrana semilu- naris, resting on the pessulus, developed between the bronchi. In the space corresponding to the interval between the last tracheal and first bronchial rings, an accumulation of the fibrous and elastic tissues of the lining membrane of the bronchus gives rise to a well -developed vocal cord. ON THE TRACHEA IN THE " RATITE " BIRDS. 241 Rliea macrorliynclia closely resembles, in all respects, R. americana in the structure of its syrinx. In the single specimen I have by me, the P ' Z> ,?- J 881 tympanic box is anteriorly marked by four sutures on one side, and by three only on the other, as sometimes happens in R. americana (as, e. n a sort of delta of two small rivers, the Capibaribe and the Beberibe, which here flow into the sea. A little to the north of Recife is the old town of Olinda, situated on a hill of perhaps 200 feet or so above the sea, and commanding an extensive view of the flat and marshy country to the south which immediately surrounds Recife. Looking inland from this, the ground is seen to rise gradually, and then becomes more or less forest^covered, these low hills running to the south and west, and continuing into the hilly country which runs thence, more or less parallel with the coast, both north and southwards. Recife itself, now probably the second city of the Brazilian Empire, with a population of about 90,000 souls, consists of three towns, con- nected with each other by excellent iron bridges. That with the port is Recife proper ; the other two are called San Antonio and Boa Vista. Ibb, 1881, p.' 310. ' * I will not venture to ascribe any specific name to this bird, seeing the difficulty that attaches to its correct determination. 244 ELEVEN WEEKS IN NORTH-EASTEKN BRAZIL. Northwards there lies between Eecife and Olinda a low, mangrove- covered, swampy tract, separated from the sea by a beach of sand and shingle, whilst to the south lies Cocoa-nut. Island and more swampy country. Towards the west lies the suburb called Boa Vista ; and here, and extending more or less to Caxanga (an outlying village celebrated for its pine-apples, with which Eecife is connected by a street-railway), are situated the villas and houses of the more wealthy inhabitants. These are generally surrounded by gardens, often well kept and stocked with all kinds of tropical plants, native or otherwise. Here, in a quarter called Estancia, I found excellent accommodation at a boarding-house kept by two American ladies, and tenanted chiefly by Englishmen engaged in business in Eecife. As the house stood in a large garden of its own, with numerous fruit-trees, and abutted on a considerable tract of marshy and little-cultivated ground, I determined on ma'king this my head quarters, and after safely passing my baggage through the Custom House, set to work on the birds and insects. There are some considerable patches of wood on the outskirts of the town in this direction, and numerous more or less deserted gardens, orangeries, and pieces of marshy ground, in which birds were fairly abundant, though in the town itself excepting Urubiis (Cathartes atratus), a stray Humming-bird or two, Swallows (Hirundo leucorrhoa), and " Lavenderas " (Fluvicola dimacura), which last are to be seen everywhere and are very tame, like Eobins not a bird is to be seen. No regular forest is met with till near Caxanga, about 8-10 miles from Eecife, where the country becomes hilly and covered with thick wood, which, in places, is, I believe, undoubted virgin forest, though most of this has been cleared and replaced by second-growth (capoeira) of varying size and thickness. Unfortunately the weather was not all favourable to collecting during my stay in Eecife, the rainy season, which usually, I was told, ceases about the end of July, lasting on more or less for another month *. As Ibis, 1881, the soil here is, as nearly universally elsewhere in Brazil, a thick red p. 316. clay, the roads and by-paths remained almost impassable, rain falling heavily nearly every day for some hours. In the " Gymnasium " of San Antonio is a small museum, with a decent, though badly named, series of birds and Mammalia. Most of the * The dry and hot weather (which also is the season for yellow fever on the coast) in Pernambuco commences about September and continues till March. November, January, and February are usually about the hottest months. May, June, and July are all very wet months, on the coast at least. The heat, even during the hot season, is never very great ; during my stay, the ordinary temperature was about 7S-80 F. in the shade, and about 8-10 cooler at night. The thermometer rarely falls below 65 even on the coldest nights, and at that temperature one begins to shiver in the tropics and want blankets ! Further information on the climate of Pernambuco will be found in a paper by M. Beringer in the ' Annuaire ' of the French Meteorological Society, vol. xxvi. p. 28 (1878). ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 245 birds, however, are either from Para or Bio, comparatively few from Pernambuco itself. I noticed two specimens of Rhea macro rhyncha (of which more below) and an Ara spixi, said to be from Angola ! Amongst the Mammalia I saw some good specimens of the big Armadillo (Prio- dontes), which were said to be from the Sertoes of the interior. After being in Recife for about ten days, an opportunity occurred of making a flying visit to Goyanna, a town situated near the coast about fifty miles north of Recife, and a great emporium of the sugar-trade. As there is a decent road the whole way, which passes by Olinda and Iguarassu, and the weather was not at all settled, we decided to and drive. I was thus enabled to see something of the general features of the country, though there was little chance of shooting birds. Between the two towns the country rises somewhat, the more elevated parts being pretty generally covered with forest, often thick, whilst the lower slopes of the hills, and the moister bottoms between them, are nearly uniformly cleared or planted with sugar, some of the fields being of enormous extent. Birds were plentiful, especially in the more wooded parts ; and I now saw Jacamars and Parrots alive and wild for the first time, as well as " Sangre de Boi" (Ramphoccelus bmsilius) and many other birds not Ibis, 1881, to be found in the immediate vicinity of Recife. After about three weeks' stay at Estancia, I paid a week's visit to Cabo, a station about twenty miles from Recife on the Recife and Sao Francisco Railway, and the head quarters of the staff of that Company. Mr. Wells Hood, the general manager of the line, with whom I had gone out from England, possesses a capital residence here, and was kind enough to entertain me during my visit to Cabo. Here the country, which is generally flat so far, begins to rise in low, rounded hills, of no great elevation, which are covered, on their tops and steeper slopes, with the remains of the virgin forest. Unfortunately the weather during my stay at Cabo was exceedingly bad. It rained continuously for about three days, which resulted in a general flood of all the low-lying ground in the vicinity. Hence my collection of birds did not increase much, though I believe from what I saw that Cabo would in more favourable weather be a good locality. On August 12 I returned to Estancia, making excursions thence to Caxanga and other places in the vicinity. Having pretty well exhausted the neighbourhood of Recife by this time, on August 181 started for a trip to Parahyba do Norte, the capital of the next province to the north of Pernambuco, in company with my friend Mr. C. A. Craven, of the Recife Gas Company, whose acquaintance I had made in Recife, and whom I found much interested in the natural history of the country. Parahyba is about ten hours' run up the coast, and I found the steamers belonging to the Brazilian Steam Navigation Company by no means worthy of the evil reports I had heard of them. They are fine, well-built boats, receiving a heavy government subsidy for each trip made. By their means communication is kept up between 246 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. the imperial capital and the capitals of the more northern provinces of the empire up to Para. Parahyba is situated in reality only about four to five miles from the sea-coast, on a river which is navigable for these steamers nearly up to. the town. The river, however, turns off consider- ably to the north at about the point where the town is situated, so that Ibis, 1881 , it is a trip of some ten to fifteen miles up the river from the bar at its p * " mouth to where the steamer stops. The country is low, and the river is fringed on each side with mangrove- swamps, behind which the forest- covered country, which rises towards the interior, appears. On the mud-banks exposed at low tide many white Egrets (Ardea candidissimct) might be seen, as well as tens of thousands of a large and brightly coloured land-crab, with vermilion white-tipped claws, which gave quite a bright appearance to the scene. A railway, the Conde d'Eu, has just been commenced at Parahyba, to run inwards for about fifty miles, with the object of developing the sugar business. The inaugural fetes which celebrated the turning of the first sod had just terminated when we arrived, and the English engineers charged with the construction of the line were now the most important and popular personages in the town. Their then chief, Mr. A. M. Rymer Jones, a son of the well-known naturalist lately deceased, was kind enough to entertain me at the house they occupied, and he and his companions made us very much at home during our stay there. The country round Parahyba is flat, but rather thickly covered with forest, which extends from near the town to near the sea. I succeeded in securing the services of a Brazilian " Cagador "to shoot and show the way about. Though the number of birds I got did not at all equal the anticipations I had formed from, his glowing accounts of the abundance of all kinds of beasts and birds around Parahyba, I nevertheless got a considerable number of new ones, and had several very enjoyable excursions with him and some of my English friends. Besides the thick forests, nearer the town there is a good deal of scrub and bush-covered country, where small birds were rather plentiful. In the forests, indeed, these were far less abundant than in the more open parts ; and several times I walked for miles along tracts in the high and thick forests scarcely seeing or hearing a bird of any kind. " Antonio/' however, assured us that at the proper season of the year, i. e. when the fruits were ripe, these forests abounded with " Tocanos," "Trocas" (Columba speriosa), " Grallegas " (Columba rufina), and many other birds of which I saw Ibis, 1881, nothing. Antonio himself was armed, like most Brazilians, with an P' 319 ' ancient muzzle-loader of Trench make ; it was quite uncertain whether or not this weapon would go off when needed. Usually it missed fire three or four times in succession, by which time the bird aimed at had generally been prudent enough to retire out of range. Hence he did not increase my bag very much, though his astonishment at the shooting- ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 247 powers of my own gun, a double-barrelled central-fire of 16 bore, was immense, a successful shot being invariably greeted with much gesticu- lation of delight and loud remarks of " Espingarda 6oa, espingarda ingleza" &c. After a very pleasant week at Parahyba I returned to Eecife by the steamer on the 24th. The * Espirito Santo,' which had come from Para, had on board as miscellaneous an assortment of passengers as I have ever travelled with, and it would be difficult to say whether there were more parrots or slaves on board ; of the latter we had at least 200, on their way south to Kio to be sold for the coffee-plantations. Besides the parrots, chiefly Chrysotis cestiva, there were a lot of other birds and beasts, including a nice and tame Layothrix and some electric eels. Of the birds the most noticeable was an Icterus chrysocephalus, said to have come from the Rio Negro. This I bought, and kept alive in Recife, but unfortunately it died on its way home just as we got to Lisbon. I never saw the species alive in Europe nor elsewhere in Brazil. When I left England I hoped to be able to go overland from Eecife to the great waterfall Paulo Affonso, the "Niagara of Brazil," on the" S. -Francisco river. However, the state of the roads up to the present time, as well as the difficulty of getting an interpreter (my own know- ledge of Portuguese being very rudimentary), had prevented my making a start as soon as I had hoped. At length I succeeded in getting hold of a man who would do, and a day or two after my return from Parahyba, started with him from Eecife. I had also endeavoured to get some one to skin and shoot, but in this was unsuccessful, the only man I could hear of wanting terms for his services which were quite unreasonable. The Eecife and S.-Francisco railway runs for about 70 miles in a S.S.W. Ibis, 1881, direction towards the river from which it derives its name. From its p ' terminus at Una (or Palmares) another line of about the same length is now in progress, continuing it on to Garanhuns, which is situated about halfway in a straight line between Eecife and the Paulo Affonso. It was originally intended to have continued the line to the river above the falls near Boa Vista, and so to have brought down all the traffic of the upper part of the S.-Francisco river to the port of Eecife. This, however, has proved too expensive for the government, and the " Prolongamento," as it is called, is now destined to stop at Garanhuns. There is therefore but little chance at present of the Eecife and S.-Francisco railway ceasing to be a misnomer. A line, however, has been made and opened from the river above the falls near Tacaratu to Piranhas, situated below them, so that the traffic that was to have come to Eecife now goes down to Penedo and Maceio at the mouth of the river. The line of railway after leaving Cabo passes through a country similar to that which I have described as commencing there. The cultivation of sugar is general, and it is only on the tops of the hills, which are more 248 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. or less rounded, hummocky, and low, the highest being perhaps 700-800 feet in height, that any extent of the virgin forest is left. In some places along the line patches of quite open country may be seen, which are covered with grass, without trees or under-growth, and in general character a good deal resemble our south downs. I at first thought they were natural, but afterwards found out that they were inclosed spaces, used for horses and cattle. The grazing has evidently prevented them from becoming covered with a thick growth of capoeira, which always covers the hills where these have been cleared for sugar and afterwards allowed to lie fallow for a time. The destruction of the forests is still going on, as new ground is continually cleared by burning and cutting away the undergrowth for more sugar, so that in a few years there will, if this goes on, be Jittle trace of the old forests left. Ibis, 1881, At Palmares the railway ceases, and henceforward all travelling and p. 621. traffic has to be done on horseback, there being no roads in the interior worthy of the name. The earthworks of the " Prolongamento " are now nearly complete, only a few of the deeper cuttings and a tunnel or two being unfinished. The line of railway now forms the chief road to the interior ; but at this time, after the end of the rainy season, the stiff red clay had become worked up, in most places, into the most frightful mud conceivable, so that the horses were often up to their knees in it, and the rate of progression in consequence was a walk. At Palmares I was fortunate enough to fall in with the engineer-in- chief of the first section, Dr. Abel, a most pleasant and well-educated Brazilian gentleman. He too was going up country with the paymaster, so that I had the advantage of his company and escort (two Brazilian troopers) for the first part of my ride. As far as Barra do Jangada (a small village situated on the river Pirangi, which falls into the Una near Palmares), about thirty miles from Palmares, the country retains much the same features, though it gradually rises towards the interior. The hills perhaps are higher, and in some places, as around Catende, still pretty thickly covered with " matto " (the Brazilian term for the virgin forest), there being less sugar- cultivation here than nearer the coast. Towards Barra do Jangada cotton appears for the first time, a sure sign of the increasing elevation of the country. Biding along in this way I had no opportunity of shooting, but from the saddle I saw many birds already seen or secured. The " Sangre de Boi," however, disappeared soon after leaving Catende, and I saw no more of it as we approached the Sertoes. Another day's ride brought us to Quipapa, the most important town between Una and Graranhuns. After leaving Barra the country gets decidedly more hilly and open, and the forest begins to disappear, though many blackened and dead trunks of old forest trees standing on the higher hills show that this is due in large part to man's action. The soil is still clayey, restiog on ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 249 solid rock, apparently granite or gneiss, which in some places on the shoulders of the hills is left quite bare in great rounded patches. These at first suggest glacial action ; but I am inclined to think that in reality Ibis, 1881, they are merely the beds of old streams which formerly flowed down over them before the disappearance of the forests on the hills around had reduced the rainfall, and so caused their drying-up. I had no chance of shooting before getting to Quipapa, though between that town and Barra I fell in with a fine specimen of the much-dreaded Jararaca (Trigonocephalus brasilieusis ?), a rather rash attack on which resulted in nothing further on my part than a narrow escape from being bitten. Quipapa is distant 12 leagues from Palmares, on the Pirangi, and is at an elevation of about 1450 feet above the sea. The mean annual tempe- rature is about 72 0> 5 F., the maximum being about 92 and the minimum 62. For these details I am indebted to my friend Mr. H. E. Weaver, an English engineer who resides there, and who is chief of the second section of the " Prolongamento." He entertained me most hospitably at his house for several days, and aided me greatly in obtaining specimens of all kinds, as well as in other ways. There is no high forest very close to Quipapa, though there are still patches of it on the higher hills here, as elsewhere. The lower slopes, where not cleared for sugar, are covered with a rather thick growth of brushwood, in which, particularly along the river, birds were rather abundant. The weather too had now become markedly finer ; in fact, since leaving Eecife, hardly a drop of rain had fallen. Soon after leaving Una I had made up my mind that any idea of getting to the S. Francisco in the limited amount of time (about a month) now at my disposal must be abandoned, as I was due at Cam- bridge by the middle of October. I very much regretted having to give up the Paulo Affonso, but getting there and back in a month would have entailed continual travelling, and I should have had no chance whatever of collecting. I therefore determined to go no further than Garanhuns or thereabouts, staying en route at various places to collect. At Quipapa I remained till September 6, and then went on a few miles to a Brazilian friend living at Vista Alegre, two houses in a valley off the main line of the railway. From here I went, after a couple of nights, Ibis, 1881, to Macuca, where I found a most hospitable (if somewhat primitive) P e welcome from Mr. J. "Watt, also an English engineer employed on the Prolongamento. The country here much resembles that around Quipapa, but there is less forest and sugar and more capoeira. I continually added new birds to my list, and no doubt if I had had any assistance could have much increased the number both of species and specimens. But I had to do all my skinning myself ; the Brazilians, though they talked much, did little, and that chiefly in snakes and lizards : my inter- preter was useless for any purpose but to interpret (I doubt if he had ever fired a gun in his life), and my English friends were too busy with 250 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. their professional duties to be able to spare much time to shoot ; more- over, there were no guns available except those of the natives, and, as I have already said, the capabilities of these weapons as firearms were small. After a few days most agreeably spent at Macuca I went on to G-aranhuns, the termination of the " Prolongamento," distant about 80 miles from Una. Garanhuns is situated in the zone of country called the *' Agreste," that intervenes between the forest-clad " Matto," which extends inwards 60 to 70 miles from the sea-shore, and the open, elevated country, or Sertoes (pronounced " Sertongs "), of the interior. The " Agreste " zone participates to some extent in the features of both " Matto " and " Sertad " ; the forests have not altogether disappeared, but are smaller in size and of a different character ; the climate is much drier, and the vegetation lower and more scrubby in character. In the Sertoes, I am told (for I did not actually get into the real Sertoes country), the vegetation becomes still more low and scrubby, and the aspect of the country generally arid and stony. There is little water, and cultivation is confined chiefly to the ridges of hills that intersect the general level of the plateaux of 3000-5000 feet forming the Sertoes. In the height of the dry season many of the shrubs and trees lose their leaves. The growth of cotton and the raising of stock are the two great industries pursued in the Sertoes, which probably p.' 324. ' extend over nearly the entire area of the interior of the province of Pernambuco. The distance from Macuca to Garanhuns is about 33 miles ; after leaving Canotinho, about an hour and a half's riding from Macuca, the aspect of the country begins to alter visibly. The soil becomes sandy, and the vegetation generally lower and more scrubby, with patches of forest in places. Great Cacti, too, some 40-50 feet high, and forming large trees in some places, become conspicuous features in the landscape, and two or three species of Begonias also appear. In bird-life the "Salta Caminho" (Zonotrichia pileata) for the first time appears, hopping about the sandy roads, and marking the changed nature of the country. Garanhuns is a large village (although called a city) of perhaps more than 2000 inhabitants, and lies at an elevation of about 3000 feet above the sea. The country round is hilly, though none of the hills attain any great elevation ; these are pretty uniformly covered with a thick scrub of low bushes and aromatic herbs, with, in some places, small patches of " matto." There is little water. The temperature is noticeably cooler than nearer the coast, though sufficiently hot when the sun shines; indeed, on account of its dry soil and rather bracing atmosphere, Garanhuns is acquiring some celebrity in Pernambuco as a sanitarium, during the dry season, for the residents in the lower parts. I was most hospitably entertained, during my week's stay at Garanhuns, ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 251 by Senhor Doutor Jose Aloes Lima, the Juiz Municipal, who most kindly placed an empty house at my disposal, where I slept and kept my apparatus. The country round Garanhuns seemed to be rather rich in birds ; but partly from the thickness of the scrub, which in some cases was nearly impenetrable, and partly, I think, from the recent occurrence of a prolonged " Secca? or drought, during which everybody who could went out and shot small birds indiscriminately, thereby rendering them very shy, I failed to get several species I saw there and did not elsewhere meet with. A more prolonged stay would, I feel sure, have added numerous species to my lists. I also believe that Garanhuns would prove a very rich station for a botanist, judging from what I saw of its Ibis, 1881, flora during my rambles after birds or insects. p> Garanhuns is the principal town of a considerable district, and every Saturday a fair takes place there, which is largely attended by the "Matutos" or peasants of the country round. I was told this fair would be a capital chance of obtaining animals and birds from the country people who come in to attend it ; and I therefore decided to stay a couple of days to witness it, rather than going on to S. Bento, in the Sertoes, about 35 miles north of Garanhuns, and returning thence by a different route to Macuca, as I had originally intended. However, the fair, though it certainly gave me an excellent chance of seeing " the natives" (and, perhaps I should add, of their seeing me), produced nothing, or next to nothing, in the way of " bichos" a most convenient term used in Brazil for denominating all and any animals from an elephant to a blackbeetle. I managed, however, to pick up a live " Ema " (Rhea macrorhyncha), of which more anon, at Garanhuns, as well as a lot of Tinamus, sundry Hawks, Guans, and other live birds, so that when I left I had a regular caravan of living animals, which necessitated my taking on an extra horse or two and man for their safe conveyance to Palmares. I finally left Garanhuns on September 19th, and returning by the same way as I came, stopped en route a night at Macuca and two days at Quipapa, and reached Recife September 24th. A few days were spent in packing up and settling things generally ; and on September 29th I left, with my live animals, which had now increased to about 35 in number, in the Royal Mail Steamer * Neva/ and arrived at Southamp- ton October 15th. Before concluding this account of my trip, I ought to return my best thanks to the numerous gentlemen in Brazil who did all in their power to help me, and especially to my friends Mr. Wells Hood of Cabo, who most kindly procured me numerous valuable introductions, to Messrs. W. Elliott and C. A. Craven of Eecife, to Mr. Curling of Parahyba, as well as to Messrs. Weaver, Watt, Abel, and the other engineers of the " Prolongamento," and to Dr. Lima of Garanhuns. The total number of species of birds of which I obtained or observed Ibis, 1881, p. 326. 252 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. specimens during my trip was 116. In the following list they are treated of in systematic order, according to the nomenclature of Messrs. Sclater and Salvin's ' Nomenclator Avium Neotropicalium,' unless other- wise stated. I am much indebted to Mr. Sclater for having kindly gone through and named the greater number of my birds for me, whilst Mr. Salvin was good enough to give me the names of the few others. As far as can be judged from the results of a short trip like mine, the avifauna of Pernambuco is essentially South-east Brazilian, with few, if any, Amazonian forms. It would appear from my observations on the birds, and from the general features of the country, that Pernambuco is far less rich in birds than either Bahia or Para, the comparative poverty no doubt being due to the long time that the country, at least near the sea, has been colonized, and to the consequent destruction of the primaeval forests. Moreover, as the forest only forms a comparatively narrow zone along the coast, with a dry and elevated " campos " country behind, there has been no possible retreat towards the interior for the original inhabitants of the coast forests, and many of the most characteristic forms have, in consequence, disappeared or become scarce. Of course, my collections do not represent any thing like the total number of species to be found in Pernambuco ; but I think the above conclusion will be confirmed by further collections from that district. From what I heard of the nature of the country, Ceara, and most likely Maranhao as well, must probably be included within the limits of the South-east Brazilian fauna, so that the boundary between it and the Amazonian province must lie still further north-west on the coast, whilst in the interior it may correspond to the watershed between the Tocantins and the Paranahyba. The following is an account of the birds of which I obtained or observed specimens. The specimens are mostly deposited in Mr. Sclater's collection. Ibis, 1881, ! TUEDUS FUMIGATUS. p. 327. j am no ^. q u it e certain as to the correctness of the above name, though it is probably this species that occurs in this part of Brazil, as the only specimen of this Thrush that I shot fell into a swollen stream and was lost. Two I bought alive also escaped. The " Sabia," as it is called by the Brazilians, is very much esteemed by them as game, and therefore relentlessly shot down ; hence it has become a very shy bird, at least in the neighbourhood of towns. It has a rather pleasant song, and is also on this account sought after by the natives, who keep it as a cage-bird very commonly. I saw caged specimens in nearly every place I visited from Parahyba to Garanhuns, but only a few times observed it in its native state at Cabo and Recife, so can say nothing further on its habits. ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 253 2. TURDUS RUFIYENTRIS. I found this Thrush common all over the districts I visited, except in the immediate neighbourhood of Recife. In its habits it much resembles the common Thrush of England (T. musicus), spending a good deal of its time on the ground in pursuit of its food. It is usually to be seen in paths in the lower second growth, or in the clearings for railways, or on the line itself, and is not found in the thick forests. The Brazilians call this Thrush, as well as the preceding species, " Sabia," and esteem it highly for eating-purposes. Hence probably it has become rare near Recife, and shy elsewhere in the neighbourhood of towns. Eyes brown ; beak greenish yellow, the upper mandible greyer ; feet dirty flesh. 3. POLIOPTILA LEUCOGASTRA. I first met with this elegant little bird near Parahyba, and subse- quently saw it frequently in the interior between Quipapa and Garanhuns. It goes about in small companies of two or three, and is a most active little creature, in almost perpetual motion from twig to twig, the meanwhile constantly flipping its tail up and down. Eyes brown. 4. DOXACOBIUS ATRICAPILLUS. Ibis, 1881, p. 328. This bird I first observed from the train on the railway between Cabo and Una, frequenting the marshy bottoms of the valleys. I subse- quently saw it at Cabo, and found it more or less abundant in suitable situations all along my route thence to Macuca. It is a very noisy bird, with a loud chattering cry. It flies about in small companies of three or four, and is found among the marshy vegetation that grows along the banks of the stream. The bird is a very conspicuous one, both owing to its noisy cry and the habit it has of fluttering its short and rounded wings, when the white bar at the base of the primaries forms a very much more striking mark than would be imagined from the skins. I heard the name " Casaca do Couro," signifying " Leathern Jacket," applied to this bird by a Brazilian friend who had paid some attention to animals ; but whether it is the same bird as that mentioned by Capt. Burton (' Highlands of Brazil,' ii. p. 316) under the same name, and noticed by him on account of its remarkable nest, I do not know. I never saw Donacobius nesting. As mentioned by Burrneister (Thiere Bras. ii. p. 130) there is a narrow naked space, about an inch long, on the neck of this bird, behind the angle of the jaw, which shows conspicuously in the shot bird. It is coloured bright chrome-yellow (Burmeister says "fleischroth "), and with the bright yellow irides makes a freshly shot Donacobius a far more beautiful object than one that is 254 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. skinned. This brightly coloured nude space is probably present in both sexes, as the only specimen I procured was a female. I do not at present recall any precisely similar case of ornamentation by a bright nude skin-space on this part of the neck in any other bird certainly not in any other Passerine. The feet are grey. 5. TEOGLODTTES FUEYTJS. This is the common Wren of the country, and is very abundant everywhere in the neighbourhood of houses or gardens, though it is not much of a forest-bird. It has a remarkably strong song for such a small bird, and may often be seen perched on the roofs of the houses of the villages in the early morning, carolling. Eor its notes it is, I think, on the whole the best singing-bird I heard whilst in Brazil. Eyes brown ; feet flesh-coloured. 6. BASILEUTEEUS AUEICAPILLTJS (Sw.) *. I shot a single female specimen of this bird, the only one I saw, in the depths of some high forest near Quipapa. 7. CTCLOEHIS ALBIYENTRIS. I found this curious bird rather common nearly all over the country I visited, though nowhere abundant. It is found amongst the vegetation of the more open parts, usually singly, and seems to be a very quiet bird, hopping about from leaf to leaf of the bush or tree it is in, and not uttering any cry ; at least I never remarked any. The irides are beautiful bright orange-yellow ; the strangely shaped bill has the upper mandible dark flesh-coloured, the lower pale bluish slate. The feet are pale dirty fleshy. 8. HlKTTNTK) LEUCOEEHOA. This Swallow I found very common in Recife, where it might be seen flying about in numbers in some of the streets, as well as over the rivers which separate the various parts of the town. I also observed it at Parahyba ; but in the interior it seems to disappear, and be replaced by the Atticora next mentioned. 9. ATTICOEA CYANOLEUCA. I did not bring home any specimens of this Swallow, the only one I shot having been too much damaged to skin ; I have, however, little doubt that this is the species I met with, as I continually saw it in numbers, and was able to examine it often through my field-glasses. It was very abundant at Cabo, and might be seen there sitting in numbers, * Cf. Berlepsch, Ibis, 1881, p. 240. ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 255 particularly in the morning, on the telegraph-wires of the railway opposite Mr. Hood's house ; I also saw it at Parahyba and Graranhuns, perched on the roofs and eaves of the churches, and therefore not to be shot at with impunity. In Recife, on the other hand, I never saw it at Ibis, 1881, all, though the last species, as already mentioned, abounded there. 10. STELGIDOPTEETX RUFICOLLIS. This Swallow I found common in numerous places from Recife and Parahyba on the coast inland as far as Macuca. It perches freely, and may be often seen along the roads and railway, where there are cuttings Eyes brown. 11. DACKES CAYANA. I only rarely met with this species, once near Caxanga, and another time near Eecife, where I came across a small flock of three or four in an old, overgrown garden some two miles from Estancia ; of these only one was a full-plumaged male. I also saw one or two near Parahyba. Irides red-brown ; beak blackish brown, with the base of the mandible fleshy ; legs fleshy, the claws greyer. 12. DACNIS PLTJMBEA. I only met with this bird in the garden at Estancia, and there only saw it a few times. It hops about the trees and bushes in a systematic sort of way, going from leaf to leaf in search of small insects and other food, which it picks up off the leaves. I did not observe any full- plumaged male. Eyes (in the female) greyish brown ; legs dirty flesh-coloured ; beak pale fleshy, with the culmen broadly darker, horny black. 13. C(EEEBA CYANEA. Only once did I come across this bird a single specimen in immature plumage that I saw in the garden at Estancia. 14. CEETHIOLA CHLOEOPYGA. This little bird is one of the very commonest in those parts of Brazil I was in, being most abundant in all the gardens near Eecife, and almost equally so elsewhere in the neighbourhood of houses, though sometimes seen in the wilder parts. It assiduously visits all the shrubs that may happen to be in flower in any particular spot, collecting from the blossoms its meal of insects, mixed, no doubt, with the nectar of the Ibis > 1881 i flowers. It has a weak, though rather pleasing, song of a few notes, P< the last note being considerably more powerful than those that precede 256 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. it. It is known by the Brazilians as " G-uarratan," a name, however, which it shares with the Euphonia and some other Tanagers. Eyes brown. 15. EUPHONIA VIOLACEA. This violet-and-yellow Tanager I found sparingly round Eecife, and also at Parahyba, both in gardens and in the vicinity of high forest. I did not, unfortunately, see enough of its habits to throw any light upon the raison d'etre of the peculiarly developed stomach of this genus, a feature first observed by the late Dr. Lund, and lately redescribed and figured by myself *. This bird is kept commonly as a cage-bird by the Brazilians, who call it " Gruarratan," a name, as already observed, also applied to several other small brightly plumaged birds. 16. CALLISTE FASTUOSA. This very beautiful Tanager, to my mind one of the finest of the beautiful genus it belongs to, is believed to be peculiar to the province of Pernambuco, from which skins are occasionally received by the dealers in Paris and elsewhere. It is a species often seen, too, alive in the larger Zoological Gardens of Europe, though no naturalist seems to have yet met with it in the wild state. It does not appear to be common in Pernambuco at least I only met with it twice, once near Macuca, where I shot a female out of some bushy capoeira, and again at Quipapa, where I saw what I believe was this species in the virgin forest. The bird, however, was perched at a great height from the ground, in the topmost branches of a large tree, and only the brilliant orange-yellow of its rump was visible. Whilst staying at Cabo, a freshly shot adult of this bird was also brought to me to skin, so that probably Ibis, 1881, it is also to be found considerably nearer to the coast than the localities I P- 332> saw it in. Eyes brown. 17. CALLISTE FESTIVA. I only saw this beautiful Oalliste once, when I fell in with a small party of it in a patch of virgin forest near Quipapa, and succeeded in shooting a fine male. Eyes brown. 18. CALLISTE FLAVA. This beautiful, though peculiarly coloured, bird is, perhaps with the exception of Tanagra cana, the commonest Tanager in the provinces I * P.Z.S. 1880, pp. 143-147, " On the Structure of the Stomach in certain Genera of Tanagers." ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 257 visited. I met with it everywhere from Recife to Garanhuns ; and though never seen in numbers, it appeared to be fairly abundant. It frequents chiefly gardens or plantations of fruit-trees, but I have also seen it in thick forest country. It was abundant in the garden at Estancia, frequenting the orange-trees, sapotis (Achras sapotd), and other fruit- bearing plants ; and I have also met with it feeding on the flowering shrubs of the virgin forest. It goes about either singly or in small companies, and most of the specimens seen are either immature or females. The adult males are usually met with singly, though I have seen three perched close together in the same tree. I failed in iny endeavours to bring living specimens to England, though I got one as far as St. Vincent. Eyes brown ; feet lead-grey. 19. TANAGRA SATACA. This Tanager is abundant all over the country. It is found, like the last species, in small flocks of four or five, and is common near the vicinity of houses and gardens wherever there are fruit-trees. At Estancia it abounded, visiting the orange- and lemon-trees, also the sapotis, mamans (Carica papaya}, and other fruits in season. It did not apparently mix with T. palmarum, and left the cocoa-palms to the latter species, preferring the lower and bushy trees. The Brazilian name is " Sayau," a name, however, which they also apply to T. palmarum. Eyes brown. 20. TANAGEA PALMAEUM. Ibig This Tanager is very abundant near the coast, and may be seen in P- 333 ' numbers quite close to Recife. In the interior it is less common, though I saw it once or twice near Macuca. The specific name is very appro- priate, as the bird frequents the cocoanut-palms, flying in small flocks from one tree to another, and settling about the root of the " crown," where it probably finds abundant food in the shape of insects and spiders. Eyes brown ; feet leaden grey. 21. RAMPHOCXELUS BRASILIUS. I first met with this splendid bird on the road between Iguarassu and Olinda, and subsequently found it abundantly, in favourable situations, nearer Recife, as well as at Parahyba and all along the line of railway as far as Catende. It seems, however, an essentially low-country bird, and as the country rises in the interior disappears. This bird goes about like several of the other Tanagers, in small parties, composed chiefly of 258 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. immature or female birds, so that the number of those seen in the gorgeous crimson and black dress of the adult male is comparatively small. It is always to be found in the low bushes and vegetation that grow about the lower slopes and bottoms of the valleys in the neighbourhood of water, and is never, according to my observation, found in gardens or the virgin forests. It has a quick, rather loud, sharp, chirping note, of a single syllable, repeated several times in sharp succession, which one soon gets to recognize. The Brazilian name is " Sangre de Boi," i. e. ox's blood, from the brilliant crimson of the plumage of the male. Irides orange-brown. 22. TACHYPHCXNTJS MELALETJCUS. This Tanager is widely spread over the province of Pernambuco, where I met with it at nearly all the places I visited, from Caxanga to G-aranhuns ; and I also obtained it at Parahyba. It is usually seen singly or in pairs, the black male with the chestnut female. The stomach of one I examined contained insects. Eyes brown. Ibis, 1881, 23. NEMOSIA PILEATA. P' ' I obtained this species of Nemosia in the garden at Estancia, where, however, I only observed it once or twice. The only other place I met with it was at Cabo, where I once saw two or three in Mr. Hood's garden. Eyes bright yellow ; legs horny yellow, with the claws horny. 24. NEMOSIA FTJLVESCENS. I first met with this Tanager at Quipapa, where it was not uncommon in the low bushy capoeira near the river. It was usually seen in small companies of three or four, hopping about amongst the leaves like a Dacnis. I also saw it at Garanhuns, but not nearer the coast. Eyes brown ; beak and legs grey. 25. SALTATOB, MAGNUS. I only once met with this bird, which I shot in thick and high forest some miles from Parahyba. Eyes brown. 26. ORCHESTICUS CAPISTRATUS. This curious Bullfinch-like Tanager I first met with near Vista Alegre, about halfway between Quipapa and Macuca. I subsequently saw it at both those places, as well as at Garanhuns ; indeed it seems rather a common bird in this part of the province. It is nearly always seen singly near, but not in, high forest, and perches in the larger trees that ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 259 rise above the bushes and undergrowth of the capoeira. It appears to be not at all shy, and is easily shot. The sexes are similar. Eyes reddish brown ; feet black-grey. 27. ORCHESTICUS ATER. This bird was rather abundant round Parahyba in the neighbourhood of the forest, and in its habits resembles the last species. At Vista Alegre I found both species together in the same locality, but beyond that point it seems to be replaced by 0. capistratus. At Parahyba I obtained a specimen which is in all probability the Ibis, 1881, young bird of this species, though it was the only one seen in that P* ^* plumage. It is of a dark olive-green above, lighter below, with the fore- head, chin, breast, and angle of the wing greenish yellow. It is probably 0. ater in this phase of plumage that was described by Sclater as Tanagra olivina (cf. Sclater, P.Z.S. 1881, p. 213). I bought a single living specimen of this bird in a shop in Eecife, and brought it safely to London, where it is still living in the Zoological Society's Gardens. I never before saw it living in Europe. Eyes reddish brown ; the beak leaden grey, with the tip broadly black (in the adult). 28. GUIRACA CYANEA. The Blue Grosbeak I first saw on the road between Iguarassu and Olinda, and I subsequently met with it at most of the places I stayed at as far as Parahyba and Garanhuns. It frequents low bushy ground, and is usually seen singly or in pairs. The Brazilians call it " Azulin." Eyes brown. 29. ORYZOBORUS TORRIDUS. I only twice met with this little Einch, which I saw singly near Cabo and Parahyba. An allied species (0. maximiliani*!) is kept as a cage-bird by the Brazilians, who give high prices for them, they being much esteemed for their song, which is supposed to be only surpassed by that of the " Patitiva." They call it " Bicuda," from its large beak. 30. AMATJROSPIZA UNICOLOR. Whilst staying at Quipapa a Brazilian servant of Mr. Weaver's brought me a fresh specimen of this rather rare little Einch, which he had obtained near the town. This was the only specimen I saw. Eyes brown; feet fleshy grey. 31. SPERMOPHILA NIGRO-AURANTIA. I obtained this species at Recife, frequenting the same localities as the 82 260 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. IbiS 'l?f 81 ' next two, but it appeared to be less common. I also saw, and shot, what I believe was a specimen of this bird near Vista Alegre, but did not succeed in finding it. The Brazilians call it " Caboclo," a name applied to the tamed aboriginal Indians in Pernambuco. It may sometimes be seen in Recife in cages with crowds of sundry other Spermophilce, Canaries (Sycalis\ Cardinals (Paroaria), &c. 32. SPEEMOPHILA GUTTTJEALIS. This little Spermopliila was very abundant in the garden at Estancia, frequenting the reedy and marshy parts, where it congregates in small flocks, feeding on the seeds of the grasses, sedges, and other similar plants. I also saw it abundant afterwards at Quipapa, as well as in the low bush- covered country round Garanhuns, so that it is by no means confined to the sea-board or even to the neighbourhood of water. It is often kept as a cage-bird. Eyes brown. 33. SPEEMOPHILA HTPOLEUCA. This species of Spermophila is also common and widely distributed, frequenting grassy or open places, and often coming into gardens. It appears to feed mainly on grass-seeds, and is social in its habits. The females are brown. Called by the Brazilians " Papa Cupim," i. e. grass- eater. This name it shares with S. gutturalis. Eyes brown ; bill (in the male) fleshy red. A closely allied species (S. plumbea, distinguished easily by its smaller and black beak) I never succeeded in identifying for certain in a wild state, though it is greatly esteemed by the Brazilians as a cage-bird. They call it " Patitiva de Parahyba " (those caught at that place being supposed to be particularly excellent songsters) and often pay considerable prices for good singers. The song is loud for the size of the bird and rather pretty, though monotonous, and not at all comparable to a Nightingale's, or even a Red Cardinal's (Cardinal). 34. VOLATimA JACAEINA. This little Pinch, though not very common as a rule, I found widely Ibis, 1881, spread, extending from Parahyba and Recife on the coast to Quipapa and p. 337. yista Alegre in the interior. The adult males may often be seen singly, and they have a peculiar habit of selecting some particular twig on a bush or small tree as a pet perch. Here they sit for a long time, twittering out a little song of a few notes, and then jumping vertically up some little way in the air, and, turning a somersault, alighting in the same place. I have watched them on several occasions repeat this performance a number of times consecutively, continuing, in fact, till they were disturbed. They would then fly off to some other place, and go on with ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 261 their performance. The females and young birds are brown, and these seem to be more sociable, going about in small flocks. The Brazilians call the bird " Saltado." The eyes are brown. 35. PAEOAEIA LARVATA. The Red-headed Cardinal I found common at Parahyba, and again saw it in the neighbourhood of Garanhuns, so that it occurs all over the district I traversed. It is usually seen singly or in pairs in the more or less cleared and open ground near cultivation. Many dozens are brought into the market at Recife to sell as cage-birds. The Brazilians call it " Gallo do campina." 36. ZONOTEICHIA PILEATA. This bird marks the approach of the traveller, as I have already mentioned, to the Sertoes of the interior. I never once saw it on the coast, or anywhere in the " matto " zone, though on passing Canotinho and getting on to the sandy soil of the interior, it almost at once becomes abundant. Round Garanhuns it was very common, hopping about the highroads, often two or three together, and very tame. The Brazilian name is " Salta Caminho," or " Road-Hopper." 37. COTUENICULITS MANIMBE. I only got this bird at Caxanga, where it seemed rather common, frequenting the heaps of rubbish left near the railway station. Eyes brown. 38. CHEYSOMITRIS TAEEELLI. Ibig Of this pretty little bird I obtained a living specimen at Parahyba, P- which is now alive in the Zoological Gardens. I subsequently saw one near Garanhuns, and a pair near some forest close to Quipapa. The Brazilian name was, I understood, "Pinta Silva" (? Pintasilgo= Gold- finch). 39. SYCALIS FLAVEOLA. The " Brazilian Canary " is a very abundant bird in the parts of Brazil visited by me, being found from the coast, at Parahyba and Recife, to the interior. Large flocks of it, sometimes containing one or two hundred individuals, may be seen in suitable localities, which are usually the more or less cleared grounds in the neighbourhood of engenhos, or sugar- farms. In these places it frequently consorts with numbers of the little green Tapacu Parrakeets (Psittacula passenna). A specimen which I shot near Cabo " towered " in the air as wounded Partridges and other 262 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. birds often do, to a height of fifty or sixty feet or more, and then dropped down dead ; on examination I found it had been shot through the brain. The " Canario " is a very common cage-bird with the Brazilians, nearly every house having one or more pairs. 40. CASSICUS PEESICUS. This is one of the commonest and most characteristic birds of the country near the coast, where it is very abundant, and may be seen commonly, even in the neighbourhood of Recife, nearly anywhere where cocoa-palms grow. It usually goes about in small parties of about four or five, which keep up, when perched, a continuous chattering, often leading to their discovery before being seen themselves. Towards even- ing they seem to collect in larger parties, as at that time numbers might often be seen returning homewards, always flying in the same direction, and usually making for a clump of palms, on which, no doubt, to pass the night. In the interior it is much less common, and I often went several days without seeing one. The Brazilians call it " Sheshou," and keep it often in cages. Irides deepish blue. Ibis v 1881, 41. ICTEETJS TIBIALIS. This bird I first observed at Quipapa, where it was not uncommon in the vicinity of the town, flying about in small companies of twos and threes. I afterwards found it at Macuca and Garanhuns, and saw a single specimen in the garden at Cabo a few days before I sailed, though I did not see the species at all during my previous stay there. The bird is also found at San Lorenzo, a village about 20 miles west of Eecife, as a living specimen I bought in Eecife came from there. The bird is not rarely to be seen caged in the houses of the Brazilians, who call it "Sheshou de Bananeira," to distinguish it from the common " Sheshou" (Cassicus persicus). It is also sometimes called " Soldado," or soldier. I succeeded in bringing three specimens alive to London, two of which are still living in the Zoological Gardens, where it has not before, I believe, been exhibited alive. Eyes reddish brown ; feet bluish grey. 42. MOLOTHETTS BONABIENSIS. I never met with this species in the wild state, but saw several in cages in Eecife, Quipapa, and elsewhere, and was told that it was found in the neighbourhood. 43. LEISTES STJTEECILIAEIS. This Eed- breasted Hangnest I only saw at Cabo. Here it was abun- dant in the open, down-like fields that are found on the engenJios where ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 263 the forest has been cleared and subsequently inclosed for the use of cattle and horses. I only once or twice saw the f ull-plumaged birds ; all the others were immature, and these kept in large flocks like Starlings, feeding, like them, largely on the ground on the insects and other crea- tures always present where cattle are feeding. This is, I believe, the most northern locality from which the species has yet been recorded. It is replaced further north by a representative (L. guianensis} which has no white supercilia. 44. APHOBUS CHOPI. A single female skin that I brought home is apparently referable to a small form of this species, which I found rather abundant at one or two localities, Vista Alegre and Macuca, in the interior of Pernambuco. ibis, 1881, Though local, the bird was common where it occurred, flying about in P- 34 - large flocks, like Starlings, in the neighbourhood of sugar-plantations. They were rather wary and not easily approached. The Brazilians called it " Arumara." Eyes brown. 45. FLUVICOLA CLIMACURA. This bird is, I consider, the most characteristic of the country of all the species met with during my trip. Save in the thick forest, it may be seen nearly everywhere, even in the busiest parts of the town of E/ecife, close to the sea, and everywhere it is conspicuous alike by its tameness and its sharply contrasted colours. To the structure of a Tyrant-bird it unites the habits of a Wagtail and a good deal of the appearance of a Saxicola. It spends a great part of its time on the ground, running swiftly, like a Wagtail, after the insects which it puts up, and seizing them as they rise from the ground. It is by no means afraid of man, coming up to within a few yards of the houses, and mixing freely with the poultry and dogs of the establishment. Usually it is seen in twos or threes, but never, so far as my experience goes, in flocks or larger parties. Frequently two may be seen perched on the top of a wall or house, " standing up " to each other, with fluttering wings, spreading tails, and outstretched necks, chattering away vigorously at each other the whole time. It is nearly the only bird that is not shot or eaten by the Brazilians. They call it the " Lavendera," or Washerwoman, from a legend of its having formerly performed those functions to the Virgin Mary. Hence they hold it almost an act of sacrilege to kill one, and think very much the same of a man who shoots a " Lavendera " as we do in England of one who kills a Robin. In consequence of this immunity from destruc- tion, the Lavendera is, as I have already said, exceediugly tame and familiar everywhere, and even nests close to the houses. One or two nests that I observed were built in low bushes, and composed of a loose fabric of 264 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. Ibis, 1881, grass-stems, vegetable fibres, &e. Both tlie birds seemed to take part in p. 341. the construction of the nest, and made a great business of it, though it was apparently done in a desultory sort of way, and as much for pleasure as business. At least two nests I observed closely were never finished, and apparently ultimately abandoned. 46. ABTJKDINICOLA LEUCOCEPHALA. This bird I only observed near the sea-coast round Eecife, at Caxanga, and near Parahyba. It frequents the margins of ponds or rivers, as its name well indicates, and is usually seen solitarily, though near Caxanga I came across three or four together on the edge of the same pool, a family party of parents and young birds, the latter being distinguishable by their less-defined colouring. Eyes brown ; beak blackish horny, the lower mandible at base (normally) yellowish ; legs and claws black, the soles paler. 47. MACHETOBKCS EIXOSA. I obtained this bird at Eecife, and subsequently saw it at Cabo. Its habits resemble those of Myiozetetes, from which, indeed, I did not, at the time of getting my specimen, distinguish it. [ 48. TODIROSTETJM CINEEETTM. This little Tody -like Tyrant I found rather common from Eecife and Parahyba to Garanhuns. It is usually seen singly, but I have sometimes seen two or three together, chasing each other and fighting furiously, like Humming-birds. It hops about nimbly from twig to twig of the par- ticular bush or shrub it is in, in search of insects, and does not, as far as 1 saw, select a particular perch from which to dart off at any passing insect, like so many of its congeners. 49. EUSCAKTIIMUS GTTLAKIS. I only met with this bird once or twice round Garanhuns, where it occurred usually singly, actively hopping and creeping about the thick scrub which is so prevalent there, very much in the same way as the last species. Irides brown. Ibis, 1881 50. SEEPHOPHAGA SUBCEISTATA. I shot a single specimen of this bird in the thick scrub near Garanhuns. 51. PHTLLOMTIAS SEMIFTJSCA. This little bird was abundant in gardens round Eecife, resembling in its habits Elainea pagana. 52. ELAINEA PAGANA. This Tyrant I also met with commonly, from the sea-coast to Gara- nhuns. In its habits it resembles the other larger Tyraimidse, but has ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 265 no loud call-note. It is common in gardens, and has rather a pleasant, low, warbling-like song of a few notes ; when the bird sings, its throat- feathers are considerably puffed out. The elongated head-feathers form a sort of crest, which is erected when the bird is excited. Eyes brown. 53. MTIOZETETES SIMILIS. This species of Tyrant-bird is very abundant in those parts of Brazil I visited. In its habits it resembles Pitangus sulphuratus, being usually seen singly or in pairs, perched in conspicuous positions on the projecting or topmost branches of the trees, and flying off from them in pursuit of its prey. It is common in gardens, even in the neighbourhood of houses, and, compared with the Pitangus, it is a silent bird, without the charac- teristic notes of that species. The Brazilians, however, do not discrimi- nate between the two, and call both alike " Bentivi." Eyes greyish brown. 54. PITANGUS STTLPHTJKATUS. This bird is one of the very commonest and most characteristic of the country, being seen nearly everywhere, and from its habits most con- spicuous, even to the casual observer. The " Bentivi," as it is called, from its note, almost always selects some prominent twig or branch of a tree on which it perches, and from that post of vantage flies off after any passing insect ; then, having captured it, it returns to its favourite spot to repeat the process, varied only by continually uttering its loud, some- Ibis, 1881, what plaintive, but screaming cry, ben-ti-vi, ben-ti-vi, &c. Frequently **' two may be seen together, but they are not at all gregarious in their habits. Where these birds are common, as in most gardens in the neigh- oourhood of houses, the ear soon gets to recognize their continued ben-ti- vis as a pleasant evidence of the Neotropical fauna. The flight is weak and undulating and never long sustained. 55. HlBTJKDINEA BELLICOSA. The first specimen of this curiously coloured Tyrant I saw was just after landing at Parahyba, where it was perched on the roof of one of the houses close to the river. Subsequently I saw it again several times, both there and at Quipapa, in similar positions, but being always in the towns, and on houses, or, more frequently, churches, I was unable to get a shot at one, for fear of consequences in the shape of a " row " with the police or other authorities. From its post of vantage it flies off after passing insects, and after capturing them returns to its former perch, in the manner of other Tyrannidse. I was rather surprised to meet a pair of the same birds a few days afterwards at Macuca, both of which I got. They were met with in capoeira, a long way from any houses, perching in the larger trees which rose above the thick scrub and bushes below. Eyes brown. 266 ELEVEN WEEKS IN NORTH-EASTERN BRAZIL. 56. MYIOBIUS N^EYIUS. I shot a single specimen of this Tyrant in a patch of high forest near Macuca. Eyes straw-yellow. 57. MTIOCHANES CINEEETJS. I only got this species twice, in the high trees of the forest near Quipapa and at Macuca. It appears to be solitary in its habits. Irides brown ; feet black ; upper mandible blackish brown, lower pale orange. 58. MYIAECHUS TYEAKNTJLUS. Ibis, 1881. I first met with this bird near Quipapa, and afterwards obtained several specimens between there and Garanhuns. It is a quiet soli- tary bird, which usually I met with perched on the sides of paths or tracts through the brushwood, and was not shy. Irides brown. 59. TYEAjrNTJS MELANCHOLICUS. This is nearly as common as Pitangus 9ulpkur